article
stringlengths 0
745k
| abstract
stringlengths 0
92.7k
|
|---|---|
in this supplementary material section we will first provide the details of our experimental setup and then discuss the validity of some crucial assumptions of the well known theories of the iqhe .
namely , first we will briefly discuss the assumptions on different boundary conditions and topological aspects of the quantized hall samples within the single - particle ( sp ) theories .
the main discussion is on the mapping between the momentum space representation of the edge states to real space representation .
this discussion clarifies the importance of boundary conditions , where the normalisation condition is the only physical restriction . here , instead of a detailed mathematical description
we will make use of some schematic presentations which are common in the qhe discussions . however , to lift the confusion between different theories we used distinguishing colours to discriminate single - particle and screening theory concepts , e.g. incompressibility , edge states , etc .
our first discussion is based on the single - particle theory , where incompressible ( namely energy gapped region or so to say insulating ) bulk is essential for qhe and the incompressible regions are denoted by blue colour .
we denote the sp chiral edge states by solid lines , which differ by colours depicting different filling factors , e.g. red corresponds to @xmath40 edge state where spin degree of freedom is neglected . on the other hand ,
consistent with the main text , we denote compressible 2des by green and incompressible 2des by black , within the screening theory .
yellow regions denote the contacts , whereas white corresponds to electron depleted ( etched ) regions for both theories .
first , we aim to clarify that , using periodic boundary conditions for finite size hall bars is questionable which then makes topological arguments prone to break .
second we would like to highlight that , considering quantum capacitance in calculating impedance or current modifies both the sp and screening theory results .
and third , we explicitly show that our sample geometry is quite different then the well known corbino and the anti - hall bar geometries . in this paper
our aim is to show that the bulk of the hall bar is not incompressible through out quantum hall plateaus by direct transport measurements .
our samples are defined on high purity gaas / algaas wafers , produced at braun center for submicron research .
the hall bars are 10 @xmath30 wide and 40 @xmath30 m long , defined by chemical etching .
hall contacts are designed like a lizard " such that they are not effected by the inner contacts .
we embedded two inner contacts to the bulk of the hall bar utilising air bridge technique , each inner contact has approximately 1 @xmath30m@xmath65 area and they are 7.5 @xmath30 m apart from each other . to measure hall effect we excite the 2deg by imposing a constant ac voltage between the source and drain contacts at 8.54 hz by a lock - in amplifier ( sr 850 , with an internal impedance of 10 m@xmath66 + 25 pf ) .
a 10 m@xmath66 resistance is placed between inner contact a and source contact , whereas a 1 k@xmath66 resistance is placed between inner contact b and drain contact .
we measure hall voltage ( both x and y components ) between contacts 2 - 3 or 1 - 4 . to check whether the hall voltage is affected by inner contacts we also reversed source and drain contacts and observed similar features .
similar to hall voltage measurements , we excite the inner contacts by a lock - in amplifier at 11.5 hz , followed by a 10 m@xmath66 resistance before contact a and 1 k@xmath66 resistance after contact b. we measure the potential difference between the inner contacts , while a low - noise preamplifier is utilized to filter high frequency noise ( @xmath67 30 hz ) .
all the signals go through a room temperature rc circuit to filter noise above 92 khz .
we used 300 ms time constant and swept the magnetic field by a 0.1 t / min rate .
the measurements are repeated for many thermal cycles and different samples defined on the same wafer .
all the results were consistent . once the landau hamiltonian is solved using coulomb ( also called landau ) gauge , i.e. translational invariance in @xmath68 direction and open boundary conditions in @xmath69 direction ( @xmath70 ) , the solution yields plane waves in the current ( @xmath68 ) direction as if as a free electron and harmonic oscillator wave functions in the other direction .
such a choice of gauge and boundary conditions can be utilized to describe a homogeneous hall bar that extends to infinity in both directions . however , to describe a more realistic hall bar one usually assumes infinite walls in @xmath69 direction , which modifies the related wave - functions by parabolic cylinder functions approaching to simple harmonic oscillator solutions away from the boundaries . however , these preferences of boundary conditions yield the problem of normalisation . in order to overcome the normalisation problem one assumes periodicity in momentum along the current direction similar to bloch wave function describing electrons in a crystal . as the hall bar is not periodic in real space
, one can not impose periodicity in @xmath68 , hence , periodicity in @xmath71 is assumed .
this assumption yields to the well known description of thouless @xcite which explains quantised hall effect in terms of chern numbers in _ momentum _ space , where one simply counts the berry flux encircled and describes transport utilising the kubo formalism .
such an approach is well justified only if the system is in electrostatic equilibrium , i.e. if no external current is imposed , or might be reasonable if the qhe can be treated within the linear response regime , if it can be handled within this regime at all .
* cylindrical geometry .
* in the earliest attempt to elucidate iqhe , laughlin proposed that if a 2des resides on the surface of a cylinder and a time dependent perpendicular magnetic field penetrates the surface , then to protect the geometrical phase only an elementary charge can be moved from one end of the cylinder to the other end when the flux changes by one quantum . in this description periodicity in real space is assumed , however , the boundary conditions are not specified in the perpendicular direction .
in contrast , in the approach of thouless the periodicity is imposed in momentum space @xmath71 , where the 2des is defined on a flat surface . ]
the first case can be ruled out in our experiments since we excite the system by an external voltage .
the second case is also ruled out , since the excitation energy ( voltage ) is comparable with the landau gap , hence the system is far away from linear response . in
contrary to periodicity in momentum space , laughlin assumed periodicity in real space considering a cylinder where instead of an imposed current he assumed a radial magnetic field which changes adiabatically by time ( cf .
[ fig : add1 ] ) , leading to a phase which is protected by the topology resulting in conductance quantisation .
@xcite this approach also for sure does not entirely describe the experimental realisation of the qhe , at least for our experiments .
both of the descriptions of the qhe implicitly require a bulk incompressibility , namely an energy gap induced by the magnetic field .
this is somewhat similar to 3 + 1 d topological insulators where an energy gap opens due to crystal structure and provides an incompressible ( insulating ) bulk .
@xcite however , note that this gap is in the energy dispersion , namely when one plots the relation between the energy and momentum there opens a gap due to symmetries of the crystal .
similarly , for an infinite hall bar with periodic boundary conditions one can obtain such a gapped energy dispersion in momentum space , which is usually mapped to real space without taking care of the boundary conditions of the physical devices . for a finite width and length ( flat ) hall bar assuming periodic boundary conditions is not appropriate .
hence , bulk incompressibility which is essential to describe qhe does not apply to our physical system at hand . in connection with 3 + 1 d topological insulators
, we should note that in this case the system is periodic in ( crystal ) momentum space and hence the energy dispersion presents the well appreciated topologically " protected energy gap . in contrast , qhe does not provide the periodicity in momentum space once realistic , i.e. physical , boundary conditions are considered . [
[ real - space - momentum - space - duality - of - halperin - bttiker - edge - states ] ] real space - momentum space duality of halperin - bttiker edge states ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ * the corbino and hall bar geometries . * the quantised hall effect measurements are usually performed either on the rotationally periodic corbino * ( a ) * geometry or on the translationally non - invariant hall bar * ( b ) * geometry in real space . for a corbino disc , inner and outer contacts ( yellow regions )
are kept at different electrochemical potentials , where edge states ( red solid lines ) are parallel to contacts and the conductance between these contacts is measured .
in contrary , for a hall bar geometry edge states are perpendicular to the contacts , where contacts are kept at different electrochemical potentials , here 0 and @xmath30 .
the bottom panel in * ( b ) * presents the filling factor ( or electron density ) distribution along the cut denoted by dashed lines in the upper panel .
one clearly sees the unrealistic electron distribution which for sure can not guarantee electrostatic equilibrium .
note that , the electron depleted regions near the edges are neglected , hence the related capacitances , which is consistent with the single - particle theories . ] the first important manifestation of boundary conditions on the transport properties of a 2des subject to high magnetic fields is proposed by halperin in early 80 s .
@xcite in his pioneering work he imposed the above mentioned infinite wall boundary conditions in the radial direction @xmath72 , which essentially bends the landau levels in the close vicinity of boundaries .
utilising periodicity in azimuthal ( @xmath73 ) direction ( both in momentum and real space ) , one can map the energy dispersion ( i.e. energy versus momentum ) to real space ( i.e. energy versus @xmath72 ) and obtain the geometry known as the corbino geometry , as shown in fig . [
fig : add2]a . in the original work
it is stated that this is a slight modification " of laughlin s cylindrical geometry ( see fig . [ fig : add1 ] ) , which is not reasonable from topological point of view in real space , since the genus numbers are different ( for laughlin s case it is 1 , since periodicity is also assumed in @xmath74 which defines essentially a torus , and for halperin s case it is 0 ) .
we should also note that , due to the absence of edges in laughlin s geometry , the current is uni - directional , whereas at a corbino edge states are present and carry chiral current .
for a corbino geometry , in contrast to a hall bar ( cf .
[ fig : add2]b ) , there are no source and drain contacts , however , the edge states exist due to boundaries where all the system is in equilibrium .
namely , no external current can be and is imposed . in this geometry
, one can measure the conductance between the inner and outer contacts if contacts are kept at different electrochemical potentials .
in such a system the electrical measurements essentially results in a quantised hall conductance .
it is also stated that in a real experiment , the measured hall potential @xmath75 is the sum of an electrostatic potential @xmath76 and the difference in fermi levels @xmath77 ( electrochemical potential difference in our notation ) .
the edge current is then only a _ fraction _ of the total hall current , given by ( e_f^(2)-e_f^(1))/ev n r_c_cc / e^2,[eq : halperin]where @xmath12 is the capacitance per unit length of the edge states , and @xmath78 is a number of unity . " . here , @xmath79 is an integer determined by the filling factor @xmath5 at the bulk and fermi energy lies in between the energies @xmath80 of two landau levels @xmath81 and @xmath82 , in the interior of the sample .
the electrochemical potentials near the boundaries ( @xmath83 and @xmath84 , at @xmath85 and @xmath86 , respectively , where the contacts reside ) are also supposed to lie in the interval @xmath87 , which implicitly assumes linear response .
then the total current carried by the edge states between @xmath83 and @xmath84 is given by @xmath88 .
now , we should clarify couple of points with the properties of contacts and their equilibration with edge states : first , in halperin s approach it is implicitly assumed that the capacitances @xmath89 and @xmath90 shown in fig .
[ fig : add2]a are neglected , which imposes that the contacts are in electrochemical equilibrium with the edge states , namely the scattering between the edge states and contacts is possible pointing a compressible edge .
hence , the contribution to hall current in real experiments only come from the bulk capacitance , which is quantised if finite tdos at @xmath4 is assumed .
if the compressible edge assumption is lifted , then the total capacitance would be the sum of @xmath89 , @xmath91 and @xmath90 .
for @xmath92 the total capacitance would yield an edge current approximately @xmath93 , which is not quantised .
here we assumed @xmath94 . for higher bulk
filling factors , such an approach implicitly assumes that the inner edge states are in equilibrium with each other similar to the outer edge states , which again requires that the edges are compressible , i.e. scattering between inner or outer edge states is possible .
this assumption then makes the current quantisation questionable . as discussed in the main text and details given in the following subsection
, capacitance is composed of the classical and the quantum counterparts .
once , the region between two edge states ( or contacts ) is insulating ( namely incompressible ) then the quantum capacitance vanishes for an ideally pure system .
hence the hall current mentioned above reads to zero , i.e. impedance diverges . in real experimental devices
there are potential fluctuations due to disorder which yield localised states as discussed by halperin , which then results in conductance quantisation for a finite magnetic field interval , where bulk incompressibility is still preserved .
furthermore , once again the main assumption is the incompressibility of the bulk , i.e. the impedance between the inner and the outer contacts reads to infinity ( or relatively high compared to conductance ) hence the conductance measured between these contacts should be quantised . to be explicit , if one wants to move an electron from one contact to the other one has to get across @xmath79 edge states ( namely pass through metal - topological insulator boundary @xmath79 times ) and has to pay an amount of energy that corresponds to @xmath79 ( berry ) flux quanta , yielding to quantisation . in fig .
[ fig : add2 ] we present both the hall bar and the corbino geometries also depicting edge states " .
note that , in the corbino geometry edge states are parallel to the contacts whereas for the hall bar geometry they are perpendicular to the contacts .
hence , the impedance is infinite ( or huge for real devices at finite @xmath62 ) for the corbino geometry , however , for the hall bar capacitance vanishes ( or much smaller then the resistance ) considering a bttiker " contact , i.e. transmission from the contact to edge state is unity .
it is important to note that , the mathematical mapping between the momentum and real space representations of the edge states makes sense only if the periodicity is preserved , which is the case for the corbino geometry .
however , when a finite size hall bar is considered such a mapping becomes questionable , at least for all geometries .
couple of years later than the work of halperin , bttiker developed a transport theory based on the landauer formalism to describe the iqhe .
@xcite in his work also the finite size of the hall bar is taken into account by imposing a confinement potential in @xmath69 direction , which essentially varies smoothly on quantum mechanical length scales .
although yielding to similar results with halperin , the smooth confinement potential allows one to use simple harmonic oscillator solutions also close to the boundaries .
in addition , the scattering probability between edge states at the same boundary is suppressed by the fact that the edge states are relatively far apart from each other both in real space and in momentum space . utilising the periodicity in momentum space in equilibrium one
can then map the momentum space representation of edge states to real space and draw the well known bttiker " edge states for a hall bar as shown in fig .
[ fig : add2]b .
this picture is modified if an external current is imposed by applying a voltage difference @xmath95 between the source and drain contacts , yielding an electrochemical potential energy difference @xmath96 .
of course in this non - equilibrium situation assuming periodic boundary conditions become questionable , where one end of the hall bar is kept at potential @xmath97 . turning back to our discussion on the capacitances and calculating the actual hall current
one can still utilise eq .
[ eq : halperin ] .
note that the edge states at the same sides have the same electrochemical potential , hence , @xmath89 and @xmath90 vanishes and the only contribution to capacitance comes from @xmath91 . once again , for an ideal 2des capacitance vanishes yielding zero hall current pointing a back - scattering free transport throughout the plateau interval .
however , even a small amount of tdos below @xmath4 would yield a finite capacitance , hence a finite hall current which would result in deviations from perfect quantisation . by the above discussion
, we have shown that the momentum - real space duality of both halperin and bttiker edge states becomes questionable since such a duality strongly depends on the imposed boundary conditions and the symmetries of the sample .
in addition , once the quantum capacitance is taken into account , which is directly proportional to tdos at @xmath4 , impedance between probe contacts also strongly depends on the edge state configurations , namely whether the edge states are perpendicular or parallel to imposed current .
such geometrical and topological aspects of so called qhe samples are usually undermentioned in the well known theories .
* screening theory expectation of a corbino disc . *
the inner and outer contacts are decoupled by a bulk incompressible region denoted by black . however ,
if the magnetic field is reduced the picture essentially does not change drastically . at a lower @xmath3 field ,
the bulk incompressible region will split into two encircling incompressible rings near the edges and since the capacitances are connected in series the contacts would be still decoupled . ] in a simplistic treatment , inverse capacitance can be described as the amount of energy to add a charge to a system , given by @xmath98 .
classically , one can approximate the capacitance of a sheet as shown in fig .
[ fig : add3 ] as @xmath99 , where @xmath100 is the dielectric constant , @xmath101 is the perimeter of the ring , @xmath102 is the width of the incompressible strip and @xmath103 is the distance between the gate and the 2des .
@xcite on the other hand , the quantum capacitance per area is solely dependent on the thermodynamical density of states given by @xmath104 .
since these two capacitances are connected in series , the total capacitance is given by @xmath105 .
the classical capacitance is finite except that the width of the incompressible region is infinite , however , the quantum capacitance becomes zero if there exists an incompressible strip between the inner and outer contacts at zero temperature and for an ideally clean system , i.e. no level broadening due to impurities .
therefore , the capacitance is dominated by the quantum counterpart if there exists an incompressible strip decoupling the contacts .
note that , the contacts and the compressible regions are in electrochemical equilibrium , since they both behave like a metal .
furthermore , since scattering is suppressed exponentially along the incompressible ring and is quantised across the ring , resistance is much smaller than the total capacitance therefore impedance reads z 1/d_t(e_f ) .
hence , for a corbino geometry the bulk incompressibility is the guarantee of conductance quantisation within the single particle picture .
however , when interactions are taken into account we observe that it is not necessary to have an incompressible bulk and only a single incompressible strip is sufficient to decouple inner and outer contacts .
the next discussion is on the thermodynamical definition of incompressibility which is a statistical quantity .
one usually defines a system to incompressible if the ratio between the change in number of particles and the change in electrochemical potential presents a discontinuity .
note that , electrochemical potential is a statistical quantity which makes sense physically only if there are sufficient number of particles within the system considered .
therefore , assigning incompressibility to a strip is bounded by the number of particles within the strip , which is essentially determined by the fermi wavelength at zero temperature .
hence , once the strip width becomes small or comparable with the fermi wave length the insulating behaviour of the strip is destroyed .
such a situation can be considered as a leaky capacitor , as stated in the main text . in addition ,
when the strip width becomes even smaller than the magnetic length , which is the quantum mechanical length scale , then it is possible to tunnel across the incompressible strip by quasi - inelastic scattering mechanisms .
[ [ what - to - expect - within - the - bttiker - edge - state - picture - between - inner - contacts ] ] what to expect within the bttiker edge state picture between inner contacts ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ * the relation between @xmath22 and @xmath23 as a function of @xmath3 field . * based on the above discussion concerning capacitances one expects to observe an infinite @xmath23 whenever the hall voltage is constant since the bulk should be incompressible . due to @xmath106 at realistic samples
the impedance becomes finite .
in addition , at finite temperatures the sharp features both at @xmath22 and @xmath23 are smeared out .
however , the constant hall voltage and high impedance intervals should still overlap . ] as discussed above , once the qhe is well developed namely scattering is suppressed by the incompressible bulk the impedance between contacts should diverge for an ideally pure 2des .
hence , if hall voltage presents quantisation @xmath23 should diverge simultaneously .
however , this is true only at zero temperature and for an ideally pure 2des .
once , effects of finite temperature and tdos at the fermi energy is considered then the hall plateau shrinks together with the high impedance interval .
let us first consider @xmath107 however assume a non - ideal 2des , i.e. with impurities leading to level broadening , then the amplitude of the impedance ( therefore @xmath23 ) is bounded from up by the tdos at @xmath4 .
[ fig : add4 ] depicts the upper boundary of @xmath23 at zero temperature by the thick horizontal dotted line , whereas ideally pure 2des case is shown by the solid ( blue ) line . once the impurity scattering is taken into account at finite temperature the curves are modified , which are depicted by the thin broken lines for @xmath22 and by the thin solid lines for @xmath23 . based on above arguments
one can conclude that our experimental findings do not coincide with the results expected from bttiker type edge state theory , up to our knowledge .
we show the expected edge state distribution and corresponding capacitances in fig .
[ fig : add5 ] .
the scattering between edge states out of the plateau interval is depicted by @xmath10 denoting the resistance . here
one can see once more the geometrical difference between the corbino geometry ( one hole ) and our device ( two holes ) . in the latter
both contacts are encircled by the edge states where excess current is perpendicular to the contacts , whereas in the former one the outer edge state is encircled by the outer contact and there is no current imposed between the two contacts .
the quality of our sample guarantees that the electrochemical equilibration between the edge states is strongly suppressed given the fact that the distance between inner contacts is 3.5 @xmath30 m , the area of contacts being 1 @xmath30m@xmath65 and the physical dimensions of the our device is ( @xmath108 ) @xmath109 @xmath30m@xmath65 . in our sample the mean free path , the localisation and the equilibration lengths are sufficiently large to suppress scattering between edge states both encircling the contacts and the ones along the edges of the sample . * the edge state distributions and capacitances within the sp theories .
* sketches present edge states considering different average filling factors , * ( a ) * @xmath40 and * ( b ) * @xmath110 , where also the capacitances between the edge states and contacts are presented by inner ( white ) symbols . similar to halperin s discussion of a corbino geometry , the capacitance between contacts are determined by the bulk incompressible region , however , here the excitation is perpendicularly imposed with respect to edge states .
interestingly , even at the transition between @xmath110 to @xmath40 plateau the inner edge states should provide a high capacitance .
namely one can expect scattering between @xmath110 edge states , which would yield a huge impedance even at the transition intervals . ]
* sketch of an anti - hall bar .
* once an anti - hall bar is embedded " into a hall bar , one expects to have two independent edge state configurations which are decoupled by the incompressible region in between .
here it is essential to realise that , the edge states of the anti - hall bar connect the source and drain contacts and do not encircle entirely any of the contacts . ] modifying the topology of a hall bar has been realised some 18 years ago by r. mani @xcite and recently modelled by oswald and his co - workers within the non - equilibrium network model .
@xcite at a first glance our experimental system and the so called anti - hall bar ( ahb ) geometry seems to resemble each other , as shown in fig .
[ fig : add6 ] . however , there are couple of major differences , first at an ahb geometry there is a single etched region at the bulk of the sample . in this
configuration edge states percolate around the ahb and contacts are connected by the edge states .
however , in our situation we have two electron depleted regions , i.e. inner contacts , which are not connected by incompressible strips .
hence for the ahb geometry , impedance is mainly dominated by the resistive counterpart , whereas for our geometry the transport between contacts is dominated by the capacitive counterpart throughout the plateau interval .
another important difference is the properties of our sample , namely its size and the mobility .
different from the actual ahb sample , our samples are defined on a high purity wafers which strongly suppress scattering between edge states , either between the inner ones or between the inner and outer ones .
[ [ finite - density - of - states - below - the - fermi - energy - indirect - proof - of - a - single - incompressible - strip - and - the-1.35-mv - value ] ] finite density of states below the fermi energy , indirect proof of a single incompressible strip and the 1.35 mv value ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ * the arrhenius plot of max(@xmath23 ) . *
the maximum of the voltage drop between contacts a and b considering different filling factors .
one can see that , the activated behaviour of the well developed integer states coincide nicely , whereas for fractional states the gap value is scattered depending on the filling factor . ] from the above discussions one can clearly see that the potential difference measured between contacts a and b , should diverge or at least become huge at certain @xmath3 intervals within the quantised hall plateau .
consequently , the maximum value of @xmath23 in certain magnetic field intervals is attributed to finite tdos at the fermi energy .
we also observe that , this maximum value is not bulk filling factor dependent and is not quantised .
the non - quantisation is clarified by the temperature dependent measurements . at base temperature and at the lowest inner excitation voltage ( @xmath50 mv )
, we observe that max(@xmath23)@xmath111 mv for the plateaus @xmath112 and 2/3 , whereas it seems that for the other quantised hall states only a spike is measured .
now let us consider a situation where we imposed an excitation at very low frequencies ( practically dc ) and let us assume that the an infinite resistor ( impedance ) is placed between the inner contacts at zero temperature , then we would expect to measure 2 mv , i.e. one has basically 2 to 1 voltage divider .
hence , the maximum value indicates that the bulk is not an infinite resistor , however , it highly decouples inner contacts due to finite tdos at @xmath4 . in the next experimental investigation we measured @xmath23 as a function of temperature and observed that max(@xmath23 ) decreases by increasing @xmath62 .
the decrease of the maximum value is a direct consequence of finite tdos at @xmath4 and is a measure of the activation gap , indirectly . in fig .
[ fig : add7 ] , we show the arrhenius plot of max(@xmath23 ) and see the commonly observed activated behaviour .
we also checked the value of max(@xmath23 ) at higher excitation voltages and see the same ratio between the max(@xmath113 ) , indicating that this maximum value is not due to a cutoff voltage of our electronic setup .
both of these experimental results clarify that the bulk resistance ( in fact voltage difference at fixed excitation current ) is not quantised .
one can also claim that there exists only a single incompressible encircling strip with a given ( integer or fractional ) filling factor , since if there existed more than one strip max(@xmath23 ) would present a stepwise behaviour at each integer filling factor plateau .
we would like to thank j. jain and k. von klitzing for initiating the idea of embedding inner contacts .
also acknowledges itap - marmaris for providing a perfect scientific environment , where the final version of the paper was edited .
this work is supported by tbitak under grant ( 112t248 and 211t164 ) and istanbul university scientific projects department iu - bap:6970 and 22662 .
a. m. chang , `` a unified transport theory for the integral and fractional quantum hall effects : phase boundaries , edge currents , and transmission / reflection probabilities , '' _ solid state commun .
_ , vol .
74 , p. 871
, 1990 .
a. yildiz , d. eksi , and a. siddiki , `` the consequences of bulk compressibility on the magneto - transport properties within the quantized hall state , '' _ journal of the physical society of japan _ , p. accepted , 2013 .
m. e. suddards , a. baumgartner , m. henini , and c. j. mellor , `` scanning capacitance imaging of compressible and incompressible quantum hall effect edge strips , '' _ new journal of physics _ , vol .
14 , p. 083015
, aug . 2012 .
f. dahlem , e. ahlswede , j. weis , and k. v. klitzing , `` cryogenic scanning force microscopy of quantum hall samples : adiabatic transport originating in anisotropic depletion at contact interfaces , '' _ phys .
b _ , vol . 82 , pp . 121305+ , sept .
2010 .
b. i. halperin , `` self - consistent local - equilibrium model for density profile and distribution of dissipative currents in a hall bar under strong magnetic fields , '' _ phys .
b _ , vol . 25 , p. 2185 | since the experimental realisation of the integer quantised hall effect in a two dimensional electron system subject to strong perpendicular magnetic fields in 1980 , a central question has been the interrelation between the conductance quantisation and the topological properties of the system .
it is conjectured that if the electron system is described by a bloch hamiltonian , then the system is insulating in the bulk of the sample throughout the quantised hall plateau due to magnetic field induced energy gap .
meanwhile , the system is conducting at the edges resembling a 2 + 1 dimensional topological insulator without the time - reversal symmetry .
however , the validity of this conjecture remains unclear for finite size , non - periodically bounded real hall bar devices .
here we show experimentally that the close relationship proposed between the quantised hall effect and the topological bulk insulator is prone to break for specific magnetic field intervals within the plateau evidenced by our magneto - transport measurements performed on gaas / algaas high purity hall bars with two inner contacts embedded to bulk .
our data presents a similar behaviour also for fractional states , in particular for 2/3 , 3/5 and 4/3 .
the mutual relation between the integer quantised hall effect @xcite ( iqhe ) and topology is a reoccurring theme .
@xcite salient features of the iqhe are the precise hall conductance measured as integer multiples of the conductance quanta @xmath0 ( @xmath1 is the elementary charge and @xmath2 is the planck constant ) accompanied by zero longitudinal resistance at certain magnetic field intervals .
the robustness of the iqhe against material systems points to a universal origin , which is claimed to be the topology of the system @xcite and relies on a key argument : the bulk of the two dimensional electron system ( 2des ) is incompressible .
@xcite for the iqhe , the incompressible state is a direct consequence of the quantising magnetic field @xmath3 localising the electrons when the fermi energy @xmath4 falls a landau gap .
hence , due to the lack of available states the bulk is insulating , while the number of landau levels below @xmath4 determines the filling factor @xmath5 . in the case of the fractional quantised hall effect
the energy gap emanates from many - body interactions , based on exchange and correlation effects , and @xmath5 assumes some particular fractional numbers .
@xcite quite generally , the bulk insulating region is assumed to be incompressible and provides a scattering free region between the probe contacts .
the rationale behind the relation between the incompressible topological bulk insulator and the conductance quantisation stems from the mathematical map between the landau hamiltonian and the bloch hamiltonian defined on a periodic system , which can be well understood physically in terms of the berry phase .
@xcite each time the field is increased by one magnetic flux quantum @xmath6 @xmath7 , an electron is adiabatically transferred from one edge of the cylinder to the other edge ( cf . supplementary material fig.4 ) , keeping the geometrical phase protected . in the standard picture , the charge transport along the edges is described by the single particle , i.e. , non - interacting , edge states which form due to level bending imposed by the boundary conditions . the transport is regarded as non - local and dissipationless .
@xcite assuming periodic boundaries , the hall conductance @xmath8 is a topological invariant , namely the chern number , which can be elegantly proven using the kubo formalism . in the case of the iqhe with @xmath9
the filling factor @xmath5 is the chern number .
@xcite within the single - particle approaches , adding two inner contacts to the bulk of the system modifies the topology .
hence , conductance quantisation should be severely affected , unless the bulk insulating state remains unchanged . to probe whether the bulk remains incompressible
one can impose an external excitation between the inner contacts and measure the potential difference as a function of magnetic field .
different from typical hall experiments , here one does not only measures the resistance @xmath10 but measures the impedance @xmath11 to explore the contribution from the capacitance @xmath12 .
the impedance is evaluated as @xmath13 , @xmath14 being the frequency of the ac excitation . here
, the ( quantum ) capacitance is given by @xmath15 , and @xmath16 is the thermodynamic density of states ( tdos ) at @xmath4 .
since , the bulk is incompressible due to @xmath17 throughout the conductance plateau , the capacitance vanishes and impedance diverges to infinity for an ideally pure 2des at @xmath18 . * the experimental setup and measured voltages at base temperature . * * a , * the sem image of the hall bar defined on high purity gaas / algaas wafer , together with the measurement setup . a 26 mv@xmath19 ac voltage excitation is imposed between source - drain contacts ( s and d , at 8.54 hz ) ) @xmath20 and inner contacts ( a - b , 4 mv@xmath19 at 11.5 hz ) @xmath21 , simultaneously .
the hall potential @xmath22 is measured between contacts 1 - 4 or 2 - 3 , whereas longitudinal voltage difference is measured between contacts 1 - 2 or 3 - 4 .
* b , * the hall potential measured between contacts 1 - 4 ( broken ( black ) line ) and @xmath23 ( thick solid ( blue ) line ) as a function of magnetic field @xmath3 , while imposing ac excitations between contacts a - b and s - d
. we also checked that there is no correlation between @xmath23 and @xmath20 by observing the measured potential differences without one of the excitations .
shaded ares depict the constant @xmath23 subintervals observed for @xmath24 and 2/3 , which are zoomed in * c , d , e * , respectively .
the vertical long - lines indicate the plateau intervals , whereas their relative centres are indicated by vertical short - lines with circles .
the constant voltage intervals are determined by a percentage error of % 1 , corresponding to an absolute error less than @xmath25 . ] in contrast to the above single particle description of the iqhe , it is proposed that the direct coulomb interaction modifies the electronic distribution and the bulk becomes compressible also within the plateau , i.e. the bulk behaves like a metal with high density of states at @xmath4 .
@xcite in this situation , the incompressible states ( or commonly called strips ) reside at the edges of the sample , due to homologous level bending of the 1d edge - states . under these conditions
, it is predicted that one can drive an excitation between the inner contacts and measure a finite impedance , without disturbing the perfect quantisation .
@xcite inspired by this proposal , it is interesting to determine the effects of inner contacts on the hall conductance when imposing an excitation between them . it is possible to find few similar attempts in the literature to clarify the incompressibility of the bulk , however , most of the experiments utilise indirect measurement techniques , such as capacitive coupling , scanning force microscopy or single electron transistors , @xcite where the samples usually are either low purity ( @xmath26 @xmath27/vs ) and/or large in dimensions ( @xmath28 m ) . here
we investigate directly the bulk transport properties of a topologically modified hall bar using our tlm 400 dilution fridge equipped with a 20 tesla super - conducting magnet at low temperatures ( @xmath29 mk ) .
the potential difference between the inner contacts is measured simultaneously with a standard hall characterisation while an ac excitation voltage is imposed .
we find that the bulk of the hall bar is not entirely insulating all along the conductance plateaus .
this observation is in perfect agreement with the predictions of the interaction theory of the iqhe , however , challenges the single particle theories , which incontestably assume an insulating bulk in the plateau interval regardless of the magnetic field strength .
the experimental findings will lead to reexamine the existing theories of the iqhe from a more general topological point of view and open new a horizon in investigating the relation between topological insulators and quantised hall effect .
* experiments at base temperature * in fig.[fig:1]a we show the sem image of our 10 @xmath30 m wide hall bar together with the sketch of the experimental setup .
the 2des is created in a high purity ( @xmath31 @xmath27/vs , after illumination ) gaas / algaas heterostructure residing 130 nm below the surface .
the transport experiments are repeated for three different thermal cycles , at dark and after illumination considering two different hall bars to eliminate specific sample properties .
we found that all measurements present very similar salient features .
fig.[fig:1]b depicts measured @xmath22 ( broken lines ) between contacts 1 and 4 , where a 26 mv root mean square @xmath20 excitation is imposed at base temperature @xmath32 mk ( electron temperature @xmath33 80 mk ) , corresponding to an excitation current of 0.4 na , sufficiently small to prevent heating effects .
the thick solid line depicts @xmath23 , where couple of spikes for @xmath34 t is observed , each corresponding to an integer @xmath5 .
similar spike - like features are also observed in the intervals @xmath35 t @xmath36 t and @xmath37 t @xmath38 t for which one can assign fractional states with @xmath39 11/7 , 4/3 , 3/5 , etc . in contradistinction to mentioned spikes , for @xmath40 , 1 and 2/3 we observe a constant @xmath23 ( highlighted by shaded ares ) in a subinterval of the corresponding hall plateau .
the related filling factors are enlarged at the bottom panel , fig .
[ fig:1](c)-(e ) .
we attribute this behaviour to a well established bulk incompressible region as we will explain in the following .
even grippingly , on both sides of the plateaus @xmath23 takes values similar to that of which are completely out of the plateau intervals .
a homologous behaviour is also observed for the fractional state 2/3 , as shown in fig .
[ fig:1]e .
the variation of @xmath23 is of primary importance pointing a compressible bulk in spite of a simultaneously measured quantised hall plateau .
after a first glance at the experimental data , the questions are : why is @xmath23 constant only for a specific @xmath3 subinterval and is it quantised ?
what is the mechanism behind the observed asymmetry with respect to the centre of the plateau and how the quantised @xmath22 and @xmath23 compare with longitudinal potential difference ? in the next section we will answer these questions within the screening theory of the iqhe .
* discussion * we base our model on the formation of the compressible and the incompressible strips resulting from electron - electron interactions and the confinement . since the non - interacting single particle picture yields a stepwise density distribution violating the electrostatic equilibrium ( cf . supplementary material fig.5 ) , it is propounded that the 2des comprises local insulator - like incompressible regions surrounded by metal - like compressible regions .
strips with integer @xmath5 emanate from direct coulomb ( hartree ) interaction , whereas , the strips assuming fractional filling factors account on many - body interactions .
taking into account interactions and lateral confinement results in the fact that the incompressible region(s ) can reside at the bulk , as well as along the edges .
the former is called the bulk incompressible region , whereas the later is called the edge incompressible strip . the spatial distribution and the widths of the incompressible strips
are determined by the external @xmath3 field , the density gradient , the temperature and most importantly by the energy gap , as clarified both analytical and self - consistent numerical calculations .
@xcite assuming a constant @xmath4 , the evolution of the incompressible regions with changing @xmath3 field can be summarised as follows : let us start with a situation where the field is sufficiently high and only the lowest landau level is partially filled , hence , the average filling factor is less than 2 ( neglecting the spin degeneracy ) . due to the lateral confinement ,
the spatial distribution of the electron density is inhomogeneous such that it reaches its highest value close to the centre ( at the bulk ) whereas gradually decreases towards the edges and vanishes at the boundaries .
therefore , @xmath41 is larger than that of at the edges .
* graphical sketches of the electron density together with the measured voltages . *
central figure : the measured voltages @xmath22 ( solid thin ( black ) line ) , @xmath42 ( solid thick ( red ) line ) and @xmath23 ( broken ( blue ) line ) .
vertical lines are guide to eye to separate different cases depicted in * a - f*. in graphical sketches , colour gradient depicts the electron density distribution and boxes denote contacts ( blue : probe contacts - red : inner contacts - yellow : source and drain contacts . ) .
black regions identify incompressible regions , whereas dashed lines correspond to thermodynamically compressible strips . *
b - f * , the spatial distribution and widths of incompressible regions together with corresponding electric circuit of each case . here
@xmath43 and @xmath44 denote the contact and compressible bulk resistances respectively , while @xmath45 depicts the bulk capacitance and @xmath46 the capacitance of the encircling strips . ]
once the @xmath3 field is decreased , the fermi energy falls into the landau gap at the bulk yielding @xmath47 , hence the bulk becomes incompressible .
this resembles the single - particle description of the iqhe , where the @xmath22 is quantised and @xmath42 vanishes .
further decrease of the field results in a situation where the incompressible regions shift towards edges and become narrow , due to the electron density gradient .
the transport properties and evolution of strips along the edges are the well understood both experimentally and theoretically,@xcite and guarantee quantised @xmath22 together with vanishing @xmath42 by the thermodynamical and the electrical decoupling of the probe contacts .
namely , scattering between contacts is prohibited by the incompressible regions between them .
here we used the fundamental principle of thermodynamics in determining the ( in)compressibility of the strips : thermodynamical quantities are physically meaningful only if there are sufficiently large number of particles , which dictates a lower bound for the length scales . at @xmath18 this length is the fermi wavelength , @xmath48 , since only the electrons at @xmath4 contribute to current .
therefore once the width of the incompressible strip becomes comparable or less than @xmath48 , assigning incompressibility to strips become meaningless , physically .
hence , it is thermodynamically reasonable to expect that at a certain @xmath3 field the width of incompressible strips become narrower than @xmath48 and scattering between neighbouring compressible regions and thereby the probe contacts is possible . in this situation
, @xmath42 becomes finite and @xmath22 deviates from its quantised value .
the schematic presentation of the above discussion is depicted in fig.[fig:2 ] , with increasing @xmath3 from ( a ) to ( f ) where the filling factor distribution of a hall bar is shown which comprises two inner contacts .
the electron density gradient is depicted by the colour gradient , whereas the incompressible regions are shown by black .
similar to edges , incompressible strips form in the close vicinity of inner contacts , where the broken lines denote the case where the strips become leaky " both thermodynamically and electrically .
* the temperature dependency of the inner voltage difference * * a , * the 12 tesla wide span of @xmath23 for various temperatures , while excitations are @xmath49 mv and @xmath50 mv fixed to sufficiently small voltages to prevent additional heating .
* b - d , * the zoomed in filling factors together with the corresponding hall plateaus . ] in the following discussion we will assume spin degeneracy initially , since the mechanism to elucidate experiments is independent of the particular nature of the energy gap .
we start our investigation with a situation where the lowest landau level is partially occupied , i.e. , the 2des is completely compressible as depicted in fig .
[ fig:2]f .
the central panel in fig .
[ fig:2 ] shows the measured potential difference between the inner contacts @xmath23 ( broken ( blue ) line ) , the hall potential @xmath22 ( solid thin line ) and the potential difference between contacts 3 and 4 ( thick ( red ) solid line ) , @xmath42 . in the graphical demonstrations ( a - f ) the electron density variation is depicted by the colour gradient ( green ) , which vanishes ( white ) near both the inner ( red square boxes ) and the source - drain contacts ( yellow , rectangular boxes ) .
the black coloured areas correspond to incompressible ( constant density ) regions .
it is worth to note that , sketching the density gradient and its behaviour near the contacts is well justified both experimentally and theoretically .
@xcite we see that for @xmath51 ( case f ) @xmath22 and @xmath42 increase ( almost ) linearly , whereas , @xmath23 considerably small .
this behaviour suggests that the bulk of the 2des acts as a poorly conducting metal , as predicted .
we explain the behaviour of @xmath23 by modelling the bulk as serially connected resistances , composed of the inner contact resistance(s ) @xmath43 and the resistance of the 2des between contacts @xmath44 . decreasing the magnetic field , results in the quantisation of @xmath22 and an increase in @xmath23 , whereas @xmath42 remains non - zero ( case e ) .
we elucidate these behaviours as follows : the region between contacts 1 and 4 becomes incompressible and decouples the hall contacts and yielding a perfect quantisation .
however due to the density gradient induced by the inner contacts , the entire bulk is not incompressible .
we model this situation by a resistance and a capacitor ( @xmath45 ) connected in parallel . since the capacitance approaches zero due to @xmath52 , all the measured potential results from @xmath44 and @xmath43 .
in addition there are regions between the contacts 3 and 4 where scattering takes place , yielding a non - zero @xmath42 .
notice that , @xmath42 approaches zero in the e@xmath53 case since the resistance along the leaky incompressible strip is reduced compared to a fully compressible bulk , as in case e@xmath54 . lowering @xmath3
further , results in formation of a bulk incompressible region , hence insulating state , spread all over the sample : case d. here , qhe is well developed and @xmath23 becomes remarkably large and is bound by a cutoff voltage , which we attribute to finite tdos at @xmath4 .
namely , since @xmath55 , where @xmath56 is the excess current , the impedance reads @xmath57 constant . at a lower @xmath3 the bulk incompressible region
splits into strips , two of which reside along the edges and the other two encircling the inner contacts .
the edge strips decouple hall contacts and , simultaneously , the encircling strips decouple the inner contacts due to their approximately infinite capacitance , resulting in a constant @xmath23 , known as the corbino effect .
@xcite note that , from the plotted measurements one can not quantitatively determine the boundary between case ( c ) and ( d ) , however , when we discuss the temperature effects this transition will become clear .
the cardinal situation is where the encircling strips become thermodynamically compressible , i.e. similar to a leaky capacitor depicted by broken lines as depicted in case e@xmath53 , whereas the edge strips are still well decoupling as shown in fig .
[ fig:2]b by solid strips . in short ,
the evanescent encircling strips can be considered as leaky capacitors and the impedance between inner contacts is modelled similar to that of case ( f ) . on the other hand ,
the edge strips are sufficiently wide , due to smoother edge density profile compared to that of in front of the inner contacts , and therefore decouple hall contacts yielding a plateau .
the edge strips connect contacts 3 - 4 , hence , @xmath42 is still zero .
the situation in case ( a ) is trivial , where the edge strips also become transparent to hall voltage and iqhe fades .
now , we are obliged to answer questions asked at the end of the previous section .
@xmath23 is constant only in a certain @xmath3 subinterval since the inner contacts are only decoupled by the incompressible bulk ( case d ) or by the encircling incompressible strips ( case c ) for a specific @xmath3 interval and the observed asymmetry is a direct result of incompressible state distribution depending on @xmath3 .
namely at low field side it is due to encircling strips whereas at high field side it emanates from the bulk incompressible region .
our model also covers the @xmath58 and 2/3 plateaus , by lifting the assumption on spin effects and taking the energy gap to be the zeeman gap for @xmath58 and the many - body gap for @xmath59 .
interestingly , for all other visible plateaus we do not observe the saturated @xmath23 , which suggests that the 2des does not comprise an entire incompressible bulk .
note that , for @xmath60 and 4 the maximum value of the spikes hits the constant value @xmath61 1.35 mv .
as the incompressibility conditions of the edge or encircling strips are critical to determine the transition from case ( c ) to ( d ) , it is worth to investigate the effect of temperature on the observed features , by which we can also test our constant tdos argument .
one expects that , an increase in @xmath62 will result in decrease of the saturation value of @xmath23 which indirectly measures the available states at @xmath4 .
in addition , the incompressible regions will shrink in their widths at higher temperatures , hence the @xmath3 interval that we observe saturated @xmath23 will also decrease .
this will allow us to determine qualitatively the transition between case ( c ) and ( d ) .
fig .
[ fig:3 ] plots the temperature dependence of @xmath23 together with @xmath22 .
the experimental findings strongly follow our expectations , such that the saturation value decreases ( @xmath63 ) meanwhile subinterval shrinks till the thermal energy overcomes the energy gap .
remarkably , we can identify the largest incompressible bulk at @xmath64 for @xmath40 from the maxima of @xmath23 , where the slope traces the spatial extend of the bulk incompressible region .
notice that , the encircling strips decay faster due to thermal broadening of the tdos , which makes easy for electrons to scatter across the incompressible strips , hence become leaky .
similar line of argumentation also holds for @xmath58 , however , for the focused fractional state @xmath59 the maxima shifts to the high field edge of the plateau interval .
we attribute this effect to different activation behaviour of the many - body gap .
temperature dependency of @xmath23 completes our experimental investigation .
* conclusion * we have reported on the bulk transport measurements of a geometrically modified hall bar .
on one hand we experimentally evidenced that it is possible to observe quantized hall effect even if the bulk is not entirely incompressible , however , we simultaneously showed that it is also possible to observe quantized conductance if an incompressible strip resides at the edges preventing scattering between the hall contacts . on the other hand , the incompressible bulk based topological theories are well justified at certain magnetic intervals also for finite size and non - periodically bound systems .
in addition we showed that , the incompressible strips narrower than the thermodynamical length scales are prone to become leaky in the plateau - to - plateau transition intervals . for the iqhe , remarkably , there exists only a single incompressible strip with a given integer filling factor , in contradiction to single particle theories .
nevertheless , note that the berry flux enclosed in real space comprised by the incompressible strip equals to the number of fully occupied landau levels , hence , the chern number is still determining the hall conductance .
new experiments based on the screening theory outlined here would be significantly important to determine interrelation between the topological insulators and quantized hall effect which even could be extended to non - abelian states . |
dwarf galaxies are at the focus of a major cosmological problem affecting the @xmath3cdm scenario : the number of dark matter subhalos around milky way - type galaxies predicted by @xmath3cdm simulations is much larger than the number of observed satellite dwarf galaxies @xcite .
most proposals to overcome this problem stem from the idea that the smallest halos would have formed very few stars or failed to form stars at all , and that gas would have been removed in an early epoch . in this way , the lowest mass sub - halos would be either completely dark , and thus undetectable , or extremely faint .
two main mechanisms are usually invoked as responsible of the smallest sub - halos failing to have an extended star formation history ( sfh ) : heating from cosmic ultraviolet ( uv ) background radiation arising from the earliest star formation in the universe @xcite and internal sn feedback following the first star formation episodes in the host dwarf galaxy .
the cosmic uv background raises the entropy of the intergalactic medium around the epoch of reionization , preventing baryons from falling into the smallest sub - halos and it can also heat and evaporate the interstellar medium of larger sub - halos which have managed some star formation .
the former would never form stars while the latter would presently show only a very old stellar population .
recent high resolution simulations of dwarf galaxy formation show that the cosmic uv radiation field can also still suppress star formation , even when it can not evaporate the gas from the halos , by simply preventing gas from becoming dense enough to form molecular clouds @xcite , verifying a previous proposal by @xcite .
it has also been proposed that ram pressure stripping in the diffuse corona of the host massive galaxy could very rapidly remove the ism already heated by the cosmic uv even over a large range of dwarf galaxy masses @xcite .
however , such a mechanism would become dominant later , during the main accretion phase of typical milky way - sized halo , at @xmath4 , in principle allowing star formation to extend for at least a couple of gyr beyond the epoch of reionization .
although consensus exists on the important role played by the two former mechanisms , less clear is the mass range of the affected sub - halos ( e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
the fact that most or all of the recently discovered ultra - faint dwarfs ( ufds ) appear to host only a small population of very old stars , points to them as possible fossils of this process , but some of the classical dsph galaxies may also be affected .
besides heating the gas , uv photons produce the global cosmic reionization .
the redshift at which the universe was fully reionized was @xmath5 , as obtained from the presence of the gunn - peterson trough in quasars , although there is increasing evidence that this process was inhomogeneous @xcite . according to models , the minimum circular velocity for a dwarf halo to accrete and cool gas in order to produce star formation is in the range of @xmath6 to @xmath7 kms@xmath8 , which corresponds to a total mass of @xmath9 m@xmath2 .
however , while most dwarf galaxies in the local group show circular velocities below this range and dynamical masses smaller than @xmath10 m@xmath2 ( see e.g. , * ? ? ?
* ) , many of them have cmds that have been interpreted as indicating the presence of star formation activity extended well beyond the reionization epoch , even in old dsph galaxies .
two main mechanisms have been proposed to overcome this problem .
the first one is that dwarf halos could have been much larger in the past and have lost a significant amount of mass due to tidal harassment @xcite .
this scenario is further supported by detailed simulations of the tidal interaction between satellites and the host which includes also the baryonic component and ram pressure stripping @xcite . however , counter - arguments exists pointing to dwarf halos being resilient to tidal harassment @xcite .
the second one is that a self - shielding mechanism would be at work , protecting the gas in the central denser regions of the dwarf galaxy , where gas can be optically thick to the impinging radiation field .
the first mechanism is robust , since it is a natural consequence of hierarchical accretion as dwarf satellites move on highly eccentric orbits , suffering strong tidal shocks from the host potential .
the second mechanism is more subtle , since models neglect a related , competitive effect , namely that the local uv radiation from the primary galaxy or nearby proto - clusters could have been much higher than the mean cosmic ionizing flux at @xmath11 @xcite .
models by show that milky way satellites with total luminosities @xmath12 are very unlikely to be true fossils of the reionization epoch , and that they probably are the result of hierarchical build - up from smaller halos extending beyond the reionization epoch . more specifically , conclude that the simulated properties of true fossils , i.e. , those which have not undergone any merging events after reionization , agree with those of a subset of the ultra - faint dwarf satellites of andromeda and the milky way .
also , they found that most classical dsph satellites are unlikely true fossils , although they have properties in common with them : diffuse , old stellar populations and no gas .
we note that , while it would seem natural to associate ufds to reionization fossils , alternative explanations for their origin have recently appeared in the literature in which at least a fraction of them could be remnants of the oldest , most heavily stripped population of galaxy satellites accreting at @xmath13 onto the milky way halo @xcite
. used the fraction of star formation produced before reionization as a test to distinguish true fossil galaxies from non - fossils , defining the former as those having produced at least 70% of their stars by @xmath14 .
this criterion was defined after the analysis of the theoretical simulations and has the advantage of allowing a thorough comparison of models and observations . in turn , @xcite have obtained the sfhs of 38 local group dwarf galaxies with stellar masses in the range @xmath15 , finding that only five of them are consistent with forming the bulk of their stars before reionization and that only two out of the 13 predicted true fossils by show a star formation quenched by reionization .
however , it should be noted that the results of @xcite are affected by limited time resolution at early ages while the predictions by are free of the these effects .
these effects are expected to significantly modify the sfhs obtained from observational data , as can be seen in the simulations shown by e.g. , @xcite or @xcite among others . as a consequence of this , either observational effects should be simulated in theoretical models or they should be taken into consideration when comparing observational results with the theoretical models .
the second is the objective of this paper . in this paper , we discuss how the temporal resolution limitations of a sfh derived from a cmd might be accounted for . in particular , the goal of this paper is to obtain an estimate of the maximum fraction of mass , _ consistent with the observations _
, which has been converted into stars by @xmath14 and of the time by which 70% of the baryonic mass has been converted into stars in each galaxy .
these values can be directly compared with the predictions of any theoretical model of star formation in dwarf galaxies and we do this comparison , for illustrative purposes , with some of the most recent ones , currently available in the literature .
we have used high resolution sfhs obtained for a set of isolated local group dwarfs by the lcid collaboration ( * ? ? ? * and references therein ) .
our approach opens a new window to the possibility of testing the effects that cosmic ultraviolet ( uv ) radiation and internal sn feedback have on the early sfh of dwarf galaxies ( whether or not combined with self - shielding or other effects ) .
this is an exploratory paper in which we provide details of the method and apply it to a sample of very well studied dwarf galaxies .
environmental effects , such as tides and ram pressure , affect the mappings between present - day and pre - infall mass satellites and between baryon content and star formation .
therefore , we specifically concentrate on a sample of relatively isolated faint dwarfs .
the chosen sample is diverse enough to contain both dirrs , transition dwarfs and dsphs . in future works
, the method will be applied to an extended sample including ultra faint dwarfs .
the structure of the paper is as follows . in section [ secdat ]
observational data are described . in section [ secdeconvol ] ,
the proposed method is explained and applied to a sample of four local group dwarf galaxies .
sfhs from section [ secdeconvol ] are compared in section [ secmod ] with a representative set of state - of - the - art theoretical models of galaxy formation .
the results are discussed in section [ secsum ] , together with the main conclusions . as with the previous lcid papers ,
cosmological parameters of @xmath16 kms@xmath8mpc@xmath8 , @xmath17 , and a flat universe with @xmath18 are assumed @xcite .
for this work , regarding the observational material , we have used the sfhs of the galaxies cetus @xcite , tucana @xcite , lgs-3 @xcite , and phoenix @xcite , obtained by the lcid collaboration .
data were obtained with the acs and wfpc2 onboard hst .
the sfhs used in this paper were derived using the iac method , based on the suite of codes iac - star / iac - pop / minniac . for a more detailed analysis ,
we have divided the three galaxies with larger field coverages , cetus , tucana , and lgs-3 , into two regions : an inner one , within one scale length from the center , and an outer one located beyond two scale lengths .
the properties of the galaxies are summarized in table [ t1 ] .
the total mass in stellar objects ( @xmath19 ) , @xmath20 luminosity ( @xmath21 ) , velocity dispersion ( @xmath22 ) , and metallicity ( @xmath23 $ ] ) are given .
@xmath21 , @xmath22 , and @xmath23 $ ] are from @xcite .
the @xmath19 values are calculated scaling @xmath21 with the mass - luminosity relation obtained from the sfh solution of each galaxy .
the @xmath21 values range from @xmath24 to @xmath25 , bracketing the limiting value obtained by for galaxies likely to be true fossils .
robust sfhs are derived from the cmds of resolved stellar populations , but they are still affected by several sources of uncertainty that limit time resolution . in short , these sources are of three kinds : ( 1 ) those affecting data , ( 2 ) those linked to physical properties , and ( 3 ) those inherent to the methodology used to derive the sfh . limitations of the first kind are related to photon counting statistics , defective flat - field corrections , or psf sampling , among others .
sources of the second kind are mainly distance to the object ( contributing to blending and crowding ) , background and foreground contamination , and differential reddening . the term _ observational effects _ has often been used by our team to refer to these two kinds combined and we will adopt it here from now on .
they result in limiting the photometry depth and completeness , which in turn vary with stellar colors and magnitudes .
they are modeled ( e.g. , with artificial star tests ) and accounted for when synthetic populations are computed .
effects of the third kind refer to the robustness of the method and include the accuracy of the stellar evolution libraries from which synthetic populations are simulated as well as the way in which the best solution is reached .
other items closely related to the physics of the problem like the age - metallicity degeneracy in the cmd are also involved .
all these effects combined result in time resolution limitations and age inaccuracies in the final sfh solutions . to derive the sfh from the cmd of a resolved stellar population one makes a reliable simulation of observational effects ( 1 and 2 above ) in the cmd of one or several synthetic stellar populations .
these cmds are in turn compared with the observational cmd to obtain the sfh .
and @xcite made an analysis of the effects of this process on the time resolution . to a good approximation
it can be assumed that all together , the effects on the sfh are similar to a temporal shift plus a convolution with a gaussian function , @xmath26 .
* it should be noticed that average ages of older populations tend to be biased to younger values .
this is mainly the effect of the time limit imposed by the method , that does not allow ages older than the age of the universe in the solution . *
the function @xmath26 can be obtained according to the following recipe ( see * ? ? ?
first , a single burst stellar population with no age dispersion ( or a very small one ) is computed with the age at which @xmath26 is searched .
second , the observational effects obtained for the galaxy in the artificial star tests are simulated .
third , the sfh of this synthetic population is derived .
the result can be taken as @xmath26 .
the @xmath26 function can be used to partially remove the observational effects from the sfh derived for a galaxy . to do so
, one can proceed parametrically .
we do this by : ( i ) computing a large number of model sfhs of given shape , peak , and duration ; ( ii ) convolving them with @xmath26 , and ( iii ) selecting those producing results compatible with the observationally derived sfh .
the result of this inverse procedure is not the real , free of errors sfh .
the problem we are facing is in fact a bias - variance tradeoff one ( see e.g. , * ? ? ?
our objective is removing bias , but the intrinsic variance of the problem remains .
nevertheless it provides a better sfh for comparing to the theoretical models of galaxy formation .
this is especially true for the oldest ages , where the limitations on temporal resolution are greatest .
specifically , we proceed as follows for each observed galaxy field .
first , we obtain @xmath26 functions for a range of input ages from 5.0 to 13.0 @xmath27 and we select the one whose resulting peak age is closest to the galaxy sfh peak age .
input ages of the selected @xmath26 functions range from 11.0 ( inner lgs-3 field ) , to 13.0 gyr ( outer cetus and tucana fields ) .
second , we have computed a large number of sfhs shaped according to gaussian functions of amplitude @xmath28 , mean age @xmath29 , and standard deviation @xmath22 .
we will refer to these trial model sfhs simply as _
trial models_. values of @xmath29 have been sampled from 6.0 to 13.5 gyr with step of 0.1 gyr . in turn
, @xmath22 has been sampled from 0.1 to 6 gyr with steps of 0.1 gyr and with the additional criterion that the resulting functions are truncated for age larger than 13.5 gyr .
trial models include a simple but realistic simulation of the metallicity distribution of the real galaxy for the model age . to this
, a gaussian metallicity distribution is used .
the mean metallicity is the metallicity observationally obtained for the galaxy at age @xmath29 and the sigma is the metallicity dispersion obtained for the galaxy and the same age .
each trial model shape has been computed a total of 201 times , each one with a different value of @xmath28 .
a total of about @xmath30 trial models have been computed for each galaxy field . in the third step
, each trial model has been convolved with the selected @xmath26 function .
we will refer by _ convolved models _ to the results of these convolutions .
the fourth step has been to select all the convolved modles that are compatible with the sfh obtained for the galaxy field within the error intervals of the latter , including its integral .
to apply this criterion , only the age interval @xmath31 gyr has been considered .
the final step has been to average all the compatible convolved modles and , in turn , all the trial models originating them .
figure [ f0 ] shows a summary of the models , including the ones selected for the case of the outer field of the tucana galaxy .
general results are given in figure [ f1 ] , which shows that , except for the case of phoenix , the average of the _
good _ trial models peaks at older ages and is narrower than the convolved models or the observational sfhs . following this ,
the averages of the good trial models will be used to make direct comparisons to theoretical models of galaxy formation .
the gaussian functions used above are clearly a simplified representation of the sfh .
it is not expected to reproduce extended sfhs or those in which recursive star formation bursts are going on for an extended period of time .
indeed , figure [ f1 ] shows that intermediate ages for phoenix are not well reproduced .
nonetheless it can still be a good approximation for old stellar populations which is the purpose of this exercise . to check for the dependence of the solutions with the trial model shape
, we have repeated the same experiments using in turn triangular and step functions . in all the cases results
are similar and do not significantly change the conclusions of our work .
since we are interested here in the fraction of stars formed before a given time , it is better to use the cumulative sfh . we define it as @xmath32 , where @xmath33 is the age assumed for the onset of the star formation and @xmath34 is the star formation rate
figure [ f2 ] shows the observed and the good trial model cumulative sfhs for all the galaxy fields .
the latter is the information needed to make a direct comparison with theoretical models of galaxy formation .
in figure [ f3 ] the average of the good trial models of the galaxies are compared with a few , representative theoretical models of galaxy formation , namely models 2 , 7 , and 22 by @xcite , model _ doc _ by @xcite , and two models by @xcite representative of old and old+intermediate age dwarf galaxies .
the latter are those given by the authors by the green curves in their figure 5 .
@xcite have presented high - resolution hydrodynamical simulations of the formation and evolution of isolated dwarf galaxies including the most relevant physical effects , namely , metal - dependent cooling , star formation , feedback from type ii and ia sne , uv background radiation , and internal self - shielding .
models 2 and 7 include uv background radiation and internal self - shielding together with sne feedback .
model 22 includes only sne feedback , i.e. , it is representative of the case that uv background radiation has no effect on the sfh .
@xcite have carried out fully cosmological , very high resolution , @xmath3cdm simulations of a set of field dwarf galaxies . model _
doc _ corresponds to a virial mass of @xmath35 .
finally , @xcite have used the cosmological hydrodynamical simulation of the local group carried on as part of the clues project . in all the cases ,
the reader is referred to the source papers for details
. some relevant properties of the model galaxies are summarized in table [ t1 ] .
the total mass in stellar objects ( @xmath19 ) , @xmath20 luminosity ( @xmath21 ) , velocity dispersion ( @xmath22 ) , and metallicity ( @xmath23 $ ] ) are given .
@xmath19 is provided by @xcite and @xcite , respectively . regarding @xmath21 ,
the values given by @xcite are listed for their models . for the @xcite models , the @xmath21
are obtained with iac - star using the model sfhs provided by these authors as input .
no data are given for the @xcite models , since these authors do not provide them .
figure [ f3 ] shows that the average of the cumulative good trial models of cetus , tucana , and lgs-3 lay between models sm2 and sm7 ( both including reionization ) , on the one side , and models sm22 ( without reionization ) and bl-1 ( average of their oldest models ) , on the other side .
furthermore , the outer parts of cetus and tucana show a reasonable correspondence with sm7 , indicating that reionization could have played a role in quenching the star formation in those regions . however , according to this figure , the inner regions of both galaxies plus lgs-3 ( inner and outer ) and phoenix seem difficult to reconcile with the @xcite models with reionization . a good correspondence exists between the inner region of lgs-3 and models bl-1 and sm22 , while no good match is found for phoenix with any of the models considered here .
finally , models bl2 , doc , and bashful are , by far , too young to reproduce any of the observed galaxies and are likely to be more suitable for much younger galaxies , like ic-1613 ( see * ? ? ?
have defined true fossil galaxies as those having formed all or most of their stars before the reionization era , at @xmath14 .
they used the cumulative fraction of star formation produced before reionization as a test to distinguish true fossil galaxies from non - fossils , defining the former as those having produced at least 70% of their stars at @xmath14 , which , for the cosmological parameters adopted here ( see above ) , corresponds to approximately @xmath36 gyr ago .
table [ t2 ] gives , for each galaxy field and model , the age at which @xmath37 and the value of @xmath38 for redshift @xmath14 .
columns 2 and 3 list the values corresponding to the observed sfhs . columns 4 and 5 give those corresponding to the average of the good trial models .
errors have been obtained from the error bands shown in figure [ f2 ] .
two main conclusions can be obtained from table [ t2 ] .
first , the age for which the cumulative sfh , @xmath38 reaches 70% is increased by observational effects by @xmath39 on average , while the value of @xmath38 at @xmath14 is increased by a factor of @xmath40 , on average , although it can be larger than two in some cases .
this shows that working with the average of good trial models is useful if the earliest evolution of dwarf galaxies is sought .
second , taken at face value , none of the four galaxies analyzed here fulfill the criterion by of having @xmath41 at @xmath14 , to be considered true fossils from the pre - reionization era , although the outer regions of cetus and tucana might marginally qualify , within a 2@xmath22 error interval .
in addition , also concluded that galaxies with @xmath42 are unlikely true fossils , while those with @xmath43 remain reasonable candidates to be fossils of the first galaxy generation .
however , three out of the four galaxies considered here have @xmath44 ( tucana , lgs-3 , and phoenix ) and are not true - fossils .
in summary , we have presented a method to address the limitations of the temporal resolution of the early sfhs of galaxies derived from deep cmd modeling .
we have applied the method to the analysis of the duration of the early star formation activity in a sample of local group dwarf galaxies with the purpose of testing whether or not they are true fossils of the pre - reionization era . for a study of this kind to be accurate , the affects of limited time resolution need to be accounted for . for testing for fossils , using sfhs directly derived from deep cmd modeling may produce biased results due to limited time resolution .
we have shown elsewhere @xcite that the sfh derived for a synthetic , single age and metallicity stellar population can be simulated by a gaussian whose sigma depends on age ( @xmath26 ) .
the method presented here consists in computing a large number of stellar populations with gaussian sfhs and metallicity distribution similar to the on obtained for the galaxy for the same age interval ( that we call _ trial models _ ; about @xmath30 per galaxy in this case ) of varying mean , sigma , and amplitude , and selecting all those that , convolved with @xmath26 , produce results compatible with the optimal sfh directly derived from observations .
the average of the good trial models is an improved approach to the real sfh of the galaxy at the very earliest times . in general , the ages of the averages of the good trial models are about @xmath45 gyr older than the optimal sfh solutions for the case of predominantly old galaxies .
we have applied our method to four local group dwarfs : cetus , tucana , lgs-3 , and phoenix and we have compared our results with predictions made by a number of recent cosmological hydrodynamical simulations for the formation and evolution of dwarf galaxies . a relatively sharp exhaustion of the star formation is necessary at an early epoch , close to @xmath14 , to account for the sfh obtained for the outer region of tucana and cetus .
the inner part of both galaxies , as well as lgs-3 and , more clearly , phoenix , are compatible with models predicting more extended star formation activity for dwarf galaxies . however
, none of the galaxies and fields studied here , except perhaps the outer regions of cetus and tucana , fulfill the criterion to be considered as fossil remnants of the pre - reionization era . as a cautionary note , at present
, it can not be excluded that cetus and tucana were affected by environmental effects to some extent .
their distances to the massive spirals of the lg , while placing them outside the virial radius of either today , is still small enough to admit scenarios in which these galaxies are on nearly radial orbits and they may have already had a close pericenter passage with their host @xcite .
if they are bound satellites , their relatively large velocity dispersion , @xmath46 km / s ( see table 1 ) , would make them comparable to the brightest dsph satellites such as fornax , and implies their halos could have been much more massive before tides began to prune them substantially .
if this were the case , it would be likely that their halo mass before infall was high enough to place them clearly above the threshold mass for reionization to play a role , strengthening further the conclusion that they can not be reionization fossils .
the authors thanks the anonymous referee for her / his comments and suggestions , that help to improve the paper clarity .
this work has been funded by the economy and competitiveness ministry of the kingdom of spain ( grants aya2010 - 16717 and aya2013 - 42781-p ) and by the instituto de astrofsica de canarias ( grant p3/94 ) .
aparicio , a. , & gallart , c. 2004 , , 128 , 1465 aparicio , a. , gallart , c. , & bertelli , g. 1997 , , 114 , 680 aparicio , a. , & hidalgo , s. l. 2009 , , 138 , 558 becker , r. h. , fan , x. , white , r. l. , et al .
2001 , , 122 , 2850 becker , g. d. , bolton , j. s. , madau , p. , et al .
2015 , , 447 , 3402 bentez - llambay , a. , navarro , j. f. , abadi , m. g. , et al . 2015 , , 450 , 4207 bovill & ricotti 2011a , , 741 , 17 bovill & ricotti 2011b , , 741 , 18 bullock , j. s. , kravtsov , a. v. , & weinberg , d. h. 2000 , , 539 , 517 gnedin , n. y. 2000 , , 542 , 535 gnedin , n. y. , & kravtsov , a. v. 2006 , , 645 , 1054 grebel , e. k. , & gallagher , j. s. iii 2004 , , 610 , l89 hastie , t. , tibshirani , r. , friedman , j. 2009 _ the elements of statistical learning _ , springer - verlag , new york .
hidalgo , s. l. , aparicio , a. , martnez - delgado , d. , & gallart , c. 2009 , , 705 , 704 hidalgo , s. l. , aparicio , a. , skillman , e. , et al .
2011 , , 730 , 14 hidalgo , s. l. , monelli , m. , aparicio , a. , et al .
2013 , , 778 , 103 iliev , i. t. , moore , b. , gottlber , s. et al .
2011 , , 413 , 2093 kauffmann , g. , white , s. d. m. , & guiderdoni , b. 1993 , , 264 , 201 kazantzidis , s. , mayer , l. , mastropietro , c. , et al .
2004 , , 608 , 663 kazantzidis , s. , okas , e. l. , callegari , s. , et al .
2011 , , 726 , 98 klypin , a. , kravtsov , a. v. , valenzuela , o. , & prada , f. 1999 , , 522 , 82 komatsu , e. , dunkley , j. , nolta , m. r. , et al .
2009 , , 180 , 330 kravtsov , a. v. , gnedin , o. y. , & klypin , a. a. 2004 , , 609 , 482 loeb , a. , & barkana , r. 2001 , , 39 , 19 mac low , m .- m . , & ferrara , a. 1999 , , 513 , 142 mayer , l. , kazantzidis , s. , mastropietro , & c. , wadsley , 2006 , , 445 , 738 mayer , l. , 2010 , advances in astronomy , art .
278434 mcconnachie , a. w. 2012 , , 144 , 4 monelli , m. , hidalgo , s. l. , stetson , p. b. , et al . 2010a , , 720 , 1225 monelli , m. , gallart , c. , hidalgo , s. l. , et al . 2010b ,
, 722 , 1864 moore , b. , ghigna , s. , governato , f. , et al . 1999 , , 524 , l19 pearrubia , j. , navarro , j. f. , & mcconnachie , a. w. 2008 , , 673 , 226 ricotti , m. , & gnedin , n. y. 2005 , , 629 , 259 sawala , t. , scannapieco , c. , maio , u. , & white , s. 2010 , , 402 , 1599 schaye , j. 2001 , , 562 , l95 shen , s. , madau , p. , conroy , c. , et al . 2014 , , 792 , 99 skillman , e. d. , hidalgo , s. l. , weisz , d. r. , et al .
2014 , , 786 , 44 sobacchi , e. , & mesinger , a. 2015 , , 453 , 1843 spitler , l. r. , romanowsky , a. j. , diemand , j. , et al .
2012 , , 423 , 2177 susa , h. , & umemura , m. 2004 , , 610 , l5 tomozeiu , m. , mayer , l. , & quinn , t. 2015 , arxiv:1506.02140 weisz , d. r. , dolphin , a. e. , skillman , e. d. , et al .
2014a , , 789 , 147 weisz , d. r. , dolphin , a. e. , skillman , e. d. , et al . 2014b , , 789 , 148 ) the used models .
the light grey region shows the sfh of the galaxy including the 1@xmath22 error interval estimate .
the dark grey region shows the average of convolved - sfhs compatible with empirical results .
the wideness shows the 1-@xmath22 dispersion .
thick , black line shows the average of the good trial models .
thin , black lines show the 1-@xmath22 dispersion while thin dashed lines are the envelope of all the good trial models . , width=529 ] error interval estimate .
dark grey regions show the average of convolved models compatible with empirical results .
the wideness of these regions show the 1-@xmath22 dispersion .
thick , black lines show the average of the good trial models .
thin , black lines show the 1-@xmath22 dispersions . ,
width=529 ] ) compared with the same for a number of theoretical models of galaxy formation .
numbers in the figure correspond to 1 : tucana - outer , 2 : cetus - outer , 3 : tucana - inner , 4 : lgs-3-outer , 5 : cetus - inner , 6 : lgs-3-outer , and 7 : phoenix - inner . regarding to models , sm2 , sm7 and sm22
are respectively models 2 , 7 and 22 by @xcite .
model `` doc '' is by @xcite .
models `` bl1 '' and `` bl2 '' are models 1 and 2 by @xcite .
, width=529 ] ccccc cetus & 7.17 & 2.58 & @xmath47 & @xmath48 + tucana & 1.66 & 0.54 & @xmath49 & @xmath50 + lgs-3 & 2.08 & 0.94 & @xmath51 & @xmath52 + phoenix & 1.29 & 0.78 & & @xmath53 + s2 & 0.96 & 0.43 & 7.3 & @xmath54 + s7 & 10.02 & 4.50 & 9.1 & @xmath55 + s22 & 2.58 & 1.34 & 7.1 & @xmath56 + shen13-doc & 34.0 & 34.04 & & + shen13-bashful & 115.0 & 135.52 & & + bl-1 & & & & + bl-2 & & & &
+ cccccccc cetus inner & @xmath57 & @xmath58 & @xmath59 & @xmath60 + cetus outer & @xmath61 & @xmath62 & @xmath63 & @xmath64 + tucana inner & @xmath65 & @xmath66 & @xmath67 & @xmath68 + tucana outer & @xmath69 & @xmath70 & @xmath71 & @xmath72 + lgs-3 inner & @xmath73 & @xmath74 & @xmath75 & @xmath76 + lgs-3 outer & @xmath77 & @xmath78 & @xmath79 & @xmath80 + phoenix inner & @xmath81 & @xmath82 & @xmath83 & @xmath84 + sm2 & & & @xmath36 & @xmath85 + sm7 & & & @xmath86 & @xmath87 + sm22 & & & @xmath88 & @xmath89 + shen13-doc & & & @xmath90 & @xmath91 + shen13-bashful & & & @xmath92 & @xmath93 + bl - old & & & @xmath94 & @xmath95 + bl - young & & & @xmath96 & @xmath97 + | the analysis of the early star formation history ( sfh ) of nearby galaxies , obtained from their resolved stellar populations is relevant as a test for cosmological models .
however , the early time resolution of observationally derived sfhs is limited by several factors .
thus , direct comparison of observationally derived sfhs with those derived from theoretical models of galaxy formation is potentially biased .
here we investigate and quantify this effect .
for this purpose , we analyze the duration of the early star formation activity in a sample of four local group dwarf galaxies and test whether they are consistent with being true fossils of the pre - reionization era ; i.e. , if the quenching of their star formation occurred before cosmic reionization by uv photons was completed .
two classical dsph ( cetus and tucana ) and two dtrans ( lgs-3 and phoenix ) isolated galaxies with total stellar masses between @xmath0 to @xmath1 m@xmath2 have been studied .
accounting for time resolution effects , the sfhs peak as much as 1.25 gyr earlier than the optimal solutions .
thus , this effect is important for a proper comparison of model and observed sfhs .
it is also shown that none of the analyzed galaxies can be considered a true - fossil of the pre - reionization era , although it is possible that the _
outer regions _ of cetus and tucana are consistent with quenching by reionization . |
starting from the work of deutsch @xcite , it has been known for some time now , that inertial particles in a random velocity field undergo clustering phenomena .
these behaviors are observed in numerical simulation of turbulence as well ( see e.g @xcite and references therein ) and are thought to give an important contribution to coalescence phenomena , e.g. in the process of rain formation @xcite .
both in the case of random fields and of real turbulence , spatial inhomogeneity of the statistics , though contributing to particle segregation @xcite , does not appear to be an essential factor .
random velocity fields with various statistical properties , have been used to model inertial particles in turbulent flows , both in the presence of gravity @xcite and in its absence @xcite .
simplified models disregarding in part or in total the spatial structure of the velocity field have been introduced as well ( see e.g. @xcite ) , to cope with the difficulty of the analytical treatment of the multi - particle statistics . in the simplest instance , an inertial particle in a turbulent flow is characterized by the relaxation time of its velocity relative to the fluid : the stokes time @xmath0 , where @xmath1 is the particle radius , @xmath2 is the ratio of the particle to fluid density ( assumed large ) and @xmath3 is the kinematic viscosity of the fluid @xcite . experimental data @xcite and numerical simulations @xcite both indicate that clustering is stronger for particles with @xmath4 of the order of the kolmogorov time scale .
this observation , and the fact that stokes times for most atmospheric aerosols of interest lie in this range @xcite have motivated substantial analytical effort in the small stokes time regime , with smooth velocity fields mimicking turbulence at the kolmogorov scale @xcite . some analysis has been carried on also for stokes times in the turbulent inertial range @xcite , but numerical simulations @xcite indicate that concentration fluctuations do not follow simple scaling rules . due to the relative ease of analytical treatment ,
another regime which attracted attention is that of inertial particles in a smooth random velocity field , with correlation time @xmath5 much shorter than the geometric scale @xmath6 obtained from the correlation length @xmath7 and the amplitude @xmath8 of the velocity field fluctuations @xcite .
however , this is a short - time correlated regime that is very different from the situation in realistic turbulent flows : @xmath9 . the analysis of random velocity field models has allowed to identify at least two clustering mechanisms , expected to be present also in real turbulence .
the first one , originally proposed in @xcite , is preferential concentration of heavy ( light ) particles in the strain ( vortical ) regions of the flow .
this effect has been recently observed also in the case of particles with very small inertia @xcite .
the second mechanism , already present in one dimension ( 1d ) @xcite , is that of the particles catching one another in their motion , as they slip with respect to the fluid . in the case of smooth velocity fields
, this leads to the formation of caustics in the instantaneous concentration field , which , in turn , act as back - bones for the clustering process @xcite .
the clustering itself is a long - time process , which can be described in terms of the lyapunov exponents of the particle pair dynamics @xcite .
focusing on the second mechanism , it appears that caustics formation tends to become maximum for @xmath10 @xcite , and that , at the same time , clustering becomes weaker . for larger @xmath4
, one expects that the particles be scattered by the velocity fluctuations they cross in their motion @xcite as if undergoing brownian diffusion , resulting in vanishing particle correlations and in clustering destruction .
it is to be stressed that the interest of this limit is by no means academic , as particles in turbulent flows for which @xmath4 lies in the inertial range ( or above ) , will see smaller vortices precisely in this way .
purpose of this paper is to understand in detail how and under which conditions the uncorrelated limit described in @xcite is achieved .
it will appear that clustering destruction occurs in a very non trivial way , requiring consideration , among the other things , of how particle correlations decay at scales comparable with or above @xmath7 .
we are going to show that , provided we are away from the short correlation time regime @xmath11 , clustering at @xmath12 will not be dominated by the small - separation particle pair dynamics ( small with respect to @xmath7 ) described in @xcite .
particle pairs remaining close long enough for a lyapunov exponent approach to be appropriate , are still present , but their contribution to concentration fluctuations is negligible . the key mechanism for the destruction of concentration fluctuations will appear to be that , at large @xmath13 and finite @xmath14 , only particles with increasingly small relative velocities ( but not small enough for a local theory on the lines of @xcite to be valid ) have a chance to be correlated .
this is to be contrasted with the picture in @xcite , of correlation fluctuations disappearing as caustics occupy in the above limit larger and larger portions of space , and with the one in @xcite of saturation to the space dimension of the particle distribution correlation dimension . neither in @xcite nor in @xcite , however , was a quantitative prediction on clustering decrease provided .
the present analysis will also allow to show that , for large @xmath13 and rather generic ( not too large ) values of @xmath15 , clustering is associated with smaller typical relative inter - particle velocities , compared to what would be observed in the absence of correlations .
thus , there are circumstances under which , clustering may hinder rather than enhance coalescence phenomena .
this paper is organized as follows .
the main definitions and approximations will be presented in section ii , following by and large the notation of @xcite . in section iii
, the pair particle dynamics will be analyzed in the large @xmath4 limit , identifying the relevant time and velocity scales for clustering . in section iv
, the relation between particle - particle velocity depletion and clustering will be established , and heuristic estimates for the concentration correlation will be obtained , starting from the fokker - planck equation for the distribution of the relative particle velocity and coordinate . in section
v , a proof for the dominance of slowly approaching particle pairs in cluster generation will be provided .
section vi contains the conclusion .
consider an ensemble of inertial particles transported by a 1d zero mean , gaussian random velocity field @xmath16 with correlation u(x , t)u(0,0)=_u^2f(t/_e)g(x / r _ ) , [ eq1 ] where @xmath17 , the function @xmath18 is assumed to be smooth and to have all necessary moments , and @xmath19 .
the particles are immaterial , so that they can cross without interaction , and their velocity @xmath20 is taken to obey the stokes equation : v=_s^-1[-v(t)+u(x(t),t ) ] .
[ eq2 ] introduce the stokes and kubo numbers : s=_s/_e , k=_u_e / r_[eq3 ] and choose units so that @xmath21 ; therefore : _
e = s^-1 , r_=(ks)^-1 .
[ eq4 ] in the regime @xmath22 , the particle velocity will change little on the lifetime of a fluid fluctuation and eq .
( [ eq2 ] ) could be approximated by a langevin equation .
this , independently of @xmath23 , i.e. of the random field being short - time correlated or not .
it is possible to substitute into eq .
( [ eq2 ] ) : u(x(t),t)(2_p(v))^1/2(t ) , [ eq5 ] with @xmath24 white noise : @xmath25 and _ p(v)= _ 0^ t [ eq5.01 ] the correlation time of the fluid velocity sampled by a particle moving at speed @xmath20 . in the above formula
, the averages are calculated along trajectories , with @xmath26 indicating the condition on @xmath20 .
the average on the trajectory in the denominator , due precisely to the presence of concentration fluctuations , does not necessarily coincide with the space average @xmath27 . from eq .
( [ eq1 ] ) , we can put @xmath28 , provided @xmath29 , with @xmath30 the particle velocity variance . if @xmath28 , we can estimate from eqs .
( [ eq2 ] ) and ( [ eq5 ] ) : ^2_vs^-1 , [ eq5.1 ] which is satisfied if @xmath31 .
following the same line of reasoning ( see appendix ) , it is possible to show that prescription problems in the definition of the white noise arise only at @xmath32 .
notice that @xmath31 and @xmath22 imply @xmath33 , where @xmath34 is precisely the force magnitude parameter introduced in @xcite .
this condition is easily satisfied unless one chooses to work in a frozen turbulence regime @xmath35 .
turning to the relative motion of particle pairs , let us introduce difference variables @xmath36 where 1 and 2 label members of a particle pair , and indicate @xmath37 . in the regime @xmath22 , starting from eq .
( [ eq2 ] ) , we can approximate the equation for the relative motion of particles with the langevin equation : = -+b(r ) , r=. [ eq7 ] for @xmath38 , we have for the noise square amplitude @xmath39 , from eq .
( [ eq1 ] ) : @xmath40[u_2(0)-u_1(0)]|\nu , r\rangle\d t \nonumber \\ \nonumber \\ & \simeq 4s^{-1}[1-g(r / r_\v ) ] .
\label{eq8}\end{aligned}\ ] ] the decoupling of the difference variables @xmath41 from the center of mass variables @xmath42 and @xmath43 descends basically from the condition @xmath44 .
in order to determine how concentration fluctuations are generated , we need to understand the particle pair dynamics at small separations . in the large @xmath45 limit , we expect that the velocities of colliding particles be uncorrelated , however , this condition turns out not to be automatically satisfied .
a condition for the particles to be weakly correlated , despite being close , is that the stokes time be much longer than the time spent at @xmath46 , so that the velocity of the two particles will be the result of many contributions by uncorrelated regions of the fluid ( see fig .
[ corrfig0 ] ) . under these conditions
, particles will move ballistically at scale @xmath46 , and they will cross at velocity @xmath47 , after a time @xmath48 .
the weak correlation condition is therefore @xmath49 , i.e. : = k^2s1 .
[ eq9 ] @xmath50 and @xmath51 @xmath52 .
the exit time to the right is @xmath53 .
the small rectangles are space - time correlated regions for the fluid , of size @xmath7 and duration @xmath5 . the time @xmath4 quantifies the particle memory .
the particle approach in @xmath50 is governed by an @xmath54 negative lyapunov exponent @xcite .
, width=283 ] the parameter @xmath55 is the same as in the theory in @xcite , who considered in detail the small @xmath55 regime . for @xmath56 ,
the time spent by a pair of particles at @xmath46 before collision , would be longer than @xmath4 , and the particles would have memory only of that last portion of their history , when they were at @xmath57 ; therefore , @xmath58 could not be treated , at the time of crossing , as a difference of uncorrelated velocities @xcite .
notice that the conditions @xmath56 and @xmath22 can be realized only if @xmath59 , corresponding to a short - time correlated random velocity field . in this regime
, @xmath55 plays the role of an effective stokes number ( see in particular @xcite ) , in which the place of @xmath5 is taken by @xmath60 , that is the diffusion time of a ( non - inertial ) tracer across a distance @xmath7 .
notice also that the condition @xmath61 , i.e. @xmath62 , is not sufficient by itself to guarantee weak correlation , that in fact is not fulfilled in the range @xmath63 , where @xmath56 . back to the large @xmath55 range , although most particles will be weakly correlated , those approaching at speed @xmath64 may stay at @xmath65 sufficiently long to end up being strongly correlated .
it is then important to understand whether an effect like clustering , which depends on pair - correlation , is due to the effect of the many weakly correlated particle pairs , or to the one of the few strongly correlated ones .
we will answer this question in sec .
v. for the moment , we identify the important time and velocity scales involved in the dynamics of slowly approaching particle pairs .
we rescale variables : t = t / t _ , = / _ , r = r / r _ , [ eq10 ] where t_=(2)^-1/3 , _ = 2 ^ 1/3 ^ -1/6_v [ eq11 ] and @xmath7 is given in eq .
( [ eq4 ] ) . in the new variables , at short separations , the noise amplitude ( [ eq8 ] ) can be taylor expanded and eq .
( [ eq7 ] ) becomes : = -(2)^-1/3+r ; = .
[ eq12 ] thus , @xmath66 and @xmath67 are in some sense the time and velocity scales at the correlated scale @xmath7 .
notice that the condition for eq .
( [ eq12 ] ) to be meaningful as a stochastic differential equation is @xmath68 , i.e. again @xmath33 . to understand better the meaning of @xmath66 and @xmath67 ,
consider the separation of a pair of particles such that @xmath69 and @xmath70 .
this is equivalent to looking at the approach of a particle pair , with conditions imposed on the velocity at the moment of crossing , rather than at a previous initial time . in the initial phase
, the separating particles will move at almost constant relative velocity , and @xmath71 .
the variation of @xmath72 will be @xmath73 so that @xmath74 .
the initial phase will last until @xmath75 , i.e. until @xmath76 , when @xmath77 .
we then see that @xmath78 , i.e. @xmath79 is the minimum relative velocity to which the approximation of constant relative velocity at the correlated scale applies , and @xmath66 is the associated exit time : @xmath80
. the following phase will be determined by the linear nature of eq .
( [ eq12 ] ) ; for @xmath81 , the drift @xmath82 can still be disregarded , and we expect @xmath83 . after a time of the order of @xmath84 , the particles will have moved apart by @xmath85 , i.e. they will be moving out of a correlated region for @xmath86 . in this final part of the separation process , eq .
( [ eq12 ] ) will cease to be valid , with @xmath72 obeying instead @xmath87 . for @xmath88 ,
@xmath58 will behave like a brownian motion : @xmath89 , while @xmath90 will undergo a richardson - like diffusion process : @xmath91 . in conclusion ,
we have two groups of particle pairs . 1 . the great majority , for which , at the moment of crossing : @xmath92 ; their motion in a correlated region can be approximated as ballistic and their permanence time is @xmath53 .
a smaller fraction , for which , at the moment of crossing : @xmath93 , with @xmath94 ; their dynamics in a correlated region is determined by fluctuations of @xmath16 at scale @xmath95 and their permanence time grows logarithmically as @xmath96 : @xmath97 .
the fact that for @xmath51 , @xmath98 , has the consequence that only particles meeting at exponentially small relative velocities @xmath99 will have a chance to stay within a correlated region for a time @xmath100 . an exponentially low level of molecular diffusion is thus sufficient to destroy these pairs . even in the absence of molecular diffusion , however , the relative motion of these pairs is characterized in the limit @xmath101 by a positive lyapunov exponent @xcite and no long - time contributions to clustering are present .
the remaining particle pairs can be shown not to produce singular contributions to the two - particle probability density function ( pdf ) for @xmath58 @xcite .
we can then estimate the fraction of particle pairs staying close for a time @xmath102 , i.e. the fraction of particles crossing at speed @xmath103 ( provided of course @xmath104 ) , as @xmath105 this , again , is exponentially small for @xmath100 and large @xmath55 , while the fraction of particle pairs that are merely not ballistic , is the much larger value @xmath106 .
the situation in the small @xmath55 regime is very different , as eq .
( [ eq12 ] ) leads to instantaneous equilibration of the particle velocity : @xmath107 and the ballistic particle fraction , associated with caustics formation , disappears @xcite .
hence , @xmath108 and the particles behave for @xmath56 as a mono - disperse phase ( although with velocity not locally equal to that of the fluid , as it happens instead in the small @xmath45 limit @xcite ) . at @xmath109
, all particle pairs are therefore co - moving .
we have seen at the end of section ii that in the regime @xmath22 , @xmath38 , the difference variables @xmath110 decouple from the center of mass ones .
this means that at equilibrium , the two - particle pdf will be in the form @xmath111 where @xmath112 is the length of the domain for @xmath113 .
( unless ambiguous , we do not use subscripts to identify pdf s referring to different stochastic variables ) .
multiplying by @xmath114 , with @xmath115 the total number of particles in the domain @xmath112 and integrating over @xmath116 , we obtain the expression for the concentration correlation : n(r)n(0)=|n^2(,r):=|n^2[1+f(r ) ] , [ eq14 ] where @xmath117 is the mean concentration .
the quantity @xmath118 gives the strength of the concentration fluctuations .
the difficulty of the problem , compared with the @xmath56 limit , is the finite width in @xmath58 of the pdf @xmath119 . in the @xmath56 limit , small @xmath120 implied small @xmath58 , a long permanence at that particular @xmath120 and the possibility of linearizing the fokker - planck equation in @xmath120 along the lines of @xcite . in our case
this is not possible , and we must consider the full fokker - planck equation , which , from eqs .
( [ eq7 ] ) and ( [ eq8 ] ) , reads , at steady state : _ r(,r)- _
= b(r)_^2(,r ) [ eq15 ] ( see @xcite for another example of kinetic treatment of the particle pair statistics in turbulent flows )
. taking moments of the fokker - planck equation ( [ eq15 ] ) allows to draw important conclusions on the connection between clustering and decrease of the relative velocity variance .
the first two moments of eq .
( [ eq15 ] ) read : ll _
r(|r(r))=0 , + _ r(^2|r(r))+|r(r)=0 , [ eq16.01 ] where @xmath121 .
we can impose boundary conditions to eq .
( [ eq16.01 ] ) at @xmath122 , where the solution , corresponding to uncorrelated particles , is known .
the first moment equation tells us that @xmath123 is constant ; but , at @xmath124 correlations are absent ; therefore @xmath125 and @xmath126 ; thus @xmath127 for all @xmath120 . substituting into the second one ,
we get @xmath128 , and , using again @xmath125 and @xmath129 , we get : _ ^-2 ^ 2|r=[1+f(r)]^-1 .
[ eq16.02 ] in other words , clustering in 1d is necessarily associated with velocity variance decrease .
notice that this result rests only on the applicability of the fokker - planck equation ( [ eq15 ] ) , which requires @xmath130 , but is otherwise independent of @xmath55 .
an idea of how @xmath131 goes to zero as @xmath132 can be obtained treating the function @xmath18 in eq .
( [ eq8 ] ) as a perturbation .
this is motivated by the observation ( see previous section ) that most particles spend very little time at distances @xmath65 , and by the assumption ( verified in @xcite ) that the remaining particles produce only a correction to @xmath133 .
to lowest order , the particles are therefore independent random walkers with velocity correlation time @xmath4 .
a further simplification is obtained assuming that the scale of variation of @xmath119 remains @xmath134 , as in the unperturbed case .
we can then write @xmath135 , with _ 0(,r)=^-1_0()=(- ) [ eq16.1 ] the solution of eq .
( [ eq15 ] ) with @xmath136 set to zero , and estimate @xmath137 , @xmath138 for some constant @xmath139 . substituting into eq .
( [ eq15 ] ) : _ 1 ~b_0g(r / r_)_0 .
[ eq16.2 ] a new length @xmath140 , the distance travelled by a particle in a stokes time , thus enters the problem . on the scale of @xmath140 , the term @xmath18 in eq .
( [ eq16.2 ] ) behaves like a dirac delta and we find : _
1(,r)~^-1 ^ -1/2(-s^1/2|r|)_0 ( ) .
[ eq17 ] substituting into eq .
( [ eq14 ] ) would lead to concentration fluctuations with @xmath141 variance and @xmath142 correlation length .
the analysis carried on in the previous section did not answer the question of which particle pairs are responsible for the production of the concentration fluctuations at large @xmath55 .
we are going to show here that concentration fluctuations are indeed produced by slowly approching particle pairs .
we prove that the contrary is impossible and assume therefore that the dominant contribution is due to particles approaching ballistically at scale @xmath7 , which spend a time @xmath53 at that scale . in order to show that the assumption of dominance of ballistic particles can not be true , it is sufficient to calculate @xmath143 as the stationary solution of the evolution equation ( r;0)=r(r;0|,r;-t)(,r;-t ) , [ eq17.1 ] where @xmath144 is a generic non - equilibrium pdf and @xmath145=@xmath146=@xmath147=@xmath148 is a transition pdf for the dynamics described by eqs .
( [ eq7]-[eq8 ] ) . if we take @xmath149 and @xmath46 , for most particles [ i.e. for typical values of @xmath150 , we will have : |r(-t)||r(0)-(-t)t|r _ , [ eq17.2 ] which descends from the fact that , for @xmath151 and @xmath92 , @xmath58 can be approximated as constant .
but eq . ( [ eq17.2 ] )
tells us that the dominant contribution to the integral in eq .
( [ eq17.1 ] ) for @xmath152 ( the equilibrium pdf ) , is from values of @xmath153 for which @xmath154 , where @xmath155 is the spatially homogeneous pdf of eq .
( [ eq16.1 ] ) . for the ballistic particle pairs ( which we are assuming to dominate the statistics )
, we can calculate the correction to ballistic motion , writing @xmath156 , where , using eq .
( [ eq7 ] ) : ll r_0(t)&=r(-t)+(-t)(t+t ) , + r_1(t)&_-t^tb(r_0 ( ) ) ( ) .
[ eq18 ] using eq .
( [ eq18 ] ) , we can express the transition pdf @xmath157 in eq .
( [ eq17.1 ] ) in terms of the equivalent pdf @xmath158 and , using the first of eq .
( [ eq18 ] ) , in terms of the conditional pdf @xmath159 . substituting into eq .
( [ eq17.1 ] ) , with @xmath152 and @xmath160 , we obtain : @xmath161 where use has been made of the relation @xmath162
. we can taylor expand eq .
( [ eq19 ] ) around @xmath163 and we obtain @xmath164 where the average is on @xmath165 ( in terms of @xmath155 ) and @xmath166 . using eqs .
( [ eq18 ] ) and ( [ eq8 ] ) , we can write : @xmath167 where the remaining average is on @xmath58 ; setting @xmath168 , we see from eq .
( [ eq8 ] ) that the time integral is dominated by @xmath169 and we obtain the final result ( r=0)~^-1r_v^-2b(0)(r_/||)^3|||>r_/t , [ eq20 ] where the conditional average to right hand side is peaked at @xmath170 and diverges for @xmath171 .
thus , ballistic pairs do not dominate the dynamics . a calculation taking into account small velocities in consistent way @xcite , in turn ,
would lead to convergent expressions for @xmath119 and allows to show that the contribution to concentration fluctuation is concentrated at @xmath172 , as could have been guessed from the analysis in section iii .
the large stokes number limit of inertial particles in a random velocity field is plagued by several subtleties . as pointed out in @xcite ,
when this limit is reached , the particle dynamics becomes dependent on a new parameter defined in terms of the stokes and kubo numbers @xmath45 and @xmath23 [ see eq .
( [ eq3 ] ) ] , namely : @xmath173 .
this becomes apparent in the case of short - time correlated velocity fields , such as the kraichnan model @xcite , in which @xmath174 , and the role of the stokes number is played indeed by @xmath55 [ from eq .
( [ eq3 ] ) , @xmath45 is trivially equal to infinity ] .
thus , as pointed out in @xcite , the large stokes number asymptotics is really the large @xmath55 one of eq .
( [ eq9 ] ) .
the point of this paper is that , for @xmath51 , the concentration dynamics is not governed by particle pairs staying close long enough for a local theory on the lines of @xcite to work . while , for small @xmath55 , the particle phase is locally mono - disperse in velocity @xcite , for @xmath175 , the particle velocity distribution has finite width and only very few such almost co - moving particle pairs exist ( see discussion at the end of section iii ) .
this property is expected to hold also in more than 1d .
small relative velocities still appear to be important : as suggested in @xcite , the particles behave like a gas in thermodynamic equilibrium , but the correlations originate , rather than from random collisions , from particle encounters at small relative velocities , as made clear by the analysis leading to eq .
( [ eq20 ] ) . a qualitative argument in section iii , confirmed by the analysis in @xcite
, suggests that the only particle pairs that have a chance to be correlated are those travelling at relative velocity @xmath176 , that is the minimum for ballistic relative motion at scale @xmath7 .
however , these velocities are still @xmath177 larger than those producing the long time dynamics needed for clustering in the @xmath56 limit .
the concentration fluctuation variance appears to decay rather slowly , like @xmath178 at large @xmath55 [ see eq .
( [ eq17 ] ) ] . this decay rate could be interpreted as the product of the concentration fraction @xmath179 of the particles staying close long enough for their relative velocity to be modified in some way , and their permanence time @xmath180 ( in units @xmath4 ) at separation @xmath65 [ see eq .
( [ eq11 ] ) ] .
this mechanism of concentration fluctuation production leads to an important modification of the relative velocity distribution .
as particle clusters are the result of particle pairs with small relative velocities , the total velocity pdf will be weighed more to small relative velocities .
this results in a depletion of the relative velocity variance , quantified in the exact relation ( [ eq16.02 ] ) , that becomes substantial at @xmath181 .
it is to be ascertained whether this property is preserved in more than 1d .
summarizing , we have the following situation : extrapolating the small @xmath55 theory of @xcite to @xmath182 , predicts that clusters become less and less singular , making eventually a transition from fractal to space - filling objects . extrapolating the opposite way around
, the present large @xmath55 theory predicts residual ( not singular ) concentration fluctuations , becoming of the order of the mean concentration as @xmath183 , due to more and more particle pairs contributing to the fluctuations .
it is reasonable to expect that the situation at @xmath181 will be a combination of the two pictures . in @xcite , it was suggested that clustering is not central in determining collision rates of relevance e.g. in rain formation , as instead are caustics formation and a finite width of the velocity distribution .
if the present picture continues to hold in more than 1d , it would result in an even stronger statement , that is , at large @xmath55 , clustering and modifications to the velocity distribution work against one another .
this is in some way analogous to the situation at @xmath109 , of clustering dominance accompanied by caustics disappearance and vanishing of collision rates @xcite ( see also @xcite ) .
another feature of concentration fluctuations at large @xmath55 , is the strong scale separation between their decay length and the correlation length of the fluid : the first is by a factor @xmath184 larger than the second [ see eq . ( [ eq17 ] ) ] .
the concentration fluctuation decay length is basically the distance travelled by a particle in a stokes time and is independent of the fluid correlation structure .
taking into account also the @xmath178 variance decay , smaller vortices may thus contribute in a non negligible way to the concentration fluctuations of particles with stokes time in the turbulent inertial range .
the langevin approximation to eq .
( [ eq2 ] ) provided by ( [ eq5 ] ) is the lowest order in an expansion in inverse powers of the stokes number @xmath45 .
let us calculate the next order .
given initial conditions at @xmath185 , eq .
( [ eq2 ] ) can be integrated to give : @xmath186 and we can write @xmath187 with @xmath188 and @xmath189 following an approach similar to @xcite , the perturbation expansion is carried on with respect to @xmath190 and the first correction to @xmath191 is given by , for @xmath192 : @xmath193 where @xmath194 and @xmath195 .
taking @xmath196 , we find that the second piece to right hand side of eq .
( a1 ) behaves like a wiener increment ; using eq .
( [ eq5.01 ] ) : @xmath197 ^ 2\big\rangle=2\tau_p(v(0))t\ ] ] and we recover eq .
( [ eq5 ] ) .
notice that the substitution @xmath198 avoids a definition of the noise in implicit form and @xmath24 in eq .
( [ eq5 ] ) is automatically defined in the it prescription @xcite .
the last term in eq .
( a1 ) contains a drift correction in the form , from eq .
( [ eq1 ] ) : @xmath199 equation ( a1 ) takes then the final form , in the limit @xmath200 : @xmath201v\d
t+(2\tau_p(v))^{1/2}\d w \eqno(a2)\ ] ] where @xmath202 and @xmath203 is the wiener increment .
the noise amplitude is defined up to @xmath204 terms associated with the fluctuating part of the second line of eq .
( a1 ) .
thus , the first order correction to the langevin equation for @xmath191 is an @xmath32 renormalization of the stokes time , whose sign depends on the profile of the correlation function for @xmath205 : a strictly positive correlation corresponds to a decrease of the stokes time , while oscillations in the correlation may lead to an increase .
this would lead to a decrease ( an increase ) of the drift induced on the particle by an external force .
notice that the drift correction @xmath206 provided in eq .
( a3 ) does not necessarily coincide with what would be expected from a change from stratonovich to it prescription @xcite .
for instance , adopting for the random field correlation the explicit expression @xmath207 , would lead to twice what would be obtained interpreting the substitution @xmath208 in @xmath209 as a stratonovich to it prescription change .
indicating by @xmath210 the stratonovich differential and writing @xmath211 : @xmath212 we find in fact , from eq .
( [ eq5.01 ] ) : @xmath213 .
m. cencini , j. bec , l. biferale , g. boffetta , a. celani , a.s .
lanotte , s. musacchio and f. toschi dynamics and statistics of heavy particles in turbulent flows , jot * 7 * , 1 ( 2006 ) r.a .
particle - turbulence interactions in atmospheric clouds ,
fluid mech . * 35 * , 183 ( 2003 ) this would be true also if the particles approached with speed @xmath214 , like in the @xmath51 case ; it is then necessarily true also in the present case , where @xmath64 ( see discussion at the end of the section ) . a simple alternative way to see how caustics disappear for @xmath109 : from eq .
( [ eq12 ] ) , @xmath215 , while the velocity required for a ballistic flight to @xmath216 is @xmath217 .
the probability of such a flight would be : @xmath218 \sim \exp(-c/\epsilon)$ ] , with @xmath219 a constant .
apart near caustics , where the distribution is a superposition of mono - disperse jets .
y. zhou , , a.s .
wexler and l .- p . wang on the collision rate of small particles in isotropic turbulence .
ii . finite inertia case , phys .
fluids * 10 * , 1206 ( 1998 ) | we analyze the behavior of an ensemble of inertial particles in a one - dimensional smooth gaussian velocity field , in the limit of large inertia , but considering a finite correlation time for the random field .
the amplitude of the concentration fluctuations is characterized by slow decay at large inertia and a much larger correlation length than that of the random field .
the fluctuation structure in velocity space is very different from predictions from short - time correlated random velocity fields , with only few particle pairs crossing at sufficiently small relative velocity to produce correlations .
concentration fluctuations are associated with depletion of the relative velocity variance of colliding particles .
v *o * = cmr10 at 24truept = cmbx10 at 15truept |
understanding fermion masses and mixings is one of the outstanding problem of present day particle physics .
the idea of quark mixing phenomena was initiated by cabibbo in 1963 @xcite , subsequently generalized to two generations by glashow , illiopoulos , maiani @xcite and finally to three generations by kobayashi and maskawa @xcite .
this has been tested to a great accuracy and is well accommodated by the standard model ( sm ) .
recently , flavor mixing has also been observed in the case of neutrinos implying the existence of non zero , non degenerate neutrino masses necessitating the need to look beyond sm .
also , one has to go beyond the sm in order to understand the pattern of quark masses and mixing parameters as in the sm the quark mass matrices are completely arbitrary . in view of the relationship of fermion mixing phenomena with that of fermion mass matrices , the understanding of the above mentioned issues of flavor physics essentially implies formulating fermion mass matrices . while on the one hand , grand unified theories ( guts ) have provided vital clues for understanding the relationship of fermion mass matrices between quarks and leptons , on the other hand , horizontal symmetries @xcite have given clues for the relationship between different generation of fermions .
ideas such as extra dimensions @xcite have also been invoked to understand the flavor puzzle .
unfortunately , at present it seems that we do not have any theoretical framework which provides a viable and satisfactory description of fermion masses and mixings .
the lack of a convincing fermion flavor theory from the ` top down ' perspective necessitates the need for formulating fermion mass matrices from a ` bottom up ' approach .
the essential idea behind this approach is that one tries to find the phenomenological fermion mass matrices which are in tune with the low energy data and can serve as guiding stone for developing more ambitious theories . in this context ,
initially several _ anstze _ @xcite were suggested for quark mass matrices .
one of the successful _
anstze _ incorporating the `` texture zero '' approach was initiated by fritzsch @xcite . _
a particular texture structure is said to be texture @xmath0 zero , if it has @xmath0 number of non - trivial zeros , for example , if the sum of the number of diagonal zeros and half the number of the symmetrically placed off diagonal zeros is @xmath0_. the detailed plan of the article is as follows . in section ( [ tsmm ] ) ,
we discuss some of the broad features pertaining to quark and lepton texture specific mass matrices .
the relationships of the fermion mass matrices and mixing matrices have been presented in section ( [ form ] ) .
present status of the quark and neutrino mass and mixing parameters have been given in section ( [ inputs ] ) .
the details pertaining to texture 6 , 5 , 4 zero quark and lepton mass matrices have respectively been presented in sections ( [ quarkmm ] ) and ( [ lepmm ] ) . finally , in section ( [ summ ] ) we summarize and conclude .
the mass matrices , having their origin in the higg s fermion couplings , are arbitrary in the sm , therefore the number of free parameters available with a general mass matrix is larger than the physical observables .
for example , if no restrictions are imposed , there are @xmath1 real free parameters in the two @xmath2 general complex mass matrices , @xmath3 and @xmath4 , which in the quark sector need to describe ten physical observables , i.e. , six quark masses , three mixing angles and one cp violating phase .
similarly , in the leptonic sector , physical observables described by lepton mass matrices are six lepton masses , three mixing angles and one cp violating phase for dirac neutrinos ( two additional phases in case neutrinos are majorana particles ) .
therefore , to develop viable phenomenological fermion mass matrices one has to limit the number of free parameters in the mass matrices . in this context , it is well known that in the sm and its extensions wherein the right handed fields in the lagrangian are su(2 ) singlets , without loss of generality , the mass matrices can be considered as hermitian .
this immediately brings down the number of real free parameters from 36 to 18 , which however , is still a large number compared to the number of observables . to this end ,
fritzsch @xcite initiated the idea of texture specific mass matrices which on the one hand imparted predictability to mass matrices while on the other hand , it paved the way for the phenomenology of texture specific mass matrices . to define the various texture specific cases , we present the typical fritzsch like texture specific hermitian quark mass matrices , for example , m_u= ( 0 & a _ u & 0 + a_u^ * & d_u & b_u + 0 & b_u^ * & c_u ) , m_d= ( 0 & a _ d & 0 + a_d^ * & d_d & b_d + 0 & b_d^ * & c_d ) , [ nf2zero ] where @xmath5 and @xmath6 correspond to up and down mass matrices respectively . it may be noted that each of the above matrix is texture 2 zero type with @xmath7 and @xmath8 , where @xmath9 .
the texture 6 zero fritzsch mass matrices can be obtained from the above mentioned matrices by taking both @xmath10 and @xmath11 to be zero , which reduces the matrices @xmath5 and @xmath6 each to texture 3 zero type .
this fritzsch _ anstze _ @xcite as well as some other _ anstze _ @xcite were ruled out because of the large value predicted for @xmath12 due to the high ` t ' quark mass .
further , a few other texture 6 zero mass matrices were analyzed by ramond , roberts and ross @xcite revealing that these matrices were again ruled out because the predicted value of @xmath12 came out to be much larger than the available data at that time .
they also explored the question of connection between phenomenological quark mass matrices considered at low energies and the possible mass patterns at the gut scale and showed that the texture structure of mass matrices is maintained as we come down from gut scale to @xmath13 scale .
this important conclusion also leads to the fact that the texture zeros of fermion mass matrices can be considered as phenomenological zeros , thereby implying that at all energy scales the corresponding matrix elements are sufficiently suppressed in comparison with their neighboring counterparts .
this , therefore , opens the possibility of considering lesser number of texture zeros . besides ramond
, roberts and ross @xcite , several authors @xcite-@xcite then tried to explore the texture 5 zero quark mass matrices .
fritzsch - like texture 5 zero matrices can be obtained by taking either @xmath10 = 0 and @xmath14 or @xmath15 and @xmath11 = 0 in eq .
( [ nf2zero ] ) , thereby giving rise to two possible cases of texture 5 zero mass matrices pertaining to either @xmath3 or @xmath4 being texture 3 zero type while the other being texture 2 zero type .
these analyses reveal that texture 5 zero mass matrices although not ruled out unambiguously yet are not able to reproduce the entire range of data .
further , the issue of the phases of the mass matrices , responsible for cp violation , was not given adequate attention in these analyses .
as an extension of texture 5 zero mass matrices , several authors @xcite-@xcite carried out the study of the implications of the fritzsch - like texture 4 zero mass matrices .
it may be noted that fritzsch - like texture 4 zero mass matrices can be obtained by considering both @xmath3 and @xmath4 , with non zero @xmath16 in eq .
( [ nf2zero ] ) , to be texture 2 zero type .
although from the above mentioned analyses one finds that texture 4 zero mass matrices were able to accommodate the quark mixing data quite well , however it may be noted that these analyses assumed ` strong hierarchy ' of the elements of the mass matrices as well as explored only their limited domains .
further , in the absence of any precise information about cp violating phase @xmath17 and related parameters , again adequate attention was not given to the phases of the mass matrices .
recent refinements in quark mixing data motivated several authors @xcite-@xcite to have a re - look at the compatibility of fritzsch like texture 4 zero mass matrices with the quark mixing data . in particular , using assumption of ` strong hierarchy ' of the elements of the mass matrix defined as @xmath18 , having its motivation in the hierarchy of the quark mixing angles several attempts @xcite-@xcite were made to predict the value of precisely known parameter sin@xmath19 .
unfortunately , the value of sin@xmath19 predicted by these analyses came out to be in quite disagreement with its precisely known value .
a somewhat detailed and comprehensive analyses of texture 4 zero quark mass matrices for the first time was carried out by xing and zhang @xcite , in particular they attempted to find the parameter space available to the elements of mass matrices .
their analysis has also given valuable clues about the phase structure of the mass matrices , in particular for the strong hierarchy case they conclude that only one of the two phase parameters plays a dominant role .
subsequently , attempts @xcite have been made to update and broaden the scope of the analysis carried out by xing and zhang @xcite , in particular regarding the structural features of the mass matrices having implications for the value of parameter sin@xmath19 . in the leptonic sector
, one would like to mention that the observation of neutrino oscillations has added another dimension to the issue of fermion masses and mixing .
in fact , the pattern of neutrino masses and mixings seems to be vastly different from that of quarks . at present
, the available neutrino oscillation data does not throw any light on the neutrino mass hierarchy , which may be normal/ inverted and may even be degenerate .
further , the situation becomes complicated when one realizes that neutrino masses are much smaller than charged fermion masses as well as it is not clear whether neutrinos are dirac or majorana particles .
the situation becomes more complicated in case one has to understand the quark and neutrino mixing phenomena in a unified manner . in this context , to understand the pattern of neutrino masses and mixings , texture zero approach has also been tried with good deal of success @xcite-@xcite . an early attempt to formulate lepton mass matrices was carried out by frampton , glashow and marfatia @xcite , wherein assuming a complex symmetric majorana mass matrix and considering seven possible texture 2 zero cases , they carried out the implications of these for the neutrino oscillation data .
thereafter , several attempts were made using texture specific lepton mass matrics to explain the pattern of neutrino masses and mixings .
in particular , for normal hierarchy of neutrino masses , fukugita , tanimoto and yanagida @xcite carried out an analysis of fritzsch - like texture 6 zero mass matrices .
similarly , zhou and xing @xcite also carried out a systematic analysis of all possible texture 6 zero mass matrices for majorana neutrinos with an emphasis on normal hierarchy of neutrino masses .
recently , @xcite for all possible hierarchies of neutrino masses , for both majorana as well as dirac neutrinos , detailed analyses of fritzsch - like texture 6 , 5 and 4 zero mass matrices was carried out . from the above discussion
, one finds that texture specific mass matrices are able to accommodate the quark as well as neutrino mixing data .
this brings into fore the issue of quark - lepton unification , advocated by smirnov @xcite , by considering similar structures for quark and lepton mass matrices . keeping this in mind as well as in view of absence of any theoretical justification for fritzsch - like mass matrices
, recent attempts @xcite were made to consider non fritzsch - like mass matrices for quarks as well as neutrinos .
in the sm , the quark mass terms for three generations of quarks can be expressed as _ u_l m_u q_u_r + q_d_l m_d q_d_r , [ mc ] where @xmath20 and @xmath21 are the left - handed ( right - handed ) quark fields for the up sector @xmath22 and down sector @xmath23 respectively .
@xmath3 and @xmath4 are the mass matrices for the up and the down sector of quarks . in order to re - express above equation in terms of the physical quark fields , one can diagonalize the mass matrices by the following bi - unitary transformations v_u_l^m_u v_u_r = m_u^diag ( m_u , m_c , m_t ) , [ v11 ] v_d_l^m_d v_d_r = m_d^diag ( m_d , m_s , m_b ) , [ v21 ] where @xmath24 are real and diagonal , while @xmath25 and @xmath26 etc . are complex unitary matrices .
the quantities @xmath27 etc .
denote the eigenvalues of the mass matrices , i.e. the physical quark masses .
using eqs .
( [ v11 ] ) and ( [ v21 ] ) , one can rewrite ( [ mc ] ) as _
u_lv_u_lm_u^diag v_u_r^ q_u_r + _
d_l v_d_l m_d^diag v_d_r^ q_d_r , which can be re - expressed in terms of physical quark fields as ^phys_u_l m_u^diag q^phys_u_r + ^phys_d_lm_d^diag q^phys_d_r , where @xmath28 and @xmath29 and so on .
the mismatch of diagonalizations of up and down quark mass matrices leads to the quark mixing matrix , referred to as the cabibbo - kobayashi - maskawa ( ckm ) @xcite matrix given as v_ckm = v_u_l^ v_d_l . [
1mix ] the ckm matrix expresses the relationship between quark mass eigenstates @xmath30 which participate in the strong q@xmath31q and q@xmath32 interactions and the interaction eigenstates or flavor eigenstates @xmath33 which participate in the weak interactions and are the linear combinations of mass eigenstates , for example , ( d + s + b ) = ( v_ud & v_us & v_ub + v_cd & v_cs & v_cb + v_td & v_ts & v_tb + ) ( d + s + b ) , [ 1vckm]where @xmath34 , @xmath35 etc .
describe the transition of @xmath36 to @xmath37 , @xmath36 to @xmath38 respectively , and so on .
in view of the relationship of the mixing matrix with the mass matrix , a knowledge of the elements would have important implications for the mass matrices .
the , by definition , is a unitary matrix , hence can be expressed in terms of three real angles and six phases . out of the six phases , five
can be re - absorbed into the quark fields , therefore , one is left with only one non - trivial phase which is responsible for cp violation in the sm .
there are several parameterizations of the sm , however the most commonly used parameterization is the standard parameterization given by particle data group ( pdg ) @xcite .
the pdg representation of the is given as v_ckm = ( c_12 c_13 & s_12 c_13 & s_13e^-i + -s_12 c_23 - c_12 s_23 s_13e^i & c_12 c_23 - s_12 s_23s_13e^i & s_23 c_13 + s_12 s_23 - c_12 c_23 s_13e^i & - c_12 s_23 - s_12c_23 s_13e^i & c_23 c_13 ) , [ 1ckm ] with @xmath39 and @xmath40 .
the angles @xmath41 and @xmath42 can be chosen to lie in the first quadrant , whereas the quadrant of @xmath17 has physical significance , therefore can not be fixed . in the pdg representation
, sin@xmath43 implies the existence of cp violation . a precise measurement of elements and the parameters characterizing the pdg representation e.g. , mixing angles and cp violating phase @xmath44 will undoubtedly have implications for the mass matrices .
the observation of neutrino oscillation phenomenon which essentially implies the flavor conversion of neutrinos is similar to the quark mixing phenomenon .
this possibility of flavor conversion was originally examined by b. pontecorvo and further generalized by maki , nakagawa and sakata @xcite . the emerging picture that neutrinos are massive and therefore mix has been proved beyond any doubt and provides an unambiguous signal of np .
in the case of neutrinos , the generation of masses is not straight - forward as they may have either the dirac masses or the more general dirac - majorana masses .
a dirac mass term can be generated by the higgs mechanism with the standard higgs doublet . in this case , the neutrino mass term can be written as @xmath45 where @xmath46 = @xmath47 , @xmath48 , @xmath49 and @xmath50 , @xmath51 , @xmath52 are the flavor eigenstates .
@xmath53 is a complex @xmath54 dirac mass matrix .
the mass term mentioned above would also be characterized by the same symmetry breaking scale such as that of charged leptons or quarks , therefore , in this case very small masses of neutrinos would be very unnatural from the theory point of view . on the other hand , the neutrino might be a majorana particle which is defined as is its own antiparticle and is characterized by only two independent particle states of the same mass ( @xmath55 and @xmath56 or @xmath57 and @xmath58 ) . a majorana mass term , which violates both the law of total lepton number conservation and that of individual lepton flavor conservation , can be written either as @xmath59 where @xmath60 and @xmath61 are complex symmetric matrices .
a simple extension of the sm is to include one right handed neutrino in each of the three lepton families , while the lagrangian of the electroweak interactions is kept invariant under @xmath62 gauge transformations .
this can be shown to lead to dirac - majorana mass terms which further lead to the famous seesaw mechanism @xcite for the generation of small neutrino masses , e.g. , m_=-m_d^t(m_r)^-1m_d , [ seesaweq2 ] where @xmath53 and @xmath61 are respectively the dirac neutrino mass matrix and the right - handed majorana neutrino mass matrix .
the seesaw mechanism is based on the assumption that , in addition to the standard higgs mechanism of generation of the dirac mass term , there exists a beyond the sm mechanism of generation of the right - handed majorana mass term , which changes the lepton number by two and is characterized by a mass @xmath63 .
the dirac mass term mixes the left - handed field @xmath64 , the component of a doublet , with a single field @xmath65 . as a result of this
mixing the neutrino acquires majorana mass , which is much smaller than the masses of leptons or quarks .
similar to the quark sector , the lepton mass matrices can be diagonalized by bi - unitary transformations and the corresponding mixing matrix obtained , known as pontecorvo - maki - nakagawa - sakata ( pmns ) or lepton mixing matrix @xcite , is given as v_pmns= v^_l_l v__l .
the pmns matrix expresses the relationship between the neutrino mass eigenstates and the flavor eigenstates , e.g. , ( _ e + _ + _ ) = ( v_e1 & v_e2 & v_e3 + v_1 & v_2 & v_3 + v_1 & v_2 & v_3 ) ( _ 1 + _ 2 + _ 3 ) , [ nm1 ] where @xmath66 , @xmath67 , @xmath68 are the flavor eigenstates , @xmath69 , @xmath70 , @xmath71 are the mass eigenstates and the @xmath2 mixing matrix is the leptonic mixing matrix @xcite . for the case of three dirac neutrinos , in the standard pdg parametrization @xcite , involving three angles @xmath72 , @xmath73 , @xmath42 and the dirac - like cp violating phase @xmath74
the mixing matrix has the form @xmath75 with @xmath76 , @xmath77 . in the case of the majorana neutrinos
, there are extra phases which can not be removed .
therefore , the above matrix takes the following form ( e^i _ 1/2 & 0 & 0 + 0 & e^i _ 2/2 & 0 + 0 & 0 & 1 ) , [ nmm ] + where @xmath78 and @xmath79 are the majorana phases which do not play any role in neutrino oscillations .
to carry out any analysis regarding exploring the compatibility of fermion mass matrices with the recent data , one needs to keep in mind the experimental constraints imposed by the relationship between mass matrices and their corresponding mixing matrices . to facilitate our discussion in this regard , we first present the status of relevant data in the quark as well as in the lepton sector . in the quark sector ,
the most important constraints are provided by the directly observed quantities , such as masses of the quarks , elements , cp violating phase @xmath17 and @xmath80 , etc .. the quark masses relevant for the present work are the `` current '' quark masses at @xmath13 energy scale @xcite , e.g. , m_u = 1.27^+0.5_-0.42 mev , m_d = 2.90^+1.24_-1.19 mev , m_s=55^+16_-15 mev , + m_c=0.619 0.084 gev , m_b=2.89 0.09 gev , m_t=171.7 3.0 gev .
[ qmasses ] the light quark masses @xmath81 , @xmath82 and @xmath83 are usually further constrained by using the following mass ratios @xcite m_u / m_d = 0.553 0.043 , m_s / m_d = 18.9 0.8 .
[ ratios ] the analysis of mass matrices yields the ckm mixing matrix elements which can then be compared with mixing matrix obtained from a rigorous data based analysis . for ready reference as well as for the sake of readability , we reproduce here the ckm matrix as per pdg 2010 @xcite , at 95@xmath84 c.l .
as v_ckm = ( 0.974280.00015 & 0.22530.0007 & 0.00347_-0.00012^+0.00016 + 0.22520.0007 & 0.97345_-0.00016 ^+0.00015 & 0.0410_-0.0007^+0.0011 + 0.00862_-0.00020^+0.00026 & 0.0403_-0.0007^+0.0011 & 0.999152_-0.000045^+0.000030 ) [ vckm ] . similarly , the precisely measured values @xcite of the cp violating parameter sin@xmath19 , the jarlskog rephasing invariant parameter @xmath85 and the cp violating phase @xmath17 respectively are 2=0.673 0.023 , j=(2.91_-0.11^+0.19)10 ^ -5 , = ( 73_-25^+22)^. [ sin ] adopting the three neutrino framework , several authors @xcite-@xcite have presented updated information regarding neutrino mass and mixing parameters obtained by carrying out detailed global analyses . the latest situation regarding these parameters at 3@xmath86 c.l .
is summarized as follows @xcite , m_21 ^ 2 = ( 7.05 - 8.34 ) 10 ^ -5 ev^2 , sin^2_23 = 0.36 - 0.67 , sin^2_13 0.056 .
[ s13 ] for the sake of completion as well as for ready reference , we present the following pmns matrix determined by taking into account the neutrino oscillation data by garcia _
_ @xcite at 3@xmath86 c.l . as = ( 0.79 - 0.86 & 0.50 - 0.61 & 0.00 - 0.20 + 0.25 - 0.53 & 0.47 - 0.73 & 0.56 - 0.79 + 0.21 - 0.51 & 0.42 - 0.69 & 0.61 - 0.83 ) .
& class i & class ii + a & @xmath87 & @xmath88 + b & @xmath89 & @xmath90 + c & @xmath91 & @xmath92 + d & @xmath93 & @xmath94 + e & @xmath95 & @xmath96 + f & @xmath97 & @xmath98 + as mentioned earlier , texture 6 zero fritzsch mass matrices have already been ruled out @xcite .
therefore , for the sake of completion , we would like to discuss all possible non fritzsch - like combinations of texture 6 zero hermitian mass matrices as well . before counting all possibilities , in view of non zero masses of quarks ,
these matrices have to satisfy the conditions @xmath99 and @xmath100 .
one can easily check that in case of texture 3 zero mass matrices we arrive at 20 different possible texture patterns , out of which 8 are easily ruled out by imposing these conditions .
the remaining 12 possible textures break into two classes , as shown in table ( [ t1 ] ) , depending upon the equations these matrices satisfy .
for example , six matrices of class i , mentioned in table ( [ t1 ] ) , satisfy the following equations c = m_1- m_2 + m_3 , a^2 + b^2 = m_1m_2 + m_2m_3 -m_1m_3 , a^2 c = m_1m_2m_3 .
[ classi ] similarly , in case of class ii , all six matrices satisfy the following equations c + d= m_1
-m_2 + m_3,a^2 -c d = m_1m_2 + m_2m_3 - m_1m_3 , a^2 c = m_1m_2m_3 .
[ classii ] the subscripts u and d have not been used as these are valid for both kind of mass matrices .
it may be added that these classes are also related through permutation symmetry @xcite .
matrices @xmath5 and @xmath6 each can correspond to any of the 12 possibilities , therefore yielding 144 possible combinations which in principle can yield 144 quark mixing matrices .
these 144 combinations can be put into 4 different categories , e.g. , if @xmath5 is any of the 6 matrices from class i , then @xmath6 can be either from class i or class ii yielding 2 categories of 36 matrices each .
similarly , we obtain 2 more categories of 36 matrices each when @xmath5 is from class ii and @xmath6 is either from class i or class ii .
the 36 combinations in each category further can be shown to be reduced to groups of six combinations of mass matrices , each yielding same ckm matrix .
thus , the problem of exploring the compatibility of 144 phenomenologically allowed texture 6 zero combinations with the recent low energy data is reduced only to an examination of 4 groups each having 6 combinations of mass matrices corresponding to the same ckm matrix .
detailed analysis @xcite of these four groups of texture 6 zero mass matrices reveals that all possible combinations of texture 6 zero are ruled out as these are not able to reproduce the ckm element @xmath12 .
considering now the case of texture 5 zero mass matrices which consist either of @xmath5 being 2 zero and @xmath6 being 3 zero or vice versa .
texture 3 zero possibilities have already been enumerated , therefore we consider only the possible patterns of texture 2 zero mass matrices . after taking into consideration the trace and determinant conditions mentioned earlier
, one can check that there are 18 possible texture 2 zero patterns .
these textures further break into three classes , detailed in table ( 2 ) , depending upon the diagonalization equations satisfied by these matrices , however , it can be shown that the classes iv and v essentially reduce to texture 3 zero patterns . we are therefore left with only class iii of texture 2 zero matrices that needs to be explored for texture 5 zero combinations .
all matrices of this class satisfy the following equation c + d = m_1- m_2 + m_3,a^2 + b^2 - c d = m_1m_2 + m_2m_3 -m_1m_3,a^2 c = m_1m_2m_3 .
[ classiii ] considering class iii of texture 2 zero mass matrices along with different patterns of class i and class ii of texture 3 zero mass matrices we find a total of 144 possibilities of texture 5 zero mass matrices , in sharp contrast to the case if we had considered the classes iv and v also yielding 432 possibilities . keeping in mind the hierarchy of the elements of the ckm matrix , we observe that out of 144 cases , we are again left with only 4 such groups of texture 5 zero mass matrices leading to mixing matrix having hierarchical structure as that of ckm matrix . & class iii & class iv & class v + a & @xmath101 & @xmath102 & @xmath103 + b & @xmath104 & @xmath105 & @xmath106 + c & @xmath107 & @xmath108 & @xmath109 + d & @xmath110 & @xmath111 & @xmath112 + e & @xmath113 & @xmath114 & @xmath115 + f & @xmath116 & @xmath117 & @xmath118 + a detailed analysis of these texture 5 zero mass matrices have been carried out @xcite which shows that interestingly only one possibility , corresponding to the usual fritzsch - like texture 5 zero mass matrix where @xmath5 is of texture 2 zero and @xmath6 is of texture 3 zero type , appears to be viable .
also it may be added that the viability of this combination depends on the light quark masses used as inputs . as is well known , fritzsch - like texture 4 zero mass matrices are compatible with the quark mixing data , however , there are two issues which need to be addressed in this context .
firstly , one needs to carry out a detailed analysis of the various possible non fritzsch - like texture 4 zero mass matrices .
in fact , the texture two zero possibilities , presented in table ( 2 ) , result into 324 texture 4 zero possibilities , the analysis of such a large number of possibilities is yet to be carried out .
the other issue is whether we can consider ` weakly ' hierarchical mass matrices to reproduce ` strongly ' hierarchical mixing angles .
this issue has been explored in a detailed manner @xcite and in the present work we reproduce some of the essential details regarding this . in this context
, on the one hand , many authors @xcite-@xcite have shown that on using strong hierarchy of the elements of the mass matrices , texture 4 zero mass matrices appear to be incompatible with the recent value of sin@xmath19 . while on the other hand , recently @xcite extension of the parameter space of the elements of these matrices has been carried out to include the case of ` weak hierarchy ' amongst them along with the usually considered ` strong hierarchy ' case and thereby they have been shown to be compatible with the parameter sin@xmath19 .
it may be noted that although hierarchy of the elements of the mass matrices has been considered often while carrying out the analysis , however it has been explicitly defined only in @xcite .
the various relations between the elements of the mass matrices , given in eq .
( [ nf2zero ] ) , @xmath119 @xmath120 essentially correspond to the structural features of the mass matrices including their hierarchies .
as is usual the element @xmath121 takes a value much smaller than the other three elements of the mass matrix which can assume different relations amongst each other , defining different hierarchies .
for example , in case @xmath122 it would lead to a strongly hierarchical mass matrix whereas a weaker hierarchy of the mass matrix implies @xmath123 .
it may also be added that for the purpose of numerical work , one can conveniently take the ratio @xmath124 characterizing strong hierarchy whereas @xmath125 implying weak hierarchy .
the analysis carried out by @xcite incorporates the quark masses and their ratios mentioned in eqs .
( [ qmasses])and ( [ ratios ] ) as well as by imposing the constraints given in eqs . ( [ vckm ] ) and ( [ sin ] ) .
further , full variation has been given to the phases associated with the mass matrices @xmath126 and @xmath127 , the parameters @xmath128 and @xmath129 have been given wide variation in conformity with the hierarchy of the elements of the mass matrices e.g. , @xmath130 for @xmath131 .
the extended range of these parameters allows the calculations for the case of weak hierarchy of the elements of the mass matrices as well . to begin with , in fig .
1(a ) @xmath132 versus @xmath133 has been plotted .
a look at the figure reveals that both @xmath134 as well as @xmath133 take values from @xmath135 , which interestingly indicates the ratios being almost proportional .
also , the figure gives interesting clues regarding the role of strong and weak hierarchy .
in particular , one finds that in case one restricts to the assumption of strong hierarchy then these ratios take large values around @xmath136 .
however , for the case of weak hierarchy , the ratios @xmath132 and @xmath133 take much larger number of values , in fact almost the entire range mentioned above , which are compatible with the data .
= 1.55 in in fig .
1(b ) , the plot of @xmath126 versus @xmath127 has been presented .
interestingly , the present refined inputs limit the ranges of the two phases to @xmath137 and @xmath138 . keeping in mind that full variation has been given to the free parameters @xmath128 and @xmath129 , corresponding to both strong as well as weak hierarchy cases , it may be noted that the allowed ranges of the two phases come out to be rather narrow .
in particular , for the strong hierarchy case one gets @xmath139 , whereas for the case of weak hierarchy @xmath127 takes almost its entire range mentioned above .
also , the analysis indicates that although @xmath140 , still both the phases are required for fitting the mixing data . as a next step , the role of the hierarchy defining parameters @xmath10 and @xmath11 has been emphasized . to this end , in fig .
1(c ) @xmath141 versus @xmath142 has been given , representing an extended range of the parameters @xmath128 and @xmath129 .
a closer look at the figure reveals both @xmath143 as well as @xmath142 take values @xmath144 . the lower limit of the range i.e. when the ratios @xmath143 and @xmath142 are around @xmath145 corresponds to strong hierarchy amongst the elements of the mass matrices , whereas when the elements have weak hierarchy then these ratios take a much larger range of values . from this one
may conclude that in the case of strongly hierarchical elements of the texture 4 zero mass matrices , one has limited compatibility of these matrices with the quark mixing data , whereas the weakly hierarchical ones indicate the compatibility for much broader range of the elements .
having discussed the texture specific quark mass matrices , in the light of quark lepton unification hypothesis advocated by smirnov @xcite , one would also like to know the status of texture 6 zero fritzsch as well as non fritzsch - like lepton mass matrices .
a detailed analysis of texture 6 zero mass matrices have been carried out by several authors @xcite . in particular , for normal hierarchy of neutrino masses zhou and xing @xcite have carried out an analysis of all possible fritzsch as well as non fritzsch - like texture 6 zero lepton mass matrices .
their analysis has been extended further to include inverted hierarchy and non degenerate scenario of neutrino masses @xcite . as already shown for the case of quarks in section([quarkmm ] ) , there are a total number of 144 possible cases of texture 6 zero mass matrices . for the case of lepton mass matrices
, there are 6 cases for each of the 144 combinations corresponding to normal/ inverted hierarchy and degenerate scenario of neutrino masses for majorana neutrinos as well as dirac neutrinos , leading to a total of 864 cases .
the analysis carried out in @xcite reveals several interesting points . in particular , their investigations for dirac neutrinos show that there are no viable texture 6 zero lepton mass matrices for normal/ inverted hierarchy as well as degenerate scenario of neutrino masses . for the case of majorana neutrinos for texture 6 zero lepton mass matrices ,
again all the cases pertaining to inverted hierarchy and degenerate scenario of neutrino masses are also ruled out . assuming normal hierarchy of majorana neutrinos , the analysis reveals that out of 144 , only 16 combinations are compatible with current neutrino oscillation data at @xmath146 c.l .. similar to the case of texture 6 zero lepton mass matrices , the implications for different hierarchies in the case of texture 5 zero lepton mass matrices have also been investigated for both majorana and dirac neutrinos @xcite .
for the two types of neutrinos , corresponding to normal/ inverted hierarchy and degenerate scenario of neutrino masses 360 cases each have been considered for carrying out the analysis , making it a total of 2160 cases . for majorana neutrinos with normal hierarchy of neutrino masses , out of the 360 combinations ,
67 are compatible with the neutrino mixing data .
most of the phenomenological implications of combinations of different categories are similar , however , still these can be experimentally distinguished with more precise measurements of @xmath42 and @xmath73 . interestingly , degenerate scenario of majorana neutrinos is completely ruled out by the existing data . in the case of inverted hierarchy ,
24 combinations out of 360 are compatible with the neutrino mixing data . for dirac neutrinos with normal hierarchy of neutrino masses , as compared to majorana cases , out of 360 only 44 combinations are compatible with neutrino mixing data .
interestingly , 6 combinations out of 44 can accommodate degenerate dirac neutrinos .
for inverted hierarchy , 24 combinations are compatible with the existing data .
like the case of quarks , the number of viable possibilities for the case of texture 4 zero lepton mass matrices is also quite large , so in the present work we have discussed the essentials of recent detailed analyses @xcite regarding only the fritzsch - like texture 4 zero mass matrices . in particular , they have investigated the implications of different hierarchies of neutrino masses on these matrices for both majorana and dirac neutrinos .
interestingly , at 3@xmath86 c.l . , their analysis rules out both inverted hierarchy and degenerate scenario of neutrino masses for the two types of neutrinos .
we reproduce here the essentials of their arguments .
basically , they have plotted the parameter space corresponding to any of the two mixing angles by constraining the third angle by its experimental values , mentioned in eq .
( [ s13 ] ) , while giving full allowed variation to other parameters . for ready reference we present these graphs in fig .
( [ fig1 ] )
. these plots immediately reveal that the inverted hierarchy is ruled out at 3@xmath86 c.l .. for example , from fig .
( [ fig1]a ) , ( [ fig1]b ) and ( [ fig1]c ) , for majorana neutrinos , one can note that the plotted parameter space of the angles has no overlap with their experimentally allowed @xmath146 region .
similar plots pertaining to dirac neutrinos also rule out inverted hierarchy of neutrino masses .
= 2.8 in their analysis also shows that for majorana or dirac neutrinos the cases of neutrino masses being degenerate , characterized by either @xmath147 or @xmath148 corresponding to normal and inverted hierarchy respectively , are again ruled out . considering degenerate scenario corresponding to inverted hierarchy , fig .
( [ fig1 ] ) can again be used to rule out degenerate scenario at 3@xmath86 c.l . for majorana neutrinos .
it needs to be mentioned that while plotting these figures the range of the lightest neutrino mass is taken to be @xmath149 , which includes the neutrino masses corresponding to degenerate scenario , therefore by discussion similar to the one given for ruling out inverted hierarchy , degenerate scenario of neutrino masses is ruled out as well . coming to degenerate scenario corresponding to normal hierarchy
, one can easily show that this is ruled out again . to this end , in fig .
( [ th12vm - md ] ) , by giving full variation to other parameters , the plot of the mixing angle @xmath72 against the lightest neutrino mass @xmath150 have been presented .
( [ th12vm - md]a ) corresponds to the case of majorana neutrinos and fig .
( [ th12vm - md]b ) to the case of dirac neutrinos . from the figures one can immediately find that the values of @xmath72 corresponding to @xmath151 lie outside the experimentally allowed range , thereby ruling out degenerate scenario for majorana as well as dirac neutrinos at 3@xmath86 c.l .. = 2.8 in
after ruling out the cases pertaining to inverted hierarchy and degenerate scenarios , we come the normal hierarchy cases for majorana as well as dirac neutrinos . for both types of neutrinos , these yield viable ranges of neutrino masses , mixing angles @xmath72 , @xmath73 and @xmath42 , jarlskog s rephasing invariant parameter in the leptonic sector @xmath152 and the dirac - like cp violating phase in the leptonic sector @xmath153 the analysis reveals several interesting points . for both dirac or majorana neutrinos ,
the viable range of the lightest neutrino mass @xmath150 is quite different , in particular the range corresponding to dirac neutrinos is much wider at both the ends as compared to the majorana neutrinos .
therefore , a measurement of @xmath150 could have important implications for the nature of neutrinos .
also , one finds that the lower limit on @xmath42 for the dirac case is considerably lower than for the majorana case , therefore a measurement of @xmath42 would have important implications for this case .
the different cases of dirac and majorana neutrinos do not show any divergence for the ranges of jarlskog s rephasing invariant parameter .
fritzsch - like texture specific mass matrices have provided important clues for understanding the pattern of quark mixings and cp violation . likewise , in the leptonic sector also texture specific mass matrices are useful in explaining the pattern of neutrino masses and mixings . to tackle the larger issue of quark and lepton mixing phenomena together , it is perhaps desirable to take into account the quark - lepton unification hypothesis @xcite
this immediately brings forth the issue of finding the simplest texture structure at the leading order , compatible with the quark and lepton mixing phenomena .
further , in the absence of any theoretical justification for fritzsch - like mass matrices , one also needs to consider non fritzsch - like mass matrices for quarks as well as leptons . in the present work ,
we have given an overview of all possible cases of fritzsch - like as well as non fritzsch - like texture 6 and 5 zero fermion mass matrices , for details see @xcite .
further , for the case of texture 4 zero fritzsch - like quark mass matrices , the issue of the hierarchy of the elements of the mass matrices and the role of their phases have been discussed , details can be found in @xcite .
furthermore , the case of texture 4 zero fritzsch - like lepton mass matrices has also been discussed with an emphasis on the hierarchy of neutrino masses for both majorana and dirac neutrinos , elaborate analyses presented in @xcite .
these analyses reveal several interesting results . in principle , for the case of quarks @xcite , there are 144 combinations of texture 6 zero mass matrices whereas in the case of texture 5 zero matrices one can arrive at 360 combinations .
interestingly , all the texture 6 zero combinations are completely ruled out whereas in the case of texture 5 zero mass matrices the only viable possibility looks to be that of fritzsch - like matrices which shows only limited viability , depending upon the light quark masses used as input .
further , for the case of texture 4 zero quark mass matrices @xcite , including the case of ` weak hierarchy ' along with the usually considered ` strong hierarchy ' case , one finds that the weakly hierarchical mass matrices are able to reproduce the strongly hierarchical mixing angles .
also , both the phases having their origin in the mass matrices have to be non zero to achieve compatibility of these matrices with the quark mixing data .
the same number of combinations have been investigated for the neutrino mixing data considering normal/ inverted hierarchy and degenerate scenario of neutrino masses for majorana as well as dirac neutrinos .
texture 6 zero in the case of leptons results into 864 cases to be analyzed @xcite .
interestingly , all the possibilities pertaining to normal/ inverted hierarchy and degenerate scenario of neutrino masses for dirac neutrinos and inverted hierarchy as well as degenerate scenarios in the case of majorana neutrinos are ruled out .
normal hierarchy of neutrino masses for majorana neutrinos results into 16 combinations out of 144 which are in accordance with the neutrino oscillation data .
texture 5 zero entails considering 2160 cases to ascertain their compatibility with the neutrino mixing data @xcite .
interestingly , one finds that texture 5 zero lepton mass matrices can accommodate all hierarchies of neutrino masses . in the case of normal hierarchy , 67 combinations for majorana neutrinos and 44 combinations for dirac neutrinos are compatible with the neutrino mixing data .
there are 6 combinations , out of 44 , which can accommodate degenerate dirac neutrinos .
further , in case of inverted hierarchy , 24 combinations are compatible both for majorana as well as dirac neutrinos . for the fritzsch - like texture 4 zero neutrino mass matrices @xcite ,
analysis pertaining to both majorana and dirac neutrinos for different hierarchies of neutrino masses reveals that for both types of neutrinos , all the cases pertaining to inverted hierarchy and degenerate scenarios of neutrino masses are ruled out at @xmath146 c.l . by the existing data . for the normal hierarchy cases ,
one gets viable ranges of neutrino masses , mixing angle @xmath154 , jarlskog s rephasing invariant parameter @xmath152 and the cp violating dirac - like phase @xmath153 .
interestingly , a measurement of @xmath150 and mixing angle @xmath42 could have important implications for the nature of neutrinos . in conclusion
, we would like to remark that on the one hand there is a need to take the analysis of texture specific mass matrices towards completion .
for example , besides carrying out the analysis of texture 4 zero non fritzsch - like fermion mass matrices , one has to consider texture 3 zero cases also , the latter corresponding to general mass matrices after carrying out weak basis rotations . on the other hand , one may also consider breaking the hermiticity condition perturbatively as has been done recently @xcite and to go into its detailed implications .
similarly , the issue of phases of mass matrices and their relationship with the cp violating parameters also needs a careful look and detailed investigations .
the authors would like to thank the organizers of the international conference on flavor physics in the lhc era " held at singapore in november 2010 for giving an opportunity to present some aspects of the present work .
the authors would also like to thank the participants for several interesting discussions .
g.a . would also like to acknowledge dst , government of india for financial support and the chairman , department of physics for providing facilities to work in the department . c. d. froggatt and h. b. nielsen , _ nucl .
b _ * 147 * , 277 ( 1979 ) ; y. nir and n. seiberg , _ phys .
b _ * 309 * , 337 ( 1993 ) ; m. leurer , y. nir and n. seiberg , _ nucl .
b _ * 420 * , 468 ( 1994 ) ; l. e. ibanez and g. g. ross , _ phys . lett .
b _ * 332 * , 100 ( 1994 ) ; g. altarelli and f. feruglio , arxiv:1002.0211 .
n. g. deshpande , m. gupta and p. b. pal , _ phys .
d _ * 45 * , 953 ( 1992 ) ; p. s. gill and m. gupta , _ pramana _ * 45 * , 333 ( 1995 ) ; _ ibid .
_ _ j. phys .
g _ * 23 * , 335 ( 1997 ) ; _ ibid .
d _ * 56 * , 3143 ( 1997 ) ; m. randhawa and m. gupta , _ phys .
* 63 * , 097301 ( 2001 ) .
k. kang and s. k. kang , _ phys .
d _ * 56 * , 1511 ( 1997 ) ; p. s. gill and m. gupta , _ phys .
d _ * 57 * , 3971 ( 1998 ) ; m. randhawa , v. bhatnagar , p. s. gill and m. gupta , _ phys .
d _ * 60 * , 051301 ( 1999 ) ; h. nishiura , k. matsuda and t. fukuyama , _ phys .
d _ * 60 * , 013006 ( 1999 ) ; k. matsuda , t. fukuyama and h. nishiura , _ phys .
d _ * 61 * , 053001 ( 2000 ) ; k. kang , s. k. kang , c. s. kim and s. m. kim , _ mod .
a _ * 16 * , 2169 ( 2001 ) ; c. giunti and m. tanimoto , _ phys .
d _ * 66 * , 113006 ( 2002 ) ; m. frigerio and a. yu .
smirnov , _ nucl .
b _ * 640 * , 233 ( 2002 ) ; p. f. harrison and w. g. scott , _ phys . lett .
b _ * 547 * , 219 ( 2002 ) ; e. ma , _ mod .
a _ * 17 * , 2361 ( 2002 ) ; _ ibid . _
* 66 * , 117301 ( 2002 ) ; z. z. xing , _ phys .
b _ * 530 * , 159 ( 2002 ) ; _ ibid . _
b _ * 539 * , 85 ( 2002 ) ; a. kageyama , s. kaneko , n. simoyama and m. tanimoto , _ phys . lett . b _ * 538 * , 96 ( 2002 ) ; w. l. guo and z. z. xing , hep - ph/0211315 ; m. randhawa , g. ahuja and m. gupta , _ phys .
d _ * 65 * , 093016 ( 2002 ) ; m. frigerio and a. yu .
smirnov , _ phys .
* 67 * , 013007 ( 2003 ) ; k. s. babu , e. ma and j. w. f. valle , _ phys
b _ * 552 * , 207 ( 2003 ) ; w. l. guo and z. z. xing , _ phys . rev
. d _ * 67 * , 053002 ( 2003 ) ; s. kaneko and m. tanimoto , _ phys .
b _ * 551 * , 127 ( 2003 ) ; b. r. desai , d. p.
roy and a. r. vaucher , _ mod .
lett . a _ * 18 * , 1355 ( 2003 ) ; k. hasegawa , c. s. lim and k. ogure , _ phys
d _ * 68 * , 053006 ( 2003 ) ; m. honda , s. kaneko and m. tanimoto , _ jhep _ * 0309 * , 028 ( 2003 ) ; g. bhattacharyya , a. raychaudhuri and a. sil , _ phys .
d _ * 67 * , 073004 ( 2003 ) ; p. f. harrison and w. g. scott , _ phys .
b _ * 594 * , 324 ( 2004 ) ; m. bando , s. kaneko , m. obara and m. tanimoto , _ phys . lett .
b _ * 580 * , 229 ( 2004 ) ; o. l. g. peres and a. yu .
smirnov , _ nucl .
b _ * 680 * , 479 ( 2004 ) ; c. h. albright , _ phys . lett .
b _ * 599 * , 285 ( 2004 ) ; j. ferrandis and s. pakvasa , _ phys . lett .
b _ * 603 * , 184 ( 2004 ) ; s. t. petcov and w. rodejohann , _ phys
d _ * 71 * , 073002 ( 2005 ) ; s. s. masood , s. nasri and j. schechter , _ phys .
d _ * 71 * , 093005 ( 2005 ) ; r. dermek and s. raby , _ phys . lett .
b _ * 622 * , 327 ( 2005 ) ; f. plentinger and w. rodejohann , _ phys . lett .
b _ * 625 * , 264 ( 2005 ) ; s. dev and s. kumar , _ mod . phys .
* 22 * , 1401 ( 2007 ) ; s. dev , s. kumar , s. verma and s. gupta , _ nucl .
b _ * 784 * , 103 ( 2007 ) ; _ ibid . _
d _ * 76 * , 013002 ( 2007 ) .
p. minkowski , _ phys .
b _ * 67 * 421 ( 1977 ) ; t. yanagida , in _ proceedings of the workshop on unified theory and the baryon number of the universe _ , eds . o. sawada and a. sugamoto ( kek , tsukuba , 1979 ) , p. 95 ; m. gell - mann , p. ramond and r. slansky , in _ supergravity _ , eds . f. van nieuwenhuizen and d. freedman ( north holland , amsterdam , 1979 ) , p. 315 ; s.l . glashow , in _ quarks and leptons _
, eds . m. l@xmath155vy _ et al . _
( plenum , new york , 1980 ) , p. 707 ; r. n. mohapatra and g. senjanovic , _ phys . rev .
lett . _ * 44 * , 912 ( 1980 ) . | texture specific fermion mass matrices have played an important role in understanding several features of fermion masses and mixings . in the present work ,
we have given an overview of all possible cases of fritzsch - like as well as non fritzsch - like texture 6 and 5 zero fermion mass matrices .
further , for the case of texture 4 zero fritzsch - like quark mass matrices , the issue of the hierarchy of the elements of the mass matrices and the role of their phases have been discussed .
furthermore , the case of texture 4 zero fritzsch - like lepton mass matrices has also been discussed with an emphasis on the hierarchy of neutrino masses for both majorana and dirac neutrinos . |
although strong and electromagnetic decays of @xmath0 have been extensively studied for several decades , both experimental and theoretical investigations of weak decays of @xmath0 are much behind . due to smallness of the strength of weak interaction ,
the weak decays of the @xmath0 are rare processes .
sanchis - lonzano suggested to search for these rare decays whose sum of branching ratios were estimated to be at the order of @xmath4 @xcite .
such processes hardly drew much attention because the database was far from reaching such accuracy .
thus , for a long time , few further researches on this topic were done .
thanks to the progress of accelerator and detector techniques , more accurate measurements may be carried out , thus the interest on weak decays of @xmath0 has been revived .
the bes collaboration indeed starts to measure some rare weak decays of @xmath0 and eventually sets an upper bound on the branching ratio of @xmath5 at order of @xmath6 by using @xmath7 @xmath0 database @xcite .
the forthcoming upgraded besiii can accumulate @xmath8 @xmath0 per year @xcite , which makes it marginally possible to measure such weak decays of @xmath0 , at least one may expect to observe not - null such events .
thus , more careful theoretical investigation on these decays seems necessary .
indeed , the weak decays of heavy quarkonium like @xmath0 offer an ideal opportunity of studying non - perturbative qcd effects , because such systems contain two heavy constituents of the same flavor .
the situation is quite different from that for heavy mesons which contain only one heavy constituent , and the non - perturbative effects might be attributed to the light flavor , thus the heavy quark effective theory ( hqet ) applies .
moreover , for the weak decay of a vector meson , the polarization effect may play a role to probe the underlying dynamics and hadron structure @xcite .
the weak decay of @xmath0 is realized via the spectator mechanism that the charm quark ( antiquark ) decays and the antiquark ( quark ) acts as a spectator .
the characteristic of the decay modes is that the final state contains a single charmed hadron .
the theory of weak interactions has been thoroughly investigated and the effective hamiltonian at the quark level is perfectly formulated .
the main job of calculating the rates of the semi - leptonic decays of @xmath0 is to properly evaluate the hadronic matrix elements for @xmath9 , namely the transition form factors which are obviously governed by non - perturbative qcd effects .
the main aim of this work is to calculate the @xmath10 form factors in the qcd sum rules .
the weak decay of heavy quarkonium has been studied by virtue of heavy quark spin symmetry @xcite . in that framework , the transition form factors of a heavy quarkonium to heavy pseudoscalar and vector mesons are parameterized by a universal function @xmath11 in analog to the isgur - wise function for the heavy meson transitions .
however , the non - recoil approximation @xmath12 was used in ref.@xcite , which would bring up uncontrollable uncertainties to the estimation of decay widths .
it seems helpful to re - investigate these processes based on a more rigorous theoretical framework .
motivated by the arguments , in this work we will calculate the form factors for heavy quarkonium @xmath0 decays into a pseudoscalar or vector meson in the qcd sum rules . as a matter of fact , many authors have tried to evaluate the transition form factors for the heavy meson and quarkonium system in various approaches , such as the simple quark model @xcite , light - front approach @xcite , the qcd sum rules @xcite , the perturbative qcd approach @xcite and etc .
the qcd sum - rule approach , which is rooted in the quantum field theory and fully relativistic , is considered to be one of the effective tools for analyzing hadronic processes @xcite . besides evaluation of hadron spectra
, the qcd sum - rule technique has been applied to calculate the pion electromagnetic form factor at intermediate momentum transfer @xcite , various weak decay channels @xcite , the coupling constant of the strong interaction @xcite and even to determine the light cone distribution amplitudes of hadrons @xcite .
the advantage of this method is that the non - perturbative qcd effects are included in a few parameters such as the quark- and gluon - condensates which have evident physical meaning @xcite .
after this introduction , we will firstly display the effective hamiltonian relevant to the semi - leptonic decays of @xmath0 to @xmath13 , and the sum rules for form factors in section [ the standard procedure ] .
the wilson coefficients of various operators which manifest the perturbative qcd effects are also calculated in this section with the help of operator product expansion ( ope ) technique .
the numerical analysis on the form factors are performed in section [ numerical results ] .
the decay rates of semi - leptonic decay @xmath14 and a comparison of our results with that obtained based on other approaches are presented in section [ decay rate ] . in the last section
we draw our conclusion .
for the semi - leptonic decays @xmath16 , the effective weak hamiltonian is given by @xmath17 in this subsection , we calculate transition form factors of @xmath23 in the qcd sum rules . here
we present the formulations for @xmath24 transition explicitly , while the expressions for @xmath25 can be obtained by simple replacements of @xmath26 and @xmath27 quark to @xmath28 quark . following the standard procedure of the qcd sum rules @xcite ,
we write the three - point correlation function for @xmath0 to @xmath30 as @xmath31 where the current @xmath32 represents the @xmath0 channel ; @xmath33 is the weak current and @xmath34 corresponds to the @xmath35 channel . in terms of the following definitions , @xmath36 we can insert a complete set of hadronic states with the quantum numbers the same as @xmath0 and @xmath37 to achieve the hadronic representation of the correlator ( [ correlator pseudoscalar ] ) @xmath38 obviously the concerned lowest hadronic states are @xmath0 and @xmath39 , while the terms with `` higher states '' represent contributions coming from higher excited states and continuum . using the double dispersion relation , the contributions of excited states and continuum can be expressed as @xmath40 where @xmath41 denotes the integration region in the @xmath42 plane .
@xmath43 is the spectral density at the hadron level .
the subtraction terms are polynomials of either @xmath44 or @xmath45 , which should disappear after performing the double borel transformation @xmath46 , with @xmath47 on the other side , we calculate the correlation function at the quark level by using the ope as @xmath48 where each coefficient contains contributions from both perturbative part and the non - perturbative part whose effects manifest in several typical condensates , @xmath49 with @xmath50 denoting the contributions to the correlation functions from dimension 0 , 3 , 4 , 5 , ... operators . by the quark - hadron duality
, one may match the two different representations of the correlation function and perform the double borel transformation on variables @xmath44 and @xmath45 , then we get the sum rules for the form factors @xmath51 . \label{a0 pseudoscalar}\end{aligned}\ ] ] the three - point correlation function of @xmath0 to @xmath53 is @xmath54 where the current @xmath55 denotes the @xmath56 channel , and @xmath57 , @xmath58 are defined as in the above subsection .
one the one hand , inserting the hadron states , the correlation function is written as @xmath59 on the other hand , the correlation function at the quark level is formulated as @xmath60 where each coefficient @xmath61 includes contributions from both perturbative and nonperturbative parts , and is written explicitly as @xmath62 again , equating the correlation functions calculated in these two frameworks and performing the borel transformations on both sides , we derive the form factors of @xmath63 as @xmath64,\\ \tilde{a}_2(q^2)&=&{m_{d^{*}_s}^4 - 2(q^2+m_{\psi}^2)m_{d^{*}_s}^2 + ( q^2-m_{\psi}^2)^2\over 4(m_{d^{*}_s}^2-m_{\psi}^2 ) m_{\psi}f_{\psi}m_{d^{*}_s}f_{d^{*}_s}}e^{m_{\psi}^2 / m_1 ^ 2}e^{m_{d^{*}_s}^2 / m_2 ^ 2}m_1 ^ 2 m_2 ^ 2 \hat{\mathcal{b}}(f_5+f_6 ) , \\ \tilde{a}_3(q^2)&=&{m_{\psi}^2-m_{d^{*}_s}^2 \over m_{\psi}f_{\psi}m_{d^{*}_s}f_{d^{*}_s}}e^{m_{\psi}^2 / m_1 ^ 2}e^{m_{d^{*}_s}^2 / m_2 ^ 2}m_1 ^ 2 m_2 ^ 2 \hat{\mathcal{b}}(f_1-f_5 ) , \\
\tilde{a}_4(q^2)&=&{{m_{\psi}}^2-m_{d^{*}_s}^2\over m_{\psi}f_{\psi}m_{d^{*}_s}f_{d^{*}_s}}e^{m_{\psi}^2 / m_1 ^ 2}e^{m_{d^{*}_s}^2 / m_2 ^ 2}m_1 ^ 2 m_2 ^ 2\hat{\mathcal{b}}(f_4+f_6 ) , \\ \tilde{v}_1(q^2)&=&-{1 \over 2 m_{\psi}f_{\psi}m_{d^{*}_s}f_{d^{*}_s}}e^{m_{\psi}^2 / m_1 ^ 2}e^{m_{d^{*}_s}^2 / m_2 ^ 2}m_1 ^ 2 m_2 ^ 2\hat{\mathcal{b}}(f_{9}+f_{12 } ) , \\ \tilde{v}_2(q^2)&=&{1
\over 2 m_{\psi}f_{\psi}m_{d^{*}_s}f_{d^{*}_s}}e^{m_{\psi}^2 / m_1 ^ 2}e^{m_{d^{*}_s}^2 / m_2 ^ 2}m_1 ^ 2 m_2 ^ 2\hat{\mathcal{b}}(f_{9}-f_{12 } ) , \\ \tilde{v}_{3}(q^2)&=&{m_{d^{*}_s}^2 - { m_{\psi}}^2 \over 2 m_{\psi}f_{\psi}m_{d^{*}_s}f_{d^{*}_s}}e^{m_{\psi}^2 / m_1 ^ 2}e^{m_{d^{*}_s}^2 / m_2 ^ 2}m_1 ^ 2 m_2 ^ 2\hat{\mathcal{b}}(f_{14}+f_{15 } ) , \\
\tilde{v}_{4}(q^2)&=&{1\over 2 m_{\psi}f_{\psi}m_{d^{*}_s}f_{d^{*}_s}}e^{m_{\psi}^2 / m_1 ^ 2}e^{m_{d^{*}_s}^2 / m_2 ^ 2}m_1 ^ 2 m_2 ^ 2 \hat{\mathcal{b}}[(f_{14}-f_{15})q^2+(f_{14}+f_{15 } ) ( m_{d^{*}_s}^2 - { m_{\psi}}^2 ) ] , \\
\tilde{v}_5(q^2)&=&{1 \over m_{\psi}f_{\psi}m_{d^{*}_s}f_{d^{*}_s}}e^{m_{\psi}^2 / m_1 ^ 2}e^{m_{d^{*}_s}^2 / m_2 ^ 2}m_1 ^ 2 m_2 ^ 2\hat{\mathcal{b}}f_{11 } , \\ \tilde{v}_6(q^2)&=&{1 \over m_{\psi}f_{\psi}m_{d^{*-}}f_{d^{*}_s}}e^{m_{\psi}^2 / m_1 ^ 2}e^{m_{d^{*}_s}^2 / m_2 ^ 2}m_1 ^ 2 m_2 ^ 2\hat{\mathcal{b}}f_{7}. \label{v6 vector}\end{aligned}\ ] ] in this subsection we calculate the wilson coefficients which are defined above . to guarantee sufficient theoretical accuracy ,
the correlation functions are required to be expanded up to dimension-5 operators , namely quark - gluon mixing condensate .
the dimension-6 operators , such as the four quark condensates , are small and further suppressed by @xmath66 , so can be safely neglected in our calculations .
the diagrams which depict the contributions from the perturbative part and nonperturbative condensates are shown in fig .
[ wilson coefficients graph ] .
the first diagram results in the wilson coefficient of the unit operator ; the second diagram is relevant to the contribution of quark condensate , where the heavy - quark condensate is neglected . the wilson coefficient of the two - gluon condensate operator is obtained from fig .
[ wilson coefficients graph](c - h ) .
the last two diagrams fig .
[ wilson coefficients graph](i - j ) stand for the contribution of quark - gluon mixing condensate . in this work ,
all of the wilson coefficients are calculated at the lowest order in the running coupling constant of strong interaction . [ cols="^,^,^ " , ]
with the form factors derived above , we can perform calculations on partial widths of the semi - leptonic decays of @xmath0 . the relevant ckm parameters are directly taken from the particle data book @xcite : @xmath67 for the semi - leptonic decays @xmath68 ( @xmath69 ) , the differential partial decay rate is written as @xmath70 where @xmath71 ; @xmath72 and @xmath73 are the momenta of @xmath74 and @xmath75 respectively ;
@xmath76 is square of the transition amplitude after integrating over the angle between the @xmath74 and @xmath77 .
the upper and lower bounds for @xmath78 are given as @xmath79 the transition amplitude @xmath80 for @xmath81 reads as @xmath82 finally we get the branching ratios of the semi - leptonic decays as @xmath83 where we have combined various uncertainties in the form factors discussed in last section to determine the final error tolerance in our theoretical calculations .
our predictions are much below the present experimental upper bounds @xcite : @xmath84 , @xmath85 . a few remarks are presented in order .
firstly , the sum of the branching fractions of semi - leptonic decays of @xmath0 whose final state includes @xmath86 , @xmath56 , @xmath87 and @xmath88 and their charge conjugate channels can reach as large as @xmath89 , which is expected to be marginally observed at besiii .
secondly , it is worthwhile to point out that the decay rates for the dominant semi - leptonic weak decays of @xmath0 obtained in ref .
@xcite were about @xmath90 , which are two times greater than that calculated in this work .
this discrepancy can attribute to the heavy quark spin symmetry and the non - recoil approximation used in ref .
@xcite , also to the different methods used to estimate the non - perturbative form factors . , while we adopt the qcd sum rules to calculate the form factors in this work . ]
thirdly , the ratio of @xmath91 is about 2 times larger than the value calculated in @xcite , where the assumption of heavy quark spin symmetry and non - recoil approximation were adopted .
forth , the ratios @xmath92 and @xmath93 should be equal to @xmath94 under the su(3 ) flavor symmetry limit .
our numerical calculations show that @xmath95 and @xmath96 , which implies large effect of su(3 ) symmetry breaking .
the charmonium @xmath0 meson can decay via the strong and electromagnetic interactions , thus weak decays of @xmath0 should be very rare unless there is new physics beyond the standard model to make a substantial contribution .
if such weak decays can be measured by the future experiments with sizable branching ratios , it would be a clear signal for new physics . to make the new physics signal clearly distinguishable from the standard model , a careful study of weak decays in sm
is needed . in this work , we calculated the form factors of weak transitions of @xmath97 in terms of the qcd sum rules . with the form factors , we estimate the branching ratios of the semileptonic weak decays of @xmath0 and find that the sum of the branching ratios corresponding to the dominant modes is about @xmath98 which may be marginally measured by bes iii .
the qcd sum rules approach possesses uncontrollable errors as large as 20 to 30% , was confirmed by our numerical results .
moreover , due to a coulomb - type correction in heavy quarkonium decay ( or @xmath99 ) , which may manifest as the ladder structure in the loop - triangle ( as a part of multi - loop diagrams ) , the spectral function needs to be multiplied by a finite renormalization factor @xcite . this would bring up another kind of uncertainty .
it is expected that this kind of correction can give birth to the twice multiplication of the form factors at the maximal momentum transfer . however , in this work , we calculate both the three - point qcd sum rules of the weak transition form factors and the two - point sum rules for the decay constant of @xmath0 to the same order of @xmath100 . then , it is expected that most uncertainties due to the coulomb - like corrections are canceled in our calculations ; therefore , the coulomb - like corrections for the @xmath0 channel are not included in our calculations . as for the heavy - light mesons , there are no corrections in the power of inverse velocity for it , since the light quark moves relativistically .
therefore , one should explore the sum rules for both three - point and two - point correlation functions up to next - to - leading order in strong coupling constant so that coulomb - like corrections to the heavy - light mesons can be canceled effectively .
moreover , an explicit calculation of coulomb - like corrections to the heavy - light vertex in triangle diagram is still not available now , which can be left for further considerations .
one can trust the numerical results to a certain accuracy , at least the order of magnitude is reliable . with these form factors
we may continue to estimate the rates of non - leptonic weak decays of @xmath0 as long as the factorization theorem is proved
. that would be the contents of our next work @xcite .
the branching ratios of semi - leptonic weak decays of @xmath0 are very small in sm , even though their strong decay modes are ozi - suppressed .
our numerical results indicate that even with a large database which will be collected by the bes iii , such weak decay modes may only be marginally observed .
therefore we lay our expectation on our bes iii colleagues and hope them to provide sufficiently large database to make this challengeable field more fruitful .
this work is partly supported by national science foundation of china under grant no.10475085 , 10625525 and 10475042 .
the authors would like to thank t.m .
aliev , p. colangelo , t. huang , v.v .
kiselev , n. paver , m.z .
yang , f.k .
guo , y.l .
shen and w. wang for helpful discussions .
in this appendix , we would like to show the explicit expressions of wilson coefficients appearing in eq.([v pseudoscalar]-[a0 pseudoscalar ] ) after the borel transformation . as mentioned before ,
only perturbative part contributes to the correlation function for @xmath0 decays to @xmath102 at the leading order of heavy quark mass expansion and the @xmath100 expansion series .
namely , eq . ( [ fi expansion ] ) can be written as @xmath103 with @xmath104 being the heavy charm quark here .
the @xmath105 can be related to @xmath106 defined in eq .
( [ spectral density 1 ] ) by @xmath107 or @xmath108 the lowest bound of @xmath109 , i.e. , @xmath110 can be determined by the eq .
( [ integral region ] ) as @xmath111 , \label{s1l}\end{aligned}\ ] ] according to the landau equation @xcite .
the obvious forms of @xmath112 are @xmath113,\nonumber \\ \rho_{2}^{pert}(s_1,s_2,q^2)&=&- { 3 s_1 \over 2 \pi^2 \lambda^{5/2 } } \bigg\ { ( m_c - m_q)s_1 \bigg[6 m_c^4- 6(2 m_q^2 + s_1-s_2)m_c^2 + 6 m_q^4 + ( s_1-s_2)^2 \nonumber \\ & & + 6 m_q^2(s_1-s_2)\bigg]+m_c [ 2(m_c - m_q)(2m_c+m_q)-s_1+s_2]\lambda \nonumber \\ & & + q^2[2(m_c - m_q)s_1(3m_c^2 -3 m_q^2 -s_1 + 2 s_2 ) + m_c \lambda + ( m_c - m_q)s_1 q^2]\bigg\},\nonumber \\ \rho_{4}^{pert}(s_1,s_2,q^2)&= & { 3 \over 4 \pi^2 \lambda^{5/2 } } \bigg\{[-m_c \lambda^2+(m_c - m_q)(2s_2 m_c^2 + s_1(2 m_q^2 + s_1-s_2 ) ] \lambda\nonumber \\ & & + 2 ( m_c - m_q ) s_1 [ 3(s_1+s_2)m_c^4 \nonumber \\ & & -2 ( 3(s_1+s_2)m_q^2+(s_1-s_2)(s_1 + 2 s_2))m_c^2+(s_1-s_2)^2 s_2 + 3m_q^4 ( s_1+s_2)\nonumber \\ & & + 2 m_q^2 ( s_1-s_2)(s_1 + 2 s_2 ) ] + ( m_c - m_q)q^2 [ 2s_1 ( s_2 ^ 2 + ( -2 m_c^2 + 2 m_q^2+s_1)s_2 \nonumber \\ & & + ( m_c^2-m_q^2)(-3 m_c^2 + 3m_q^2 + 4 s_1))-(2 m_c^2+s_1)\lambda -4 s_1 ( m_c^2 -m_q^2 + s_2)q^2]\bigg\},\nonumber \\ \rho_{5}^{pert}(s_1,s_2,q^2)&=&{3 \over 8 \pi^2 \lambda^{3/2 } } \bigg\{\lambda ( m_q s_1 + m_c s_2 -m_c q^2 ) - 2 ( m_c - m_q)[\lambda m_c^2 \nonumber \\ & & + ( m_c^2-m_q^2)s_1 ( m_c^2-m_q^2 -s_1+s_2)+s_1 ( m_c^2-m_q^2+s_2)q^2]\bigg\}.\end{aligned}\ ] ] in this appendix , we adopt the notion @xmath114 for the convenience of writing .
similarly , we will display the forms of wilson coefficients emerged in the eq.([a1 vector]-[v6 vector ] ) after performing the borel transformation . as been discussed in the text ,
only perturbative part contributes to the three - point function , @xmath116 the relationship between @xmath117 , and @xmath118 are given as @xmath119 or @xmath120 where the definition of @xmath121 is given in eq.([s1l ] ) .
the manifest expressions of @xmath122 are displayed as @xmath123\},\nonumber \\
{ \rho'}_{4}^{pert}(s_1,s_2,q^2)&=&\frac{3s_1}{4\pi^2\lambda^{5/2 } } \{-4s_2m_c^4 - 2(-4s_2m_q^2+s_1 ^ 2 + 3s_2 ^ 2 - 4s_1s_2-\lambda)m_c^2 + 4m_qm_c\lambda-4m_q^4s_2 \nonumber\\ & & -(s_1 + 3s_2)((s_1-s_2)^2-\lambda)+2m_q^2 ( s_1 ^ 2 - 4s_1s_2 + 3s_2 ^ 2-\lambda)+q^2(4m_c^4 + 4(s_2 - 2m_q^2)m_c^2\nonumber\\ & & + 4m_q^4 + 3(s_1+s_2)^2 -4m_q^2s_2-\lambda+q^2(2m_c^2 - 2m_q^2 - 3s_1-s_2+q^2))\},\nonumber \\ { \rho'}_{5}^{pert}(s_1,s_2,q^2)&=&\frac{1}{4\pi^2\lambda^{5/2 } } \{-(s_1+s_2-q^2)[3(m_c^2-m_q^2+s_2)q^4 - 6((2s_1+s_2)m_c^2+s_2(s_1+s_2)\nonumber\\ & & -m_q^2 ( 2s_1+s_2))q^2 - 3(2s_1m_c^4-(4s_1m_q^2 + 3s_1 ^ 2+s_2 ^ 2-\lambda)m_c^2\nonumber\\ & & + 2m_qm_c\lambda+2m_q^4s_1 + m_q^2(3s_1 ^
2+s_2 ^ 2-\lambda)+s_2(s_1 ^ 2-s_2 ^ 2+\lambda))]-2s_2[3s_1(-2(m_c^2-m_q^2)^2 \nonumber\\ & & + s_1 ^ 2-s_2 ^ 2 - 4(m_c^2-m_q^2)s_2)-3(2m_c^2+s_1)\lambda+3s_1 q^2(q^2 - 2(s_1+s_2))]\},\nonumber \\ { \rho'}_{6}^{pert}(s_1,s_2,q^2)&=&\frac{3}{4\pi^2\lambda^{5/2 } } \bigg\{-2s_1(3s_1+s_2)m_c^4 + 2\bigg[s_1(2m_q^2+s_1-s_2)(3s_1+s_2)-(2s_1+s_2 ) \lambda\bigg]m_c^2\nonumber\\ & & -4m_qs_1m_c\lambda + s_1\bigg[-2(3s_1+s_2)m_q^4 + 2(-3s_1 ^ 2 + 2s_1s_2+s_2 ^ 2+\lambda)m_q^2 + ( s_1-s_2)^2(s_1+s_2)\nonumber\\ & & -(s_1 + 3s_2)\lambda\bigg]+q^2\bigg[2s_1m_c^4 + 2(\lambda-2s_1(m_q^2 + 2s_1))m_c^2\nonumber\\ & & + s_1(2m_q^4 + 8s_1m_q^2-(s_1+s_2)(3s_1 + 5s_2)+\lambda ) + s_1(2m_c^2 - 2m_q^2 + 3s_1 + 5s_2-q^2)q^2\bigg]\bigg\},\nonumber \\ { \rho'}_{7}^{pert}(s_1,s_2,q^2)&=&-\frac{3}{16\pi^2\lambda^{5/2 } } \bigg\{(2(m_c - m_q)^2 - 2s_2)\lambda^2 -2\bigg[s_2(-s_1+s_2-q^2)+(m_c^2-m_q^2)(s_1+s_2-q^2)\bigg]\nonumber\\ & & \times(s_1-s_2+q^2)\lambda-8\bigg[(m_c^2-m_q^2+s_2)(s_1+s_2-q^2)-2s_1s_2\bigg ] \bigg[((s_1+s_2-q^2)^2\nonumber\\ & & -4s_1s_2)m_c^2+s_1(m_c^2-m_q^2+s_2)^2+s_1 ^ 2s_2-s_1(m_c^2-m_q^2+s_2 ) ( s_1+s_2-q^2)\bigg]\bigg\},\nonumber \\ { \rho'}_{9}^{pert}(s_1,s_2,q^2)&=&-\frac{3}{8\pi^2\lambda^{5/2 } } \bigg\{-((m_c - m_q)^2-s_2)\lambda^2-\bigg[s_2(-s_1+s_2-q^2)+(m_c^2-m_q^2 ) ( s_1+s_2-q^2)\bigg]\nonumber\\ & & \times(s_1-s_2+q^2)\lambda+2\bigg[\lambda m_c^2+(m_c^2-m_q^2)s_1(m_c^2-m_q^2-s_1+s_2)+s_1(m_c^2-m_q^2+s_2)q^2\bigg]\lambda\nonumber\\ & & -4\bigg[(m_c^2-m_q^2+s_2)(s_1+s_2-q^2)-2s_1s_2\bigg]\bigg[((s_1+s_2-q^2)^2 - 4s_1s_2)m_c^2 \nonumber\\ & & + s_1(m_c^2-m_q^2+s_2)^2+s_1 ^ 2s_2-s_1(m_c^2-m_q^2+s_2 ) ( s_1+s_2-q^2)\bigg]\bigg\},\nonumber \\ { \rho'}_{11}^{pert}(s_1,s_2,q^2)&=&-\frac{3s_1}{8\pi^2\lambda^{5/2 } } \bigg\{-\lambda^2-(-2m_c^2 + 2m_q^2+s_1-s_2-q^2)[-2(m_c - m_q)^2 -s_1+s_2+q^2]\lambda \nonumber\\ & & -4\bigg[((s_1+s_2-q^2)^2 - 4s_1s_2)m_c^2+s_1(m_c^2-m_q^2+s_2)^2+s_1 ^ 2s_2\nonumber\\ & & -s_1(m_c^2-m_q^2+s_2)(s_1+s_2-q^2)\bigg](-2m_c^2 + 2m_q^2+s_1-s_2-q^2)\bigg\},\nonumber \\ { \rho'}_{12}^{pert}(s_1,s_2,q^2)&=&-\frac{3}{8\pi^2\lambda^{5/2 } } \bigg\{s_1\lambda^2-s_1(-2m_c^2 + 2m_q^2+s_1-s_2-q^2)(s_1+s_2-q^2)\lambda\nonumber\\ & & + 2\bigg[\lambda m_c^2+(m_c^2-m_q^2)s_1(m_c^2-m_q^2-s_1+s_2)+s_1(m_c^2-m_q^2+s_2)q^2\bigg]\lambda \nonumber\\ & & -4s_1\bigg[((s_1+s_2-q^2)^2 - 4s_1s_2)m_c^2+s_1(m_c^2-m_q^2+s_2)^2+s_1 ^ 2s_2 \nonumber\\ & & -s_1(m_c^2-m_q^2+s_2)(s_1+s_2-q^2)\bigg ] ( -2m_c^2 + 2m_q^2+s_1-s_2-q^2)\bigg\},\nonumber \\ { \rho'}_{14}^{pert}(s_1,s_2,q^2)&=&-\frac{3}{4\pi^2\lambda^{7/2 } } \bigg\{-2m_c^2(s_1+s_2-q^2)\lambda^2-s_1\bigg[4s_1s_2(-2m_c^2 + 2m_q^2+s_1-s_2-q^2)\nonumber\\ & & -(s_2(-s_1+s_2-q^2)+(m_c^2-m_q^2)(s_1+s_2-q^2))(s_1+s_2-q^2)\nonumber\\ & & + 3(m_c^2-m_q^2+s_2)(s_1+s_2-q^2)(2m_c^2 - 2m_q^2-s_1+s_2+q^2)\bigg]\lambda \nonumber\\ & & + 4s_1\bigg[4s_2\{(s_1+s_2-q^2)^2+s_1s_2\}s_1 ^ 2 - 3(m_c^2-m_q^2+s_2)\{(s_1+s_2-q^2)^2
\nonumber\\ & & + 6s_1s_2\}(s_1+s_2-q^2)s_1 + 2(m_c^2(s_1+s_2-q^2)^4 + \{6s_1(m_c^2-m_q^2+s_2)^2 \nonumber\\ & & -2m_c^2s_1s_2\}(s_1+s_2-q^2)^2 + 2s_1 ^ 2s_2(3(m_c^2-m_q^2+s_2)^2 - 4m_c^2s_2))-2(m_c^2-m_q^2\nonumber\\ & & + s_2)\{3((s_1+s_2-q^2)^2 - 4s_1s_2)m_c^2 + 5s_1(m_c^2-m_q^2+s_2)^2\ } ( s_1+s_2-q^2)\bigg]\bigg\},\nonumber \\ { \rho'}_{15}^{pert}(s_1,s_2,q^2)&=&-\frac{3}{4\pi^2\lambda^{7/2 } } \bigg\{-2m_c^2(s_1+s_2-q^2)\lambda^2-s_1(4s_1s_2(-2m_c^2 + 2m_q^2+s_1-s_2-q^2)\nonumber\\ & & -\bigg[s_2(-s_1+s_2-q^2)+(m_c^2-m_q^2)(s_1+s_2-q^2)\bigg](s_1+s_2-q^2)\nonumber\\ & & + 3(m_c^2-m_q^2+s_2)(s_1+s_2-q^2)(2m_c^2 - 2m_q^2-s_1+s_2+q^2))\lambda\nonumber\\ & & + 4\bigg[-10s_2 ^ 2(s_1+s_2-q^2)s_1 ^ 3 + 12s_2(m_c^2-m_q^2+s_2)((s_1+s_2-q^2)^2+s_1s_2)s_1 ^ 2 \nonumber\\ & & -3(2s_2((s_1+s_2-q^2)^2 - 4s_1s_2)m_c^2+(m_c^2-m_q^2+s_2)^2((s_1+s_2-q^2)^2 + 6s_1s_2 ) ) ( s_1\nonumber\\ & & + s_2-q^2)s_1 + 2(m_c^2-m_q^2+s_2)(m_c^2(s_1+s_2-q^2)^4 + 2s_1((m_c^2-m_q^2+s_2)^2\nonumber\\ & & -m_c^2s_2 ) ( s_1+s_2-q^2)^2 + 2s_1 ^ 2s_2((m_c^2-m_q^2+s_2)^2 - 4m_c^2s_2))\bigg]\bigg\}.\end{aligned}\ ] ] [ perturbative 2 ]
in this appendix , we would like to collect the sum rules for the decay constants of @xmath0 and @xmath124 for the completeness of the paper .
the decay constant of @xmath0 in the two - point qcd sum rules can be written as @xcite where the nonrelativistic approximation for the gluon condensate has been adopted for the convenience of performing the borel transformation .
it is observed that the gluon condensate has tiny effect on the results of the form factors and hence are neglected in the sum rules of charmed mesons .
the non - perturbative condensates used in the evaluation of the sum rues can be grouped as @xmath126 where @xmath127 and the subscript @xmath128 denotes the flavor of quarks .
based on the two - point sum rules of @xmath0 and the parameters showed above , we can derive the decay constant of @xmath0 as @xmath129 , where we have combined the uncertainties from the variations of the borel masses and threshold value for @xmath0 channel .
the sum rules for the decay constant of @xmath130 can be given by @xcite @xmath131 \lambda^{1/2}(s , m_c^2,m_q^2 ) e^{-{s \over m^2 } } \\ & & + \left(-m_c + { m_q
\over 2}+{m_q m_c^2 \over 2m^2}\right ) e^{-{m_c^2 \over m^2 } } \langle 0 | \bar{q } q |0 \rangle
-{m_c \over 2m^2 } ( 1- { m_c^2 \over 2m^2 } ) e^{-{m_c^2 \over m^2 } } \langle 0|\bar{q } \sigma \cdot g q|0\rangle , \nonumber\end{aligned}\ ] ] from which we can arrive at the decay constants of pseudoscalar charmed mesons as @xmath132 and @xmath133 .
the decay constants of vector charmed mesons @xmath134 in the framework of qcd sum rules can be calculated as @xcite @xmath135 e^{-{s \over m^2 } } \nonumber \\ & & + \left\{\left[-(m_c + { 8 \over 3 } m_q ) + { 1 \over 2}{m_q m_c^2 \over m^2}\right ] e^{-{m_c^2 \over m^2 } } + { 2m_q ( 4 m^2-m_c^2)\over 3 m^2 } \right\ } \langle 0|\bar{q } q|0\rangle \nonumber \\ & & + { m_c^3
\over 4 m^4 } e^{-{m_c^2 \over m^2 } } \langle 0|\bar{q } \sigma \cdot g q|0\rangle,\end{aligned}\ ] ] from which we can achieve the decay constants of vector charmed mesons as @xmath136 and @xmath137 .
m. terentev , sov .
* 24 * , 106 ( 1976 ) ; v. berestetsky and m. terentev , @xmath138 . *
24 * , 547 ( 1976 ) ; * 25 * , 347 ( 1977 ) ; p. chung , f. coester , and w. polyzou , phys . lett .
* b 205 * , 545 ( 1988 ) ; w. jaus , phys . rev . *
d 41 * , 3394 ( 1990 ) ; * 44 * , 2851 ( 1991 ) ; * 60 * , 054026 ( 1999 ) . c. ji , p. chung and s. cotanch , phys . rev . *
d 45 * , 4214 ( 1992 ) ; h.y .
cheng , c.y .
cheung and c.w .
hwang , phys .
d 55 * , 1559 ( 1997 ) [ arxiv : hep - ph/9607332 ] ; h.y .
cheng , c.k .
chua and c.w .
hwang , phys .
d 69 * , 074025 ( 2004 ) [ arxiv : hep - ph/0310359 ] ; c.w .
hwang and z.t .
wei , j. phys .
* g 34 * , 687 ( 2007 ) [ arxiv : hep - ph/0609036 ] ; c.d .
l , w. wang and z.t .
wei , phys . rev . *
d 76 * , 014013 ( 2007 ) [ arxiv : hep - ph/0701265 ] .
keum , h.n .
li , and a.i .
sanda , phys .
b 504 * , 6 ( 2001 ) [ arxiv : hep - ph/0004004 ] ; c.d .
l , k. ukai , and m.z .
yang , phys . rev . *
d 63 * , 074009 ( 2001 ) [ arxiv : hep - ph/0004213 ] ; t. kurimoto , h.n .
sanda , phys .
* d 65 * , 014007 ( 2002 ) [ arxiv : hep - ph/0105003 ] ; phys . rev .
* d 67 * , 054028 ( 2003 ) [ arxiv : hep - ph/0210289 ] ; z.t .
wei , m.z .
yang , nucl . phys . *
b 642 * , 263 ( 2002 ) [ arxiv : hep - ph/0202018 ] ; c.d .
lu and m.z .
yang , eur . phys .
j. * c 28 * , 515 ( 2003 ) [ arxiv : hep - ph/0212373 ] .
chernyak and a.r .
zhitnitsky , phys . rep . * 112 * , 173 ( 1984);v.l .
chernyak , a.a .
ogloblin and i.r .
zhitnitsky , z. phys . * c 42 * , 569 ( 1989 ) ; p. ball , v. m. braun , y. koike and k. tanaka , nucl . phys . * b 529 * , 323 ( 1998 ) [ arxiv : hep - ph/9802299 ] ; p. ball , v. m. braun , nucl . phys .
* b 529 * , 323 ( 1998 ) [ arxiv : hep - ph/9810475 ] ; p. ball , jhep * 9901 * , 010 ( 1999 ) [ arxiv : hep - ph/9812375 ] ; h.y .
cheng , c.k .
chua and k.c .
yang , phys . rev . *
d 73 * , 014017 ( 2006 ) [ arxiv : hep - ph/0508104 ] ; c.d .
l@xmath139 , y.m .
wang and h. zou , phys . rev . *
d 75 * , 056001 ( 2007 ) [ arxiv : hep - ph/0612210 ] . | within the standard model , we investigate the semi - leptonic weak decays of @xmath0 .
the various form factors of @xmath0 transiting to a single charmed meson ( @xmath1 ) are studied in the framework of the qcd sum rules .
these form factors fully determine the rates of the weak semi - leptonic decays of @xmath0 and provide valuable information about the non - perturbative qcd effects .
our results indicate that the decay rate of the semi - leptonic weak decay mode @xmath2 is at order of @xmath3 . |
during the last few years , our understanding of the structure , organization and evolution of magnetic fields in the very quiet sun ( the regions outside active regions and the network ) has become increasingly clear .
magnetic fields in the quietest areas of the sun are relatively weak and organized at small spatial scales , which yields weak polarization signals that are difficult to observe . until very recently , the general picture of the structure of its magnetism was rather rough : a `` turbulent '' disorganized field @xcite .
it is now clear that even in very quiet areas , magnetic fields may organize as coherent loops at granular and subgranular scales ( @xmath21000 km ; * ? ? ?
* ) , that these small loops are dynamic @xcite , that they pervade the quiet solar surface and may even connect with upper atmospheric layers @xcite .
yet , this picture is still incomplete .
for example , we lack a complete mapping of the full magnetic field vector on extended fields of view , because the linear polarization signals are intrinsically weak ( they are second order on the transverse magnetic field component ) , and high spatial resolution maps on linear polarization are rather patchy @xcite , which has led to incomplete ( and sometimes , physically problematic ) characterizations of the topology of the field @xcite . here we look for and trace small magnetic loops on extended regions of the quiet sun observed with the highest spatial resolution .
loop - like structures are a natural configuration of the magnetic field due to its solenoidal character . while they can be traced as single , individual , coherent entities , they characterize the magnetic field at large , and their statistics and evolution may shed some light on the origin of the very quiet sun magnetism , in particular , on the operation or not of local dynamo action @xcite . on the other hand ,
the organization of the field at small scales affect the organization of the magnetic field at larger scales and in higher atmospherics layers @xcite , and dynamics @xcite .
we find evidence for the small scale loops appearing rather irregularly , as in bursts and clumps .
moreover , wide regions of the very quiet sun show very low magnetic activity and no apparent sign of organized loops at the detection level of the instruments .
these extremely quiet ( _ dead calm _ ) regions are an intrinsic characteristic of the statistical distribution of these events .
this paper is devoted to the analysis of disk center quiet sun observations obtained with the imax instrument @xcite onboard the sunrise balloon borne observatory @xcite .
imax is a fabry - prot interferometer with polarimetric capabilities at the fei line at 5250.2 .
we analyze two different data sets .
both have the same properties except that they trace different regions of the quiet sun and were observed at different times , being times series of 22 and 31 min duration .
they consist of five filtergrams taken at @xmath640 , @xmath680 and + 227 m from the fei 5250.2 line center .
the field of view is 46.8@xmath7 ( [email protected] mm@xmath1 ) , 20 times larger than the one observed by @xcite .
the spatial resolution is of about [email protected] - [email protected]@xmath9 .
the time cadence is 32 s ( note that it is 28 s in * ? ? ?
* ) , allowing a noise level of @xmath10 and @xmath11 i@xmath12 in the circular and linear polarization , respectively ( i@xmath12 being the continuum intensity ) .
we look for small magnetic loops in the data set : coherent structures that appear as a dipole in the longitudinal magnetogram ( i.e. , adjacent positive and negative patches in stokes-@xmath13 maps ) , and a linear polarization patch between them ( see figure 1 ) , and that remain identifiable for several ( at least 2 ) time frames ( @xmath141 minute ) .
we looked for small magnetic loops by direct visual inspection .
a systematic search was performed in both field - of - views ( fovs ) at all times , for these structures by one of the authors ( mjmg ) , recording the position and evolution at different times of every single event .
the method was validated by selecting a small area ( @xmath15 mm@xmath1 ) and independently looking for such structures by a different observer ( rms ) . in this control region ,
the first observer found @xmath16 loops , the second one @xmath17 .
@xmath18 of them were found by both observers . from those values ,
the actual number of loops @xmath19 on the area may be estimated by the laplace ratio ( or lincoln - petersen index ) for population estimates : @xmath20 ( e.g. , * ? ? ?
* ) ; a better , less biased , estimate being @xmath21 @xcite .
the variance of this estimate is @xmath22=4.9 $ ] @xcite .
these formulae have been used in the literature to estimate the size of ecological populations ( e.g. , * ? ? ?
* ; * ? ? ?
* ) , and the number of errors in a message @xcite.the main difficulty for applying the method to our case is that the objects to be counted ( the loops as defined in the first paragraph of this section ) may not be unambiguous . to guarantee that this condition is fulfilled ( i.e. , that both observers were identifying and counting the same objects )
, both observers went again over all the structures found in the area by both , and they had to agree over all the events to be counted as loops .
from this analysis we conclude that the total number of magnetic structures that we find in the whole dataset could be understimated by @xmath335% .
we note that the most clear cases ( those with the strongest polarization signals , lasting longer , and clearly isolated from neighbouring magnetic patches ) were often found by both observers in the control area .
the 35% discrepancy is due to those events found by one observer but not the other ; these cases correspond to the most subtle ( often bordering the detection limit ) events .
we identify 497 small magnetic loops emerging in the observed regions .
taking into account the total time of observation and the spatial area covered , this amounts to an emergence rate of small dipoles of 0.25 loops h@xmath4 arcsec@xmath5 , rising up to 0.34 loops h@xmath4 arcsec@xmath5 when we apply the correction for undetected ( although present in the data ) loops given above .
these values are one order of magnitude larger than the previous estimation found by @xcite , but compatible with mart ' inez gonzlez ( 2007 ) and mart ' inez gonzlez et al .
there are several reasons for the discrepancy .
first , the former study covered a relatively small area and , if the emergence is not strictly uniform ( as we shall discuss below ) , the rate can be greatly understimated .
this fact was already pointed out in @xcite , where it was noted that there was evidence for preferential emergence regions .
the reason why the studies in the near - ir and this letter are in agreement is because , either having a better zeeman sensitivity ( mart ' inez gonzlez et al .
( 2007 ) and mart ' inez gonzlez et al .
( 2010 ) used the more sensitive most importantly to linear polarization spectral line of fe i at 1.5 @xmath23 m ) or having a better spatial resolution makes both studies more sensitive , inducing the identification of more ( weaker and/or smaller ) structures in the field .
both opposite polarity feet and the linearly polarized bridge between them were tracked during the full loop phase . in most cases ( 60% of the events ) , the loop collides or merges to some degree with a neighbouring structure and we could not trace further its individual evolution . this is a common case because at our level of sensitivity and at such spatial resolution , most of the pixels in the fov show circular polarization , and many , both linear and circular . for the remaining 40% of the cases
, the loop individual history could be traced beyond ( and before ) their complete loop phase ( i.e. , when both three polarization patches are seen simultaneously ) . in nearly half of these cases ( 56% ) , both footpoints and linear polarization appear and/or disappear simultaneously because the structure falls below the detection level of the instrument ( or it submerges ) .
we named this population of loops as `` low - lying '' ( see * ? ? ? * ) .
this population of loops is very particular of the quiet sun . in 42% of the cases ,
linear polarization precedes the detection of both footpoints and then disappears before them too .
@xcite computed the line - of - sight velocity of the loops using the stokes @xmath13 zero - crossing shift and verified that all the loops having this very same time evolution were rising @xmath24-loops .
we have no reasons for thinking that the loops found in imax are a different population .
however , we can not compute the stokes @xmath13 velocity in the imax data sets analyzed in this paper , hence , the identification of these loops as rising @xmath24-loops can be put in doubt . in just 3 instances ( 2% ) , we found that first the opposite polarities appear , then the linear polarization between them , and everthing disappears , wich could be interpreted as the emergence of a u - loop , or , more probably ( considering the local evolution of the flow ) , a submergence of an @xmath24-loop .
figure [ ejemplos ] display four examples of small loops found in the imax data .
note that , for the sake of clarity , we have only drawn the contours of interest , avoiding circular and linear polarization patches adjacent to the loop structure .
the time runs from top to bottom , the time cadence being irregular .
the first and third columns represent typical loops in which the linear polarization disappears at some point in the evolution of the loop while the footpoints stay in the photosphere ( probably rising @xmath24-loops ) .
these two loops have , however , some peculiarities that differentiate them .
the loop in the first column appears at the border of an expanding granule .
as the granule expands , the entire loop is dragged by the plasma flow .
the loop in the third column contains a linear polarization signal with a gap in between .
these two peculiarities are a consequence of the spatial resolution of the imax data that allows us to trace the dynamics of the linear polarization .
the loop in the second column of fig .
[ ejemplos ] has another linear polarization patch with a curious dynamics .
it seems that , somehow , the negative footpoint is disconected from the positive one ( i.e. , the loop breaks ) and that it connects somewhere else to the right of the positive footpoint .
of course , it is just a visual impression .
the example in the fourth column is an example of two loops appearing very close in time and in space .
they also disappear more or less at the same time , hence , one could think this is an evidence of a sea - serpent magnetic field line .
note also that the rightmost loop rotates .
all the observed dipoles are smaller than @xmath25 mm ( center - to - center distance between the two opposite polarity patches ) , becoming increasingly more abundant at smaller scales , with most of the observed dipolar structures being @xmath26 mm ( figure 2a ) .
this is barely three times the spatial resolution limit of imax , which suggests that the detection and our statistics might be limited by the instrument .
the tilt angle of these dipoles is uniformly distributed , meaning that it does not follow the hale s polarity law .
this last result is consistent with @xcite and even with the behaviour of the smallest ephemeral regions .
figure 2b displays the distance between footpoints with respect to the separation between them at the initial time .
therefore , positive values mean that the footpoints separate with time and negative ones indicate that the footpoints approach each other . in average
, the distance grows linearly with time with a velocity of @xmath27 km s@xmath4 , comparable to typical granular values .
this indicates that the loops passively follow the granular flows , as expected from weak magnetic features @xcite .
the magnetic properties of these small dipoles are represented in fig . 2c and 2d .
they are obtaining inverting the data in the weak field approximation following @xcite .
the magnetic flux has been computed in the area containing the observed signal .
the frontier has been defined by eye ( as an isocontour of magnetic flux density ) and hence is slightly different for the different structures .
the magnetic flux density is the mean value of the magnetic flux densities in this same region .
the magnetic flux of the loops can be explained with a power law using the exponent found by @xcite .
this means that the population of small dipoles follows the population of magnetic fields in the quiet sun . but
looking at the histogram of magnetic flux densities , the loops are located at the end tail of the histogram ; it is mostly in the range 10 - 20 mx cm@xmath4 i.e .
, 10 - 20 g if the magnetic field were uniformly distributed and volume filling .
this is compatible with the results of @xcite who state that , in the quiet sun , the larger the signal the larger the degree of organization of magnetic fields .
although small scale loops are found all over the observed areas , their spatial distribution does not seem to be completely uniform ( fig .
it can be observed that , at some locations , loops appear repeatedly and succesively as in bursts , forming clusters , a behaviour that has been noticed before by mart ' inez gonzlez & bellot rubio ( 2009 ) , who pointed out that , often , the appearence of loops made it more likely that new ones were later detected nearby . on the other hand ,
extended areas seem to be noticeably empty of such events , as if voids appeared in the distribution .
however , this may be deceiving since voids are also formed even in strictly uniform distributions of points @xcite .
a quantitative analysis to determine if these voids are statistically significant was performed .
this requires on the first place , an unambiguous definition of ` void ' , a non - trivial task in itself ( see e.g. , * ? ? ?
* ; * ? ? ?
* and discussions therein ) .
we adopted the simplest definition here and considered only voids of circular shape : the largest empty circle that can be fitted in a given region of the point field equivalently , an empty circle limited by three points of the distribution .
if the loops appeared uniformly on the solar surface , then the probability of finding a void of area between @xmath29 and @xmath30 within the field of view ( square surface of area @xmath31 ) would be ( see appendix ) : @xmath32 where @xmath33 , @xmath34 is the surface density of points , and @xmath35 is the total number of voids espected in the fov , which is given by equation ( a1 ) .
for the two data sets studied here @xmath36 mm ( which is slightly smaller than the nominal fov because we have excluded the apodized exterior area ) , and @xmath37 mm@xmath5 . note that we only need a single value of @xmath34 since the number of loops detected in both data sets are very similar ( i.e. , 248 and 249 events ) .
figure 4 shows the number of voids per unit area in both datasets and for the corresponding poisson distribution . except for the smallest areas ,
the distribution of large circular voids in the first dataset is not significantly different from the uniform one . on the second data
set , however , apart from the overabundance of small scale voids , large circular voids seem to be significatively more abundant than a strictly uniform distribution would suggest .
actually , for the parameters found for these observations , the probability of finding a circle with an area lager than @xmath38 mm@xmath1 ( equivalently , diameter larger than @xmath39 mm ) is very low ( @xmath40 ; see the inset plot in fig .
we conclude , therefore , that there is statistical evidence for the two voids marked in figure 3a to be real and not due to chance .
moreover , the oberabundance of small voids is interpreted in terms of a clumpy structuring ( see how loops appear in clumps , like a gurgle phenomena , in fig 3 ) . in order to relate the distribution of loops ( and the voids ) to the global magnetism in the observed area , fig .
3 shows the integrated longitudinal ( in black and white ) and transverse ( in yellow ) magnetograms for the two observations .
when dealing with the linear polarization , one has to remind that it is a biased estimator of the transversal field component . in the plot ,
this bias is statistically partially removed as follows : we compute the bias for a percentile 95 when the observations are pure noise ( * ? ? ?
* see ) ( note that this bias value depend on each pixel , i.e. , on the actual intensity profile ) .
this means that the `` real '' transverse magnetic will be below this bias value with a probability of 95% .
we have decided to put all the values smaller than this bias to 0 .
figure 3 shows only the statistically significant patches of linear polarization appearing at all the observed times .
the positions of the loops do not seem to be clearly correlated with the longitudinal magnetogram , but the voids encircled by the loops show less magnetic activity than other areas in the fov . considering the correlation with linear polarization , it suggests that most of the linear polarization signals that are detected are associated to loop structures embeded in the formation region .
it is known that even in very quiet areas of the sun magnetic fields may organize naturally forming loops at granular scales . in this study
we extended this observation to the smallest spatial scales observable ( 100 - 1000 km ) , finding an increasing number of loops at smaller scales up to the resolution limit .
this finding suggests that the organization of magnetic fields might continue beyond that limit .
we can not reconstruct the complete magnetic field topology because 1 ) the finite spatial resolution of our observations is ( perhaps inherently ) above the organization scale of the magnetic fields , and 2 ) we lack linear polarimetric sensitivity , which gives us only fragmentary information on the transversal ( horizontal ) component of the magnetic fields . due to these limitations the loop structures that we observe are biased towards relatively large and relatively strong with respect to the magnetic flux density in the neighbouring areas .
we found evidence that the loops thus detected are not randomly distributed on the solar surface , but rather that they may appear in bursts , and that they are noticiable absent from extended areas which are , also , only weakly magnetized .
it is not yet clear what is the nature of the magnetic fields in the quiet sun what are the fundamental physical mechanisms involved in their generation and evolution .
the presence of these _ dead calm _ areas in the quiet sun ( and small scale loops hotspots ) represent an important constrain on the origin of magnetic fields in the very quiet sun and on the dominant dynamic and magnetic mechanisms taking place there .
it is thought that the magnetism of the quiet sun can be the result of the emergence of underlying organized magnetic fields @xcite or the dragging of the overlying canopy fields @xcite .
it would then be necessary to understand why there are emergence hotspots and dead calm areas .
another possibility is that they are just recycling of the decay of active regions as they diffuse and migrate to the poles .
but it seems unlikely that such random walking would lead to the kind of organized structures and to the spatial patterns reported here .
it is also possible that they are linked to some type of dynamo action taking place in the solar surface @xcite .
it seems now clear that coherent velocity patterns are a requisite for dynamo action to take place @xcite .
the most obvious coherent velocity pattern in the solar surface is granulation .
if this velocity pattern is involved in some dynamo action , it is reasonable that it forms coherent magnetic structures ( such as the loops ) , although these does not need to be organized at the same granular scales ; it could well be that they form intermittent patterns as the ones observed here . actually , theoretical considerations @xcite and laboratory experiments @xcite support the idea that the onset of turbulent dynamo action may be highly intermittent and bursty .
finally , it could just be that these small scale loops represent the far tail of a continous range of structures from a global dynamo , just lying at the other end from sunspots .
their spatial statistics would then reflect the velocity patterns on the last ( shallowest ) layers of magnetic field emergence .
future models that we construct to understand the generation of magnetic fields in the very quiet sun have to explain the spatio - temporal coherences that we report .
further work is needed to extend these results in larger areas of the sun and along the solar cycle .
we specially thank dr .
juan betancort rijo for very helpful discussions on the determination of the statistics of voids , which have improved the paper and strengthen the conclusions
. we also thank valent ' in mart ' inez pillet for helpful discussions .
eliot hijano enjoyed a summer research grant at the instituto de astrof ' isica de canarias .
financial support by the spanish ministry of economy and competitiveness through projects aya2010 - 18029 ( solar magnetism and astrophysical spectropolarimetry ) and consolider - ingenio 2010 csd2009 - 00038 is gratefully acknowledged .
we derive equation ( 1 ) adapting to our case some arguments of the strategy by @xcite .
the probability that three points @xmath41 distributed at random ( uniformly ) on an square with area @xmath42 , have coordinates between @xmath43 and @xmath44 , @xmath45 and @xmath46 ( @xmath47 and @xmath48 ) , is @xmath49 .
with the change of variables @xmath50 , @xmath51 , we may express the probability for the center of circumcircle of the three points to lie between @xmath52 and @xmath53 , @xmath54 and @xmath55 , its radius between @xmath56 and @xmath57 , and the azimutal angles of each point between @xmath58 and @xmath59 , then , as @xmath60 , where @xmath61 is the determinant of the jacobian matrix .
the probability of one such a circle to lie between the bounds of the large square and its radius to be between @xmath56 and @xmath57 is @xmath62 , where @xmath52 and @xmath54 have been integrated betwen @xmath56 and @xmath63 , and @xmath58 between 0 and 2@xmath64 , taking into account that the integral of the angular part of @xmath65 is @xmath66 .
alternatively , the probability density of the area @xmath33 of such a circle is @xmath67 . on the other hand , in a homogeneous poisson field with density @xmath68
( @xmath69 points randomly distributed over an area @xmath70 ) , the probability of finding @xmath71 points in a region of area @xmath29 is @xmath72 .
therefore , the probability of the circumcircle of three points to be void is @xmath73 .
finally , @xmath69 points determine @xmath74 ( @xmath75 ) different triplets ( hence , possible circles ) , and the number of void circles with areas between @xmath29 and @xmath30 within the square bounds is @xmath76 .
the total number of void circles is : @xmath77,\ ] ] where @xmath78 and @xmath79 is the error function @xcite .
the probability of voids of area @xmath29 with our constrains is then the number of voids of area @xmath29 divided by the total number : @xmath80 we specially thank dr .
juan betancort rijo for very helpful discussions on the determination o f the statistics of voids , which have improved the paper and strengthen the conclusions .
the c ontribution of eliot hijano to this work was in the frame of the summer grants offered by the instituto de astrof ' isica de canarias .
financial support by the spanish ministry of economy and competitiveness through projects aya2010 - 18029 ( solar magnetism and astrophysical spectrop olarimetry ) and consolider - ingenio 2010 csd2009 - 00038 is gratefully acknowledged . , j. m. , pearce , f. , foster , c. , platen , e. , brunino , r. , neyrinck , m. , basilakos , s. , fairall , a. , feldman , h. , gottlber , s. , hahn , o. , hoyle , f. , mller , v. , nelson , l. , plionis , m. , porciani , c. , shandarin , s. , vogeley , m. s. , & van de weygaert , r. 2008 , , 387 , 933 , s. , beeck , b. , pietarila , a. , schssler , m. , solanki , s. k. , martnez pillet , v. , bonet , j. a. , del toro iniesta , j. c. , domingo , v. , barthol , p. , berkefeld , t. , gandorfer , a. , knlker , m. , schmidt , w. , & title , a. m. 2010 , , 723 , l149 , r. , & tsuneta , s. 2009 , in astronomical society of the pacific conference series , vol .
415 , the second hinode science meeting : beyond discovery - toward understanding , ed .
b. lites , m. cheung , t. magara , j. mariska , & k. reeves , 132 , v. , del toro iniesta , j. c. , lvarez - herrero , a. , domingo , v. , bonet , j. a. , gonzlez fernndez , l. , lpez jimnez , a. , pastor , c. , gasent blesa , j. l. , mellado , p. , piqueras , j. , aparicio , b. , balaguer , m. , ballesteros , e. , belenguer , t. , bellot rubio , l. r. , berkefeld , t. , collados , m. , deutsch , w. , feller , a. , girela , f. , grauf , b. , heredero , r. l. , herranz , m. , jernimo , j. m. , laguna , h. , meller , r. , menndez , m. , morales , r. , orozco surez , d. , ramos , g. , reina , m. , ramos , j. l. , rodrguez , p. , snchez , a. , uribe - patarroyo , n. , barthol , p. , gandorfer , a. , knoelker , m. , schmidt , w. , solanki , s. k. , & vargas domnguez , s. 2011 , , 268 , 57 , f. , berhanu , m. , monchaux , r. , aumatre , s. , chiffaudel , a. , daviaud , f. , dubrulle , b. , bourgoin , m. , odier , p. , plihon , n. , pinton , j .- f . ,
volk , r. , fauve , s. , mordant , n. , & ptrlis , f. 2008 , physical review letters , 101 , 074502 | we analyze two regions of the quiet sun ( @xmath0 mm@xmath1 ) observed at high spatial resolution ( @xmath2100 km ) in polarized light by the imax spectropolarimeter onboard the sunrise balloon .
we identify 497 small - scale ( @xmath3400 km ) magnetic loops , appearing at an effective rate of 0.25 loop h@xmath4 arcsec@xmath5 ; further , we argue that this number and rate are underestimated by @xmath330% . however , we find that these small dipoles do not appear uniformly on the solar surface : their spatial distribution is rather filamentary and clumpy , creating _
dead calm _ areas , characterized by a very low magnetic signal and a lack of organized loop - like structures at the detection level of our instruments , that can not be explained as just statistical fluctuations of a poisson spatial process .
we argue that this is an intrinsic characteristic of the mechanism that generates the magnetic fields in the very quiet sun .
the spatio - temporal coherences and the clumpy structure of the phenomenon suggest a recurrent , intermittent mechanism for the generation of magnetic fields in the quietest areas of the sun . |
a lot of observational information is now known about the structure of dense cores as progenitors of stars within the molecular clouds ( e.g. , evans 2010 ) . theoretically , we expect that most of these cores will collapse into protostars , but the details of evolutions are less evident .
clearly , the evolution heavily depends upon the effects of local density , pressure , rotation , magnetic fields , the presence or absence of nearby stars and protostars , and upon the other physical phenomena .
the most commonly used picture of low - mass star formation is inside - out collapse ( shu 1977 ) in which it begins at the center of a singular isothermal sphere and an expanding wave of infalling matter propagates outward with the speed of sound . up to now
, different models for core collapse and formation of protostars have been presented , in each the effects of some physical phenomena investigated ( e.g. , mckee and ostriker 2007 ) .
the velocity maps of molecular cloud cores , which are built by using the velocity measurements along the line of sight , show a small velocity gradient across each core . measuring this global velocity gradient , with assumption that core has a uniform rotation and
follow a rigid - body rotation law , can be used to deduce the core angular velocity ( e.g. , goodman et al .
the large size of initial collapsing core implies that even the modest initial rotational velocities will cause the infalling matters to land first on a rotationally supported disk rather than a pressure - supported proto - star ( e.g. , hartmann 2009 , nejad - asghar 2011 ) . in the simplest analysis of rotating collapse , we assume that pressure forces are negligible and so the problem can be analyzed by using ballistic trajectories . the results for collapse of a spherically symmetric cloud in uniform ( solid - body ) rotation
were initially worked out by ulrich ( 1976 ) , with subsequent extension to disk formation by cassen and moosman ( 1981 ) , and to collapsing singular isothermal sphere by terebey , shu and cassen ( 1984 ) .
mendoza , tejeda and nagel ( 2009 , hereafter mtn ) have recently generalized this idea by construction a steady analytic accretion flow model for a finite rotating gas cloud .
they show that the streamlines and density profiles deviate considerably from ones calculated by ulrich , and for all relevant astrophysical situations , the assumption of a finite cloud radius needs to be used .
the observations indicate that magnetic energy in the molecular clouds is comparable to the gravitational energy ( e.g. , crutcher 1999 ) .
moreover , the magnetic fields are also theoretically believed to play an important role in gravitational collapse of the molecular cloud cores .
they provide pressure support against the gravity and carries away angular momentum prior to and during the collapse of cores to form accretion disks , jets and protostars ( e.g. , machida 2010 ) .
although , in a molecular cloud core , the spatial configuration of magnetic field lines is not simple ( e.g. , whittet 2005 ) , but polarimetry observations of young stellar objects suggest that circumstellar thin disks around the young stars have approximately aligned perpendicular to the magnetic fields ( e.g. , pereyra et al . 2009 ) . here , we consider a simple initial configuration in which the magnetic fields are assumed to be parallel with rotational axis of core .
since the molecular cloud cores are lightly - ionized , the ambipolar diffusion in which the magnetic field is frozen into the charged species and drifts along with them through neutrals , is an important mechanism ( e.g. , adams 2009 ) .
thus , we consider the effect of magnetic fields directly on charged particles , while the neutral species feel them indirectly via the collisions with ions . in this way , as an extension to the work of mtn , we investigate the effect of axial magnetic field on streamlines of infalling ions and neutral particles .
for this purpose , formulation of problem and equations of motion are given in section 2 .
in section 3 , trajectories of infalling ions and neutral particles are investigated and effect of the magnetic fields on the accretion disk radius is studied .
finally , section 4 is allocated to summary and conclusions .
the molecular cloud cores are lightly - ionized gases with small fraction of ions with density @xmath0 and electrons with density @xmath1 .
if the volumetric rate of electron - ions recombination took place in gas phase , it would be proportional to @xmath2 . in a simple steady state approximation
, the volumetric rate of electron - ions recombination @xmath3 would be equal to the volumetric rate of neutrals ionization via cosmic rays @xmath4 .
although , the actual ionization balance in the molecular clouds is more complicated , elmegreen ( 1979 ) and umebayashi and nakano ( 1980 ) showed that the relation @xmath5 is a good approximation . here , we suppose for fiducial purpose that @xmath6 , where @xmath7 and @xmath8 is a dimensionless free parameter which represents the deviations from calculations in the steady state approximation . in the magnetized cloud cores , the mean velocity @xmath9 of neutrals will not generally equal to the mean velocity @xmath10 of ions and @xmath11 of electrons .
although , the positive ions and electrons feel the magnetic forces in opposite direction , but the induced electric field will generally cause they move in ambipolar motion , so that in the time - scales that are considered here , their mean velocities are approximately the same ( @xmath12 ) . since electrons carry much less momentum than ions , we neglect the dynamic of electrons . in this way , resistance of relative drift between ions and neutrals will be a drag acceleration arises from mutual collisions between them .
the drag acceleration exerted on neutrals by ions is @xmath13 ; the drag acceleration exerted on ions by neutrals is @xmath14 , where @xmath15 ( draine , roberge and dalgarno 1983 ) is the drag coefficient and @xmath16 is a dimensionless free parameter which represents the deviations from calculations in the steady state approximation .
the equations of motion for ions and neutral particles , in gravitational field of a central mass , are respectively as follows @xmath17 @xmath18 where @xmath19 is the proto - stellar mass at the origin , and @xmath20 and @xmath21 are ion charge and mass , respectively , as an integer multiple of the electron charge @xmath22 and the proton mass @xmath23 . here
, we assume that @xmath24 .
we measure the length and time in units of @xmath25 and @xmath26 , respectively , so that velocity unit is @xmath27 .
mass unit is appointed to @xmath28 , thus the gravitational constant is @xmath29 . in this manner
, we have @xmath30 and @xmath31 .
finally , the unit of magnetic field is chosen equal to @xmath32 . using these units ,
the equations of motion ( [ equmotioni ] ) and ( [ equmotionn ] ) can be rewritten in a dimensionless form as follows @xmath33 @xmath34 where @xmath35 is a unit vector in the direction of magnetic field , and @xmath36 and @xmath37 are two dimensionless free parameters .
choosing the axial magnetic field @xmath38 , the equations of motion in cartesian coordinate are @xmath39 @xmath40 @xmath41 @xmath42 @xmath43 @xmath44 where @xmath45 , @xmath46 , @xmath47 , and @xmath48 are the polar and azimuthal angles of ions and neutral particles , respectively , and @xmath49 and @xmath50 are their radial distances .
there are four free parameters @xmath37 , @xmath8 , @xmath16 , and @xmath51 which represent the strength of magnetic field , ionization fraction , drag coefficient , and neutral density , respectively .
as mentioned before , there are many observations that show velocity gradients in the maps of molecular cloud cores .
these velocity gradients can be used to deduce the rotation of cores , but evaluation of their angular velocity @xmath52 is considerably complicated because the velocity fields usually exhibit complex supersonic motions . here , we choose the fiducial value of @xmath53 , which in the time unit @xmath26 is @xmath54 .
the effects of initial angular momentum in core collapse and formation of accretion disks are now approximately well known ( e.g. , hartmann 2009 ) . here
, we turn our attention to the effects of magnetic fields and drag force on infalling matters .
if we assume that @xmath55 and @xmath56 , equations ( [ xnddot])-([ziddot ] ) lead to the trajectories which are contained on a plane and given by a _ conic section _
( i.e. , the results obtained by mtn ) . choosing the initial conditions @xmath57 @xmath58 for position and velocity of infalling particles
, the equations of motion ( [ xiddot])-([znddot ] ) can be solved by numerical methods such as runge - kutta .
approximately , all cores in the maps of molecular clouds seem apparently to be elongated rather than spherical ( e.g. , curtis and richer 2010 ) . since determining the exact three - dimensional shape of a core from apparent observations of the - plane - of - sky is impossible , statistical techniques have to be applied ; as the results , some works indicate a preference for prolate cores ( e.g. , gammie et al .
2003 , li et al .
2004 ) , and some others favour oblate shapes ( e.g. , curry 2002 , jones and basu 2002 ) .
theoretically , the prolate elongation of the cores may be inferred as a remnant of their origin in filaments and turbulent flows ( e.g. , hartmann 2002 ) , while the oblate shape of collapsing cores may be justified from the effects of magnetic fields and rotational motion ( e.g. , nejad - asghar 2010 ) .
the streamlines of infalling particles , in absence of magnetic field ( @xmath56 , @xmath55 ) and initial radial velocity ( @xmath59 ) , for initial spherical core ( @xmath60 ) , oblate core ( @xmath61 ) , and prolate core ( @xmath62 ) are shown in fig.[zrso ] .
the numerical plots of accretion disk radius for neutral particles by substitution @xmath63 , for a wide variety of initial parameters , are shown in fig .
[ rdisk ] .
the numerical results show that the accretion disk radius is approximately proportional to the inverse of initial radial velocity @xmath64 as was given by the analytical ballistic models of mtn .
since we want to investigate the effect of magnetic fields on trajectories of infalling matters , without losing the generality of problem , we consider an initial spherical core with assumption @xmath59 . here
we turn our attention to only one streamline with @xmath60 and @xmath65 .
the effects of axial magnetic field on this streamline are shown in fig.[zrdiffmag ] for three different values of free parameter @xmath37 with a fixed value of three parameters @xmath16 , @xmath8 and @xmath51 equal to @xmath66 , @xmath67 , and @xmath68 , respectively . as a result
, we see that the axial magnetic field causes the infalling ions land at a larger equatorial radius , and collisions of ions with neutral particles lead to the same behavior for infalling neutral matters . in fig.[rtogameps ] , the equatorial radii of ion and neutral infalling particles ( from @xmath65 ) are depicted versus @xmath16 and @xmath8 at a fixed value of magnetic field .
we see that increasing of drag coefficient @xmath16 leads to coupling of ions and neutrals so that their equatorial radius reach to each other asymptotically .
the same behavior can be seen for the effect of ionization degree @xmath8 so that increasing of this parameter asymptotically causes to approximately the same equatorial radii of ions and neutral infalling particles .
the third free parameter is neutral density @xmath51 which its effect is the same as @xmath16 but its effect on neutral particles is less .
the density of neutral particles can be evaluated by assuming that the mass infall rate , @xmath69 , is steady .
assuming that the cloud at @xmath70 is nearly spherical , the mass flow in a flow tube spanned by @xmath71 is @xmath72 the density at @xmath73 can be found by following the streamlines corresponding to @xmath74 and @xmath75 , @xmath76 in order to obtain the iso - density contours , we divide @xmath77 to @xmath78 equal segments so that @xmath79 . for two streamlines in the borders of each segment ,
values of @xmath80 , @xmath81 and @xmath82 are evaluated in the mid points . in this way
, the density at different radii and polar angles can be found from ( [ dens2 ] ) , and the iso - density curves can be depicted as typically are shown in fig .
[ isoden ] . in this figure ,
the mass accretion rate is assumed to be @xmath83 .
these iso - density contours show that the axial magnetic fields cause the infalling matters land at a larger equatorial radius .
nowadays , we have accepted that the molecular cloud cores can be gravitationally unstable under certain conditions , and they can collapse . rotational motion of the cores leads to land the infalling matters on an accretion disk around the proto - star .
the streamlines of infalling matters for initial spherical , oblate and prolate cores are depicted in fig.[zrso ] .
obviously , the centrifugal force causes to increase the initial dimension of accretion disk in oblate cores , while it decreases the disk radius in prolate ones .
next , we considered an axil magnetic field in the collapsing core . for investigation the effect of this magnetic field
, we wrote separately the equations of motion of neutral and ion infalling particles .
in general form , these equations are presented with four free parameters @xmath37 , @xmath8 , @xmath16 , and @xmath51 which represent the strength of magnetic field , ionization fraction , drag coefficient , and neutral density , respectively .
the accretion disk radius for different values of the relevant input parameters are depicted in fig.[rdisk ] . since
our goal is to study the effect of magnetic field and drag coefficient on infalling matters and disk formation , we turned our attention to only one streamline at @xmath65 ; the results are shown in fig.[zrdiffmag ] . in a steady state of molecular gas , the ions are well - coupled to neutral particles ( i.e. , @xmath84 , @xmath85 ) so that their dynamics are approximately the same . fig.[rtogameps ] shows that increasing of drag coefficient and ionization degree lead to the same behavior for ions and neutral infalling particles . in infalling case
, the frequency of collisions between ions and neutral particles eventually becomes so low that the balance can not be maintained , thus , the values of free parameters @xmath8 and @xmath16 deviate from the steady sate . in this non - equilibrium case
, ions are directly affected by magnetic field so that they land in a larger radii at equatorial plane . in the low - coupling ( @xmath86 ) and high - ionization degree ( @xmath87 )
, the dynamics of ions can affect the motions of neutral particles so that they also land in a larger radii to form the larger accretion disk as shown in fig.[zrdiffmag ] .
the same behavior for increasing of accretion disk radius via the effect of magnetic field can be deduced from iso - density contours that are depicted in fig .
[ isoden ] .
this work has been supported by grant of research and technology deputy of university of mazandaran . | the magnetic fields affect collapse of molecular cloud cores . here , we consider a collapsing core with an axial magnetic field and investigate its effect on infall of matter and formation of accretion disk .
for this purpose , the equations of motion of ions and neutral infalling particles are numerically solved to obtain the streamlines of trajectories .
the results show that in non - steady state of ionization and ion - neutral coupling , which is not unexpected in the case of infall , the radius of accretion disk will be larger as a consequence of axial magnetic field . |
we consider the convex minimization problem with linear constraints and a separable objective function @xmath1 where @xmath2 and @xmath3 are continuous closed convex ( could be nonsmooth ) functions ; @xmath4 and @xmath5 are given matrices ; @xmath6 is a given vector ; @xmath7 and @xmath8 are nonempty closed convex subsets of @xmath9 and @xmath10 , respectively . throughout
, the solution set of ( [ cp ] ) is assumed to be nonempty ; and @xmath7 and @xmath8 are assumed to be simple in the sense that it is easy to compute the projections under the euclidean norm onto them ( e.g. , positive orthant , spheroidal or box areas ) .
let @xmath11 be the augmented lagrangian function for that defined by @xmath12 in which @xmath13 is the multiplier associated to the linear constraint and @xmath14 is a penalty parameter .
based on the classic douglas - rachford operator splitting method @xcite , the alternating direction method of multipliers was proposed by gabay and mercier @xcite , glowinski and marrocco @xcite in the mid-1970s , which generates the iterative sequence via the following recursion : [ alx ] x^k+1=_x_(x , y^k,^k ) , + [ aly]y^k+1=_y_(x^k+1,y,^k ) , + [ all]^k+1=^k-(ax^k+1+by^k+1-b ) . based on another classic operator splitting method , i.e. , peaceman - rachford operator splitting method @xcite
, one can derive the following method for : [ alxp ] x^k+1=_x_(x , y^k,^k ) , + [ allb]^k+=^k-(ax^k+1+by^k+1-b ) , + [ alyp]y^k+1=_y_(x^k+1,y,^k+ ) , + [ allp]^k+1=^k+-(ax^k+1+by^k+1-b ) .
while the global convergence of the alternating direction method of multipliers - can be established under very mild conditions @xcite , the convergence of the peaceman - rachford - based method - can not be guaranteed without further conditions @xcite .
most recently , he et al .
@xcite propose a modification of - by introducing a parameter @xmath15 to the update scheme of the dual variable @xmath16 in and , yielding the following procedure : [ he1 ] x^k+1=_x_(x , y^k,^k ) , + [ he2]^k+=^k-(ax^k+1+by^k+1-b ) , + [ he3]y^k+1=_y_(x^k+1,y,^k+ ) , + [ he4]^k+1=^k+-(ax^k+1+by^k+1-b ) .
note that when @xmath17 , - is exactly the same as - .
they explained the nonconvergence behavior of - from the contract perspective , i.e. , the distance from the iterative point to the solution set is merely nonexpansive , but not contractive .
the parameter @xmath15 in - plays the essential role in forcing the strict contractiveness of the generated sequence . under the condition that @xmath18 , they proved the same sublinear convergence rate as that for admm @xcite .
particularly , they showed that - achieves an approximate solution of with the accuracy of @xmath0 after @xmath19 iterations convergence rate means the accuracy to a solution under certain criteria is of the order @xmath0 after @xmath19 iterations of an iterative scheme ; or equivalently , it requires at most @xmath20 iterations to achieve an approximate solution with an accuracy of @xmath21 . ] , both in the ergodic sense and the nonergodic sense .
note that the parameter @xmath15 plays different roles in and : the former only affects the update of the variable @xmath22 in while the latter is for the update of the dual variable @xmath16 .
hence , it is natural to choose different parameters in these two equalities . in this paper , we give such a scheme by introducing a new parameter @xmath23 in , i.e. , the dual variable is updated by the following manner : @xmath24 for convenience , we introduce the whole update scheme of the _ modified strictly contractive semi - proximal peaceman - rachford splitting method _ ( sp - prsm ) as [ equ : sp - prsm ] [ equ : sp - prsm1 ] x^k+1=_x
_(x , y^k,^k ) + 12 x - x^k_^2 , + [ equ : sp - prsm2 ] ^k+=^k-(ax^k+1+by^k - b ) , + [ equ : sp - prsm3 ] y^k+1=_y _(x^k+1,y,^k+ ) + 12 y - y^k_^2 , + [ equ : sp - prsm4 ] ^k+1=^k+-(ax^k+1+by^k+1-b ) where @xmath25 and @xmath26 are two positive semi - definite matrices . in applications , by choosing different matrices @xmath25 and @xmath26 customizing the problems structures , we can obtain different efficient methods .
our main contributions are 1 .
motivated by the nice analysis techniques in @xcite and @xcite , we proved that the sequence generated by sp - prsm is strictly contractive and thus convergent , under the requirement that @xmath27 moreover , we proved that sp - prsm is @xmath0 sublinearly convergent both in the ergodic and nonergodic sense .
note that the nonergodic convergence rate requires that @xmath28 $ ] .
we remark that the convergence of sp - prsm can unify that of several existing splitting methods . *
choosing @xmath29 and @xmath30 , @xmath31 , sp - prsm reduces to the classical admm while the convergence coincides with that of admm @xcite ; * setting @xmath32 and @xmath33 , sp - prsm covers the semi - proximal admm considered in @xcite and the corresponding convergence results ; * setting @xmath34 , and @xmath35 , @xmath31 , sp - prsm reduces to the strictly contractive prsm proposed in @xcite and the convergence of the two methods is identical .
we added a proximal term to each of the two main subproblems in updating the @xmath36 and the @xmath22 variables .
in fact , for admm , eckstein @xcite and he et al .
@xcite have already considered to add proximal terms to the subproblems for different purpose .
recently , fazel et al .
@xcite proposed to allow @xmath25 and @xmath26 to be positive semi - definite , in contrast to the positive definite requirements in the classical algorithms , which makes the algorithm more flexible .
the rest of this paper is organized as follows . in sect .
[ section : preliminaries ] , we give the optimality condition of by using the variational inequality and also list some assertions which will be used in later analysis . in sect .
[ section : convergence ] , we first give the contraction analysis of sp - prsm , and then in sect .
[ section : numerical ] , we test a variety of problems to show the efficiency of the proposed sp - prsm .
finally , we make some conclusions in sect .
[ section : conclusions ] . on may 25 ,
after attending the first author s thesis defense , prof .
he told us that they had also considered to use different stepsizes @xmath15 and @xmath23 in with @xmath30 and @xmath31 , and we got their manuscript on may 26 @xcite .
we found that they established the convergence when @xmath37 notice that covers the case when @xmath38 .
if we restrict @xmath39 in , then the corresponding relation will be @xmath40 by some simple calculations , it is easy to see that the domain defined by is a bit smaller than that defined by . by some private communications with prof .
he , we learned that they also discovered the formula .
however , to provide the intuitive understanding and the unified convergence analysis , they presented in @xcite which connects @xmath15 and @xmath41 directly .
in this section , we give the optimality condition of and some notations or relations which will be frequently used in our analysis .
let @xmath42 . throughout this paper ,
we make the following assumption . [
ass : opt ] let @xmath43 be the set whose elements are the optimal solutions of and the associating dual solutions of . throughout the paper ,
we assume that @xmath44 is non - empty . owing to the convexity of @xmath45 and @xmath46 ,
there exist two positive semidefinite matrices @xmath47 and @xmath48 such that for any @xmath49 @xmath50 where @xmath51 , @xmath52 , and for any @xmath53 @xmath54 where @xmath55 , @xmath56 . denote @xmath57 due to the convexity of @xmath58 and @xmath59 , it is easy to show that the operator @xmath60 is monotonic .
specifically , for any @xmath61 , we have @xmath62 where @xmath63 and the inequality is due to and .
as shown in @xcite , we say that @xmath64 if there holds that @xmath65 from theorem 2.3.5 in @xcite or theorem 2.1 in @xcite , we can see that @xmath44 is closed and convex , and it can be reformulated as @xmath66 let @xmath67 denote the feasible set of , namely , @xmath68 and denote @xmath69 .
let us restrict @xmath70 in and to be @xmath71 by some suitable modification of the proof of theorem 2.1 in @xcite , it is easy to show that the optimal solution set @xmath44 can be characterized as @xmath72 similarly , we say that @xmath64 if there holds that @xmath73 note that the optimality condition will be frequently used in the convergence analysis . owing to , we call @xmath74 to be an @xmath21 solution of if @xmath75 given two real matrices @xmath76 and @xmath77 of the same dimension , we define @xmath78 , where @xmath79 is the trace operator .
we use @xmath80 to denote the 2-norm of a vector .
we denote @xmath81 for @xmath82 and @xmath83 . for a real symmetric matrix @xmath84
, we mark @xmath85 ( @xmath86 ) if @xmath84 is positive semidefinite ( positive definite ) . to make the analysis more elegant , we use @xmath87 for short .
similarly , for any @xmath88 , we denote @xmath89 .
obviously , there holds that @xmath90 for any @xmath88 . for ease of the analysis ,
we define the following matrices as @xmath91 and @xmath92 we can easily verify that @xmath93 is positive semidefinite if @xmath94 and @xmath95 , and @xmath96 denote @xmath97 and define @xmath98 and @xmath99 where the dimension of zero matrix can be easily identified from the context . for any @xmath61 , there hold that @xmath100 and @xmath101 finally , we present some relations of the iterates . they will play a crucial role in the convergence analysis . with the update scheme and ,
it is easy to have @xmath102 recalling the definition of @xmath103 , with and , we conclude that @xmath104 with and , we thus have @xmath105
in this section , we first show that the sequence @xmath106 generated by sp - prsm is strictly contractive and thus can establish the convergence of the method . with the help of the contraction property , we further discuss the convergence rate in the ergodic and nonergodic sense .
we first give the contraction result as follows , and then complete its proof at the end of this section .
note that the following result will play a key role in proving the convergence of .
[ thm : wk : wk1:contraction ] let the sequence @xmath107 be generated by sp - prsm .
if we choose @xmath15 and @xmath23 according to , then for any @xmath108 , there holds that @xmath109 where @xmath110 , and @xmath111 and @xmath112 are some constants which only depend on @xmath15 and @xmath23 . to prove theorem [ thm : wk : wk1:contraction ]
, several lemmas should be established first .
the following lemma [ lemma : opt ] is mainly based on the optimality conditions of and .
[ lemma : opt ] let the sequence @xmath107 be generated by sp - prsm . if @xmath39 and @xmath113 , then for any @xmath88 , there holds that @xmath114 consider any @xmath115 and @xmath116 .
we then obtain respectively from the optimality conditions for and that @xmath117 and @xmath118 substituting and @xmath119 which follows from into the above two inequalities , respectively , we obtain the new formulations of the optimality conditions for and as @xmath120 and @xmath121 rewriting to be @xmath122 combing , and in a suitable way , and recalling the definition of @xmath123 , we see that for any @xmath70 there holds that @xmath124 \rrang + \llang w^{k+1 } - w , f(w^{k+1 } ) \rrang . \ ] ] with the relation and the definition of @xmath93 , by some easy calculations , we obtain that @xmath125 considering the definition of @xmath126 and @xmath127 , we have @xmath128 .
this , together with and indicates that @xmath129 notice that @xmath90 for any @xmath88 , we can immediately obtain from since the proof is completed .
[ lemma : vkvk1 ] let the sequence @xmath107 be generated by sp - prsm .
if @xmath39 and @xmath113 , then for any @xmath88 , there holds that @xmath130 using the identity @xmath131 with @xmath132 and @xmath133 , we have @xmath134 plugging into with and , we have . the proof is completed .
denote @xmath135 . with and
, the inequality can be expressed as @xmath136 our task now is to estimate the first term in the righthand side of the inequality in .
consider the case where @xmath137 and @xmath138 .
it is easy to see that @xmath139 .
moreover , by some tedious calculations , we conclude that @xmath140 where @xmath141 thus we obtain from and that @xmath142 which is sufficient to establish that the sequence @xmath143 is strictly contractive by choosing @xmath144 in and further to show the convergence of sp - prsm .
consider the case when @xmath145 but @xmath146 .
the matrix @xmath147 is not positive semidefinite any more . instead of establishing the strictly contractive property of the sequence @xmath143
, we consider to prove the property of another sequence related to the sequence @xmath143 . to do so , we need to give a more careful estimation of the intersection term @xmath148 .
[ lemma : axby - b ] let the sequence @xmath107 be generated by sp - prsm . if @xmath39 and @xmath113 , then for any @xmath70 , there holds that @xmath149 note that the optimality condition also holds with @xmath150 , then we have @xmath151 where @xmath152 . choosing @xmath22 to be @xmath153 and @xmath154 in and and then rearranging the obtained inequalities , respectively , we have that @xmath155 and @xmath156 summing and over the both sides and then using the relation , we obtain that @xmath157 where the second inequality is due to and the cauchy - schwarz inequality .
it is trivial to obtain from .
the proof is completed . with the help of lemma [ lemma : axby - b ]
, we are ready to have the following theorem
. let the sequence @xmath107 be generated by sp - prsm .
if we choose @xmath15 and @xmath23 according to , then for any @xmath88 , there holds that @xmath158 where @xmath159 and @xmath160 are some constants which only depend on @xmath15 and @xmath23 .
we consider three cases .
i ) . @xmath138 . with @xmath161 and @xmath162
, the inequality coincides with which has been proved . before we proceed , by combining and , we derive the useful inequality as @xmath163 ii ) .
it is not difficult to see from , and that holds with @xmath165 and @xmath166 .
@xmath167 . let us choose any @xmath168 .
note that this interval is well - defined due to the range of @xmath23 . with the cauchy - schwarz inequality , we know that @xmath169 plugging the above inequality into , we obtain that @xmath170 which with and implies that holds with @xmath171 , @xmath172 and @xmath173 the proof is completed .
finally , we end this subsection by finishing the proof of theorem [ thm : wk : wk1:contraction ] .
* proof of theorem [ thm : wk : wk1:contraction]*. consider any @xmath64 , recalling the monotonicity of @xmath123 , we thus obtain from the optimality of @xmath174 that @xmath175 with the above statement , choosing @xmath144 in , we can easily obtain .
@xmath176 adding the term @xmath177 on both sides of , we have @xmath178 which with and leads to . the proof is completed .
let the sequence @xmath107 be generated by sp - prsm .
if we choose @xmath15 and @xmath23 according to , then @xmath179 converges to an optimal solution of .
we divide the proof into three steps .
\(i ) to show that the sequences @xmath179 is bounded .
it is trivial to know from that @xmath180 , which with , and indicates that @xmath181 by some simple calculations , there holds that there exists a constant @xmath182 $ ] such that @xmath183 besides , it is straightforward to see from that @xmath184 is bounded .
this together with shows that @xmath185 are all bounded . obviously , we claim that @xmath186 is bounded .
moreover , we see that @xmath187 is bounded as @xmath188 .
notice that @xmath189 , with , this implies that @xmath190 is bounded .
recalling that @xmath191 and @xmath192 is bounded , it is safe to say that @xmath193 is also bounded . \(ii ) to show that any cluster point of the sequence @xmath107 is an optimal solution of .
let @xmath194 be a subsequence of the sequence @xmath107 and @xmath195 . since the graphs of @xmath196 and @xmath197 are both closed , taking the limit with respect @xmath198 on both sides of and using , we have that @xmath199 which means that @xmath200 is an optimal solution of .
\(iii ) to show that the sequence @xmath107 has only one cluster point .
we first replace @xmath201 with @xmath202 in the analysis of steps ( i ) and ( ii ) .
it follows from @xmath203 that @xmath204 .
owing to the monotonicity of the sequence @xmath205 , we can see that @xmath206 this together with shows that @xmath207 using again the inequality @xmath208 and @xmath209 , we have @xmath210 combing and , and using that @xmath211 and @xmath188 , we immediately have @xmath212 the proof is completed .
the rate of convergence of an algorithm can help us have a deeper understanding of the algorithm .
thus , in this section , we establish the sublinear rate of convergence of sp - prsm , in ergodic sense and nonergodic sense , respectively .
we now give the sublinear rate of convergence of sp - prsm in the ergodic sense , which is very easy due to the key inequality .
[ thm : ergodic ] let the sequence @xmath107 be generated by sp - prsm .
for any integer @xmath213 , define @xmath214 then for any @xmath88 , we have that @xmath215 it follows from and @xmath216 following from that @xmath217 for all @xmath218 , summing the above inequality from both sides and noting the notation of @xmath219 , we derive that @xmath220 since holds for any @xmath88 , we can easily have .
the proof is completed .
* remark * : consider the case when the set @xmath221 is compact , implies the @xmath0 ergodic convergence of sp - prsm .
[ lemma : decrease ] let the sequence @xmath107 be generated by sp - prsm . if we choose @xmath15 and @xmath23 according to , then there holds that @xmath222 note that also holds with @xmath223 , then we have @xmath224 choosing @xmath225 to be @xmath226 and @xmath227 , respectively , in and leads to @xmath228 and @xmath229 adding and and noting @xmath230 , following from , we obtain that @xmath231 \nn \\ \ge { } & ( 1 - \alpha - \gamma ) \beta \| r^{k+1 } - r^{k+2}\|^2 + ( 1 - \alpha)\beta \left \langle b\left[(y^k - y^{k+1 } ) - ( y^{k+2 } - y^{k+1})\right ] , r^{k+1 } - r^{k+2}\right\rangle .
\label{equ : lemma : opt : nonergodic : b3}\end{aligned}\ ] ] following the deriving process of , we have that @xmath232\right\|^2 + ( \alpha + \gamma ) \beta \|r^{k+1 } - r^{k+2}\|^2.\label{equ : lemma : opt : nonergodic : b4}\end{aligned}\ ] ] thus we conclude that @xmath233 + \|(w^{k+1}-w^{k } ) - ( w^{k+2 } - w^{k+1 } ) \|_h^2 \nn \\ \ge { } & ( 2 - \alpha - \gamma ) \beta \| r^{k+1 } - r^{k+2}\|^2 + 2(1 - \alpha)\beta \left \langle b\left[(y^k - y^{k+1 } ) - ( y^{k+2 } - y^{k+1})\right ] , r^{k+1 } - r^{k+2}\right\rangle \nn \\ & + ( 1-\alpha)\beta \left\|b [ ( y^{k+1 } - y^k ) - ( y^{k+2 } - y^{k+1})]\right\|^2 \nn \\ \geq { } & ( 1 -\gamma )
\beta \| r^{k+1 } - r^{k+2}\|^2 \geq \frac{1 - \gamma}{\alpha + \gamma } \|(w^{k+1}-w^{k } ) - ( w^{k+2 } - w^{k+1 } ) \|_g^2,\nn \ ] ] where the first inequality is due to and , the second inequality is trivial and the last inequality owes to . the proof is completed .
[ theorem : nonergodic ] let the sequence @xmath234 be generated by sp - prsm . if @xmath28 $ ] , then there holds that @xmath235 summing over @xmath218 leads to @xmath236 observe that @xmath237 $ ] , it follows from that @xmath238 , which together with can easily suggest .
the proof is completed .
for simplicity , we only consider the case when @xmath30 and @xmath31 . in this case
, we always have that @xmath239 [ ass : y ] the set @xmath240 .
[ ass : gradient_lips ] the gradient @xmath241 is lipschtiz continuous on @xmath10 with @xmath242 .
thus for any @xmath243 there holds that @xmath244 and @xmath245 [ ass : strong_convex ] the function @xmath59 is strongly convex on @xmath10 with constant @xmath246 .
thus for any @xmath243 there holds that @xmath247 since @xmath240 , it follows from that @xmath248 similarly , we obtain from that @xmath249 [ thm : linear ] let the sequence @xmath107 be generated by sp - prsm . under assumptions [ ass : y ] ,
[ ass : strong_convex ] , [ ass : gradient_lips ] , there must exists some constant @xmath250 such that @xmath251 letting @xmath252 in , with , we can obtain @xmath253 next we will estimate the two terms of the righthand side of , respectively .
firstly , with , we see that @xmath254 it follows from , and the convexity of @xmath58 that @xmath255 for any @xmath256 , consider the convex combination of and , we know from , and that there holds that @xmath257 by applying the inequality @xmath258 with @xmath259 , @xmath260 and @xmath261 , we have @xmath262 plugging into , we further have @xmath263 where @xmath264 , @xmath265 . with the definition of @xmath93 , we can easily see that there exits a positive constant @xmath266 such that @xmath267 combing , and , we see that @xmath268 where @xmath269 the proof is completed .
* remark*. let us consider as a special case , namely , @xmath270 . here ,
@xmath271 and reduces to @xmath272 . hence @xmath273 .
letting @xmath274 , we have that @xmath275 where @xmath276 , @xmath277 and the equality attains when @xmath278 .
the largest @xmath279 is @xmath280 which reduces to ( 3.16 ) in @xcite when @xmath281 .
in this section , we demonstrate the potential efficiency of our method sp - prsm by solving the @xmath282-regularized least square problem : @xmath283 where @xmath284 is the data matrix , @xmath285 is the number of the data points , @xmath286 is the number of features , @xmath287 is the vector of feature coefficients to be estimated and @xmath288 , @xmath289 is the observation vector and @xmath290 is the regularization parameter . given @xmath286 , @xmath291 and a @xmath292-sparse vector @xmath293 ( @xmath292 is the number of nonzero elements in @xmath36 over @xmath285 ) , the matlab codes for generating the data of are given as .... xbar = sprandn(n,1,p ) ; d = randn(m , n ) ; d = d*spdiags(1./sqrt(sum(d.^2))',0,n , n ) ; % normalize columns b = d*xbar + sqrt(0.001)*randn(m,1 ) ; mu = 0.1 * norm(d'*b , ' inf ' ) . .... by introducing an auxiliary variable @xmath294 , we reformulate the problem as @xmath295 we consider to apply the sp - prsm to solve . in our implementation
, we always choose @xmath296 and @xmath297 . starting from @xmath298 and @xmath299 , with some suitably chosen proximal matrix @xmath300
, the iterative scheme is given as @xmath301 where the shrinkage operator is defined as @xmath302 and @xmath303 if @xmath304 and @xmath305 if @xmath306 .
, we stop the iterative scheme when @xmath307 or the iterative counter @xmath308 . considering that the role of @xmath15 and @xmath23 in the sp - prsm are equal
, we can easily extend the convergence domain to @xmath309 note that @xcite also establish the convergence of the sp - prsm with @xmath310 and @xmath296 when @xmath311 the union domain defined by and will be considered below .
we generate the mesh grid with respect to @xmath15 and @xmath23 by equally dividing the intervals @xmath312 $ ] and @xmath313 $ ] into 10 parts , respectively .
we set @xmath314 , @xmath315 .
the corresponding scatter diagram is depicted in figure [ figure:1 ] . for consideration of space ,
only the detailed results for are shown in table [ results:1 ] . from this table
, we see that when @xmath316 and @xmath317 , the sp - prsm always perform best ; when @xmath15 and @xmath23 satisfy , the results are not so good .
+ , @xmath318 , the intervals @xmath319 $ ] and @xmath320 $ ] are both equally divided into 10 parts .
note that the ` iter'-axis and ` alpha'-axis are both in reverse order.,width=384 ] .the number of iterations taken by sp - prsm with different @xmath15 and @xmath23 for solving with @xmath314 , @xmath315 , ` - ' means that the corresponding @xmath15 and @xmath23 do not satisfy . [ cols="^,^,^,^,^,^,^,^,^,^,^,^",options="header " , ] from table 4 we can observe that we can use indefinite matrices in the subproblems and , which can further improve the efficiency of the algorithm .
in fact , under certain conditions on the problems data and suitable requirements on the matrices , we can theoretically establish its convergence . since the analysis is very similar to those in this paper ,
we do not describe them here .
in this paper , we proposed a modification of the peaceman - rachford splitting method by introducing two different parameters in updating the dual variable , and by introducing semi - proximal terms to the subproblems in updating the primal variables .
we established the relationship between the two parameters under which we proved the global convergence of the algorithm .
we also analyzed the sublinear rate convergence under ergodic and non - ergodic senses , respectively .
under further conditions of one of the objective functions , we proved the linear convergence of the algorithm .
finally , we reported extensive numerical results , indicating the efficiency of the proposed algorithm .
note that the parameters @xmath15 and @xmath23 are essential to the efficiency of the algorithm , which should be variable along with the iteration . allowing the parameter @xmath15 and @xmath23
varying with the process of the iterate may give us the freedom of choosing them in a self - adaptive manner .
suitable updating rules are among our future research tasks . on the other hand , considering that the main task in the algorithm is to solve the @xmath36-optimization problem ( e.g. , ) or
the @xmath22-optimization problem ( e.g. , ) , solving them in an inexact manner may improve the efficiency of the algorithm .
an approximate version of the proposed sp - prsm with practical accuracy criteria is also our future research topic . , _ some reformulation and applications of the alternating directions method of multipliers_. in : hager , w.w .
, large scale optimization : state of the art , pp .
115 - 134 .
kluwer academic publishers , 1994 . , _ sur lapproximation , par lments finis dordre un , et la rsolution , par pnalisation - dualit dune classe de problmes de dirichlet non linaires _ , esaim : mathematical modelling and numerical analysis - modlisation mathmatique et analyse numrique , 9 ( 1975 ) , pp . | the peaceman - rachford splitting method is very efficient for minimizing sum of two functions each depends on its variable , and the constraint is a linear equality .
however , its convergence was not guaranteed without extra requirements . very recently , he _ et al . _
( siam j. optim . 24 : 1011 - 1040 , 2014 ) proved the convergence of a strictly contractive peaceman - rachford splitting method by employing a suitable underdetermined relaxation factor .
in this paper , we further extend the so - called strictly contractive peaceman - rachford splitting method by using two different relaxation factors , and to make the method more flexible , we introduce semi - proximal terms to the subproblems .
we characterize the relation of these two factors , and show that one factor is always underdetermined while the other one is allowed to be larger than 1 .
such a flexible conditions makes it possible to cover the glowinski s admm whith larger stepsize .
we show that the proposed modified strictly contractive peaceman - rachford splitting method is convergent and also prove @xmath0 convergence rate in ergodic and nonergodic sense , respectively .
the numerical tests on an extensive collection of problems demonstrate the efficiency of the proposed method . *
key words * : semi - proximal , strictly contractive , peaceman - rachford splitting method , convex minimization , convergence rate . |
femtosecond pulses offer unique ways to coherently control photo - induced quantum dynamics of matter @xcite .
the corresponding key characteristic is their coherence over a broad spectrum .
multiphoton absorption processes in atoms and molecules , which are of fundamental scientific importance as well as applicative importance to the fields of nonlinear spectroscopy and microscopy , are among the processes that have been controlled most effectively by shaped femtosecond pulses @xcite .
the control principle is the coherent manipulation of interferences among the manifold of initial - to - final state - to - state multiphoton pathways that are induced by the pulse .
constructive interferences lead to absorption enhancement ( i.e. , increased transition probability ) , while destructive interferences lead to absorption attenuation ( i.e. , decreased transition probability ) . the interference manipulation is implemented by shaping the femtosecond pulse @xcite , i.e. , manipulating the spectral phase , amplitude , and/or polarization of its different frequency components .
hence , in order to fully utilize the coherent control potential of a given excitation scheme , the ideal line of action is shaping the pulse based on an initial identification of the different multiphoton pathways and their interference mechanism .
when such identification is not possible , a practical partial solution is to use automatic experimental optimization of the pulse shape using learning algorithms that generally considering the system as a black box @xcite .
the lesson , which can be learned from the many successful coherent control studies of multiphoton absorption conducted in the past @xcite , is that this ideal line of action is feasible and very powerful once the photo - excitation picture is available in the frequency domain .
this is possible only within the framework of perturbation theory , where a valid perturbative description in the time domain is ( fourier ) transformed to the frequency domain .
however , until recently the frequency domain has been exploited only in the weak - field regime @xcite , where the n - photon absorption is described by perturbation theory of the lowest non - vanishing order , i.e. , the n@xmath0 order .
physically it means that the n - photon absorption is coherently induced by all the possible initial - to - final state - to - state pathways of n absorbed photons . for two - photon absorption
the lowest order is the 2@xmath1 one , involving all the pathways of two absorbed photons @xcite .
the downside of being limited to the weak - field regime is the low absorption yields associated with it . for two - photon absorption they are typically below 0.1% population transfer .
recently @xcite , we have extended the powerful frequency - domain picture of femtosecond two - photon absorption to a regime of considerable absorption yields , exceeding the weak - field yields by more than two orders of magnitude .
it corresponds to intermediate field strengths , where the interfering pathways are the weak - field ( non - resonant ) pathways of two absorbed photons as well as additional four - photon pathways of three absorbed photons and one emitted photon .
the picture is based on 4@xmath0-order perturbation theory , which includes both the 2@xmath1 and 4@xmath0 orders associated , respectively , with the two- and four - photon pathways .
the relative contribution of the 4@xmath0-order absorption amplitude increases as the field strength ( pulse intensity ) increases .
this intermediate - field regime is distinguished from the strong - field regime where no perturbative description is valid .
the strong - field regime is actually the one that all the other past multiphoton control studies , which have deviated from the weak - field regime , have focused on @xcite .
our previous work @xcite has focused on the family of spectral phase patterns that are anti - symmetric around half the two - photon transition frequency ( @xmath2 ) .
we have found this family to enhance the intermediate - field two - photon absorption relative to the unshaped transform - limited pulse , when the central spectral frequency is properly detuned , to the red or to the blue ( depending on the system ) , from @xmath2 . in the present work we systematically study in detail the intermediate - field coherent control and the corresponding interference mechanisms , including their dependence on the pulse spectrum and its detuning from @xmath2 .
the absorption dynamics in the intermediate - field regime is compared with the one in the weak - field regime .
the model system is the sodium ( na ) atom . as a test case for femtosecond phase control
the study uses the family of shaped pulses having a @xmath3 spectral phase step , which in the weak - field regime allows high degree of control over the full accessible range of the non - resonant two - photon absorption .
section [ sec : theoretical ] presents and elaborates on the extended frequency - domain 4@xmath0-order perturbative theoretical description .
the na intermediate - field control results are presented in sec .
[ sec : results ] .
they include experimental results , exact non - perturbative results calculated by the numerical propagation of the time - dependent schrdinger equation , and perturbative results calculated numerically using the frequency - domain 4@xmath0-order formulation .
the formers are used to validate the latter .
then , in sec .
[ sec : discussion ] , the perturbative results are analyzed and discussed based on their corresponding frequency - domain description , which allows the identification of the interference mechanisms leading to the different intermediate - field features .
the atomic femtosecond two - photon absorption process we consider is from an initial ground state @xmath4 to a final excited state @xmath5 , which are coupled via a manifold states @xmath6 having the proper symmetry .
the spectrum of the pulse is such that all the @xmath4-@xmath6 and @xmath5-@xmath6 couplings are non - resonant , i.e. , the spectral amplitude at all the corresponding transition frequencies is zero : @xmath7 , except for the @xmath5-@xmath8 resonant coupling for which in general @xmath9 .
the corresponding excitation scheme is shown schematically in fig .
[ fig_1 ] . within the present intermediate - field regime , the time - dependent ( complex ) amplitude @xmath10 of state @xmath5 at time @xmath11 , following irradiation with a ( shaped ) temporal electric field @xmath12 , can be validly described by 4@xmath0-order time - dependent perturbation theory .
so , in general , it includes non - vanishing contributions from both the 2@xmath1 and 4@xmath0 perturbative orders : @xmath13 with @xmath14dt_{1}dt_{2 } \label{eq2:time - amp } \ ; , \\
% a_{f}^{(4)}(t ) & = & -\frac{1}{\hbar^{4}}\sum_{k , l , m}\mu_{fk}\mu_{kl}\mu_{lm}\mu_{mg } \int_{-\infty}^{t}\int_{-\infty}^{t_{1}}\int_{-\infty}^{t_{2}}\int_{-\infty}^{t_{3 } } \varepsilon(t_{1})\varepsilon(t_{2})\varepsilon(t_{3})\varepsilon(t_{4 } ) \nonumber \\ & & \times \exp\left[-i(\omega_{fk}t_{1}+\omega_{kl}t_{2}+\omega_{lm}t_{3}+\omega_{mg}t_{4})\right ] dt_{1}dt_{2}dt_{3}dt_{4 } \ ; , \label{eq3:time - amp}\end{aligned}\ ] ] where @xmath15 is the dipole matrix element between a pair of states and @xmath16 is the corresponding transition frequency .
the 2@xmath1-order term @xmath17 by itself corresponds to the weak - field regime .
the intermediate - field final amplitude @xmath18 of state @xmath5 after the pulse is over ( i.e. , @xmath19 ) can be expressed as @xmath20 with @xmath21 and @xmath22 .
the final @xmath5 population @xmath23^{2}+[\im(a_{f})]^{2}$ ] of state @xmath5 reflects the degree of two - photon absorption .
this perturbative description allows a transformation into a frequency - domain picture . within the frequency - domain framework ,
the spectral field of the pulse @xmath24 $ ] is given as the fourier transform of @xmath12 , with @xmath25 and @xmath26 being , respectively , the spectral amplitude and phase of frequency @xmath27 . for the unshaped transform - limited ( tl ) pulse , @xmath28 for any @xmath27 .
we also define the normalized spectral field @xmath29 $ ] that represents the pulse shape , where @xmath30 is the peak spectral amplitude .
this allows to clearly distinguish in the expressions given below between the dependence on the pulse intensity and the dependence on the pulse shape .
the maximal spectral intensity @xmath31 is proportional to @xmath32 ( @xmath33 ) .
different values of @xmath31 correspond to different temporal peak intensities @xmath34 of the transform - limited ( tl ) pulse .
as shown before for the weak - field regime @xcite , the 2@xmath1-order amplitude @xmath35 is given by @xmath36 where @xmath37 is the @xmath4-@xmath5 transition frequency and @xmath38 is the corresponding real effective non - resonant two - photon coupling .
it is given by @xmath39 , with @xmath40 being the carrier frequency of the pulse .
( [ eq5:tot - amp-2nd - order])-([eq6:expl - amp-2nd - order - w_fg ] ) reflect the fact that @xmath35 coherently interferes all the non - resonant two - photon pathways from @xmath4 to @xmath5 of any combination of two absorbed photons with frequencies @xmath27 and @xmath41 .
several such two - photon pathways are shown schematically in fig .
[ fig_1 ] .
the phase associated with each two - photon pathway is @xmath42 .
so , with the tl pulse all these pathways acquire zero phase and thus interfere one with the other in a fully constructive way . with a given spectrum @xmath25 ,
this leads to the maximal @xmath43 and the maximal weak - field non - resonant two - photon absorption .
the 4@xmath0-order amplitude term @xmath44 is much more complicated than @xmath35 and we have calculated it to be given by @xmath45 \ ; , \label{eq7.1:amp-4th - order } \\
a_{f}^{(4)\textrm{on - res } } & = & i \pi a^{(2)}(\omega_{fg } ) a^{(r)}(0 ) \ ; , \label{eq7.2:amp-4th - order } \\
a_{f}^{(4)\textrm{near - res } } & = & - \wp \int_{-\infty}^{\infty } d\delta \frac{1}{\delta } a^{(2)}(\omega_{fg}-\delta ) a^{(r)}(\delta ) \ ; , \label{eq7.3:amp-4th - order}\end{aligned}\ ] ] where @xmath46 is defined in eq .
( [ eq6:expl - amp-2nd - order - w_fg ] ) and @xmath47 .
\hspace{0.7 cm } % \nonumber% \label{eq10:amp-4th - order}\end{aligned}\ ] ] this set of equations reflects the fact that @xmath44 interferes all the four - photon pathways from @xmath4 to @xmath5 of any combination of three absorbed photons and one emitted photon .
several typical four - photon pathways are shown schematically in fig .
[ fig_1 ] .
each four - photon pathway can actually be divided into two two - photon parts : ( i ) a non - resonant transition of two absorbed photons @xmath27 and @xmath48 with a frequency sum of @xmath49 , and ( ii ) a raman transition of two photons @xmath50 and @xmath51 with a frequency difference of @xmath52 . the border line between these parts
is detuned by @xmath53 from either @xmath5 or @xmath4 according to whether , respectively , part ( i ) or part ( ii ) comes first ( see fig .
[ fig_1 ] ) . the @xmath54 and @xmath55 terms of @xmath44 [ eqs .
( [ eq7.1:amp-4th - order])-([eq7.3:amp-4th - order ] ) ] interfere , respectively , these on - resonant ( @xmath56 ) and near - resonant ( @xmath57 ) four - photon pathways .
the on - resonant pathways are excluded from @xmath55 by the cauchy s principle value operator @xmath58 .
the integration taking place in these terms over the corresponding pathways is expressed using the product of two parameterized amplitudes , @xmath46 and @xmath59 , which originate from the different two - photon parts of the four - photon pathways .
the amplitude @xmath46 interferes all the non - resonant two - photon transitions [ parts ( i ) above ] with transition frequency @xmath60 , while the amplitude @xmath59 interferes all the raman transitions [ parts ( ii ) above ] with transition frequency @xmath61 . as given by eqs .
( [ eq8:amp-4th - order])-([eq10:amp-4th - order ] ) , @xmath59 includes two components .
the first component is @xmath62 interfering all the raman transitions that are of non - resonant nature , with @xmath63 and @xmath64 being the @xmath4-@xmath4 and @xmath5-@xmath5 real effective non - resonant raman couplings due to all the non - resonantly coupled states @xmath6 .
they are given by @xmath65 $ ] and @xmath66 $ ] .
hence , their sum @xmath67 [ appearing in eq .
( [ eq9:amp-4th - order ] ) ] is either positive or negative , depending on the physical system and pulse spectrum .
the second component of @xmath59 is @xmath68 interfering all the raman transitions that are of resonance - mediated nature via @xmath8 , with @xmath69 being the detuning from @xmath8 ( see fig .
[ fig_1 ] ) .
these on - resonant ( @xmath70 ) and near - resonant ( @xmath71 ) transitions are interfered separately , respectively , in the first and second terms of @xmath68 [ eq . ( [ eq10:amp-4th - order ] ) ] . the cauchy s principle value operator @xmath58 excludes the on - resonant transitions from the second term .
for a given physical system and a given pulse shape @xmath72 , a non - zero @xmath35 is proportional to @xmath32 while a non - zero @xmath44 is proportional to @xmath73 , i.e. , their ratio is proportional to @xmath32 or , equivalently , to @xmath31 . for a given @xmath31 , the relative magnitude and relative sign between the real parts @xmath74 $ ] and @xmath75 $ ]
are generally determined by the pulse shape @xmath72 and by the magnitudes and signs of the different raman couplings ( @xmath76 , @xmath77 , and @xmath78 ) .
the same applies also for the imaginary parts @xmath79 $ ] and @xmath80 $ ] . in the present work , for a set of intensities @xmath31 , the final @xmath5 population @xmath81 ( i.e. , the degree of two photon absorption )
is controlled via the pulse shape @xmath72 .
three different spectra @xmath82 are being considered , with the control knobs being the various spectral phases @xmath26 .
the three spectra are chosen such that their central spectral frequency @xmath40 is of no detuning , blue detuning , or red detuning from @xmath2 , with @xmath83 for the detuned cases .
this set of spectra generally corresponds to the typical case , where @xmath84 is negligible relative to @xmath85 in the weak - field limit and becomes comparable to @xmath85 in the upper intermediate - field limit .
these spectral shifts do not change the sign of the various raman couplings and hardly change their magnitude .
they do however affect the relative amplitude associated with the different interfering pathways and , thus , their overall interference result .
for example , as shown below , in the cases studied here the spectral change from red- to blue - detuning leads to a change in the relative sign between @xmath79 $ ] and @xmath80 $ ] , and thus to a change in the nature of their interference from destructive to constructive .
in general , the dynamics and interference mechanisms discussed in this work involve intra - term as well as ( intensity - dependent ) inter - term interferences involving the multiphoton pathways of the 2@xmath1- and 4@xmath0-order perturbative terms .
the physical model system of the study is the sodium ( na ) atom @xcite .
it includes the @xmath86 ground state as @xmath4 , the @xmath87 state as @xmath5 , the manifold of @xmath88 states as the @xmath6 manifold , and the @xmath89 state as @xmath8 .
so , @xmath90 and @xmath91 .
the transition frequency @xmath92 @xmath93 corresponds to two 777-nm photons and the transition frequency @xmath94 @xmath93 corresponds to one 781.2-nm photon .
the atomic sodium is irradiated with phase - shaped linearly - polarized femtosecond pulses having an intensity spectrum of @xmath955-nm fwhm - bandwidth ( @xmath95180 fs tl duration ) centered around a wavelength @xmath96 tunable between 773 to 780 nm . as a test case , for three different values of the central wavelength
, the present study uses the family of shaped pulses having @xmath3-step spectral phase patterns . in the weak - file regime
@xcite this family allows high degree of control over the full accessible range of the two - photon absorption , from zero to the maximal level ( induced by the tl pulse ) .
each such pattern is characterized by the @xmath3-step position @xmath97 , with @xmath98 and @xmath99 .
the understanding and analysis of the intermediate - field control mechanism is conducted below using the frequency - domain picture given in eqs .
( [ eq4:time - amp])-([eq10:amp-4th - order ] ) and corresponding numerical results for the na system .
however , prior to the analysis , the extended perturbative picture and results are validated by a comparison to exact non - perturbative results that their own validity is confirmed first by a comparison to experiment .
the exact non - perturbative results have been calculated by the numerical propagation of the time - dependent schrdinger equation equation ( tdse ) using the fourth - order runge - kutta method .
the theoretically considered manifold of @xmath88-states is from @xmath100 to @xmath101 , including all the ( 1/2 and 3/2 ) fine - structure states @xcite .
experimentally , atomic sodium vapor is produced in a static chamber at @xmath102c ( na partial pressure of @xmath950.1 torr ) with 10-torr ar buffer gas .
it is irradiated at a 1-khz repetition rate with shaped femtosecond laser pulses of three different central spectral wavelengths : @xmath96=773 , 777 , and 779.5 nm .
the corresponding spectral intensity bandwidth ( fwhm ) is , respectively , 5.5 , 4.5 , and 5 nm .
the 773-nm and 777-nm spectra are gaussians , while the 779.5-nm spectrum is a modified gaussian having a slight asymmetry toward short wavelengths .
the slight change in the experimental spectral shape between the three cases results from technical limitations and is of no significance to the analysis and discussion presented below .
the laser pulses undergo shaping in a 4@xmath103 optical setup incorporating a pixelated liquid - crystal spatial light phase modulator @xcite .
the effective spectral shaping resolution is @xmath104 @xmath93 ( 0.125 nm ) per pixel .
the experiment is conducted with different pulse energies . upon focusing ,
the corresponding temporal peak intensity of the transform - limited ( tl ) pulse at the peak of the spatial beam profile @xmath105 ranges from 5@xmath10610@xmath107 to 7@xmath10610@xmath108 w/@xmath109 . following the interaction with a pulse
, the na population excited to the @xmath87 state radiatively decays to the lower @xmath100 state .
the fluorescence emitted in the decay of the @xmath100 state to the @xmath86 ground state serves as the relative measure for the excited @xmath87 population @xmath110 .
it is optically measured at 90@xmath111 to the beam propagation direction using a spectrometer coupled to a time - gated camera system .
the measured signal results from an integration over the spatial beam profile .
figure [ fig_2 ] compares the non - perturbative theoretical results ( solid lines ) with the experimental results ( squares ) for the two - photon absorption in na .
each column in the figure corresponds to a different spectral case of @xmath96=773 , 777 , or 779.5 nm with different pulse energies , i.e. , different @xmath105 .
the pulse energy increases from top to bottom within a single column .
the traces show the final @xmath87 population @xmath112 as a function of the @xmath3 phase step position @xmath97 .
each of the traces is normalized by the final @xmath87 population @xmath113 excited by the corresponding tl pulse .
the weak - field @xmath3-traces are given in the first - row panels [ ( a)-panels ] of fig .
[ fig_2 ] .
the non - perturbative theoretical results shown in fig .
[ fig_2 ] account for the experimental integration over the spatial beam profile .
each of the presented traces results from an appropriately - weighted integration over a set of calculations conducted each with a different single value of @xmath31 .
as can be seen , there is an excellent agreement between the experimental results ( `` real experiment '' ) and the non - perturbative results ( `` computer experiment '' ) .
hence , the accuracy of the latter is confirmed for the present intermediate - field excitation of na .
next , the confirmed non - perturbative calculations are used to validate the intermediate - field perturbative results and to identify the intensity limit of the intermediate - field regime for the present na excitation , i.e. , the intensity up to which no perturbative order beyond the 4@xmath0 one is needed to be included . from this point on ,
the analysis and discussion are conducted based on the theoretical results .
so , for completeness , the three intensity spectra considered theoretically are of a perfect gaussian shape with 5-nm bandwidth ( fwhm ) centered around @xmath96=773.5 , 777 , and 780 nm . as mentioned above
, these @xmath96 values correspond to blue detuning , no detuning , and red detuning of @xmath40 from @xmath114 , with @xmath115 for the detuned cases .
figure [ fig_3 ] compares the theoretical non - perturbative results ( thick gray lines ) and perturbative results calculated numerically using eqs .
( [ eq4:time - amp])-([eq10:amp-4th - order ] ) ( thick black lines ) .
the perturbative calculations use a frequency grid with a bin size equal to the experimental spectral shaping resolution @xmath116 .
the presented @xmath3-traces are given on a tl - normalized scale of @xmath117 ( left - hand y - axis ) as well as on an absolute scale of @xmath112 ( right - hand y - axis ) .
the x - axis is the @xmath3 phase step position @xmath97 . also here
, each column corresponds to a different spectral case of @xmath96 with different ( single - valued ) intensities @xmath31 , i.e. , different @xmath34 .
the @xmath31 increases from top to bottom within a single column . shown
are several examples out of the full set of results .
the weak - field traces are given in the first - row panels [ ( a)-panels ] of fig .
[ fig_3 ] .
actually , on a tl - normalized scale they are all identical and independent of @xmath96 @xcite . for comparison
, the tl - normalized weak - field trace is also given in all the other panels ( thin black line ) .
as can be seen from the figure , the perturbative results reproduce the exact non - perturbative results up to the @xmath31 that corresponds to a tl peak intensity of @xmath118 w/@xmath119 [ ( c)-panels of fig .
[ fig_3 ] ] .
this is the intensity limit of the 4@xmath0-order intermediate - field regime for the present na excitation .
as the last - row panels [ ( d)-panels ] of fig . [ fig_3 ] show , the 4@xmath0-order perturbative description is not sufficient above this intensity .
the corresponding intensity limit of the weak - field regime , where @xmath120 , is @xmath121 w/@xmath119 .
the prominent features of the intermediate - field @xmath3-traces are presented below .
as previously shown @xcite and can be seen in fig .
[ fig_3 ] [ the ( a)-panels ] , the weak - field tl - normalized @xmath3-trace is symmetric around @xmath122 ( 777 nm ) and is identical for any @xmath40 ( @xmath96 ) .
its shape is determined only by the spectral bandwidth of the pulse .
also , since it is tl - normalized , it is independent of the intensity @xmath31 [ see eq .
( [ eq5:tot - amp-2nd - order ] ) ] .
however , when deviating from the weak - field regime with the 4@xmath0 perturbative order playing a role , the tl - normalized @xmath3-trace losses its weak - field symmetry and its shape becomes dependent on both @xmath40 and @xmath31 . at a given @xmath40 ,
the degree of deviation from the weak - field shape depends on @xmath31 .
as mentioned above , in the weak - field regime the maximal non - resonant two - photon absorption is induced by the tl pulse . additionally , as can be seen in the figure ,
the same maximal weak - field two - photon absorption is also induced by the shaped pulse with @xmath122 ( 777 nm ) , i.e. @xmath123 or @xmath124 .
both pulses induce fully constructive interferences ( i.e. , zero relative phase ) between all the @xmath4-@xmath5 two - photon pathways .
conversely , in the intermediate - field regime a @xmath3-phase step at @xmath122 leads to a two - photon absorption that , in general , is different from the tl absorption . the difference is highly pronounced when @xmath40 is detuned from @xmath114 [ fig . [ fig_3 ] - columns(1 ) and ( 3 ) ] and is weakly pronounced when @xmath125 [ fig .
[ fig_3 ] - column ( 2 ) ] , with @xmath126 for a blue detuning , @xmath127 for no detuning , and @xmath128 for a red detuning .
for example , at the intermediate - field limit , with @xmath31 of @xmath118 w/@xmath119 , @xmath129 reaches a value of 0.65 , 1.15 , 2.0 for @xmath96=773.5 nm [ fig .
[ fig_3](1c ) ] , @xmath96=777 nm [ fig .
[ fig_3](2c ) ] , and @xmath96=780 nm [ fig .
[ fig_3](3c ) ] , respectively .
the corresponding degree of attenuation or enhancement over the tl absorption , i.e. , @xmath130 , increases as @xmath31 increases .
also , as @xmath31 increases , this enhancement / attenuation effect occurs ( with variable corresponding values ) over an increased range of @xmath97 around @xmath114 . in the spectrally detuned cases of @xmath131 with @xmath132 ( as is the case here ; see above ) a @xmath3 phase step that is positioned around @xmath40 , i.e. , @xmath133 , leads to intermediate - field values of the tl - normalized trace [ fig . [ fig_3 ] - columns ( 1 ) and ( 3 ) , thick solid lines ] that are systematically either attenuated or enhanced relative to the weak - field trace [ fig . [ fig_3 ] - columns ( 1 ) and ( 3 ) , thin solid lines ] .
the attenuation occurs for blue - detuned @xmath96 and the enhancement occurs for red - detuned @xmath96 . also here , the degree of attenuation or enhancement relative to the weak - field case at a given @xmath97 increases as @xmath31 increases .
in the red - detuned case , at high enough intensity @xmath31 , the corresponding two - photon absorption even exceeds the tl absorption [ see fig .
[ fig_3](3c ) ] .
an additional intermediate - field feature is the dip observed when @xmath134 ( 781.2 nm ) , which does not exist in the weak - field trace .
it occurs when the field amplitude @xmath135 at @xmath136 significantly deviates from zero .
hence , this 781.2-nm dip is very prominent for @xmath96=780 nm [ fig .
[ fig_3 ] - column ( 3 ) ] , weakly noticeable for @xmath96=777 nm [ fig .
[ fig_3 ] - column ( 2 ) ] , and hardly noticeable for @xmath96=773.5 nm [ fig .
[ fig_3 ] - column ( 1 ) ] .
lastly , it is important to mention that within the intermediate - field regime , due to the 4@xmath0-order term @xmath137 , the absolute @xmath87 population @xmath113 excited by a given shaped pulse strongly deviates from the weak - field intensity dependence of @xmath138 in all the @xmath96 cases .
specifically , it applies also to the non - detuned case of @xmath96=777 nm , even though the corresponding intermediate - field shape of the tl - normalized @xmath3-trace does not deviate much from the weak - field shape .
this can be seen in fig .
[ fig_3 ] , for example , in the absolute population values that are induced by the tl pulse ( i.e. , the asymptotes of the traces ) at different intensities .
the discussion below analyzes the coherent source for the intermediate - field features described above , using the frequency - domain picture given in eqs .
( [ eq4:time - amp])-([eq10:amp-4th - order ] ) .
the different intermediate - field features , i.e. , the deviation of the intermediate - field trace from the weak - field trace , originate from the different dependence of the 2@xmath1- and 4@xmath0-order amplitudes , @xmath139 and @xmath137 , on the pulse shape . for a given spectrum @xmath25 , the on - resonant term @xmath140 of @xmath137 [ eq .
( [ eq7.2:amp-4th - order ] ) ] is proportional to @xmath139 .
it is so since @xmath141 depends only on the spectral intensity : all the corresponding raman transitions involve two identical photons and their amplitudes in eqs .
( [ eq8:amp-4th - order])-([eq10:amp-4th - order ] ) are of the form @xmath142 .
thus , the difference in the phase dependence of @xmath139 and @xmath137 originates only from the near - resonant term @xmath55 of @xmath137 , which interferes all the near - resonant four - photon pathways of non - zero detunings @xmath57 from @xmath86 or @xmath87 ( see above ) . as eq .
( [ eq7.3:amp-4th - order ] ) shows , @xmath55 is given by a proper @xmath58-integration over all the non - zero @xmath53 values .
its phase dependence originates from the phase dependence of the integrands @xmath143 and @xmath144 .
their dependence on the spectral phase pattern @xmath26 is reflected in their values for a given @xmath53 and , more importantly , in their functional dependence on @xmath53 .
so , the @xmath58-integration yields different results for different phase patterns . due to the @xmath145 weighting
( and its sign change for @xmath146@xmath147 ) , the @xmath58-integration result is dominated by the integration over small values of @xmath147 and is highly sensitive to the degree of symmetry of the integrand @xmath148 around @xmath149 , i.e. , how different are its values for @xmath146@xmath147 .
below , the intermediate - field two - photon absorption is analyzed based on the @xmath53-dependence of @xmath143 and @xmath144 for several representative cases of the @xmath3-step position @xmath97 with the different @xmath96 .
the analysis of the amplitude @xmath144 , which is contributed by all the raman parts of the @xmath53-detuned four - photon pathways , is simplified by including its component @xmath150 [ see eq .
( [ eq8:amp-4th - order ] ) ] only for the study of the dip feature at @xmath151 ( 781.2 nm ) .
as described above , @xmath150 interferes those raman parts that are resonance - mediated via the @xmath89 state .
this line of analysis is supported by the discussion given below and by the perturbative results presented in fig .
[ fig_4 ] for the @xmath3-traces at @xmath31 of @xmath118 w/@xmath119 ( i.e. , the intermediate - field limit ) .
in addition to the real tl - normalized @xmath3-traces ( gray thick lines ) , which are also shown in the ( c)-panels of fig .
[ fig_3 ] , the figure displays the @xmath3-traces ( black thick lines ) calculated with @xmath152 , i.e. , with artificially setting @xmath150 to zero for any @xmath53 . as can be seen
, the resonance - mediated raman transitions via @xmath89 are of significance only for discussing the dip at @xmath151 ( 781.2 nm ) . for completeness
, the figure also displays the weak - field trace ( black thin lines ; shown also in fig . [ fig_3 ] ) that originates only from @xmath35 and , thus , is not affected by changes in the raman part .
the non - resonant raman term @xmath153 [ eq . ( [ eq9:amp-4th - order ] ) ] coherently integrates all the non - resonant raman amplitudes @xmath154 contributed by all the possible pairs of photons with a frequency difference of @xmath53 ( @xmath27 is scanned across the spectrum ) . with @xmath28 ( tl pulse ) or
any @xmath3 spectral phase step , the amplitude contributed by any such pair of photons is a real ( positive or negative ) quantity and , thus , so is the resulting @xmath153 .
so , with these phase patterns , the relation @xmath155@xmath156@xmath157@xmath158@xmath159 holds for any @xmath53 , i.e. , @xmath153 is symmetric around @xmath149 .
this symmetry can be seen in fig .
[ fig_5 ] that shows , as an example , the value of @xmath153 ( gray thick lines ) as a function of @xmath53 for the tl pulse and for the shaped pulses with @xmath160 and @xmath161 .
each column corresponds to a different @xmath96 ( @xmath40 ) .
the x - axis values are actually the normalized detuning values @xmath162 , with @xmath163 being the bandwidth of the intensity spectrum . as can also be seen in the figure , @xmath153 is maximal at @xmath149 , with a value that depends only on the spectral intensity @xmath164 , i.e. , is independent of @xmath26 . the other term @xmath143 [ eq . (
[ eq6:expl - amp-2nd - order - w_fg ] ) ] coherently integrates all the non - resonant two - photon transition amplitudes @xmath165 contributed by all the possible pairs of photons with a frequency sum of @xmath166 . also here , with @xmath28 ( tl pulse ) or any @xmath3 spectral phase step , the amplitude contributed by any such pair of photons is a real ( positive or negative ) quantity and , thus , so is the resulting @xmath143 .
figure [ fig_5 ] displays , together with the raman part data , the corresponding @xmath143 ( black thick lines ) as a function of @xmath162 in the different cases . the zone of small @xmath147 around @xmath149 , which is the most contributing to @xmath55 ( see above ) ,
is also indicated schematically . as can be seen from the figure ,
the @xmath53-dependence of @xmath143 around @xmath149 is different from one case to the other and depends on both the spectral phase pattern and @xmath96 ( @xmath40 ) .
each panel also shows , for comparison , the @xmath143 trace of the corresponding tl pulse ( black thin line ) . the first - row panels [ ( a)-panels ] of fig .
[ fig_5 ] correspond to the tl pulse of the different @xmath96 ( @xmath40 ) cases .
since the tl pulse induces fully constructive interferences among all the two - photon pathways that contribute to @xmath46 for any @xmath60 , the corresponding value @xmath167 is actually the maximal one ( positive and real ) for any given @xmath53 . as can be seen from eq .
( [ eq6:expl - amp-2nd - order - w_fg ] ) , @xmath167 is actually a convolution of the corresponding spectrum @xmath82 .
thus , with a gaussian spectrum around @xmath40 , it is peaked at @xmath168 , i.e. , at @xmath169 .
so , for the different cases of @xmath40 one obtains the following behavior : ( i ) for @xmath170 [ @xmath171773.5 nm , fig .
[ fig_5](1a ) ] @xmath172 and @xmath143 monotonically decreases around @xmath149 as @xmath53 increases from negative to positive values , i.e. , @xmath173 for small @xmath147 ; ( ii ) for @xmath174 [ @xmath171780 nm , fig .
[ fig_5](3a ) ] @xmath175 and @xmath143 monotonically increases around @xmath149 upon the negative - to - positive increase of @xmath53 , i.e. , @xmath176 for small @xmath147 ; ( iii ) for @xmath125 [ @xmath171777 nm , fig .
[ fig_5](2a ) ] @xmath177 and @xmath143 is symmetric around @xmath149 , i.e. , @xmath178 . the second - row panels [ ( b)-panels ] of fig .
[ fig_5 ] correspond to the shaped pulse with @xmath122 ( 777 nm ) . generally , when a @xmath3 phase step is positioned at @xmath97 , the value of @xmath179 is equal to the corresponding tl value @xmath180 due to fully constructive interferences among all the involving two - photon pathways . as for the tl pulse ,
the phases associated with these pathways are all zero . however ,
as @xmath60 deviates from @xmath181 , the value of @xmath46 gradually reduces with comparable magnitude for positive and negative deviations .
hence , a @xmath3-step at @xmath122 yields a peak of @xmath143 at @xmath149 ( with the tl value ) .
when @xmath125 [ @xmath96=777 nm ; fig .
[ fig_5](2b ) ] , this peak is a global one and @xmath143 has a perfect symmetry around @xmath149 , i.e. , @xmath178 . when @xmath131 [ @xmath171773.5 and 780 nm ; fig .
[ fig_5](1b ) and ( 3b ) ] , this peak is a local one with only an approximate symmetry of @xmath143 around @xmath56 , i.e. , @xmath182 for small @xmath147 .
based on the above analysis of @xmath153 and @xmath143 , considering also the integrand factor @xmath145 that is anti - symmetric around @xmath149 , the magnitude of @xmath183 can be compared between the tl pulse case @xmath184 $ ] and the shaped pulse case of @xmath122 @xmath185 $ ] : ( i ) in the no - detuning case of @xmath96=777 nm
@xmath186 , since the amplitudes contributed to the @xmath58-integral by the four - photon pathways of positive and negative detuning @xmath146@xmath147 are equal and , thus , cancel out each other ; ( ii ) in the red- and blue - detuning cases of @xmath96=773.5 and 780 nm @xmath187 , since the amplitudes contributed by the four - photon pathways of detunings @xmath146@xmath147 are approximately equal one to the other for the shaped pulse of @xmath122 while they are significantly different one from the other for the tl pulse . as discussed below , important is also the sign of @xmath183 relative to @xmath139 , which is determined by the pulse spectrum and by the non - resonant raman couplings sum @xmath188 .
the sign of @xmath189 relative to @xmath190 and the sign of @xmath191 relative to @xmath192 are obtained here to be the same for any @xmath96 . however , upon a blue - to - red spectral shift of @xmath96 ( i.e. , from 773.5 to 780 nm ) these signs change from positive to negative , with the sign flip occurring when passing via @xmath96=777 nm .
the sign change results from a change in the relative magnitude of the amplitudes contributed by the negatively - detuned and positively - detuned four - photon pathways , while @xmath188 keeps its sign and is effectively constant over the whole @xmath96 range considered here . for comparing the two - photon absorption induced by the tl pulse vs. the absorption induced by the shaped pulse of @xmath122
, one needs to consider the coherent amplitude addition of @xmath139 and @xmath137 . for the tl pulse and for any shaped pulse with a @xmath3-step phase pattern , @xmath139 [ eq . (
[ eq5:tot - amp-2nd - order ] ) ] is an imaginary quantity , i.e. , @xmath193 $ ] , while @xmath137 is a complex quantity .
when only @xmath62 is included in @xmath194 ( see above ) , the corresponding @xmath183 component is real while the corresponding @xmath140 component is imaginary .
due to the @xmath195 preceding factor of @xmath137 [ eq . (
[ eq7.1:amp-4th - order ] ) ] , they contribute , respectively , solely to the imaginary and real parts of @xmath137 , i.e. , @xmath196 \propto a_{4s}^{(4)\textrm{near - res}}$ ] and @xmath197 \propto a_{4s}^{(4)\textrm{on - res}}$ ] .
thus , the interferences between the pathways groups of @xmath139 and of @xmath137 actually take place between those included in @xmath139 and those included in @xmath196 $ ] , i.e. , in @xmath183 . since the following relations are satisfied here : ( i ) @xmath198 , ( ii ) @xmath199 \right|^{2 } + \left| \re\left[a_{4s}^{(4)}\right ] \right|^{2}$ ] , ( iii ) @xmath197 \propto i a_{4s}^{(2)}$ ] [ see eq .
( [ eq7.2:amp-4th - order ] ) ] , and ( iv ) @xmath200 while @xmath201 [ see eqs .
( [ eq5:tot - amp-2nd - order ] ) and ( [ eq7.1:amp-4th - order ] ) ] , the tl - normalized @xmath3-trace value @xmath129 ( real and positive ) corresponding to @xmath122 is given as @xmath202/a_{4s,*}^{(2)}\right|^{2}+ \left|\re\left[a_{4s,\textrm{step@}\omega_{4s,3s}/2}^{(4)}\right]/a_{4s,*}^{(2)}\right|^{2 } } { \left|1 + \im\left[a_{4s,\textrm{tl}}^{(4)}\right]/a_{4s,*}^{(2)}\right|^{2}+ \left|\re\left[a_{4s,\textrm{tl}}^{(4)}\right]/a_{4s,*}^{(2)}\right|^{2 } } \nonumber \\ % & = & \frac{\left|1 + k_{\textrm{step@}\omega_{4s,3s}/2 } i_{0 } \right|^{2 } + \left| \kappa i_{0 } \right|^{2 } } { \left|1 + k_{\textrm{tl } } i_{0 } \right|^{2 } + \left| \kappa i_{0 } \right|^{2 } } \nonumber \ ; , \end{aligned}\ ] ] where the pulse shape dependence enters only via the @xmath203 factors .
accounting for the magnitude of @xmath204 vs. @xmath205 and for the sign of @xmath183 relative to @xmath139 , one obtains the following behavior for the different @xmath96 cases : ( i ) for blue detuning of @xmath170 ( @xmath96=773.5 nm ) : @xmath206 , @xmath207 , @xmath208 , and thus @xmath209 ; ( ii ) for no detuning of @xmath125 ( @xmath96=777 nm ) : @xmath210 , @xmath211 , and thus @xmath212 ; ( iii ) for red detuning of @xmath174 ( @xmath96=780 nm ) : @xmath213 , @xmath214 , @xmath208 , and thus @xmath215 .
in other words , the two - photon absorption induced by the shaped pulse of @xmath122 is lower , equal , or higher than the tl absorption according to whether @xmath40 ( @xmath96 ) is blue - detuned , non - detuned , or red - detuned from @xmath114 , respectively .
one can also see that , when @xmath216 , the value of @xmath217 increases as @xmath31 increases .
this entire intermediate - field behavior is indeed the one observed in the results of fig .
[ fig_3 ] , except for a small deviation of @xmath129 from a value of one in the case of @xmath96=777 nm [ for example , in fig . [ fig_3](2c ) it reaches a value of 1.15 ] .
this deviation originates from the resonance - mediated raman term @xmath68 that is excluded from @xmath194 in this part of the analysis and is not symmetric around @xmath218 [ as @xmath153 is ] .
the third - row panels [ panels ( 1c ) and ( 3c ) ] of fig .
[ fig_5 ] correspond to the shaped pulse with @xmath219 in the detuned @xmath96 cases of 773.5 nm and 780 nm ( the corresponding case with @xmath96=777 nm is actually the one already considered in sec .
[ subsec2:discussion ] ) .
it is considered here as the representative case for the region of @xmath133 , where a systematic attenuation or enhancement of the intermediate - field tl - normalized absorption relative to the weak - field tl - normalized absorption occur [ figs .
[ fig_3](1c ) and ( 3c ) , figs .
[ fig_4](a ) and ( c ) ] .
the attenuation or enhancement correspond , respectively , to the blue ( @xmath96=773.5 nm ) or red detuning ( @xmath96=780 nm ) of @xmath40 from @xmath114 .
detailed analysis of the present case , as conducted above for the tl pulse and for the shaped pulse with @xmath220 , leads to the conclusion that the only difference here is that @xmath221 while @xmath222 .
all the other qualitative conclusions regarding the different amplitude components in terms of their magnitude , sign , and real / imaginary character are exactly the same for both the @xmath219 and @xmath122 pulses . in general , the ratio @xmath223 between the intermediate - field and weak - field tl - normalized absorption corresponding to the shaped pulse of a given @xmath97 is given by @xmath224/a_{4s,\textrm{step@}\omega_{step}}^{(2)}\right|^{2}+ \left|\re\left[a_{4s,\textrm{step@}\omega_{step}}^{(4)}\right]/a_{4s,\textrm{step@}\omega_{step}}^{(2)}\right|^{2 } } { \left|1 + \im\left[a_{4s,\textrm{tl}}^{(4)}\right]/a_{4s,\textrm{tl}}^{(2)}\right|^{2}+ \left|\re\left[a_{4s,\textrm{tl}}^{(4)}\right]/a_{4s,\textrm{tl}}^{(2)}\right|^{2 } } \nonumber \\ & = & \frac{\left|1 + k_{\textrm{step@}\omega_{step } } i_{0 } \right|^{2 } + \left| \kappa i_{0 } \right|^{2 } } { \left|1 + k_{\textrm{tl } } i_{0 } \right|^{2 } + \left| \kappa i_{0 } \right|^{2 } } \nonumber \ ; , \end{aligned}\ ] ] where the pulse shape dependence enters only via the @xmath203 factors .
as can be seen , there is a close similarity between this equation and eq .
( [ eq : r ] ) .
so , based on the above analysis conclusions , one gets the observed intermediate - field behavior for @xmath225 : ( i ) for blue detuning of @xmath170 ( @xmath96=773.5 nm ) : @xmath226 , @xmath207 , @xmath227 , and thus @xmath228 ( attenuation ) ; ( ii ) for red detuning of @xmath174 ( @xmath96=780 nm ) : @xmath229 , @xmath214 , @xmath227 , and thus @xmath230 ( enhancement ) . also , for both @xmath96 values , the value of @xmath231 increases as @xmath31 increases . in the red detuning case ( @xmath96=780 nm ) , this increase leads eventually to an intermediate - field absorption that exceeds the intermediate - field tl absorption [ fig .
[ fig_3](3c ) ] .
it worth emphasizing that the intermediate - field attenuation / enahncement effect considered here is relative to the weak - field case of the same shaped pulse , while for the pulse with @xmath122 it is considered above relative to the intermediate - field tl absorption . as discussed above with regard to fig .
[ fig_4 ] , the intermediate - field dip feature at @xmath134 ( 781.2 nm ) originates from the inclusion of the raman term @xmath68 in the complete raman term @xmath194 [ eq . (
[ eq8:amp-4th - order])-([eq10:amp-4th - order ] ) ] .
as described above , @xmath68 interferes only those raman parts that are resonance - mediated via @xmath89 .
each of them is either on resonance or near resonance with @xmath89 ( @xmath69 is the corresponding detuning ) .
so , overall , each corresponding four - photon pathway is either on- or near - resonance with @xmath87 and either on- or near - resonance with @xmath89 .
the higher is the field amplitude @xmath232 at @xmath136 the more prominent is the dip .
so , here , it is most prominent for @xmath96=780 nm [ fig .
[ fig_4](c ) ] . with the inclusion of @xmath68 ,
the terms @xmath183 and @xmath140 become complex quantities , and thus they both contribute to both the imaginary and real parts of @xmath137 .
so , the analysis of the corresponding interference mechanism is much more complicated as compared to the above , when @xmath68 is excluded .
essentially , the interference mechanism leading to the dip is very similar to the one we recently identified in weak - field resonance - mediated ( 2 + 1 ) three - photon absorption @xcite , where a @xmath3-step at @xmath136 leads to a strong enhancement in the population transfer to the @xmath89 state . here
, for example , a @xmath3-step at @xmath136 leads to a constructive add - up within the @xmath58-integral of @xmath183 [ eq .
( [ eq7.3:amp-4th - order ] ) ] between the amplitudes contributed by the four - photon pathways on - resonant with @xmath89 that are of positive detuning @xmath158@xmath147 and of negative detuning @xmath156@xmath147 from @xmath87 .
the corresponding raman transition involves the absorption of the photon @xmath233 and the emission of the photon @xmath136 .
since @xmath234 and @xmath235 are of opposite signs for the shaped pulse of @xmath134 , the sign of the detuning @xmath53 ( and of the integrand factor @xmath145 ) becomes correlated with the sign of @xmath236 , leading to the constructive add - up and resulting amplitude enhancement .
for the tl pulse , the add - up is destructive since the signs of @xmath235 and @xmath234 are the same . upon a detailed analysis , accounting also for the @xmath53-dependence of @xmath237 when @xmath151 ( 781.2 nm ) ( very similar to the one when @xmath238=780 nm [ fig .
[ fig_5](3c ) ] ) , one obtains that , for the present na excitation , the sign of the amplitude contributed to @xmath137 by the four - photon pathways with non - resonant raman parts [ i.e. , corresponding to @xmath68 ] is opposite to the sign of the amplitude contributed by those with resonance - mediated raman parts [ i.e. , corresponding to @xmath68 ] .
this leads to a dip feature rather than a peak feature .
in conclusion , coherent control of femtosecond two - photon absorption in the intermediate - field regime in analyzed in detail using a powerful frequency domain description that is based on 4@xmath0-order perturbation theory .
the two - photon absorption is coherently induced by non - resonant two - photon transitions as well as by four - photon transitions that introduce a resonance - mediated nature to the excitation .
their relative contributions to the total absorption amplitude depend on the field strength .
the corresponding interference mechanism is identified to include intra - group and inter - group interferences involving these two groups of multiphoton transitions .
the inter - group interferences lead a difference between the intermediate - field and weak - field absorption dynamics .
their constructive / destructive nature is found to depend on the detuning direction of the pulse spectrum from half the two - photon transition frequency ; it changes upon a red - to - blue detuning change .
the extended frequency - domain description and its detailed understanding serve as a basis for femtosecond control with rationally - shaped pulses in a regime of significant absorption yields , reaching population transfer in the range of 10 - 40% ( depending on the specific system and excitation scheme ) .
they also serve as a basis for future extensions to molecular systems , to other types of multiphoton processes , and to more complicated excitation schemes .
this research was supported by the israel science foundation ( grant no .
127/02 ) , by the james franck program in laser matter interaction , and by the technion s fund for the promotion of research .
the transition frequencies and dipole matrix elements are based on nist atomic spectra database ( national institute of standards and technology , gaithersburg , md ) , available at http://physics.nist.gov/asd to @xmath264 .
the four - photon transitions involve three absorbed photons and one emitted photon in any possible order , and thus can be decomposed into two parts : a non - resonant two - photon transition and a raman transition
. the border line between these two parts can be either on - resonance or near - resonance with @xmath86 or @xmath87 ( with detuning @xmath53 ) .
the raman transition itself can be non - resonant due to the @xmath265 states ( except for @xmath89 ) or on / near - resonance with @xmath89 ( with detuning @xmath69 ) . , width=642 ] -step spectral phase pattern .
the results include an integration over the experimental spatial beam profile .
the traces show the final @xmath87 population @xmath112 as a function of the step position @xmath97 .
the value of @xmath112 is normalized by the final population @xmath113 excited by the corresponding transform - limited ( tl ) pulse .
each column corresponds to a different central spectral wavelength @xmath96=773 , 777 , and 779.5 nm with a different pulse energy in each row ( increasing from top to bottom ) .
each pulse energy corresponds to a different transform - limited intensity at the peak of the spatial beam profile @xmath105 .
, width=755 ] -order perturbative theoretical results ( black thick lines ) for the two - photon absorption in na induced by the shaped pulses having a @xmath3-step spectral phase pattern . as a 4@xmath0-order calculations ,
the perturbative calculations include contributions from the 2@xmath1 and 4@xmath0 orders .
the traces show the final @xmath87 population @xmath112 as a function of the step position @xmath97 .
the right - hand y - axis scale is the absolute value of @xmath112 .
the left - hand y - axis scale corresponds to @xmath112 normalized by the final population @xmath113 excited by the corresponding transform - limited ( tl ) pulse .
each column corresponds to a different central spectral wavelength @xmath96=773.5 , 777 , and 780 nm with a different ( single - valued ) spectral intensity @xmath31 in each row ( increasing from top to bottom ) .
each @xmath31 corresponds to a different transform - limited peak intensity @xmath34 .
the weak - field traces are given in the first - row panels [ ( a)-panels ] .
for comparison , they are also shown in all the other panels ( black thin lines ) on the tl - normalized scale . ,
width=774 ] order ) for the two - photon absorption in na induced by the shaped pulses having a @xmath3-step spectral phase pattern .
the traces show the final @xmath87 population @xmath112 as a function of the step position @xmath97 .
the value of @xmath112 is normalized by the final population @xmath113 excited by the corresponding transform - limited ( tl ) pulse .
each panel corresponds to a different central spectral wavelength @xmath96=773.5 , 777 , and 780 nm .
the traces given in gray thick lines are the real traces ( shown also in fig . [ fig_3 ] ) at @xmath118 w/@xmath119 , which is the upper intensity limit of the present intermediate - field regime .
the traces given in black thick lines ( without the intermediate - field dip feature at 781.2 nm ) are traces that have been calculated at @xmath118 w/@xmath119 with artificially setting to zero the contribution from the resonance - mediated raman transitions via @xmath89 , i.e. , @xmath266 ( see text ) . for completeness , also presented are the real weak - field traces ( black thin lines ; also shown in fig .
[ fig_3 ] ) . , width=377 ] -order perturbative description .
they are discussed in the text for explaining the different intermediate - field features .
each panel shows the raman term @xmath153 ( gray thick lines ) and the two - photon transition term @xmath267 ( black thick lines ) as a function of the detuning @xmath53 ( see text ) .
the detuning values ( x - axis scale ) are actually given as the normalized values @xmath162 , with @xmath163 being the bandwidth of the intensity spectrum of the pulse .
the zone of small @xmath147 around @xmath149 , which is the most contributing to @xmath55 ( see text ) , is indicated schematically ( dashed area ) .
each column corresponds to a different central spectral wavelength @xmath96=773.5 , 777 , and 780 nm .
each row corresponds to a different pulse shape : ( a ) transform - limited ( tl ) pulse , ( b ) shaped pulse with @xmath160 ( 777 nm ) , and ( c ) shaped pulse with @xmath161 ( the corresponding central spectral frequency ) . for comparison
, each panel also shows the @xmath267 term of the corresponding tl pulse ( black thin lines ) .
, width=744 ] | coherent control of femtosecond two - photon absorption in the intermediate - field regime is analyzed in detail in the powerful frequency domain using an extended 4@xmath0-order perturbative description .
the corresponding absorption is coherently induced by the weak - field non - resonant two - photon transitions as well as by four - photon transitions involving three absorbed photons and one emitted photons .
the interferences between these two groups of transitions lead to a difference between the intermediate - field and weak - field absorption dynamics .
the corresponding interference nature ( constructive or destructive ) strongly depends on the detuning direction of the pulse spectrum from half the two - photon transition frequency .
the model system of the study is atomic sodium , for which both experimental and theoretical results are obtained .
the detailed understanding obtained here serves as a basis for coherent control with rationally - shaped femtosecond pulses in a regime of sizable absorption yields . |
in the last years it has been largely developed the study of intrinsic submanifolds inside the heisenberg groups @xmath3 or more general carnot groups , endowed with their carnot - carathodory metric structure , also named sub - riemannian . by an intrinsic regular ( or intrinsic lipschitz ) hypersurfaces we mean a submanifold which has , in the intrinsic geometry of @xmath3 , the same role like a @xmath4 ( or lipschitz ) regular graph has in the euclidean geometry .
intrinsic regular graphs had several applications within the theory of rectifiable sets and minimal surfaces in cc geometry , in theoretical computer science , geometry of banach spaces and mathematical models in neurosciences , see @xcite , @xcite and the references therein .
we postpone complete definitions of @xmath3 to section [ s : hnrecalls ] .
we only remind that the heisenberg group @xmath5 is the simplest example of carnot group , endowed with a left - invariant metric @xmath6 ( equivalent to its carnot - carathodory metric ) , not equivalent to the euclidean metric .
@xmath3 is a ( connected , simply connected and stratified ) lie group and has a sufficiently rich compatible underlying structure , due to the existence of intrinsic families of left translations and dilations and depending on the horizontal vector fields @xmath7 .
we call intrinsic any notion depending directly by the structure and geometry of @xmath3 . for a complete description of carnot groups @xcite
are recommended .
as we said , we will study intrinsic submanifolds in @xmath3 .
an intrinsic regular hypersurface @xmath8 is locally defined as the non critical level set of an horizontal differentiable function , more precisely there exists locally a continuous function @xmath9 such that @xmath10 and the intrinsic gradient @xmath11 exists in the sense of distributions and it is continuous and non - vanishing on @xmath12 .
intrinsic regular hypersurfaces can be locally represented as @xmath13-graph by a function @xmath14 , where @xmath15 , through an implicit function theorem ( see @xcite ) . in @xcite the parametrization
@xmath16 has been characterized as weak solution of a system of non linear first order pdes @xmath17 , where @xmath18 and @xmath19 , ( see theorem [ miotheorem ] ) . by an intrinsic point of view , the operator @xmath20 acts as the intrinsic gradient of the function @xmath21 . in particular it can be proved that @xmath16 is a continuous distributional solution of the problem @xmath17 with @xmath22 if and only if @xmath16 induces an intrinsic regular graph , ( see @xcite ) .
let us point out that an intrinsic regular graph can be very irregular from the euclidean point of view : indeed , there are examples of intrinsic regular graphs in @xmath23 which are fractal sets in the euclidean sense ( @xcite ) .
the aim of our work is to characterize intrinsic lipschitz graphs in terms of the intrinsic distributional gradient .
it is well - know that in the euclidean setting a lipschitz graph @xmath24 , with @xmath25 can be equivalently defined * by means of cones : there exists @xmath26 such that @xmath27 for each @xmath28 , where @xmath29 ; * in a metric way : there exists @xmath26 such that @xmath30 for every @xmath31 ; * by the distributional derivatives : there exist the distributional derivatives @xmath32 provided that @xmath33 is a regular connected open bounded set .
intrinsic lipschitz graphs in @xmath3 have been introduced in @xcite , by means of a suitable notion of intrinsic cone in @xmath3 . as consequence
, the metric definition ( see definition [ d : deflip ] ) is given with respect to the the graph quasidistance @xmath34 , ( see ) i.e the function @xmath35 is meant lipschitz in classical metric sense .
this notion turns out to be the right one in the setting of the intrinsic rectifiability in @xmath3 .
indeed , for instance , it was proved in @xcite that the notion of rectifiable set in terms of an intrinsic regular hypersurfaces is equivalent to the one in terms of intrinsic lipschitz graphs .
we will denote by @xmath36 the class of all intrinsic lipschitz function @xmath37 and by @xmath38 the one of locally intrinsic lipschitz functions .
notice that @xmath36 is not a vector space and that @xmath39 where @xmath40 and @xmath41 denote respectively the classes of euclidean lipschitz and @xmath42-hlder functions in @xmath33 .
for a complete presentation of intrinsic lipschitz graphs @xcite are recommended .
the first main result of this paper is the characterization of a parametrization @xmath37 of an intrinsic lipschitz graph as a continuous distributional solution of @xmath17 , where @xmath43 .
[ t : firsttheorem ] let @xmath44 be an open set , @xmath45 be a continuous function and @xmath46 .
@xmath47 if and only if there exists @xmath48 such that @xmath16 is a distributional solution of the system @xmath17 in @xmath33 .
we stress that this is indeed different from proving a rademacher theorem , which is more related to a pointwise rather than distributional characterization for the derivative , see @xcite .
nevertheless , we find that the density of the ( intrinsic ) distributional derivative is indeed given by the function one finds by rademacher theorem .
we also stress that there are a priori different notions of _ continuous _ solutions @xmath49 to @xmath17 , which express the lagrangian and eulerian viewpoints .
they will turn out to be equivalent descriptions of intrinsic lipschitz graphs , when the source @xmath1 belongs to @xmath50 .
this is proved in section [ s : furthereq ] and it is summarized as follows .
[ t : othertheorem ] let @xmath51 be a continuous function .
the following conditions are equivalent 1 .
[ item : distr ] @xmath16 is a distributional solution of the system @xmath17 with @xmath52 ; 2 .
[ item : lagr ] @xmath16 is a broad solution of @xmath17 , i.e. there exists a borel function @xmath53 s.t . *
@xmath54 @xmath55-a.e .
@xmath56 ; * for every continuous vector field @xmath57 ( @xmath58 ) having an integral curve @xmath59 , @xmath60 is absolutely continuous and at points @xmath61 of differentiability it satisfies @xmath62 in the statement , @xmath63 denotes the set of measurable bounded functions from @xmath64 to @xmath65 , while @xmath50 denotes the equivalence classes of lebesgue measurable functions in @xmath63 which are identified when differing on a lebesgue negligible set .
we will keep this notation throughout the paper : its relevance is remarked by examples [ ex:1 ] , [ ex:2 ] below .
[ [ outline - of - the - proofs ] ] * outline of the proofs * + + + + + + + + + + + + + + + + + + + + + + + with the intention of focussing on the nonlinear field , we fix the attention on the case @xmath66
. the variables will be denoted by @xmath67 and @xmath68 , and the subscripts @xmath69_{t}$ ] , @xmath69_{z}$ ] will denote the distributional derivatives @xmath70 , @xmath71 in the euclidean sense w.r.t . these variables .
given a continuous distributional solution @xmath72 of the pde @xmath73_{t}=w(z , t)\qquad { \rm for}\ ( z , t)\in { { \omega}},\ ] ] we first prove that it is lipschitz when restricted along any characteristic curve @xmath74 , where @xmath75 .
the proof follows a previous argument by dafermos ( see lemma [ l : daflipschitz ] ) . by a construction based on the classical existence theory of odes with continuous coefficients
, we can then define a change of variable @xmath76 which straightens characteristics .
this change of variables does not enjoy @xmath77 or lipschitz regularity , it fails injectivity in an essential way , though it is continuous and we impose an important monotonicity property .
this monotonicity , relying on the fact that we basically work in dimension @xmath78 , is the regularity property which allows us the change of variables .
as we exemplify below , we indeed have an approach different from providing a regular lagrangian flow of ambrosio - di perna - lions s theory , and it is essentially two dimensional . after the change of variables , the pde is , roughly , linear , and we indeed find a family of odes for @xmath2 on the family of characteristics composing @xmath79 , with coefficients which now are not anymore continuous , but which are however bounded . by generalizing a lemma on odes already present in @xcite , we prove the @xmath42-hlder continuity of @xmath16 on the vertical direction ( @xmath80 constant ) , and a posteriori in the whole domain .
this are the main ingredients for establishing that @xmath2 defines indeed a lipschitz graph : given two points , we connect them by a curve made first by a characteristic curve which joins the two vertical lines through the points , then by the remaining vertical segment .
we manage this way to control the variation of @xmath2 between the two points with their graph distance @xmath81 , checking therefore the metric definition of intrinsic lipschitz graphs .
the other implication of theorem [ t : firsttheorem ] is based on the possibility of suitably approximating an intrinsic lipschitz graph with intrinsic regular graphs .
a geometric approximation is provided by @xcite .
we also provide a more analytic , and weaker , approximation as a byproduct of the change of variable @xmath79 which straightens characteristics , by mollification ( see the proof of theorem [ t : conversedafermos ] ) .
we stop now for a while in order to clarify the features of the statement in theorem [ t : othertheorem ] , and why it is so important to differentiate between @xmath63 and @xmath50 .
lagrangian formulations are affected by altering the representative , as the following example stresses .
[ ex:1 ] let @xmath82 and let @xmath83 be the functions defined as @xmath84 then it is easy to verify that @xmath16 is a continuous distributional solution of @xmath85 consider the specific characteristic curve @xmath86 .
even if @xmath87 , the derivative of @xmath2 along this characteristic curve is not the right one : @xmath88 equation holds however on every characteristic curves provided we choose correctly an @xmath89-representative @xmath90 of the source @xmath1 : it is enough to consider @xmath91 notice that @xmath92 for @xmath93-a.e . @xmath94.@xmath95
( -1.5,0 )
( 5.5,0 ) node[above ] @xmath67 ; ( 2,0 )
( 2,5 ) node[right ] @xmath80 ; ( -1 , -1 ) rectangle ( 5 , 5 ) ; in .25,.75, ... ,3.25 plot[domain=:4 ] ( 1/4*(-)^2 , ) ; in 0,.5, ... ,3.25 plot[domain=:4 ] ( 1/4*(-)^2 , ) ; in -3.25,-2.75, ...
,-0.25 plot[domain=0:4 ] ( 1/4*(-)^2 , ) ; in -3,-2.5, ...
,-0.25 plot[domain=0:4 ] ( 1/4*(-)^2 , ) ; in .25,.75, ... ,3.25 plot[domain=:4 ] ( -1/4*(-)^2,-+4 ) ; in 0,.5, ... ,3.25 plot[domain=:4 ] ( -1/4*(-)^2,-+4 ) ; in -3.25,-2.75, ...
,-0.25 plot[domain=0:4 ] ( -1/4*(-)^2,-+4 ) ; in -3,-2.5, ...
,-0.25 plot[domain=0:4 ] ( -1/4*(-)^2,-+4 ) ; ( -1.5,-2.5 ) ( 5.5,-2.5 ) node[above ] @xmath67 ; ( 2,-2.5 ) ( 2,-1 ) node[right ] @xmath96 ; plot[domain=0:1.25 ] ( ^2,-2.5 ) ; plot[domain=-1.25:0 ] ( ^2,--2.5 ) ; ( 2,0)(2,4 ) ; ( -1.5,0 ) ( 5.5,0 ) node[above ] @xmath67 ; ( 2,0 ) ( 2,5 ) node[right ] @xmath80 ; ( -1 , -1 ) rectangle ( 5 , 5 ) ; in 0,0.25, ... ,3.25 plot[domain=:4 ] ( 1/4*(-)^2 , ) ; in -3.25,-3, ...
,-0.25 plot[domain=0:4 ] ( 1/4*(-)^2 , ) ; in 0,0.25, ... ,3.25 plot[domain=:4 ] ( -1/4*(-)^2 , ) ; in -3.25,-3, ...
,-0.25 plot[domain=0:4 ] ( -1/4*(-)^2 , ) ; ( -1.5,-2.5 ) ( 5.5,-2.5 ) node[above ] @xmath67 ; ( 2,-2.5 ) ( 2,-1 ) node[right ] @xmath96 ; plot[domain=0:1.25 ] ( ^2,-2.5 ) ; plot[domain=-1.25:0 ] ( ^2,-2.5 ) ; ( -1.5,0 ) ( 5.5,0 ) node[above ] @xmath67 ; ( 2,0 ) ( 2,5 ) node[right ] @xmath80 ; ( -1 , -1 ) rectangle ( 5 , 5 ) ; in 0,0.25, ... ,3.25 plot[domain=:4 ] ( 1/4*(-)^2,-+4 ) ; in -3.25,-3, ... ,-0.25 plot[domain=0:4 ] ( 1/4*(-)^2,-+4 ) ; in 0,0.25, ... ,3.25 plot[domain=:4 ] ( -1/4*(-)^2,-+4 ) ; in -3.25,-3, ...
,-0.25 plot[domain=0:4 ] ( -1/4*(-)^2,-+4 ) ; ( -1.5,-2.5 ) ( 5.5,-2.5 ) node[above ] @xmath67 ; ( 2,-2.5 ) ( 2,-1 ) node[right ] @xmath96 ; plot[domain=0:1.25 ] ( ^2,--2.5 ) ; plot[domain=-1.25:0 ] ( ^2,--2.5 ) ; before outlining theorem [ t : othertheorem ] we exemplify other features mentioned above by similar examples .
[ ex:2 ] let @xmath82 , and choose @xmath97 again @xmath98 one easily sees that characteristics do collapse in an essential way .
considering instead @xmath99 characteristics do split in an unavidable way .
therefore , while it is proved in @xcite that in example [ ex:1 ] one can choose for changing variable a flux which is better than a generic other_the regular lagrangian flow_we are not always in this case .
this is the main reason why we refer in our first change of variables to a monotonicity property .
we have now motivated the further study for the stronger statement of theorem [ t : othertheorem ] . in order to prove it ,
we consider also the weaker concept of _ lagrangian solution _ : the idea is that the reduction on characteristics is not required on _ any _ characteristic , but on a set of characteristics composing the change of variables @xmath79 that one has chosen .
exhibiting a suitable set of characteristics for the change of variables @xmath79 is part of the proof .
roughly , an @xmath100-representative @xmath101 for the source of the odes related to @xmath79 is provided by taking the @xmath80-derivative of @xmath102 , which is @xmath2 evaluated along the characteristics of the lagrangian parameterisation ; by construction , it coincides with the second @xmath68-derivative of @xmath103 .
here we have hidden the fact that we need to come back from @xmath104 to @xmath105 , a change of variable that is not single valued , and not surjective since the second derivative was defined only almost everywhere .
we overcome the difficulty choosing any value of the second derivative when present , and showing that it suffices .
however , if one changes the set of characteristics in general one arrives to a different function @xmath106 .
we have called _ broad solution _ a function which satisfies the reduction on every characteristic curve . in order to have this stronger characterization
, we give a different argument borrowed from @xcite .
we define a universal source term @xmath107 in an abstract way , by a selection theorem , at each point where there exists a characteristic curve with second derivative , without restricting anymore to a fixed family of characteristics providing a change of variables . after showing that this is well defined
, we have provided a universal representative of the intrinsic gradient of @xmath2 . in cases as examples
[ ex:1 ] , [ ex:2 ] it extends the one , defined only almost everywhere , provided by rademacher theorem .
[ [ outline - of - the - paper ] ] * outline of the paper * + + + + + + + + + + + + + + + + + + + + + + the paper is organized as follows . in section [
s : hnrecalls ] we recall basic notions about the heinsenberg groups . in section [ s :
pderecalls ] we fix instead notations relative to the pde , mainly specifying the different notions of solutions we will consider .
one of them will involve a change of variables , for passing to the lagrangian formulation , which is mainly matter of section [ s : existencepartialpar ] and it is basically concerned with classical theory on odes .
then appendix [ s : extpar ] also explains how to extend a partial change of variables of that kind to become surjective , and provides a counterexample to its local lipschitz regularity ; it is improved in @xcite showing that it is not in general of bounded variation . in section
[ s : equivalence ] we prove the equivalence among the facts that either a continuous function @xmath2 describes a lipschitz graph or it is a distributional solution to the pde @xmath108 , @xmath109 .
the further equivalencies are finally matter of section [ s : furthereq ] . with some simplification
, we can illustrate the main connections by the following papillon . as mentioned above
, there is also a connection with the existence of smooth approximations ( see the proof of theorem [ t : conversedafermos ] by mollification and @xcite with a more geometric procedure ) .
@xmath110 & { \text{def.~\ref{d : lagrsol } } } \ar@{.}[dd ] & { \text{def.~\ref{d : broadsol } } } \ar@{.}[d ] \\ * + [ f]{\text{continuous distributional } } \ar@/^0pc/[rr]_{\text{th.\ref{t : univselection}+l.\ref{l : daflipschitz } } } \ar@/^1pc/[dr]_{\text{cor.\ref{c : lipordistrlagr } ( n=1 ) } } \ar@<1ex>[dd]^{\text{l.\ref{l : distrlip } } } & & * + [ f]{\text{continuous broad } } \ar@/^1pc/[dl]^{\text{l.\ref{l : globchi } } } \ar@/^1pc/@{.>}@<3ex>[dd]_{\text{clear } } \\ & * + [ f]{\text{continuous lagrangian } } \ar@/^1pc/[ul]^{\text{cor.\ref{t : conversedafermos } } } \ar@/^1pc/[ur]_{\text{th.\ref{t : univselection } } } \ar@{.>}[dr]^{\text{clear } } \\ * + [ f]{\text{intrinsic lipschitz } } \ar@{->}[ur]|{\text{cor.\ref{c : lipordistrlagr } ( n=1 ) } } \ar@/^1pc/[uu]^{\text{l.\ref{l : lipdistr } } } & & * + [ f]{\text{continuous broad * } } \ar@/^1pc/@{.>}[ul]^{\text{l.\ref{l : globchi } } } \\ { \text{def.~\ref{d : distribsol } } } \ar@{.}[u ] & & { \text{def.~\ref{d : broad*sol } } } \ar@{.}[u ] } \ ] ] * a remark about the identification of the source terms . *
the intrinsic gradient is unique .
suppose @xmath2 is both intrinsic lipschitz and a broad solution : then by definition at points of intrinsic differentiability the intrinsic gradient must coincide with @xmath107 .
as a broad solution is also a lagrangian solution , at those points it must coincide also with @xmath111 . by rademacher theorem ,
these functions are therefore identified lebesgue almost everywhere , and by lemma [ l : lipdistr ] they are representative of the @xmath50-function which is the source term of the balance law .
+ we conclude remarking again that this identification lebesgue almost everywhere is not enough for fixing them , as both lagrangian solutions and broad solutions require the right representative of the source also on some of the points of non - intrinsic differentiability , not captured by the source term in the distributional formulation ( fig .
[ fig:1 ] ) . by rademacher theorem , instead , we have that the source term in the broad formulation not only fixed dirctly the source term in the lagrangian formulation , but also in the distributional one
. actually , it suffices the source of a lagrangian formulation in order to fix directly the one in the distributional one .
_ acknowledgements .
_ we warmly thank stefano bianchini and giovanni alberti for useful discussions and important suggestions , in particular on the subject of section [ s : furthereq ] .
* definition : a noncommutative lie group .
* we denote the points of @xmath112 by @xmath113 if @xmath114 , @xmath115 and @xmath116 , the group operation reads as @xmath117 the group identity is the origin 0 and one has @xmath118^{-1}=[-x ,- t]$ ] . in @xmath119
there is a natural one parameter group of non isotropic dilations @xmath120 $ ] , @xmath116 .
the group @xmath121 can be endowed with the homogeneous norm @xmath122 and with the left - invariant and homogeneous distance @xmath123 the metric @xmath124 is equivalent to the standard carnot - carathodory distance .
it follows that the hausdorff dimension of @xmath125 is @xmath126 , whereas its topological dimension is @xmath127 .
the lie algebra @xmath128 of left invariant vector fields is ( linearly ) generated by @xmath129 and the only nonvanishing commutators are @xmath130 = t , \qquad j=1,\dots n.\ ] ] * horizontal fields and differential calculus .
* we shall identify vector fields and associated first order differential operators ; thus the vector fields @xmath131 generate a vector bundle on @xmath3 , the so called _ horizontal _ vector bundle @xmath132 according to the notation of gromov ( see @xcite ) , that is a vector subbundle of @xmath133 , the tangent vector bundle of @xmath3 . since each fiber of @xmath132 can be canonically identified with a vector subspace of @xmath134 , each section @xmath135 of @xmath132
can be identified with a map @xmath136 . at each point
@xmath137 the horizontal fiber is indicated as @xmath138 and each fiber can be endowed with the scalar product @xmath139 and the associated norm @xmath140 that make the vector fields @xmath131 orthonormal . a real valued function
@xmath141 , defined on an open set @xmath142 , is said to be of class @xmath143 if @xmath144 and the distribution @xmath145 is represented by a continuous function .
we shall say that @xmath146 is an _
@xmath147-regular hypersurface _
if for every @xmath148 there exist an open ball @xmath149 and a function @xmath150 such that * @xmath151 ; * @xmath152 .
the _ horizontal normal _ to @xmath12 at @xmath153 is @xmath154 [ notation ] recalling that we denote a point of @xmath3 as @xmath155 , where @xmath156 , @xmath157 , it is convenient to introduce the following notations :
* we will denote a point of @xmath158 as @xmath159 , where @xmath160 , @xmath157 . if @xmath161 , for given @xmath162 , we denote @xmath163 .
when we will use this notation , we also denote a point @xmath164 as @xmath165 .
* let @xmath166 , for given @xmath167 and @xmath157 we will denote @xmath168 * let @xmath166 , for given @xmath169 we will denote@xmath170 * let @xmath166 , for given @xmath167 we will denote @xmath171 * @xmath172 .
a point @xmath173 will be identified with the point @xmath174 .
* let @xmath175 , for given @xmath176 we will denote @xmath177 .
a set @xmath8 is an @xmath13-graph if there is a function @xmath178 such that @xmath179 let us recall the following results proved in @xcite .
[ dinitheorem ] let @xmath180 be an open set in @xmath3 , @xmath181 , and let @xmath182 be such that @xmath183 .
let @xmath184 then there exist a connected open neighborhood @xmath185 of @xmath186 and a unique continuous function @xmath187 $ ] such that @xmath188 , where @xmath189 and @xmath190 is the map defined as @xmath191 and given explicitly by @xmath192 let @xmath193 , @xmath194 and define @xmath195 < r^2,\,|t - t^0|<r\right \}\ , .
\end{array}\ ] ] when @xmath66 and @xmath196 let @xmath197 following @xcite we define the graph quasidistance @xmath81 on @xmath33 .
we set @xmath198 , @xmath199 [ fidistanzadef2708 ] for @xmath200 we define @xmath201 following the notations [ notation ] we have explicitly if @xmath202 @xmath203 where @xmath204 . if @xmath66 and @xmath205 we have @xmath206 an intrinsic differentiable structure can be induced on @xmath207 by means of @xmath34 , see @xcite .
we remind that a map @xmath208 is @xmath207-linear if it is a group homeomorphism and @xmath209 for all @xmath116 and @xmath210 .
we remind then the notion of @xmath211-differentiablility .
[ defiwfdiff ] let @xmath212 be a fixed continuous function , and let @xmath28 and @xmath213 be given .
* we say that @xmath214 is @xmath211-differentiable at @xmath215 if there is a unique @xmath207-linear functional @xmath216 such that @xmath217 * we say that @xmath214 is uniformly @xmath211-differentiable at @xmath215 if there is a unique @xmath147-linear functional @xmath216 such that @xmath218 we will denote @xmath219 . if @xmath16 is uniformly @xmath211-differentiable at @xmath215 , then @xmath16 is @xmath211-differentiable at @xmath215 .
in @xcite it has been proved that each @xmath147- regular graph @xmath220 admits an intrinsic gradient @xmath221 , in sense of distributions , which shares a lot of properties with the euclidean gradient . since @xmath222 , where the vector fields @xmath223 must be understood as restricted to @xmath207 , it is possible to define the differential operators given , in sense of distributions , by @xmath224 according with the notation introduced in [ notation ] , we also denote by @xmath225 the family of vector fields on @xmath226 defined by @xmath227
the following characterizations were proved in @xcite .
the definitions of broad * and distributional solution of the system @xmath17 are recalled in section [ s : pderecalls ] .
[ miotheorem ] let @xmath44 be an open set and let @xmath45 be a continuous function .
then 1 .
the set @xmath228 is an @xmath147-regular surface and @xmath229 for all @xmath148 , where @xmath230 is the horizontal normal to @xmath12 at @xmath153 .
is equivalent to each one of the following conditions : 1 .
there exists @xmath231 and a family @xmath232 such that , as @xmath233 , @xmath234 and @xmath235 in @xmath236 in sense of distributions .
there exists @xmath231 such that @xmath2 is a _ broad * solution _ of the system @xmath17 .
3 . there exists @xmath231 such that @xmath16 is a distributional solution of @xmath17 .
@xmath16 is uniformly @xmath211-differentiable at @xmath237 for all @xmath238 .
it follows that , for given @xmath37 such that @xmath239 is @xmath147-regular , then at any point @xmath56 the differetial @xmath240 can be represented in terms of the intrinsic gradient @xmath241 .
more precisely ( see @xcite ) @xmath242 [ d : gfderivative ] let @xmath37 be a continuous function and let @xmath56 be given .
we say that @xmath16 admits @xmath243-derivative if there exists @xmath244 such that for each integral curve @xmath245 of @xmath243 with @xmath246 @xmath247 and @xmath248 .
notice that if @xmath239 is an @xmath147-regular hypersurface then @xmath16 admits @xmath243-derivative at @xmath237 for every @xmath56 for @xmath162 ( see @xcite , @xcite ) .
* introduction to the concern of this paper .
* let us now introduce the concept of intrinsic lipschitz function and intrinsic lipschitz graph .
[ d : deflip ] let @xmath249 .
we say that @xmath16 is an intrinsic lipschitz continuous function in @xmath33 and write @xmath250 , if there is a constant @xmath26 such that latexmath:[\[\label{deflip } moreover we say that @xmath16 is a locally intrinsic lipschitz function in @xmath33 and we write @xmath252 if @xmath250 for every @xmath253 .
we remark that when @xmath2 is intrinsic lipschitz , then there exists @xmath254 such that @xmath255 in particular , the graph distance @xmath256 is also equivalent to the carnot - carathodory distance restricted to the corresponding points on the graph of the lipschitz intrinsic hypersurface .
this means that @xmath2 is lipschitz continuous also in the classical sense when evaluated on any fixed integral curve of the vector field @xmath257 , while it is @xmath42-hlder on the lines where @xmath67 is fixed . in @xcite
is proved the following characterization for intrinsic lipschitz functions .
[ t : tesipinamonti ] let @xmath175 be open and bounded , let @xmath37 .
then the following are equivalent : 1 .
@xmath252 2 .
there exist @xmath258 and @xmath259 such that @xmath260 there exists @xmath261 such that 1 .
@xmath262 uniformly converges to @xmath16 on the compact sets of @xmath33 ; 2 .
@xmath264 @xmath265 ; 3 .
@xmath266-a.e . @xmath238
. moreover if ( ii ) holds , then @xmath267 @xmath55-a.e . @xmath238 .
let us finally recall the following rademacher type theorem , proved in @xcite .
[ t : rademacher ] if @xmath268 then @xmath16 is @xmath211-differentiable for @xmath55-a.e @xmath56 .
even when @xmath269 , where @xmath64 is an open subset of @xmath270 , the equation @xmath271_{t}=w(z , t)\qquad{\rm in\ } { { \omega}}\ ] ] allows in general for discontinuous solution .
however , it is the case @xmath66 of the system @xmath272 where @xmath273 and this system , by theorem [ miotheorem ] , describes an @xmath147-regular surface @xmath228 which is an @xmath274-graph .
since we want to study in the present paper intrinsic lipschitz graphs , then we do not require anymore the continuity of @xmath1 but we allow @xmath52 .
notwithstanding that , the continuity of @xmath2 remains natural .
there are a priori different notions of _ continuous _ solutions @xmath49 .
we recall some of them in this section : distributional , lagrangian , broad , broad*. all of them will finally coincide . after giving in the present sections the definitions for all @xmath275
, we will focus in the next one the analysis on the non - linear equation in the case @xmath66 , which conveys the attention on the planar case .
we will remind this reduction by adopting often the variables @xmath276 instead of @xmath277 .
the generalization to other cases @xmath193 of most of the lemmas is straightforward , because the fields @xmath278 and @xmath279 are linear .
it is not basically in lemma [ l : gfunztx ] , where we prefer taking advantage of the continuity of @xmath79 ; however , we have no reason to prove it in full generality .
we recall that in general solutions are not smooth , even if we assume the continuity see e.g. example [ e : fillholesnogloblip ] below
. the equation is then interpreted in a distributional way .
[ d : distribsol ] a continuous function @xmath49 is a _ distributional solution _ to if for each @xmath280 @xmath281 and @xmath282 we consider now different versions for the lagrangian formulation of the pde .
the first one somehow englobes a choice of trajectories for passing from lagrangian to eulerian variables , and imposes the evolution equation on these trajectories .
[ d : lagrparam ] a family of _ partial lagrangian parameterisations _ associated to a continuous function @xmath49 and to the system is a family of couples @xmath283 ( @xmath58 ) with @xmath284 open sets and @xmath285 borel functions such that for each @xmath162 * the map @xmath286 , @xmath287 is valued in @xmath64 ; * for every @xmath288 , the function @xmath289 is nondecreasing ; * for every @xmath290 , for every @xmath291 , the function @xmath292 is absolutely continuous and @xmath293 we call it a family of _ ( full ) lagrangian parameterisations _ if @xmath294 is onto the section @xmath295 for all @xmath296 .
we remark again that we emphasized in this definition the nonlinear pde of the system : a lagrangian parameterisaiton provides a covering of @xmath64 by characteristic lines for that equation .
indeed , a covering by characteristic lines of the other equations is immediately given by an expression like @xmath297 moreover , the reduction along characteristics for the linear equations , and thus the equivalence between lagrangian and distributional solution , holds with less technicality .
a family of ( partial ) parameterisations @xmath298 extends the family of ( partial ) parameterisations @xmath299 , we denote @xmath300 , if there exists a family of borel injective maps @xmath301 when @xmath302 and @xmath303 they are called equivalent .
the notion of lagrangian parameterisation given above does not consist in a different formulation for the notion of regular lagrangian flow in the sense by ambrosio - di perna - lions ( see @xcite for an effective presentation ) .
particles are really allowed both to split and to join , therefore in particular the compressibility condition here is not required , while instead we have a monotonicity property .
when we need to distinguish letters , we denote with @xmath304 functions defined on @xmath64 but possibly related to a parameterisation , with @xmath305 functions defined on @xmath306 , and with @xmath307 functions defined on @xmath64 not related to specific parameterisations .
notice that a full lagrangian parameterisation is continuous in the two variables for free : indeed , e.g. for @xmath66 , by monotonicity one has that for each @xmath61 @xmath308 by the surjectivity then equality must hold .
considering the lipschitz continuity in the other variables one gains the joint continuity in @xmath276 .
the same holds for @xmath309 provided that @xmath79 is continuous on the hyperplane @xmath310 , since the argument above gives the continuity only on the planes where @xmath311 is constant .
we do not mind about continuity in @xmath311 . before giving the notion of lagrangian solution
, we recall that a set @xmath312 is _ universally measurable _ if it is measurable w.r.t . every borel measure .
universally measurable sets constitute a @xmath313-algebra , which includes analytic sets .
a function @xmath314 is said universally measurable if it is measurable w.r.t .
this @xmath313-algebra .
in particular , it will be measurable w.r.t .
any borel measure .
+ notice that borel counterimages of universally measurable sets are universally measurable .
then the composition @xmath315 of any universally measurable function @xmath135 with a borel function @xmath316 is universally measurable
. this composition would be nasty if @xmath317 were just lebesgue measurable .
+ since restrictions of borel functions on borel sets are borel , all the terms in the following definition are thus meaningful .
[ d : lagrsol ] a continuous function @xmath51 is a _
lagrangian solution _ of if there exists a family of lagrangian parameterisations @xmath298 ( @xmath162 ) , associated to @xmath2 and , and a family of universally measurable functions @xmath318
( @xmath162 ) , such that @xmath319 , @xmath320 , @xmath321 , it holds that * the function @xmath322 is absolutely continuous and @xmath323 where @xmath324 is given in ( l.1 ) of definition [ d : lagrparam ] . * if @xmath16 admits @xmath243-derivative at @xmath56 , then @xmath325 @xmath326 .
we are going to prove in section [ ss : lagrdistr ] that @xmath2 above is a distributional solution to provided that the function @xmath327 is a pointwise representative of the source term @xmath328 .
the fact that we do not require in the definition that @xmath329 and @xmath1 represent the same function is just a notational convenience for the proofs below .
we give now the strongest notion of solution : the evolution equation is imposed on every trajectories . [
d : broadsol ] a continuous function @xmath49 is a _ broad solution _ of if there exists a universally measurable function @xmath330 such that * @xmath331 @xmath55-a.e . in @xmath33 ; * @xmath326 , for every integral curve @xmath332 of the vector field @xmath57 it holds that @xmath333 is absolutely continuous and @xmath334 * if @xmath16 admits @xmath243-derivative at @xmath56 , then @xmath335 @xmath336 .
we also remind the intermediate notion of broad * solution introduced in @xcite .
[ d : broad*sol ] a continuous function @xmath2 is a _ broad * solution _ of if for every @xmath28 there exists a map , that we will call exponential map , @xmath337\times{i_{{{\delta}}}(a_0)}\rightarrow{i_{\delta_1}(a_0 ) } \subset\omega\,,\ ] ] where @xmath338 such that , if @xmath339 for @xmath162 , then * @xmath340;\,{{\mathbb r}}^{2n})}\,,$ ] * @xmath341 * there exists a universally measurable function @xmath342 such that @xmath343 @xmath55-a.e . in @xmath33 , @xmath326
the map @xmath344 is absolutely continuous and @xmath345 * if @xmath16 admits @xmath243-derivative at @xmath56 , then @xmath346 @xmath336 .
being a distributional solution to the pde , a lagrangian solution will be in particular a broad * solution with @xmath347 .
viceversa , if the exponential map related to a broad solution satisfies the relative monotonicity property , then one can prove that the broad * solution is also a lagrangian solution , with the same @xmath111 .
one can moreover derive a procedure for constructing a lagrangian parameterisation in lemma [ l : globchi ] , where the curves @xmath348 should be replaced by the ones of the exponential map .
as mentioned , we can focus in this section on the case @xmath66
. we will give information about the generalization to the case @xmath193 in remark [ r : remarkdimnsez4 ] .
given a continuous function @xmath2 on @xmath349 , we defined what we mean by ` lagrangian parameterisation ' . here
we show that the definition is non - empty .
it is part of the very classical theory on odes with continuous coefficients covering @xmath64 with integral curves of the continuous vector field @xmath350 , see for example the first chapters of @xcite . in the definition
we of lagrangian parameterisation it is however essential that we cover the region with curves which are ordered and not with generic ones : we therefore briefly remind below one way of doing it , with also the aim being self - contained .
firstly , in lemma [ l : solode ] we show , for familiarising with notations , that if one takes the integral curves through an @xmath61-section of @xmath64 which are minimal , in the sense that any other curve through that point lies on its right side , then we get a partial lagrangian parameterisation , as well as it happens for the maximal ones . extending it to a full one will be matter of the next section .
a full lagrangian parameterisation basically amounts to an order preserving parameterisation , with a _ real _ valued parameter , of non - crossing curves through each point of the plane .
we immediately give an example of a full one ( lemma [ l : globchi ] below ) .
notice that the definition of lagrangian parameterisation concerns only classical theory on odes with continuous coefficients , as therefore lemmas [ l : solode ] , [ l : globchi ] , [ l : gfunztx ] , [ l : holderode ] , [ l : fillinode ] .
[ l : solode ] let @xmath351 be a continuous function , @xmath352 .
then there are domains @xmath353 , @xmath354 associated to the functions @xmath355 for which @xmath356 , @xmath357 are partial lagrangian parameterisaiton relative to @xmath2 . for every
@xmath358 one could consider the minimal and maximal curves satisfying on @xmath359 the ode for characteristics and passing through that point : indeed the functions [ e : gammamm ] @xmath360 are well defined , lipschitz , and because of the continuity of @xmath2 they are still integral curves ( @xcite ) . denoting by @xmath361 the relative interior of a set
, we define the domain @xmath362 where recall that @xmath363 .
since we are assuming that @xmath64 contains the origin , @xmath364 is nonempty .
the domain @xmath365 is analogous .
being @xmath366 , @xmath367 lipschitz solutions to the ode with continuous coefficients , the functions @xmath368 in the statement are @xmath369 in the @xmath61 variable for every @xmath370 fixed .
@xmath371 are jointly borel in @xmath372 by continuity in @xmath61 and monotonicity in @xmath373 , monotonicity that now we show .
notice the semigroup property : for @xmath374 , for example for , @xmath375 this yields the monotonicity of @xmath376 for each @xmath61 fixed .
indeed , if @xmath377 and @xmath378 at a certain @xmath379 , by the continuity of the curves there exists @xmath380 when @xmath381 .
but then the curve @xmath382 is a good competitor for the definition of @xmath383 , which implies @xmath384 and therefore equality .
for @xmath385 the argument is similar . in lemma
[ l : fillinode ] we show how to make a partial parameterisation @xmath79 surjective : thus we cover @xmath64 by a family of characteristic curves which includes the ones of @xmath79 . in the following lemma , w.l.o.g . in a simpler setting , we provide instead a full lagrangian parameterisation , defined at once instead of extending a given one .
[ l : globchi ] there exists a full lagrangian parameterisation associated to a continuous function @xmath386 which is also continuous on the closure @xmath387^{2}$ ] .
step @xmath388 .
let @xmath2 be a continuous function on @xmath387^{2}$ ] , we want to give a lagrangian parameterisation for his restriction to @xmath389 as we defined it on an open set .
one can assume @xmath2 is compactly supported in @xmath390\times ( 0,1)$ ] .
if not , one can extend it to a compactly supported function @xmath391 on @xmath387\times ( -1,2)$ ] : restricting the lagrangian parameterisation @xmath392 for @xmath391 , defined as described below , to the open set @xmath393 one will get a lagrangian parameterisation for @xmath2 .
the assumption of @xmath394 compactly supported in @xmath395 implies that there are two characteristics , one starting from @xmath396 and one from @xmath395 , which satisfy @xmath397 .
this means respectively @xmath398 and @xmath399 for each @xmath400 $ ] .
step @xmath78 .
after this simplification , we associate to each point @xmath401^{2}$ ] a curve @xmath402 which is minimal forward in @xmath61 , maximal backward : @xmath403 where @xmath404 , @xmath405 were defined in .
notice that the curves @xmath406 are defined on the whole @xmath387 $ ] .
let us denote by @xmath407\to[0,1],\ ( \bar s,\bar t)\in[0,1]^2\}.\ ] ] we will endow @xmath408 by the topology of uniform convergence on @xmath387 $ ] and the following total order relation @xmath409.\ ] ] let us denote by @xmath410 the closure of @xmath411)$ ] endowed with the topology of uniform convergence .
we prove in the rest of this step the following claims : @xmath412 @xmath413 @xmath414 @xmath415 i.e. for each @xmath416 @xmath417 @xmath418 $ ] .
since @xmath408 is a family of equi - lipschitz continuous and bounded functions of @xmath419)$ ] , then , from arzel - ascoli s theorem @xmath410 is compact .
let @xmath420 and let us prove that @xmath421 or @xmath422 . by definition
there are two sequences @xmath423 such that @xmath424 and @xmath425 uniformly on @xmath387 $ ] .
assume @xmath426 , then there is @xmath427 $ ] such that @xmath428 for instance , let @xmath429 and let us prove that @xmath430.\ ] ] let @xmath431 , there exists @xmath432 such that @xmath433 because @xmath434 and @xmath435 are ordered , it follows that @xmath436 @xmath418 $ ] , @xmath437 . passing to the limit as @xmath438 in the previous inequality , we get .
proof of . by contradiction , suppose that @xmath439 , with @xmath440 and @xmath441 non empty , closed sets in @xmath419)$ ] .
it is well - know that , from and , for each subset @xmath442 there exists the least upper bound ( or supremum ) and greatest lower bound ( or infimum ) of @xmath443 ; we will denote respectively by @xmath444 , @xmath445 .
thus let @xmath446 because @xmath447 and @xmath448 are closed , @xmath449 and @xmath450 . since @xmath451 , we have @xmath452 or @xmath453 .
assume , for instance , that @xmath452 . then @xmath454 @xmath418 $ ] and @xmath455 for a suitable @xmath427 $ ] .
let @xmath456.\ ] ] by definition , @xmath457 , but @xmath458 and therefore we have a contradiction since @xmath459 .
let @xmath416 , then by definition there exists a sequence @xmath460 such that @xmath461 uniformly in @xmath387 $ ] . because @xmath462,\ \forall\ h,\ ] ] passing to the limit as @xmath438 in the previous identity we get .
step @xmath463 .
let us now consider the map @xmath464 defined by @xmath465 where @xmath466 is an enumeration of @xmath467 $ ] .
notice that @xmath468 satisfies the following properties : @xmath469 @xmath470 @xmath471;\ ] ] @xmath472\to\mathcal c^*\ \text{continuous}.\ ] ] indeed and immediately follow by the definition of @xmath468 .
equality follows by and ; is a consequence of , and . eventually let us consider the map @xmath473\times[0,2]\to[0,1]$ ] @xmath474\times[0,2].\ ] ] then , from , @xmath79 is continuous .
moreover the map @xmath475 , defined by @xmath476 , turns out to be a full lagrangian parameterisation associated to @xmath477 .
[ r : remarkdimnsez4]notice that the same construction works with more variables , considering analogous characteristic curves @xmath478having the same order relation at @xmath479 frozen and the relative parameterisation given by @xmath480 with @xmath481 defined as above .
the continuity in @xmath311 however is not guaranteed .
in this section we prove theorem [ t : firsttheorem ] , without dimensional restrictions .
preliminary we highlight here two properties of continuous distributional solutions @xmath96 to the problem @xmath482 . in particular we need regularity results of the solution along the characteristics lines @xmath483 of the fields @xmath484 . in the case of the non linear field @xmath257 ,
the integral curves of @xmath485 exist by the continuity and boundedness of @xmath2 .
[ l : daflipschitz ] let @xmath175 be an open set .
a continuous distributional solution @xmath37 to @xmath17 is @xmath486-lipschitz along any characteristic line @xmath487\to { { \mathbb r}}$ ] satisfying @xmath488,\ \hat z_n\ { \rm fixed}.\ ] ] in the same way in dafermos @xcite we obtain for @xmath489 and @xmath311 fixed @xmath490 @xmath491 ^ 2\,ds\leq0\ ] ] and then @xmath492 dividing by @xmath493 and getting to the limit to @xmath494 we obtain @xmath495 the opposite inequality is obtained in a similar way integrating on the left of the characteristic .
we obtain the same result of lemma [ l : daflipschitz ] for the linear fileds @xmath496 following the same proof .
[ l : daflipschitzbis ] let @xmath175 be an open set .
a continuous distributional solution @xmath37 to @xmath17 is @xmath497-lipschitz along any characteristic line @xmath498\to { { \omega}}$ ] satisfying @xmath499 we pass now to the hlder continuity in the other variable .
[ l : holderode ] let @xmath500 be such that of all @xmath501 there are @xmath487\to { { \mathbb r}}$ ] satisfying @xmath502\\ \gamma(0)=\tau\end{cases}\ ] ] and that @xmath503 then we have @xmath504 let us denote @xmath505 let us observe that @xmath506 by contradiction , let us assume there exist @xmath507 , @xmath508 such that @xmath509 @xmath510 the idea of the proof is the following : the lipschitz condition in the hypothesis is an upper bound on the second derivative of the mentioned curves @xmath483 .
therefore , if their first derivative wants to change it takes some time in @xmath61 .
if we assume that at @xmath511 the first derivative differs at two points @xmath512 at least of the ratio , then the the relative curves @xmath513 starting from those points must meet soon .
however , at the time they meet the first derivative did not have the time to change enough to become equal , providing an absurd .
let us introduce curves @xmath514 such that for @xmath515 , @xmath516 @xmath517 @xmath518 we observe that , by our lipschitz assumption , @xmath519 is a function bounded by @xmath520 .
therefore we can represent each @xmath521 for each @xmath522 $ ] as @xmath523 @xmath524\,.\ ] ] in particular by the second equality in , for @xmath525 , @xmath526 for @xmath527 $ ] .
by we get @xmath528 or @xmath529 let us prove now that if holds then there exists @xmath530 such that @xmath531 let @xmath532 then @xmath533 , \quad { \gamma_1(s^*)\leq\gamma_2(s^*)}\,.\ ] ] indeed by and the definition of @xmath534 , @xmath535 . on the other hand by ( with @xmath536 ) , gives @xmath537 then follows .
let @xmath538:\gamma_1(s)>\gamma_2(s)\}\ ] ] then by @xmath539 and it satisfies .
if holds we can repeat the argument reversing the @xmath61-axis getting that there exist @xmath540 such that still holds .
let us prove now that @xmath541 then a contradiction and the thesis will follow . indeed , for instance , let us assume .
then by and the bound on @xmath542 @xmath543 @xmath544 @xmath545 @xmath546\,.\ ] ] therefore we get that @xmath547 and follows .
we are now able to give the proof of theorem [ t : firsttheorem ] .
we distinguish the two different implications in the following two lemmas .
[ l : distrlip ] let @xmath175 be an open set . if @xmath16 is a distributional solution of @xmath17 , then @xmath252 .
let @xmath548 for a sufficiently small @xmath549 .
for @xmath202 let @xmath550 be the vector fields given by @xmath551 define @xmath552 observe that @xmath553 and @xmath554 are well defined .
for @xmath66 , @xmath555 is not defined and we set @xmath556 and @xmath557 .
we have to show that there exists @xmath26 such that @xmath558 @xmath559 , from lemma [ l : daflipschitz ] @xmath560 . by we
can apply lemma [ l : holderode ] and obtain @xmath561(z_n'-z_n)-2\int_0^{z_n'-z_n}{{\phi}}(\exp(s{w^\phi})(b^\ast)\,ds \bigr|\vspace{0.2cm}\\ = : & \displaystyle{{d_\phi}}(b',b)^2+r_1(b',b)+r_2(b',b)+r_3(b',b ) .
\end{array}\ ] ] for the case @xmath66 we arrive to with the same line ( it is sufficient to follow the same steps `` erasing '' the term @xmath562 ) .
now we want to prove that for all @xmath563 there is a @xmath5640,{{\delta}}_0]$ ] such that , for @xmath5650,{{\delta}}_\epsilon[$ ] , @xmath566 for all @xmath567 and that there exist @xmath568 such that @xmath569 for all @xmath570 , these estimates are sufficient to conclude : in fact , choosing @xmath571 and using , , and , we get @xmath572 whence @xmath573 which is @xmath574 and then the thesis . by
we obtain @xmath575 whence follows .
observe that follows from @xmath576 if @xmath66 , and from @xmath577 if @xmath202 .
finally , for @xmath578 $ ] we can define @xmath579s;\ ] ] we have @xmath580\,dr-\bigl [ { { \phi}}\bigl(\exp(s{w^\phi})(b^\ast)\bigr)-{{\phi}}(b^\ast)\bigr]s = o(s^2)\ ] ] because @xmath581 is lipschitz .
therefore follows with @xmath582 .
[ holder ] let @xmath583 be an open set .
if @xmath584 is a distributional solution of @xmath17 in @xmath64 , then @xmath585 if @xmath252 , then @xmath16 is a distributional solution of @xmath17 in @xmath33 .
by theorem [ t : tesipinamonti ] there exist @xmath258 , such that @xmath586 uniformly converges to @xmath16 on the compact sets of @xmath33 , @xmath587 @xmath588-a.e .
@xmath56 for every @xmath265 and @xmath589 @xmath588-a.e .
@xmath56 . therefore ,
denoting @xmath590 , we have for every @xmath265 and for every @xmath591 @xmath592 @xmath593 getting to the limit for @xmath594 we obtain @xmath595 @xmath596 i.e. @xmath16 is a distributional solution of the problem @xmath17 .
in the previous section we established the equivalence between @xmath597 and @xmath598 we establish now in section [ s : distrbroad ] a characterization more related to the lagrangian formulation : we prove that one can reduce the pde @xmath599 along any integral line of the vector fields @xmath600 , @xmath601 , provided that one chooses suitably the @xmath602 representative @xmath107 of the distribution identified by @xmath1 .
we first prove in section [ ss : distrtolagr ] , as an introduction , the weaker statement that one can reduce the pde to odes along a selected family of characteristic constituting a lagrangian parameterisation .
as well , the converse holds : if the odes on characteristics are satisfied one has a distributional solution to the pde and the sources of the two formulations can be identified ; this , as recalled in lemma [ l : lipdistr ] , is known from @xcite and we prove it differently in section [ ss : lagrdistr ] below .
the conclusion of this last section will be the following .
the various notions of continuous solutions we have considered are equivalent .
the present section is an introduction to the next one , which proves the stronger statement that distributional solutions are broad solutions .
being technically simpler , this section provides a guideline for some ideas implemented next .
one can in particular notice that the proof concerns only odes . in order to avoid technicalities we focus on @xmath603 and on @xmath604^{2}$ ] .
we refer to section [ s : distrbroad ] for the stronger statement with the universal source term @xmath107 .
the function @xmath111 below relative to the lagrangian formulation has not yet been related with the rhs of : their identification will come from theorem [ t : conversedafermos ] .
[ l : gfunztx ] let @xmath2 be a continuous function . consider a lagrangian parameterisation @xmath605 and assume that @xmath606 is lipschitz in @xmath61 for all @xmath370 . then there exists a borel function @xmath607 such that for all @xmath370 @xmath608 [ c : lipordistrlagr ] if a function @xmath609 is * either an intrinsic lipschitz continuous function * or a continuous distributional solutions @xmath2 to the balance law then it is a lagrangian solution to the equation @xmath610 .
the existence of a lagrangian parameterisation has been proved in lemma [ l : globchi ] . as recalled just after the definition [ d : deflip ] , an intrinsic lipschitz continuous function
is continuous and lipschitz along characteristics .
also continuous distributional solutions to the balance law are lipschitz continuous along characteristics by lemma [ l : daflipschitz ] , following @xcite ) .
we have therefore the thesis directly by applying lemma [ l : gfunztx ] .
we remind the notation @xmath611 second derivative in @xmath612 . by assumption @xmath613
is lipschitz in @xmath61 .
being @xmath2 continuous , one can see that the subset @xmath614 of those @xmath276 where @xmath615 is twice @xmath61-differentiable is @xmath616 , and @xmath617 is a borel function on it .
moreover , by rademacher s theorem the @xmath370-sections of @xmath618 have full measure and therefore by tonelli theorem also @xmath618 has full measure . however , the function @xmath617 is defined on @xmath619 , while we are looking for a function defined on @xmath64 . a preliminary comment . in order to check that @xmath620 lifts @xmath621 to a map
@xmath111 a.e . defined on @xmath64 , which would provide our thesis
, it would be natural to show that * @xmath622 is a lebesgue measurable subset of @xmath64 with full measure ; * @xmath623 is constant on the level set of @xmath620 intersected with @xmath618 .
since @xmath624 is not lipschitz , we do not manage to prove at this point that @xmath625 has full measure .
we instead assign a specific value , @xmath186 , to the function out of @xmath625 .
this choice of the extension does not affect our claim .
we notice moreover that the second point is true in that strong form , but we show that the points of the level set of @xmath620 corresponding to more values of @xmath626 are not relevant .
analysis of @xmath622 and partial inverse of @xmath620 .
the proof of the borel measurability of @xmath622 requires some technicality : we apply a theorem due to srivastava ( @xcite , theorem 5.9.2 ) deriving that there exists a borel restriction @xmath624 which is one - to - one to @xmath622 ; then theorem 4.12.4 in @xcite , due to lusin , would provide the thesis . in order to apply the first theorem , we partition @xmath619 into the level sets of @xmath620 , which are @xmath627 .
we need also to observe that @xmath628 is borel for each open set @xmath629 . for simplicity , consider the case when @xmath624 is already a full parameterisation and thus it is continuous .
every open set @xmath629 is @xmath313-compact : thus by continuity @xmath630 is @xmath313-compact , and finally @xmath628 is @xmath313-compact .
therefore by srivastava s theorem there is a borel cross section @xmath12 for the partition : @xmath620 restricted to @xmath631 is borel , injective and onto @xmath622 . being a borel image by a one - to - one map , lusin s theorem asserts on one hand that that @xmath622 is borel , and moreover that this restriction has a borel inverse @xmath632 maybe there is a more elementary argument which allows to approximate @xmath618 with a @xmath313-compact subset @xmath633 whose @xmath370-section have full measure
. then one could as well work with @xmath633 instead of @xmath618 avoiding measurability difficulties , even without investigating the size of its image .
analysis of @xmath623 for @xmath634 .
we can define @xmath111 as @xmath635 this map satisfies the claim by the next analysis of the set where @xmath636 is multivalued .
analysis of multivalued points of @xmath637 .
the set of points @xmath638 such that there exists @xmath639 satisfying @xmath640 and @xmath641 is borel : just see it as @xmath642 the value of @xmath111 on this set is not at all relevant : we show below that for any @xmath370 the intersection of @xmath643 with the characteristics curve @xmath644}$ ] is at most countable .
+ indeed , fix a curve @xmath645 , for simplicity fix @xmath646 .
suppose @xmath647 intersects @xmath648 at @xmath536 with @xmath649 , @xmath650 fixed . by taylor s
expansion , there exists a neighborhood @xmath651 where @xmath652 there can be at most @xmath653 such values @xmath654 of @xmath61 : otherwise by the last inequality two characteristics @xmath655 relative to two @xmath656 with @xmath657 would intersect at @xmath658 and cross each other . taking the union for a sequence @xmath659 , and then the union over @xmath650 , we get the thesis .
below we show theorem [ t : univselection ] : there exists a borel function @xmath660 such that every curve @xmath648 satisfying the ode with continuous coefficient @xmath417 has time derivative lipschitz , and it has second derivative precisely @xmath661 for a.e .
the remarkable fact is that @xmath660 could be defined independently of any set of characteristic curves .
this is thus different , and stronger , from what we already proved , which is that there exists a lagrangian parameterisation @xmath79 and an associated function @xmath662 satisfying @xmath663 .
the new point is indeed that @xmath107 is a universal representative for the source term .
due to its nature this proof works for any @xmath275 with no further complication .
the only difference is that @xmath64 will be a subset of @xmath226 instead of @xmath664 .
we write it with @xmath66 only for notational convenience .
in particular we generalize here the previous lemma [ l : gfunztx ] . since the argument is more intuitive , we mention first how to construct such a souslin function @xmath660 .
we proceed then with the borel construction because it gives a better result .
the first step is to define pointwise , but in a measurable way , a function @xmath660 such that @xmath67 is a lebesgue point for the second derivative of a curve @xmath665 with @xmath666 and satisfying the ode , whenever there exists one
. therefore one applies von neuman s selection theorem ( section 5.5 in @xcite , from @xcite ) to the subset of @xmath667 defined by @xmath668 @xmath669 is borel .
it has full measure projection on @xmath670 .
components @xmath671 . the subset @xmath672 identified by the constraints @xmath673 , moreover , its projection on @xmath277 is all @xmath674 . by rademacher theorem ,
we have moreover seen in the same lemma also that the projection of @xmath669 on @xmath277 has full measure .
component @xmath675 , discretization .
in order to establish the existence ( and the value ) of the limit for the second derivative of @xmath483 at @xmath68 , it suffices to consider e.g. the sequence @xmath676 indeed , then for @xmath677 $ ] , for example at @xmath678 @xmath679 by construction however @xmath680 yielding that the existence of the limit along @xmath681 implies the existence of the limit for any @xmath682 .
notice that this would not hold choosing a generic @xmath683 instead of @xmath684 .
measurability of @xmath669 .
the further constraint in @xmath483 can be written as @xmath685 therefore , we are considering the following subset of @xmath686 : @xmath687 since the set within brackets is closed , @xmath669 is borel .
this allows to define an @xmath443-selection @xmath688 , which by construction is a measurable function assigning to every point @xmath277 an integral curve @xmath689 for the ode @xmath690 , whenever there exists a lipschitz one having @xmath68 as a lebesgue point for the right , left and total second derivative .
as well , one can define the souslin function @xmath691 the importance of this selection is due to the following theorem .
[ t : univselection ] let @xmath2 be a continuous function which is uniformly lipschitz along characteristics curves @xmath483 : @xmath692 .
then curve @xmath483
satisfying the ode @xmath692 , one has @xmath693 [ c : distrbroad ] a continuous distributional solution @xmath2 to @xmath482 is also a broad solution . before proving the theorem , for the sake of completeness
we show that one can define as well a borel fucntion , that we still denote as @xmath694 , for which theorem [ t : univselection ] still holds .
this requires a bit more work than the previous argument , and it is conceptually a little more involved : we do not associate immediately to each point ( where it is possible ) an eligible curve and its second derivative , but something which must be close to it
. we will find then with the proof of theorem [ t : univselection ] that we end up basically with the same selection .
[ l : borelapprg ] for every @xmath695 , there is a borel function defined on the @xmath277-projection @xmath696 of @xmath669 @xmath697 such that @xmath698 of and that for @xmath699 sufficiently small @xmath700 we define the borel representative of @xmath1 as the function @xmath701 we apply arsenin - kunugui selection theorem ( @xcite , or th .
5.12.1 of @xcite ) to the set @xmath702 where @xmath686 was defined in and @xmath703 immediately below that , in the same proof .
more precisely , @xmath686 is closed and @xmath704 is also closed : therefore each @xmath277-section of is @xmath313-compact .
then the hypothesis of the theorem are satisfied : it assures that the projection of on the first factor @xmath705 is borel and there exists a borel section of defied on it , which is the function in our statement .
notice that the domain fo this function , containing @xmath706 , has full lebesgue measure .
we provide now the proof of theorem [ t : univselection ] with @xmath707 either the borel or the souslin one : we consider any characteristic @xmath708 for the balance law and we prove that for almost every @xmath68 its second derivative is precisely @xmath709 .
we remind that @xmath710 is lipschitz ( lemma [ l : daflipschitz ] ) .
step 1 , countable decomposition .
the set of @xmath68 lebesgue point of @xmath711 with value different from @xmath712 can be reduced to @xmath713 if one is considering the souslin selection , clearly there is the simplification @xmath714 .
step 2 , reduction argument .
we prove that the set @xmath715 can not contain points @xmath716 with @xmath717 .
then the thesis will follow : by the previous step the set of @xmath68 where the second derivative of @xmath718 exists and it is different from @xmath712 will be countable .
therefore the second derivative of @xmath718 will be almost everywhere precisely @xmath719 .
step 3 : analysis of the single sets . by contradiction , assume that contains two such points , for example @xmath720 , @xmath721 . by definition of the set of points we are considering , the two selected curves through @xmath722 , @xmath723 , @xmath724 must intersect in the time interval @xmath725 $ ] , say at time @xmath726 . since they satisfy the ode for characteristics , where they intersect they have the same derivative . being all of them lipschitz , we have then @xmath727 comparing the lhs and the rhs , one arrives to @xmath728 however , since the times @xmath729 belong by construction to the set one has @xmath730 reaching a contradiction .
@xmath731 in this section we prove , without passing through the implicit function theorem , that if a continuous function @xmath2 satisfies the lagrangian formulation [ d : lagrsol ] of the balance law , then in the eulerian variables @xmath2 solves indeed the balance law in distributional sense .
we instead rely on a mollification procedure in the legrangian variables .
see also @xcite , where a different , pointwise approximation of the distributional solution is provided , starting from a broad * solution .
this basically shows the converse of dafermos statement in @xcite .
we notice that the fact that if a continuous @xmath732 solves the lagrangian formulation , then it is not difficult as we see in the next theorem proving that it solves a balance law in distributional sense .
the identification of the lagrangian and distributional sources however is not trivial : we indeed use the converse implication of section [ s : distrbroad ] and rademacher theorem .
we already motivated why focusing on the case @xmath66 , whereas the equation reduces to @xmath733_{t}=w(y , t).\ ] ] we give generalizations to the case @xmath193 in remark [ r : remarkdimnsection6 ] [ t : conversedafermos ] every lagrangian solution to is also a distributional solution , with source term given by the distribution identified by the function @xmath111 in the lagrangian formulation .
[ c : broaddistrib ] any continuous broad solution @xmath2 to is also a distributional solution .
let @xmath734 be a lagrangian parameterisation associated to @xmath2 , and @xmath735 such that @xmath736 we prove then that @xmath2 is a distributional solution of the balance equation @xmath737_{t}=w(y , t).\ ] ] smoothing of @xmath738 in the @xmath370-variable . consider a suitable convolution kernel @xmath739 and define the @xmath370-regularized function @xmath740 given by @xmath741 this function @xmath742 is smooth but possibly still not injective .
however , by the monotonicity of @xmath79 one has that if @xmath743 then @xmath734 is constant for @xmath744 $ ] . as a consequence ,
@xmath745 is in turn constant for @xmath744 $ ] .
the above observation allows to define consequently the approximation @xmath746 by @xmath747 notice that @xmath748 is well defined and continuous in the @xmath749-variables .
the measurable subset of points where it is not differentiable is possibly non - empty as a consequence of the non - injectivity of @xmath742 , but it is at most countable on every @xmath61-section .
@xmath748 is @xmath750 on the remaining full measure set .
since both @xmath79 and @xmath2 are continuous , the above relations immediately imply the local uniform convergence of the regularized functions : @xmath751 convergence in the @xmath749-variables of @xmath748 .
notice that @xmath752 is converging ( as a measure ) to @xmath753 , and that their variation is unifromly bounded .
the above procedure not only defines correctly the continuous functions @xmath746 , but it allows to establish their convergence in @xmath754 : indeed @xmath755 the first factor in both the integrals is uniformly convergent to zero , while @xmath756 converges to the measure @xmath757 .
the convergence of @xmath758 is then straightforward .
approximation of the source . as @xmath748 is a smooth function a.e .
, one can define an approximate source @xmath759 by @xmath760 the above relation , by the pointwise smoothness is immediately equivalent to @xmath761 since we started from a lagrangian parameterisation , the further regularity in the @xmath61 variable @xmath762 for the relative pointwise representative @xmath111 implies the relation @xmath763 in particular , the sources @xmath764 are uniformly bounded by the @xmath89 bound for @xmath111 .
moreover , for each @xmath67 fixed @xmath765 converges in all @xmath766 to @xmath767 , and thus in @xmath768 ; the convergence is clearly uniform when @xmath111 is continuous .
the lhs of equation passes to the weak limit by the @xmath754-convergence of @xmath769 to @xmath770 established above .
the same holds as well for the rhs , since @xmath771 converge in @xmath772 to a function @xmath773 .
thus @xmath774_{t}= \bar{\bar w}^{}(y , t).\ ] ] the function @xmath2 as we have seen is precisely the one in the lagrangian parameterisation . even if @xmath765 converges in all @xmath766 to @xmath767 , it is not instead trivial that above @xmath775 , lebesgue almost everywhere , holds .
we now explain why it is so .
we prove above in section [ s : distrbroad ] that if @xmath774_{t}= \bar{\bar w}^{}(y , t)\ ] ] holds then @xmath2 is also a broad solution : there is a point - wise representative @xmath776 of @xmath777 such that the broad formulation holds with @xmath778 .
however , it is then also intrinsic lipschitz continuous with intrinsic gradient @xmath776 : by rademacher theorem the intrinsic gradient @xmath779 is uniquely defined point - wise almost everywhere , and it by lemma [ l : lipdistr ] it identifies the same distribution as @xmath780 .
[ r : remarkdimnsection6]we finally remark that theorem [ t : conversedafermos ] works immediately also in higher dimensions , because for ( almost every ) @xmath781 the restriction to the plane @xmath782 of @xmath2 is still a lagrangian solution , with source term the restriction @xmath783 . for every test function
@xmath784 we have then by fubini - tonelli theorem @xmath785 = \iint_{{{\mathbb r}}^{2(n-1 ) } } dv\int_{\omega_{v } } dy_{1}dt \left [ \varphi_{y_{1 } } \phi + \varphi_{t } \frac{\phi^{2}}{2 } \right]{\restriction_}{\omega_{v } } \\ & \stackrel{\text{th.~\ref{t : conversedafermos } } } { = } -\iint_{{{\mathbb r}}^{2(n-1 ) } } dv\int_{\omega_{v } } dy_{1}dt \left[\tilde w{\restriction_}{\omega_{v}}\right ] = -\iint_{\omega}\tilde w.\end{aligned}\ ] ] considering also the linear fields we gain the implication from to in theorem [ t : othertheorem ] , and more generally that a lagrangian solution is also a distributional solution .
in the present section we deal with the issue of extending a partial lagrangian parameterisation to a ` full ' one . we construct a function @xmath734 satisfying the ode , both monotone and surjective in the @xmath370 variable , which extends a given one @xmath786 .
this is the matter of lemma [ l : fillinode ] : we recall below how to extend a solution to an ode with continuous coefficients , whose existence is a classical result . the procedure can be first understood considering example [ e : fillholesnogloblip ] below , illustrated in figure [ fig : parameterisation ] .
this deals with the simpler case of an @xmath61-independent @xmath2 , but it has all the ingredients of the general construction of lemma [ l : fillinode ]
. moreover , example [ e : fillholesnogloblip ] provides a counterexample for the following fact : even if characteristics are @xmath787 in @xmath61 with lipschitz derivative , _ it is not possible in general to extend a partial , monotone lagrangian parameterisation to a full one which is locally lipschitz continuous_. the reduction of the balance law along characteristics , which is equation , has been inserted in the text ( lemma [ l : globchi ] and theorem [ t : univselection ] ) .
here we just notice that it holds , with some @xmath788 pointwise defined in @xmath64 , for a particular lagrangian parameterisation @xmath789 .
the argument is @xmath78-dimensional .
let @xmath790 be a partial lagrangian parameterisation .
focus e.g. the attention on @xmath791 $ ] and @xmath792 valued in @xmath387 $ ] and right continuous , the general case being similar .
we fix also @xmath793 .
we construct an extension @xmath794 by a recursive procedure . for convenience ,
the induction index is given by couples @xmath795 with @xmath796 and @xmath797 .
the ordering is lexicographic , starting from the second variable : @xmath798 iff either @xmath799 or @xmath800 and @xmath801 .
the starting point is @xmath802 defined for @xmath400 $ ] , @xmath803 $ ] .
consider the dichotomous points @xmath804 , which go from @xmath805 to @xmath806 at step @xmath807 , associated to the indexes @xmath795 with @xmath796 and @xmath797 .
induction step @xmath795 , @xmath808 : general description .
assume you have been given @xmath79 defined on @xmath809\times t $ ] by a previous step .
if at @xmath810 the map @xmath811 is not onto @xmath812 $ ] we construct an extension @xmath813 such that induction step @xmath795 , @xmath808 : change of parameter set .
because of monotonicity the complementary of the image of @xmath819 is the at most countable union of disjoint intervals @xmath820 , which correspond to the discontinuities @xmath821 of this real valued map . at those parameters
@xmath822 the two characteristics @xmath823 and @xmath824 bifurcate , and at time @xmath825 their opening is an interval @xmath826 : @xmath827 define consequently the strictly increasing map opening each of those parameters @xmath828 into an interval proportional ( with factor @xmath829 ) to the hole @xmath830 that the relative characteristics leave at @xmath825 : @xmath831 & & \\ j^{h , n } & : & \tau&&\mapsto & & \tau+{{\mathcal l}}^{1}(\cup_{\tau_{k}^{}\leq \tau}i_{k}^{})/2^{2n-1 } .&&\end{aligned}\ ] ] the inequality @xmath832 holds because we are assuming that @xmath79 is valued in @xmath387 $ ] , thus @xmath833 .
the set @xmath834 will be the new space of parameters , and the injection @xmath816 will bring from the old set @xmath835 to the new one .
induction step @xmath795 , @xmath808 : extension of the parameterisation .
we just constructed a new parameter set @xmath834 and an immersion @xmath816 from the old one @xmath835 .
in particular , the lagrangian parameterisation is fixed on @xmath836 , but in order to conclude the induction step we need to define the new lagrangian parameterisation on @xmath837 .
we clearly need to respect also the monotonicity property , taking into account that part of the parameterisation is already fixed : the new characteristics that we are going to to define must not cross the old ones . for each @xmath838 , consider the @xmath787 maximal curve through @xmath839 defined at until it touches either @xmath840 or @xmath841 , in which case it goes on with the curve which has been touched .
it is a little complicated to write formally , but it is just that .
the times when this touching happens , if it ever happens , are @xmath842 with the notation that @xmath843 is empty if @xmath844 , a possible right continuous extension is then give by @xmath845 define finally on @xmath387\times t_{h , n}$ ] the lagrangian parameterisation which coincides with the previous one on the image of the old parameter set , and which is extended as described above elsewhere : @xmath846 for the second line , @xmath847 is the length of the interval @xmath848 , part of the ones added to the parameter set precisely at the @xmath795-th step
. rescaled by @xmath849 , it gives the length of the segment @xmath850 : the point @xmath839 is where we start for defining the characteristic @xmath851 that we are inserting for extending the lagrangian parameterisation .
notice that also the surjectivity property at @xmath810 is satisfied . conclusion .
let us first look at how much the domain @xmath852 grows in the extension process . since @xmath853 couples of indices have second variable @xmath275 ,
then the total size of the intervals added by those couples alltogether is at most @xmath805 : thus , setting @xmath854 , @xmath855 , \quad t_{1}\subset [ 0,3/2 ] , \quad \dots\ , \quad t_{n}\subset [ 0,2 - 2^{-n } ] , \quad \dots\ .\ ] ] the maps @xmath856 are the lagrangian parameterisations @xmath861 and @xmath863 have different domains for the second components , the space of parameters @xmath834 , @xmath864 . however , as seen in the previous step there are strictly monotone injections from each of them to the interval @xmath865 given by @xmath866 and @xmath867 . being strictly monotone ,
they are invertible if we fill the graphs at discontinuity points : we can compare these compositions , with the same second component in @xmath865 , and we find for them @xmath868 the sequence of these compositions therefore converges uniformly .
one verifies that the limit is a monotone lagrangian parameterisation @xmath79 which extends @xmath869 , with injection map @xmath860 .
being surjective and monotone , each @xmath870 is continuous . by the continuity in @xmath61
we deduce then surjectivity also at the remaining times : indeed if by absurd we had @xmath871 , we could not have @xmath872 at @xmath873 arbitrarily close to @xmath654 .
the following example introduces the extension of a lagrangian parameterisation .
it shows moreover that it is not possible in general to get a full one which is lipschitz continuous , even though characteristics below are twice continuously differentiable .
consider the very simple equation for @xmath105 in the rectangle @xmath874\times[0,1]$ ] @xmath875_{t}=w(z , t ) , \qquad \phi_{z}(z , t)=0.\ ] ] being @xmath2 dependent on one variable , we change notation and we write @xmath876 .
+ we consider the partial lagrangian parameterisation @xmath877 defined in lemma [ l : solode ] .
this would need to specify the function @xmath2 : since the construction is involved , we specify it below and the reader can immediately visualise it in figure [ fig : parameterisation ] ( left side ) , where a family of integral curves @xmath878 is drown .
step 1 : building block ( figure [ fig : parameterisation ] , right side ) .
define first a smooth function @xmath648 , for @xmath400 $ ] , which increases continuously from @xmath186 to @xmath388 .
let @xmath879 be even , strictly increasing in the first half interval from @xmath186 to its maximum .
let @xmath880 vanish at @xmath881 and be positive in @xmath882 $ ] .
for instance , consider @xmath883 step 2 : iteration step .
we define now a first sequence of points @xmath884 and intermediate ones @xmath885 where @xmath886 and @xmath887 will vanish .
the first ones are approximatively @xmath186 , @xmath888 , @xmath889 , @xmath890 , @xmath891 , @xmath892 , @xmath893 , , and each interval @xmath894 $ ] is divided into @xmath895 equal subintervals for determining the points @xmath896 . for @xmath897 , @xmath898 @xmath899
step 3 : iteration .
now we define the functions @xmath886 , @xmath887 on each subinterval @xmath900 $ ] . as a preliminary half - step
consider the rescaled smooth functions @xmath901 given by @xmath902,\ ] ] notice that each increases monotonically from @xmath903 to @xmath896 , and the first two derivatives vanish at the endpoints .
in particular , we can associate to each @xmath904 $ ] a curve @xmath901 such that @xmath905 it will be unique out of the points @xmath906 , while at these junctions there will be two such curves , with however vanishing first two derivatives . observing figure [ fig : parameterisation ] , this curve @xmath901 is just the first segment of what will be part of @xmath877 @xmath907\ni s \mapsto \gamma_{h , i}(s ) \equiv \chi_{m}(s , z^{+}_{h+1,i}).\ ] ] define then , @xmath908 these functions are clearly continuous out of the nodes @xmath906 , and they vanish there , but the continuity at the limit point @xmath909 should be checked .
it holds because @xmath910 and @xmath911 , which implies they vanish for @xmath912 .
step 4 : non - lipschitz lagrangian parameterisation .
having defined the function @xmath2 , we have already defined the partial lagrangian parameterisation @xmath877 of lemma [ l : solode ] .
we now extend it to a surjective one , but there is no way of having it lipschitz continuous , as we compute now .
different colors in figure [ fig : parameterisation ] show different steps of the extension process .
minimal characteristics starting from @xmath913 do not cover almost all the interval at @xmath914 .
adding those starting at @xmath914 , it remains to cover at @xmath915 open intervals of total length @xmath916 2 .
including at the second step all those minimal characteristics which intersect the line @xmath915 , similarly , one does not cover the whole line @xmath917 : a length @xmath918 remains to cover at both those two values of @xmath80 .
+ i. at the subsequent @xmath919-th step , at each of the @xmath920 values of @xmath921 one needs covering a length @xmath922 . in the whole process
, it must be covered a total length equal to @xmath923 any monotone , lagrangian parameterisation @xmath79 must map a disjoint family of real intervals @xmath924 with @xmath925 to the intervals @xmath926\}_{i}$ ] , respectively at the above valies @xmath927 .
however , we have just computed that for all constants @xmath686 @xmath928 ) \\ > & c\sum_{h=0,\dots,2^{i-1}-1,\ i\in{{\mathbb n}}}| i_{h , i}| , \end{aligned}\ ] ] preventing any lipschitz regularity .
indeed , the map @xmath929 can be lipschitz with some constant @xmath930 , but @xmath930 must blow up as @xmath931 . at other values of @xmath61 , this map is just @xmath932 .
notice finally that the fact that the blow - up of the lipschitz constant is caused here by a behaviour approaching the boundary @xmath933 is incidental .
indeed , one can extend the present construction also on @xmath934 for example reflecting the characteristics w.r.t . the axis @xmath935 . | in this paper we provide a characterization of intrinsic lipschitz graphs in the sub - riemannian heisenberg groups in terms of their distributional gradients .
moreover , we prove the equivalence of different notions of continuous weak solutions to the equation @xmath0_{t}=w$ ] , where @xmath1 is a bounded measurable function depending on @xmath2 . |
polynomial series are of great importance in control theory .
both continuous and discrete polynomial series are useful in approximating state and/or control variables , in modal reduction , optimal control , and system identification , providing effective and efficient computational methods @xcite . from recent years ,
the theory of control for discrete and continuous time is being unified and extended by using the formalism of time scales : see @xcite and references therein
. looking to the literature on time scales , one understands that such unification and extension is not unique .
two main directions are being followed : one uses @xmath1-derivatives while the other chooses @xmath2-derivatives instead . in this paper
we adopt the more general notion of diamond-@xmath0 derivative @xcite , and give the first steps on a correspondent theory of polynomial series .
as particular cases , for @xmath3 we get @xmath1-polynomial series ; when @xmath4 we obtain @xmath2-series . by choosing the time scale to be the real ( integer ) numbers ,
we obtain the classical continuous ( discrete ) polynomial series .
diamond - alpha derivatives have shown in computational experiments to provide efficient and balanced approximation formulas , leading to the design of more reliable numerical methods @xcite .
we claim that the combined dynamic polynomial series here introduced are useful in control applications .
here we give only very short introduction ( with basic definitions ) on three types of calculus on time scales . for more information we refer the reader to @xcite . by a time scale , here denoted by @xmath6 , we mean a nonempty closed subset of @xmath7 .
as the theory of time scales give a way to unify continuous and discrete analysis , the standard cases of time scales are @xmath8 , @xmath9 , or @xmath10 , @xmath11 . for @xmath12 , the forward jump operator @xmath13 and the graininess function @xmath14 are defined by @xmath15 and @xmath16 if @xmath17 ; @xmath18 .
moreover , we have the backward operator @xmath19 and the backward graininess function @xmath20 defined by @xmath21 and @xmath22 if @xmath23 ; @xmath24 . in the continuous - time case , when @xmath8 , we have @xmath25 and @xmath26 for all @xmath27 . in the discrete - time case , @xmath28 , @xmath29 , and @xmath30 for each @xmath31 . for the composition between a function @xmath32 and functions @xmath33 and @xmath34
, we use the abbreviations @xmath35 and @xmath36 .
a point @xmath37 is called left - scattered ( right - scattered ) if @xmath38 ( @xmath39 ) .
a point @xmath37 is called left - dense ( right - dense ) if @xmath40 ( @xmath41 ) .
the set @xmath42 is defined by @xmath43 $ ] if @xmath44 , and @xmath45 if @xmath46 ; the set @xmath47 by @xmath48 if @xmath49 , and @xmath50 if @xmath51 .
moreover , @xmath52 @xmath53 and @xmath54 . for a function @xmath32
, we define the @xmath1-derivative of @xmath55 at @xmath56 , denoted by @xmath57 , to be the number , if it exists , with the property that for all @xmath58 , exists a neighborhood @xmath59 of @xmath56 such that for all @xmath60 , @xmath61 .
function @xmath55 is said to be @xmath1-differentiable on @xmath42 provided @xmath57 exists for all @xmath56 .
the @xmath2-derivative of @xmath55 , denoted by @xmath62 , is defined in a similar way : it is the number , if it exists , such that for all @xmath58 there is a neighborhood @xmath63 of @xmath64 such that for all @xmath65 @xmath66 .
function @xmath55 is said to be @xmath2-differentiable on @xmath47 provided @xmath62 exists for all @xmath64 .
the classical settings are obtained choosing @xmath67 and @xmath68 : 1 .
let @xmath8 . then , @xmath69 and @xmath55 is @xmath1 and @xmath2 differentiable if and only if it is differentiable in the ordinary sense .
2 . let @xmath10 , @xmath11 .
then , @xmath70 and @xmath71 always exist .
it is possible to establish some relationships between @xmath1 and @xmath2 derivatives .
@xcite ( a ) assume that @xmath32 is delta differentiable on @xmath42 .
then , @xmath55 is nabla differentiable at @xmath37 and @xmath72 for all @xmath64 such that @xmath73 .
( b ) assume that @xmath32 is nabla differentiable on @xmath47 .
then , @xmath55 is delta differentiable at @xmath37 and @xmath74 for all @xmath56 such that @xmath75 .
a function @xmath76 is called rd - continuous provided it is continuous at right - dense points in @xmath6 and its left - sided limits exist ( finite ) at left - dense points in @xmath6 . the class of real rd - continuous functions defined on a time scale @xmath6 is denoted by @xmath77 . if @xmath78 , then there exists a function @xmath79 such that @xmath80
. the delta - integral is defined by @xmath81 .
similarly , a function @xmath76 is called ld - continuous provided it is continuous at left - dense points in @xmath6 and its right - sided limits exist ( finite ) at right - dense points in @xmath6 . the class of real ld - continuous functions defined on a time scale @xmath6 is denoted by @xmath82 . if @xmath83 , then there exists a function @xmath84 such that @xmath85 . in this case
we define @xmath86 .
let @xmath10 , @xmath11 , and @xmath87 .
then , one has : @xmath88 , @xmath89 .
@xcite let @xmath90 , @xmath91 , and @xmath32 .
the diamond - alpha derivative of @xmath55 at @xmath37 is defined to be the value @xmath92 , if it exists , such that for all @xmath58 there is a neighborhood @xmath59 of @xmath37 such that for all @xmath93 @xmath94\eta_{ts } + ( 1-{\mathbb{\alpha}})\left[f^{\rho}(t)-f(s)\right]\mu_{ts } -f^{{\diamondsuit_{\alpha}}}(t)\mu_{ts}\eta_{ts}\right|\leq \varepsilon|\mu_{ts}\eta_{ts}| \ , .\ ] ] we say that function @xmath55 is @xmath5-differentiable on @xmath95 , provided @xmath92 exists for all @xmath96 .
@xcite[thsheng ] let @xmath32 be simultaneously @xmath1 and @xmath2 differentiable at @xmath96 . then , @xmath55 is @xmath5-differentiable at @xmath37 and @xmath97 , @xmath98 $ ] .
the @xmath5-derivative is a convex combination of delta and nabla derivatives .
it reduces to the @xmath1-derivative for @xmath99 and to the @xmath2-derivative for @xmath100 .
the case @xmath101 has proved to be very useful in applications . for more on the theory of @xmath5-derivatives than that we are able to provide here , we refer the interested reader to @xcite .
the same idea used to define the combined derivative is taken to define the combined integral .
[ diaminteg ] let @xmath102 and @xmath103 .
then , the @xmath5-integral of @xmath55 is defined by @xmath104 , where @xmath98 $ ] . in general the @xmath5-derivative of @xmath105 with respect to @xmath37
is not equal to @xmath106 @xcite .
next proposition gives direct formulas for the @xmath5-derivative of the exponential functions @xmath107 and @xmath108 .
for the definition of exponential and trigonometric functions on time scales see , , @xcite .
[ diamexpder ] let @xmath6 be a time scale with the following properties : @xmath73 , and @xmath75 .
assume that @xmath109 and for all @xmath12 one has @xmath110 , and @xmath111 .
then , @xmath112e_p(t , t_0 ) \
, , \label{eq : a } \\ \hat{e}_p^{{\diamondsuit_{\alpha}}}(t ,
t_0)=\left[(1-{\mathbb{\alpha}})p(t)+ \frac{{\mathbb{\alpha}}p^{\sigma}(t)}{1-\mu(t)p^{\sigma}(t)}\right]\hat{e}_p(t , t_0 ) \ , , \label{eq : b}\end{gathered}\ ] ] for @xmath96 .
firstly , recall that @xmath113 and @xmath114 .
hence , @xmath115 , and then @xmath116 , from where it follows .
similarly , we have that @xmath117 , and then @xmath118 , from where holds . [ cor1 ]
let @xmath119 , @xmath120 , and @xmath121 for all @xmath12 . then , for @xmath96 , \(a ) @xmath122 @xmath123 ; \(b ) @xmath124 @xmath125 ; \(c ) @xmath126 @xmath127 ; \(d ) @xmath128 @xmath129 .
let @xmath109 , @xmath120 , and @xmath130 for all @xmath12 . then , for @xmath96 , \(a ) @xmath131 @xmath132 ; \(b ) @xmath133 @xmath134 ;
\(c ) @xmath135 @xmath136 ; \(d ) @xmath137 @xmath138 .
let @xmath6 be an arbitrary time scale .
let us define recursively functions @xmath139 , @xmath140 , as follows : @xmath141 similarly , we consider the monomials @xmath142 : they are the functions @xmath143 , @xmath140 , defined recursively by @xmath144 all functions @xmath145 are rd - continuous , all @xmath146 are ld - continuous . the derivatives of such functions show nice properties : @xmath147 , @xmath56 ; and @xmath148 , @xmath64 , where @xmath109 and derivatives are taken with respect to @xmath37 .
we have that @xmath149 for @xmath8 , @xmath150 .
finding exact formulas of @xmath145 or @xmath146 for an arbitrary time scale is , however , not easy .
from @xcite we have the following result : let @xmath151 and @xmath152
. then , @xmath153 for all @xmath140 .
next proposition gives explicit formulas for homogenous time scales with @xmath154 , @xmath155 a strictly positive constant .
for that we need two notations of factorial functions : for @xmath156 we define @xmath157 and @xmath158 with @xmath159 and @xmath160 .
let @xmath11 and @xmath10 .
for @xmath140 the following equalities hold : \(a ) @xmath161 ; \(b ) @xmath162 .
firstly , @xmath163 . next we observe that @xmath164 .
hence , by the principle of mathematical induction , ( a ) holds for all @xmath140 . since @xmath153 , ( b ) is also true .
let @xmath10 , @xmath165 . from the properties of factorial functions
it follows : 1 . if @xmath166 , then @xmath167 ; 2 . if @xmath168 , then @xmath169 . in particular ,
when @xmath170 and @xmath9 , we have : 1 . if @xmath171 , then @xmath167 ; 2 . if @xmath172 , then @xmath169 . in the next section
we need the following results .
let @xmath10 , @xmath11 , and @xmath12 .
then , for @xmath140 , the following holds : \(a ) @xmath173 ; \(b ) @xmath174 .
[ prop3012 ] let @xmath10 , @xmath11 , @xmath119 .
then , \(a ) @xmath175 for @xmath176 ; \(b ) @xmath177 for @xmath178 .
let @xmath10 , @xmath11 , and @xmath119 .
for @xmath176 , it is enough to notice that @xmath179 to prove ( a ) .
equality ( b ) is proved in a similar way : for @xmath178 , we have : @xmath180 .
@xcite let @xmath152 and @xmath140 .
then , @xmath181 for @xmath182 . as a consequence of ( [ rel3 ] ) and equalities @xmath183 and @xmath184 , the following laws of differentiation of generalized monomials follow .
[ derdor1 ] \(a ) @xmath185^{\nabla } = \sum_{j=0}^{k}(-1)^j\nu^j(t)h_{k - j}(t , t_0)$ ] ; \(b ) @xmath186^{\delta } = \sum_{j=0}^{k}\mu^j(t)\hat{h}_{k - j}(t , t_0)$ ] .
[ exdor1 ] let @xmath6 be an homogenous time scale with @xmath187 , @xmath188 .
let us recall that for @xmath189 we have @xmath8 and for @xmath170 we have @xmath9 .
then , @xmath185^{\nabla } = \sum_{j=0}^{k}(-1)^jc^jh_{k - j}(t , t_0)$ ] , @xmath186^{\delta}= \sum_{j=0}^{k}c^j\hat{h}_{k - j}(t , t_0)$ ] . for @xmath8 : @xmath185^{\nabla } = \left[\hat{h}_{k+1}(t , t_0)\right]^{\delta } = \frac{(t - t_0)^k}{k!},$ ] + for @xmath9 : @xmath185^{\nabla}= \sum_{j=0}^{k}(-1)^jh_{k - j}(t , t_0 ) = \binom{t - t_0}{0}-\binom{t - t_0}{1 } + \cdots+(-1)^k\binom{t - t_0}{k},$ ] and @xmath186^{\delta } = \sum_{j=0}^{k}\hat{h}_{k - j}(t , t_0)$ ] .
let @xmath119 .
then , \(a ) @xmath190 ; \(b ) @xmath191 ; \(c ) @xmath192 . from the definition of @xmath5-derivative and corollary [ derdor1 ]
, we have : @xmath193 .
next , @xmath194 .
[ tayth1 ] @xcite assume that @xmath55 is @xmath195 delta - differentiable on @xmath196 .
let @xmath197 , @xmath12 .
then , @xmath198 where @xmath199 .
[ tayth2]@xcite assume that @xmath55 is @xmath195 times nabla differentiable on @xmath200 .
let @xmath201 , @xmath12 .
then , @xmath202 where @xmath203 . by a polynomial real series we usually understand a series of the form @xmath204 , where @xmath205 is a given sequence of polynomials in the variable @xmath37 and @xmath206 is a given sequence of real numbers . in the continuous case
one has @xmath207 . for the time
scales we are considering in this paper , we have @xmath208 or @xmath209 , and we speak about _ generalized power series on time scales _ @xcite . [ szereg ]
let @xmath6 be a time scale and let us fix @xmath109 . by a _ @xmath1-polynomial series ( on @xmath6 , originated at @xmath210 )
_ we shall mean the expression @xmath211 , @xmath212 ; by a _
@xmath2-polynomial series ( on @xmath6 , originated at @xmath210 ) _ we mean @xmath213 , @xmath212 , where for each @xmath214 , @xmath215 .
the sequence @xmath216 is called the corresponding sequence of the series .
for any fixed @xmath217 , both type of series become ordinary number series .
if they are convergent for @xmath37 we say that the polynomial series is convergent at @xmath37 . if @xmath9 , then for each @xmath218 , @xmath219 , the number series @xmath220 is convergent because it is finite . the same situation we have when @xmath221 : @xmath222 is finite , so convergent . in @xcite and
@xcite it is proved the following : let @xmath109 .
if the power series @xmath223 with the corresponding sequence of coefficients @xmath224 is convergent at @xmath225 and @xmath219 , then the polynomial series is convergent for all values of @xmath12 such that @xmath226
. two polynomial series of the same type can be added and multiplied by scalars giving the same type of series .
we can define the @xmath1-derivative of @xmath1-polynomial series : @xmath227 .
similarly , we have the @xmath2-derivative of @xmath2-polynomial series in the form @xmath228 .
additionally , if the @xmath1-polynomial series is convergent for @xmath229 and if the @xmath2-polynomial series is convergent for @xmath230 , then their derivatives are also convergent on the same sets . from corollary [ derdor1 ]
we obtain the following result .
[ derser ] let @xmath231 , @xmath232 , and @xmath233 be a sequence such that @xmath234 for each @xmath235 .
we have :
\(a ) let @xmath236 .
then , the series @xmath237 is convergent for @xmath238 , and @xmath239 exists and it is convergent for @xmath238 .
\(b ) let @xmath240 .
then , the series @xmath241 is convergent for @xmath242 , and @xmath243 exists and it is convergent for @xmath242 .
let @xmath10 , @xmath11 .
there is no problem with convergence ( i ) in points @xmath244 for series of the first type , ( ii ) at points @xmath245 for series of the second ( `` hat '' ) type , because such series are finite . in @xcite and
@xcite one can find generalized series for an exponential @xmath246 with constant function @xmath120 : for @xmath12 and @xmath244 one has @xmath247 it follows that @xmath248 , which gives the rule @xmath249 .
as in proposition [ diamexpder ] , let us consider now a time scale with @xmath73 .
then , @xmath250 where @xmath251 if @xmath252 .
this gives that @xmath253 and then @xmath254 is as in proposition [ diamexpder ] .
in @xcite it is proved that @xmath255 for @xmath244 .
then , we have that @xmath256 .
we obtain that the @xmath1-derivative of @xmath257 with respect to @xmath37 is given by the formula @xmath258 , where @xmath259 if @xmath260 .
this gives that @xmath261 and then @xmath262 is also as in proposition [ diamexpder ] .
the diamond-@xmath0 derivative reduces to the standard @xmath1 derivative for @xmath3 and to the standard @xmath2 derivative for @xmath4 .
the same `` weighted '' type definition is proposed for the diamond-@xmath0 integral .
based on this simple idea , we introduce diamond type monomials .
let us begin with the trivial remark that for any @xmath263 we can write @xmath264 .
assume that @xmath55 is @xmath195 delta- and nabla - differentiable on @xmath196 and @xmath200 , respectively .
let @xmath265 , @xmath12 .
then , @xmath266 where @xmath267 , and @xmath268 with remainders @xmath269 and @xmath270 given as in theorems [ tayth1 ] and [ tayth2 ] .
[ diam_ser ] let @xmath6 be a time scale and @xmath109 . by a _ combined - polynomial series ( on @xmath6 , originated at @xmath210 )
_ we shall mean the expression @xmath271 where @xmath12 and @xmath98 $ ] .
if in ( [ diam_ser1 ] ) we put @xmath99 , then we have a @xmath1-polynomial series . for @xmath100
we obtain @xmath2-polynomial series .
a combined - series is convergent if both types of polynomial series are convergent . for fixed @xmath119
we get usual number series , so we can say that the series originated at @xmath210 is convergent at @xmath37 if it is convergent as a number series .
let @xmath10 , @xmath11 , and @xmath272 , @xmath273 be two real sequences with nonzero elements such that @xmath274 , @xmath275 .
then , the combined - polynomial series @xmath276 @xmath98 $ ] , is convergent for all @xmath12 . based on proposition [ prop3012 ] ,
we consider : @xmath277 , when combined - series @xmath278 ; @xmath178 , when the first part is finite the second is convergent ; @xmath176 , when we have opposite situation to the previous one .
let @xmath9 and @xmath279 .
then , @xmath280 and @xmath281 for @xmath140 .
additionally , @xmath282 for any @xmath283 .
but this series is not convergent for @xmath284 .
we have @xmath285 and @xmath286 .
the series @xmath287 is convergent for any @xmath288 . for
that let @xmath289 .
then , @xmath290 for each fixed @xmath291 .
the combined - polynomial series has the form @xmath292 and is convergent for @xmath283 .
polynomial series have been used in the literature for solving a variety of problems in control . in this paper
we define taylor series via diamond - alpha derivatives on time scales and provide the first steps on the correspondent theory .
such a theory provides a general framework that is valid for discrete , continuous or hybrid series .
we trust that the polynomial series here introduced are important in the analysis of control systems on time scales .
the first author was supported by biaystok technical university grant s / wi/1/07 ; the second author by the r&d unit ceoc , via fct and the ec fund feder / poci 2010 .
r. j. higgins and a. peterson .
cauchy functions and taylor s formula for time scales @xmath293 . in _ proceedings of the sixth international conference on difference equations _ , pages 299308 , boca raton , fl , 2004 . | _ the objective of this paper is twofold : ( i ) to survey existing results of generalized polynomials on time scales , covering definitions and properties for both delta and nabla derivatives ; ( ii ) to extend previous results by using the more general notion of diamond - alpha derivative on time scales .
we introduce a new notion of combined - polynomial series on a time scale , as a convex linear combination of delta and nabla generalized series .
main results are formulated for homogenous time scales . as an example , we compute diamond - alpha derivatives on time scales for delta and nabla exponential functions
. _ * keywords : * time scales , diamond-@xmath0 derivatives , generalized polynomials , generalized series . *
2000 mathematics subject classification : * 40c99 , 39a13 , 40a30 . |
eclipsing binary systems provide a good opportunity for studying the presence of an unresolved third body by observing their minima times because of the light - time effect ( hereafter lite ) .
it was explained by irwin and its necessary criteria have been mentioned by frieboes - conde & hertzeg and also by mayer .
presence of the third body in the system is possible only if the times of minima behave in agreement with a theoretical lite curve , the resultant mass function has reasonable value and corresponding variations in radial velocities are measured . in the last decade also
a confirmation by astrometry seems to be plausible . in each case
we have calculated new light elements of the eclipsing pair and also the parameters of the predicted third body orbit .
the tables i. and ii .
present results for each system , where @xmath0 are masses of components , @xmath1 computed period of the unresolved body , @xmath2 semiamplitude of lite , @xmath3 eccentricity , @xmath4 length of periastron , @xmath5 mass function and @xmath6 minimum mass ( for @xmath7 = 90@xmath8 ) of predicted body , respectively .
the subscript 3 and 4 denotes the parameter of the third and fourth body , respectively .
we have derived new linear light elements : + oo aql : @xmath9 + v338 her : @xmath10 + t lmi : @xmath11 + rv lyr : @xmath12 + tw lac : @xmath13 + v396 mon : @xmath14 + in the case of oo aql and t lmi also the hypothesis of the fourth body was suggested .
the results are in table ii . in the system
oo aql the components are very similar and also similar to the sun ( sp . g5v , t @xmath15 5700 k ) , so the second order variations could also be caused by the the magnetic cycles , similar to the sun ( so - called applegate mechanism , see eg .
applegate ) . in the system
v338 her also the quadratic term was applied , so the mass transfer in the system could be present . from this hypothesis ( with mass transfer parameter @xmath16 )
we have derived the mass transfer rate to @xmath17 . in a few cases here
the potential third body would be detectable in the detailed light curve analysis .
especially the cases where the third body has not negligible mass , comparing with the eclipsing pair .
regrettably we have no information about the distance of individual systems and also about the absolute magnitudes , so the determination of the angular separation of the possible third body is very uncertain .
references : [ 1 ] al - naimiy et.al . ; [ 2 ] budding ; [ 3 ] cester ; [ 4 ] budding ; [ 5 ] halbedel ; [ 6 ] yang & liu .
we have derived new lite parameters for six eclipsing binaries by means of an @xmath18 diagram analysis . in two cases ,
oo aql and t lmi , another variation was found , so there is a possibility of a presence of the fourth body in the system , or magnetic activity in them .
but we have not enough data to make a final decision .
so the consequence is , that for the confirmation of presence of the lite in these systems , we need detailed photometric , spectroscopic or astrometric data of these binaries .
this research has made use of the simbad database , operated at cds , strasbourg , france , and of nasa s astrophysics data system bibliographic services .
this investigation was supported by the czech - greek project of collaboration rc-3 - 18 of ministry of education , youth and sport and by the grant agency of the czech republic , grant no . | the period changes of six eclipsing binaries have been studied with focus on the light - time effect . with the least squares method we also calculated parameters of such effect and properties of the unresolved body in these systems . with these results we discussed probability of presence of such bodies in the systems with respect to possible confirmation by another method . in two systems we also suggested a hypothesis of fourth body or magnetic activity for explanation of the `` second - order variability '' after subtraction of the light - time effect of the third body . |
observations by @xmath3 of binary systems have so far yielded 11 transiting circumbinary planets . until recently ,
all discovered circumbinary planets have resided near to or outside the dynamical stability limit characterized by @xcite .
this finding has prompted many to study how planets form in such systems and why they seem biased towards lying at the brink of dynamical instability @xcite .
simulations and theoretical arguments by @xcite and @xcite rule out in situ formation in the inner edge of the disc near where several circumbinary planets have been observed . currently , one of the most successful models in explaining the circumbinary planet population is planetary migration . in this model , a planet ,
having formed farther out in the disc , migrates inward through viscous interactions with a gaseous disc , potentially undergoing planet - planet scattering , until it reaches near its observed location .
several previous studies applied migration to observed circumbinary planetary systems and have been able to show that for certain disc and viscous drag models , the observed planet orbits are nearly recovered ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
however , the recent discovery by @xcite of the first long - period transiting circumbinary planet complicates this picture .
the newly discovered planet candidate , koi-2939b , is a jupiter - sized exoplanet on a roughly 3 year orbit that suggestively lies in the conservative habitable zone of two short - period g dwarfs @xcite .
the existence of a such a system shows that not all circumbinary planets are driven , either through migration or other mechanisms , inward towards the dynamical stability limit .
this has important consequences for not only how planets migrate and evolve in a viscous circumbinary disc , but also for how and where these planets form within the protoplanetary disc . clearly , characterizing the influence exerted on
the planet by the protoplanetary disc is important for understanding how such systems form .
additionally , the influence of the central binary on the external disc causes the disc to evolve and potentially undergo large scale changes .
previous studies of both protoplanetary discs and accretion discs around binary supermassive black holes ( smbhs ) found that circumbinary discs can become eccentric , precess , and have density waves excited from resonances @xcite . specifically in the context of planetary systems , a hydrodynamic theory presented by @xcite showed that nonlinear coupling mediated by density waves launched at eccentric lindblad resonances causes disc eccentricity growth .
much theoretical work in this area has focused on how planetesimals can grow and evolve in circumbinary discs .
@xcite found that for planetesimals on circumbinary orbits , in situ formation proves quite difficult , suggesting that planets form far out in the disc and subsequently migrate inward .
more recently , @xcite showed that planetesimals on the most circular orbit about the central binary can attain small relative velocities , facilitating their growth . with a model that considers both
the gravity and gas drag of a precessing , eccentric circumbinary disc , @xcite found that the inner radius for @xmath4 m planetesimal growth depends on the disc eccentricity , density and precession rate .
the binary s influence on the disc is not without a cost , however , as the disc also drives changes in the binary orbital elements , which as they evolve , can change how the binary influences the disc .
early work on how binary systems interact with accretion discs by @xcite studied jupiter s interaction with the sun s protoplanetary disc .
this study showed that a satellite s orbital eccentricity could be increased through energy and momentum transfers with the disc at lindblad resonances , causing significant changes over a few thousand years .
the case of accretion discs around two objects was explored by @xcite and later by @xcite .
@xcite found that a central binary s interaction with an external accretion disc can decrease binary separation and change binary eccentricity , depending on disc structure . in the context of a jupiter to brown dwarf mass companion orbiting a central star
, @xcite found that for sufficiently massive companions , a coupling between small initial disc eccentricity and the companion s tidal potential excited an @xmath5 wave from the 1:3 eolr causing further disc eccentricity growth .
wave excitation at the 1:3 eolr can also couple with the interaction between the eccentric disc and the companion to induce orbital eccentricity growth of the companion .
many recent and past works explored the more general case of binary stars evolving under the influence of a gaseous circumbinary disc .
simulations of unequal mass binaries embedded in a protoplanetary disc by @xcite show rapid binary eccentricity growth and semimajor axis decay due to interactions with the 1:3 eolr and the viscous disc .
subsequent theoretical work by @xcite and @xcite derived relations to show how resonant and viscous interactions drive @xmath6 and @xmath7 .
binary orbital element and disc evolution has also been explored in systems with circumbinary gaseous discs containing migrating planets in simulations by @xcite who also found binary and disc eccentricity growth . characterizing disc - binary interactions
are important on much larger scales , as well .
simulations by @xcite show that eccentric supermassive black hole binaries can rapidly form from the merger between two spiral galaxies .
an external disc forms exterior to the smbh binary as interactions with the disc and external gas clouds can cause the black hole separation to decrease @xcite .
numerous studies have been conducted to explore how binary smbh - disc interactions cause binary smbh eccentricity growth and semimajor axis decay , possibly explaining the last parsec " problem allowing smbh binaries to coalesce ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
extensive work has also been made to explore accretion onto binary smbhs , how it varies with both the binary and disc properties and what impact accretion has on the dynamical properties of the system ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) . for circumbinary systems , the binary and its disc
are intertwined in non - trivial ways .
the coevolution of such systems depends strongly on resonant interactions that can impart significant changes on both the binary and disc .
previous simulations focused on studying either disc evolution or binary evolution , often making approximations such as holding the binary orbital elements fixed . in this work
we seek to explore how the coevolution of the disc and binary proceeds by allowing all particles to gravitationally interact . using the kepler 38 binary as our model system
, we present the results of n - body sph simulations of unequal mass binaries of variable initial eccentricity that probe the dynamical coupling of binary stars and a circumbinary disc .
we show that the initial eccentricity of the binary dictates how strongly the disc - binary system resonantly couples .
the strength of this coupling in turn dictates how eccentricity grows in either the disc , the binary or both and how structure forms within the disc .
the simulations described in this paper were performed in 3d using the massively parallel n - body and smooth particle hydrodynamics ( sph ) code , changa @xcite .
changa , implemented in charm++ @xcite , uses a modified version of the barnes - hut tree algorithm with hexadecapole order multiples and a node opening criterion of @xmath8 for fast and accurate calculation of gravitational forces .
the euler equations which describe the gas dynamics of the simulated circumbinary disc were solved using an sph implementation based on @xcite .
changa uses a multistepping algorithm that gives each particle its own timestep to ensure sufficient dynamical resolution @xcite .
artificial viscosity was implemented using monaghan viscosity @xcite .
the viscosity @xmath9 and @xmath10 parameters were set to 1 and 2 , respectively .
the balsara switch was used to limit shear viscosity @xcite . for additional information about the implementation and performance of changa
see @xcite .
.kepler 38 parameters adapted from @xcite . [ cols="^,^,^,^",options="header " , ] [ tab : table_2 ]
as the simulation evolves , the time varying gravitational force of the binary truncates the inner edge of the circumbinary disc and excites various lindblad resonances within the disc .
the gap , cleared quickly in about 100 binary orbits , is preserved by a balance of resonant and viscous torques within the disc @xcite . for larger gaps , corotation and lindblad resonances can fall within the evacuated region removing their influence from the system .
these resonances , especially those closest to the inner edge of the disc , can drive evolution in the binary s orbital elements @xcite and pump eccentricity in the disc @xcite .
therefore , in order to understand the subsequent dynamical evolution of both the binary and the disc , the inner disc edge structure must be understood to see which resonances may play a role .
the approximate size of the gaps found in these simulations is @xmath11 with more eccentric binaries producing larger gaps , in good agreement with the results of @xcite .
our results are also consistent with the findings of @xcite who show that for binary mass ratios above @xmath12 as is the case for our simulations , a hollow central cavity forms around the secondary within the circumbinary disc
. gaps of this size tend to remove the eccentricity - damping 1:2 corotation resonance while leaving the 1:3 eolr as the closest to the inner disc edge at a radius of @xmath13 suggesting that this resonance dominates the evolution of the binary s eccentricity and semimajor axis @xcite . to verify the absence of the 1:2 corotation resonance and prominence of the 1:3 eolr ,
the surface density profiles for simulations @xmath14 are shown in fig .
[ fig : figure1 ] after 200 years of evolution . for all @xmath15 ,
the surface density at the 1:3 eolr is at least an order of magnitude larger than at the 1:2 corotation resonance when it is present within the disc .
since the gap opens rather quickly , in of order 5 years for the kepler 38 binary , this finding supports the hypothesis that the 1:3 eolr will be the dominant resonance within the disc that will drive subsequent evolution within the system .
the structure and eccentricity of the protoplanetary disc was examined over @xmath1 binary periods in over 230 snapshots for each simulation to track how it evolves with the central binary .
a central binary excites eccentricity in the surrounding circumbinary disc via resonant gravitational interactions @xcite .
previous simulations of circumbinary discs @xcite found that disc eccentricity increases due to interactions that occur at the 1:3 eolr . to explain why the disc becomes eccentric , @xcite showed that disc eccentricity growth occurs via a parametric instability driven by coupling between the binary s tidal potential and a disc @xmath2 mode due to a small initial disc eccentricity .
this coupling excites an @xmath5 spiral wave from the 1:3 eolr that removes angular momentum from the disc at constant energy making the gas orbits eccentric .
the material at the 1:3 eolr rotates more slowly that the orbital pattern speed allowing the resonant torques to grow eccentricity in the system through the @xmath5 wave .
we therefore expect our circumbinary discs to become eccentric as well . to explore this effect in our simulations , we computed the disc eccentricity via a mass average following the prescription of @xcite @xmath16 where @xmath17 is the local surface density and the integral was evaluated out to a radius of 3 au over 50 radial bins .
the disc eccentricity in each radial bin was taken to be the mass - weighted average of the eccentricity of all gas particles within the bin assuming the particles orbit the system s barycenter .
we neglected the influence of gas pressure in this calculation .
for all simulations , disc eccentricity growth is observed and consistent with the results of the similar gaseous circumbinary disc simulations of @xcite , @xcite and @xcite . the disc eccentricity change over time for each simulation
is shown in fig .
[ fig : figure2 ] . for @xmath18 ,
significant disc eccentricity growth occurs .
after about 500 years , the disc reaches eccentricities of about 0.1 while continuing to grow linearly . for larger initial @xmath15
up to @xmath19 , less eccentricity growth occurs indicating that more eccentric binaries tend to produce less eccentric discs . for non - zero @xmath15 , the disc eccentricity grows linearly with a periodic modulation .
the period of this disc eccentricity oscillation is similar to the inner disc edge clump precession timescale discussed in section [ discstructevolution ] below suggesting that the clump impacts the disc s eccentricity modulation , but only when the binary is sufficiently eccentric .
this finding is consistent with the work of @xcite who explain that the inner disc edge precesses with a period of @xmath20 binary periods when @xmath21 .
we find that the global disc eccentricity growth is predominantly due to the inner edge of the disc becoming eccentric .
[ fig : figure3 ] shows the disc eccentricity versus radius for simulations 1 - 5 after about 520 years of evolution . for binaries with @xmath18 , disc inner edge eccentricities are of order 0.4 while more eccentric binaries tend to produce less eccentric inner edges . to understand the eccentricity growth of our circumbinary discs ,
specifically why less eccentric binaries tend to produce more eccentric discs , we turn to the theory of @xcite .
@xcite explains that a nonlinear coupling between the binary and a small initial disc eccentricity excites an @xmath5 wave from the 1:3 eolr within the disc with a resonant forcing pattern speed @xmath22 for binary orbital frequency @xmath23 .
this wave transports angular momentum outwards , driving eccentricity growth in the system .
we confirmed the presence of the @xmath5 spiral wave originating from the 1:3 eolr in our simulations via a fourier transform over azimuthal angle of the disc surface density .
this wave removes angular momentum from the disc , accounting for the increase in disc eccentricity .
disc eccentricity growth as a function of @xmath15 depends on how strongly the binary couples to the disc .
when strong coupling occurs , both the binary and inner disc edge grow to similar eccentricities while weak coupling results in the less massive of the binary or the disc inner edge developing appreciable eccentricity . for initially circular binaries , @xcite described strong coupling as occurring when the disc mass within the gap radius is comparable to the mass of the secondary . for our simulations ,
the mass of the secondary is roughly an order of magnitude larger than the mass of the entire disc , so we expect the binary to be weakly coupled to the disc resulting in significant disc eccentricity growth as is observed in our simulations ( see fig .
[ fig : figure2 ] and fig .
[ fig : figure3 ] ) . for initially eccentric binaries
, the @xcite strong coupling criterion does not apply .
instead , we note that the time - averaged orbit of an eccentric binary corresponds to an azimuthal @xmath2 mode perturbation to a circular orbit .
this @xmath2 mode couples to the @xmath2 mode of the eccentric inner disc edge , placing the system in the strong coupling regime causing the disc eccentricity and @xmath15 to grow to similar magnitudes , as is observed in our simulations ( see fig .
[ fig : figure2 ] and fig .
[ fig : figure4 ] ) .
we discuss these coupling mechanisms and their implications in more detail in sections [ discstructevolution ] and [ binaryevolution ] .
the @xmath24 and @xmath25 simulations show larger disc eccentricity than the @xmath26 case in contrast to expected behavior .
for @xmath25 , this can be understood as intermediate coupling .
the initial binary eccentricity is not low enough to conform exactly to the @xcite criterion and is not large enough to couple strongly to the disc inner edge resulting in an intermediate coupling with larger disc eccentricity growth than the @xmath26 case and also appreciable @xmath15 growth that is still less than the @xmath26 case ( see section [ binaryevolution ] ) . for the @xmath24 case
, the binary eccentricity may be large enough that higher order resonances in the disc begin to impact the evolution @xcite , potentially accounting for the system s departure from expected behavior . the gravitational influence of the binary forces several major changes within the structure of the circumbinary disc .
to explore how the structure of the disc changes with time and @xmath15 , we examined the orbits of gas particles and how they vary with distance from the binary . to accomplish this ,
two - dimensional histograms of all gas particles within a radial distance of 3 au from the barycenter were made for each snapshot .
we computed the histograms over semimajor axis , @xmath27 , and the longitude of periastron , @xmath28 , defined in this work as the sum of the argument of periastron , @xmath23 , and the longitude of the ascending node , @xmath29 , relative to that of the binary , @xmath30 , for each gas particle .
[ fig : figure5 ] and fig .
[ fig : figure6 ] shows these histograms for all particles in the disc out to 3 au for several of our simulations . in all simulations ,
a precessing overdense knot was found just outside inner edge of the disc .
the knot corresponds to a coherent precession of eccentric gas particle orbits at the inner disc edge .
the knot depicted in fig .
[ fig : figure5 ] precesses relative to the binary in the prograde sense with a period of about 20 years , or 400 binary periods , for @xmath18 . fig .
[ fig : figure5 ] shows the @xmath31 histogram for two simulations with initial @xmath15 of 0 and 0.1032 , respectively after 200 years of evolution .
precession of orbits near the binary at the inner disc edge are expected due to the binary s time - varying potential as shown in the simulations and analytic theory of @xcite .
the existence of this knot is consistent with the identification of a similar overdense lump located at the inner edge of gaseous circumbinary accretion discs about binary black holes from 2d simulations by @xcite .
an additional structure identified in the disc is a single arm ( @xmath2 ) spiral wave launched from near the inner edge of the disc as shown in fig .
[ fig : figure5 ] .
the spiral wave is an alignment of gas particle longitude of periastrons relative to the binary s . for the initial @xmath26 simulation
, the wave develops rapidly within the first 50 years , or about @xmath32 binary periods .
recent n - body simulations of circumbinary planetesimal discs by @xcite confirmed the presence of an @xmath2 wave discs about eccentric binaries .
when under the influence of an asymmetric gaseous disc potential , @xcite identified the wave as a preferential alignment of planetesimal longitude of periastrons as a function of orbital radius around the eccentric binary of the kepler 16 system , similar to the spiral wave found in this work . in our simulations ,
the spiral wave s orientation remains locked to the binary s slow prograde @xmath30 precession throughout the entire simulation , although a slight drift of a degree or so does occur .
the wave s fixed orientation relative to the binary is an important effect whose consequences will be examined more carefully in section [ binaryevolution ] .
simply put , if the wave did circulate relative to the binary , it s long term effect on the system , if any , would average out to zero , so the fixed orientation could indeed dynamically impact the system . note that in the simulation with an initial @xmath33 ( left panel ) , no spiral @xmath2 arm exists in contrast to the @xmath26 simulation ( right panel ) which shows a prominent spiral arm .
the arm is also observed in the simulation with initial @xmath24 but not with initial @xmath34 .
since the spiral arm is only observed when the binary has an appreciable eccentricity , we can infer that a coupling between binary eccentricity and the inner disc edge impacts its formation . to investigate the role of binary eccentricity in exciting the wave , simulation 3 was ran with initial @xmath25 to see if the spiral wave could be excited with an intermediate @xmath15 between the two regimes identified above . in the intermediate regime as explained in section [ disceccevolution ] , @xmath15 is not large enough to be strongly coupled to the disc while also not low enough to weakly couple to the disc to drive disc eccentricity . from the onset of the initial @xmath25 simulation , a faint @xmath2 spiral wave appeared and
gradually strengthened as shown in fig .
[ fig : figure6 ] . the wave , initially weak , became more apparent after about 900 years .
the @xmath2 spiral arm in this simulation does not become as pronounced as the one seen in the initial @xmath26 simulation , suggesting that the strength of the arm depends on @xmath15 and supporting the notion that this disc - binary system undergoes an intermediate coupling .
we again apply the theory of @xcite to understand the origin and behavior of the spiral wave .
we know from section [ disceccevolution ] that a nonlinear coupling between non - zero disc eccentricity and the binary s tidal potential excites an @xmath5 spiral density wave from the 1:3 eolr that mediates angular momentum transfer in the system .
also , we have shown that the strength of the coupling between the binary and disc , which depends on @xmath15 , determines the magnitude of the disc inner edge eccentricity .
additional structural changes within the disc proceed via a higher order coupling . in @xcite ,
the authors show that the @xmath35 density wave emitted at the 1:3 eolr can couple back through the binary tidal potential .
this additional coupling produces a time independent @xmath2 wave and an associated potential .
the extra potential from the @xmath2 wave can allow for the removal of angular momentum from the system via resonant torques .
the @xmath2 wave produced via the recoupling mechanism is precisely the @xmath2 spiral wave identified in this work .
the presence of the @xmath2 wave was reconfirmed via a fourier decomposition of the disc surface density .
since we only observe the @xmath2 spiral wave in discs around eccentric binaries , we infer that this recoupling mechanism only occurs when the disc and binary are strongly coupled .
@xcite s observation of a @xmath2 spiral wave present in their simulation of a planetesimal disc surrounding the eccentric kepler 16 binary support this argument .
since the orientation of the @xmath2 wave in our simulations remains locked to that of the binary s , it does not circulate and hence is independent of time .
also , as we will explore in section [ binaryevolution ] , the additional potential from the @xmath2 resonantly torques on the binary causing evolution in its orbital eccentricity .
next we examine why the orientation of the spiral wave remains fixed relative to the binary .
we apply the analytic theory for circumbinary orbits of @xcite to partially explain this effect .
the theory , accurate to first order in @xmath15 , decomposes the orbit of a test particle about two stars into a superposition of the circular motion of a guiding center and the radial and vertical epicyclic motion due to the non - axisymmetric components of the binary s potential .
@xcite give the equations for the precession rate of the argument of periastron , @xmath23 , and the longitude of the ascending node , @xmath29 , respectfully , to be @xmath36 @xmath37 where @xmath38 and @xmath39 are the masses of the primary and secondary stars , respectively , @xmath40 is the binary semimajor axis and @xmath41 is the radial distance from the barycenter .
since these rates are approximately equal and opposite , one expects a gas particle s longitude of periastron , as defined earlier , to remain fixed , as is observed for the spiral waves in our simulations about sufficiently eccentric binaries . in the context of gaseous circumbinary discs ,
this interpretation has a few potential shortcomings .
first , a given gas particle does not live in isolation since it feels the effects of disc self gravity and pressure gradients within the disc that can impact its orbit .
also since this theory is only linear in @xmath15 , its applicability could lessen as the binary becomes more eccentric due to interactions with the disc .
however , the binaries considered have low to moderate eccentricities and the discs are rather low - mass such that disc self gravity is negligible so the gravitational influence of the binary should dominate
. therefore , we expect this theory to still provide a decent approximate explanation for why the spiral arm remains fixed relative to the binary .
secular theory @xcite and previous simulations @xcite show that angular momentum losses to a disc change the central binary s orbital elements .
angular momentum loss occurs mainly through resonant gravitational torques at the lindblad and corotation resonances . in the case of binary stars embedded in an external disc , loss of angular momentum
can result in changes to the binary eccentricity and semimajor axis . for unequal mass binary stars with low to moderate eccentricity
, the majority of eccentricity growth is due to resonant torques at the 1:3 eolr .
this resonance dominates since these systems tend to open a gap in the disc such that the 1:3 eolr resides nearest to the inner edge of the disc while the eccentricity damping 1:2 corotation resonance lies within the evacuated region @xcite .
simulations of binaries embedded in circumbinary discs by @xcite agree with this interpretation as they identified gravitational torque density peaks responsible for binary eccentricity evolution located at the 1:3 eolr in the disc . for the simulations presented in this work , we expect a secular increase in binary eccentricity and decrease in semimajor axis as the binary loses angular momentum to circumbinary disc . as shown in fig .
[ fig : figure4 ] , binary eccentricity for initially eccentric binaries grows over the duration of the simulation .
the eccentricity growth rate , @xmath6 , seems to scale with initial @xmath15 . in the simulation with initial @xmath24 ,
however , @xmath15 increases more slowly than the initial @xmath26 case .
when the initial @xmath42 , no binary eccentricity growth occurs in contrast to the results of similar simulations of binaries embedded in an external disc by @xcite who find significant eccentricity growth with an initial @xmath18 , a discrepancy we will address later . in all cases , @xmath15 oscillates as the simulations progress .
this oscillation is due to forcing by the @xmath2 potential of the eccentric external disc @xcite . for all simulations ,
the binary semimajor axis secularly decreases where the rate of decline is lower for increasingly eccentric binaries as shown in fig .
[ fig : figure7 ] . as before
, the initial @xmath24 case defies this trend as it shows a greater semimajor axis decline than the initial @xmath26 case instead of the expected lesser decline .
additionally , the binary s longitude of periastron slowly precesses in the prograde sense less than @xmath43yr@xmath44 over the duration of the simulation similar to the results of comparable simulations by @xcite .
any changes in the binary orbital elements will be depend on the details of the binary s interactions with the disc . as discussed in section [ disceccevolution ]
, we applied the secular theory of @xcite to show that the strength of the disc - binary coupling dictates how eccentricity grows within the system .
we apply the same arguments used above to understand disc eccentricity growth to binary eccentricity evolution . for circular binaries
, we argued that the binary and disc are weakly coupled since the mass of the secondary is much greater than the mass of the entire circumbinary disc . in this weak coupling regime
, the eccentricity of the less massive of the binary - disc system grows , which in this case is the disc .
as expected , the disc develops appreciable eccentricity . from these arguments
, we expect the binary to develop very little , if any , eccentricity .
this behavior is exactly what we observe in our simulations ( see fig . [
fig : figure4 ] ) . as a function of @xmath40 from simulations with initial @xmath15 of 0.05 , 0.1032 , and 0.25
over - plotted with the analytic fit of equation [ eqn : total_deda ] with @xmath45 .
as the system evolves , time advances towards the left in this depiction . ] for simulations of discs around eccentric binaries , we argued that the disc and binary are strongly coupled through the @xmath2 modes of the eccentric binary orbit and inner disc edge orbits . in this regime , both the binary and disc eccentricities grow together and further coupling between the disc and binary can occur . additional disc - binary coupling discussed at length in section [ discstructevolution ]
lead to the excitation of a time - independent @xmath2 spiral wave and associated potential from the 1:3 eolr whose orientation remains locked to that of the binary s orbit ( see fig .
[ fig : figure5 ] ) .
since this wave remains fixed relative to the binary , it resonantly torques the binary through the 1:3 eolr , removing angular momentum from the binary s orbit , increasing @xmath15 .
if the wave instead circulated over time , its potential would time - average to zero and have no effect on the binary orbital elements .
the presence of the @xmath2 spiral wave leads to the qualitatively different binary eccentricity evolution for the @xmath26 case relative to the @xmath18 case displayed in fig .
[ fig : figure4 ] . in all simulations
, the binary semimajor axis decreases due to energy dissipation from the viscous disc .
similar n - body sph simulations of binaries embedded in a circumbinary disc by @xcite showed that for binaries with mass ratio @xmath46 and @xmath19 similar to the systems examined in this paper , resonant interactions with the surrounding disc at the 1:3 eolr drive eccentricity growth and semimajor axis decay . in this regime , we expect the binary eccentricity growth observed by @xcite since the binary and disc are strongly coupled .
with the origin of binary eccentricity evolution understood using the theory of @xcite , we seek to quantify binary orbital evolution . following the analysis of @xcite
, we applied the theory of @xcite to quantify the eccentricity and semimajor axis evolution of a binary with initial @xmath19 embedded in an external gaseous disc due to resonant interactions with the 1:3 eolr using the following relation @xmath47 where @xmath48 is the potential component corresponding to the 1:3 eolr @xcite and @xmath49 and @xmath27 are the binary eccentricity and semimajor axis , respectively . for less eccentric binaries ,
the semimajor axis and eccentricty evolution is well - described by @xmath50 where @xmath51 is the effective standard viscosity parameter @xcite .
note that @xmath51 in equation [ eqn : deda1996 ] is not in general the same as the @xmath9 viscosity parameter discussed in section [ methods_section ] .
to relate @xmath51 and @xmath9 , we use the following relation from @xcite and @xcite @xmath52 where @xmath53 is the smoothing length , @xmath54 is the disc aspect ratio and the factor of @xmath55 comes from the @xcite derivation of the coefficient for the @xcite viscosity implementation used in changa @xcite .
the @xmath56 factor arises from our use of the balsara switch which limits shear viscosity by scaling both @xmath9 and @xmath10 @xcite .
the range of @xmath56 is [ 0,1 ] .
since both @xmath53 and @xmath56 can vary between gas particles and @xmath54 can vary radially as the disc evolves , we average over the disc to get @xmath57 and @xmath58 = 0.4 . given these values
, we set @xmath45 as the approximate value for our simulations with @xmath59 .
equations [ eqn : deda ] and [ eqn : deda1996 ] , derived by @xcite via examining the balance between viscous and resonant interactions at the inner disc edge , apply in separate regimes that depend sensitively on @xmath15 .
@xcite estimate that once @xmath60 , the 1:3 eolr dominates binary eccentricity growth while below this threshold for circular binaries , no eccentricity growth occurs .
these separate regimes correspond to the weak and strong disc - binary coupling for circular and eccentric binaries , respectively , discussed in previous sections . for nearly circular binaries , @xmath61 and @xmath62 as expected for weak disc - binary coupling .
conversely for eccentric binaries , @xmath63 and @xmath64 as demonstrated above for strong disc - binary coupling .
we combine equations [ eqn : deda1996 ] and [ eqn : deda ] to model how @xmath15 and @xmath40 should evolve under the influence of an external disc following the models of @xcite and @xcite @xmath65 hence , eccentricity increases while the semimajor axis decreases .
to verify that equation [ eqn : total_deda ] is a proper model for our simulations , the observed evolution of @xmath15 and @xmath40 for simulations with initial @xmath66 0.05 , 0.1032 , and 0.25 were compared with the theoretical result of equation [ eqn : total_deda ] assuming @xmath45 .
the comparison is shown in fig .
[ fig : figure8 ] . the results of both the simulations with initial @xmath26 and @xmath24 are in good agreement with the theoretical expectations of equation [ eqn : deda ] and also in accordance with the simulation of a similar system with initial @xmath67 by @xcite .
equation [ eqn : total_deda ] can also be applied to the initial @xmath18 cases . for these simulations ,
the binary semimajor axis decreases via energy dissipation through the viscous disc while no eccentricity growth occurs as seen in fig .
[ fig : figure4 ] and fig .
[ fig : figure7 ]
. our results are consistent with the prediction of equation [ eqn : total_deda ] .
these findings indicate that equation [ eqn : total_deda ] successfully quantifies how the binary evolves in the different disc - binary coupling regimes .
the initial @xmath15 = 0.05 case proves troublesome .
equation [ eqn : total_deda ] does a poor job fitting the binary eccentricity and semimajor axis evolution .
the poor fit can be understood in the context of how the disc and binary undergo an intermediate coupling in between the strong and weak regimes .
as discussed in sections [ disceccevolution ] and [ discstructevolution ] , the binary eccentricity is not large enough to launch a prominent @xmath2 spiral wave and drive the eccentricity growth for more eccentric binaries . in between regimes
, we expect the binary eccentricity to grow weakly as we observe in our simulations .
equation [ eqn : total_deda ] succeeds for systems firmly in the weakly or strongly coupled regime but does not perform well for intermediate coupling .
this behavior has interesting consequences for the subsequent evolution of a system . for the intermediate case
as the binary eccentricity grows with time , it will eventually reach @xmath19 and will then begin to strongly couple to the gaseous disc .
as discussed previously , strong coupling launches a @xmath2 spiral wave in the disc and increases the growth rate of binary eccentricity .
also in the strongly coupled regime , we expect the disc eccentricity to decrease from higher values and be similar in magnitude to the binary eccentricity . in a similar vein for nearly circular binaries ,
we expect them to remain circular .
other simulations of binaries embedded in circumbinary discs such as those by @xcite and @xcite have found that initially circular binaries eventually develop appreciable non - zero eccentricities .
this behavior can be understood by examining equation [ eqn : total_deda ] .
if the binary is perturbed and some non - zero eccentricity develops , we would expect the binary eccentricity to grow very slowly , gradually strengthening the coupling between the disc and binary until intermediate coupling is reached and the system progresses as described above .
this picture is consistent with the results of @xcite who found the binary eccentricity began to grow on timescales longer than those explored in this work .
therefore over timescales much longer than simulated here , we would expect our initial @xmath68 case to become appreciably eccentric .
we extrapolate the results of our simulations to longer timescales and consider the consequences below .
we ran additional simulations to ensure that the 1:3 eolr did indeed dominate binary evolution and no other effect played a major role . to do this ,
a shorter simulation with an initial binary eccentricity of 0.1032 was performed with the initial disc gap radius located outside of the 1:3 eolr .
minimal binary eccentricity growth and effectively no binary semimajor axis decay occurred until the disc viscously spread inward .
additionally , the @xmath2 spiral wave observed in other simulations of sufficiently eccentric binaries also did not exist until mass was able to drift inward and accumulate at the 1:3 eolr at which point the @xmath15 and @xmath40 began to evolve .
these findings support the supposition that interactions with the 1:3 eolr drives the binary evolution as anticipated . to study what effect ,
if any , accretion has on the how the binary stars orbital elements vary , a procedure similar to that used by @xcite was performed . for a given simulation
each accretion event was tracked such that the accreted gas particle s mass and velocity components were outputted . using these events , @xmath15 and @xmath40
were evolved by adding each accreted particle to the binary imposing linear momentum and mass conservation as is done natively in changa for sink particles .
this test demonstrated that accretion had a negligible effect on @xmath15 and @xmath40 .
therefore , it is safe to assume that the evolution of the binary orbital elements is primarily driven by interactions with the external disc , in particular at the 1:3 eolr for the systems considered in this work .
previous studies have examined how varying disc properties can change how a circumbinary disc evolves .
for example , @xcite showed that disc eccentricity is sensitive to the initial disc surface density gradient and aspect ratio . here , we analyse the results of simulations that vary disc mass , gas resolution , and aspect ratio in order to examine how disc properties impact the disc - binary coevolution . to study how our results vary with disc mass
, three additional simulations were run with @xmath69 to see if varying disc mass changes binary evolution .
simulation 6 with @xmath70 , simulation 7 with @xmath71 , and simulation 8 with @xmath72 were run . for these additional simulations
, we expect the disc to be strongly coupled to the eccentric binary since the coupling only depends on the magnitude of binary eccentricity .
we do expect , however , that eccentricity growth to occur more quickly for systems with more massive discs since torque scales with the disc mass .
the results of the simulations are shown in fig .
[ fig : figure9 ] . the disc eccentricity of simulation 7 was similar to that of simulation 4 while simulation 6 showed larger disc eccentricity values .
as expected , more massive discs became more eccentric .
not depicted is the spiral arm development .
similar to the standard initial @xmath26 case , a prominent @xmath2 spiral wave quickly forms in simulations 6 , 7 , and 8 consistent with the picture that binary eccentricity determines how strongly the disc and binary couple . in all three simulations ,
the wave has the same shape and remains fixed relative to the binary except for simulation 6 which showed slight prograde precession of the spiral arm . for more massive discs , binary eccentricity grew more quickly and the binary underwent more semimajor axis decay . although not plotted , both simulations 6 , 7 and 8
are still well - described by equation [ eqn : total_deda ] and hence correspond to either faster or slower binary evolution timescales . to ensure that our simulations were sufficiently resolved , we ran two additional simulations , simulations 9 and 10 , which has decreased and increased the initial number of gas particles by a factor of 2 to @xmath73 and @xmath74 , respectively .
the results of these simulations are shown in fig .
[ fig : figure9 ] . in both simulations ,
we find the general trend of eccentricity growth and semimajor axis decay holds .
the lower resolution simulation 9 eccentricity growth is less than the fiducial simulation 4 .
in addition , the disc eccentricity does not oscillate as seen in other simulations .
this suggests that the clump which dominates the disc eccentricity does not form into a coherent structure , indicating that n@xmath75 might not be large enough to properly resolve all the physics at the disc inner edge .
we find that the @xmath2 spiral wave appears in the disc indicating that the binary strongly couples to the disc as we expect from our previous simulations .
the higher resolution simulation 10 exhibits both binary and disc eccentricity evolution that is in good agreement with the fiducial simulation 4 .
additionally , we again observed a prominent @xmath2 spiral wave within the disc that behaved identically to its simulation 4 counterpart .
one disagreement between simulation 10 and simulation 4 is that the higher resolution simulation displayed less binary semimajor axis decay .
this result is expected , however , since higher resolution n - body sph simulations will have smaller gas softening lengths , @xmath53 . as shown in equation [ eqn : alphas ] , the effective standard viscosity parameter @xmath76 . the theoretical work of @xcite estimate that @xmath77 .
we therefore expect a higher resolution run with smaller @xmath53 and hence smaller @xmath51 to exhibit less binary semimajor axis decay due to dissipation from the viscous disc .
since the main effects explored in this work , the binary eccentricity evolution and the accompanying development of disc eccentricity and structure , are in good agreement between the standard and higher resolution runs , we find that our nominal resolution of n@xmath78 is sufficient .
simulations of accretion discs around binary black holes have examined the effects of larger aspect ratio discs , mainly focusing on accretion rates .
two dimensional sph simulations of gas accretion onto binaries embedded in a circumbinary disc by @xcite found that increasing gas temperature leads to increased accretion rates onto the primary star and growth in the binary mass ratio .
simulations of discs about black hole binaries by @xcite showed that discs with aspect ratios h / r @xmath79 have enhanced accretion rates as the inspiralling gas is not supressed by the binary s gravitational torque . to examine disc aspect ratio s effect on disc - binary evolution in our simulations , we ran simulation 11 with a disc aspect ratio of @xmath80 . to initialize the disc with a larger aspect ratio , we increased the disc temperature by setting t@xmath81 in equation [ eqn : disc_temp_profile ] to 2500 k giving us about a factor of @xmath82 larger aspect ratio relative to the fiducial simulation 4 . the result of simulation 11 is presented in fig . [ fig : figure9 ] .
we found that the thicker disc resulted in greater binary semimajor axis decay and eccentricity growth compared to the fiducial simulation 4 .
the increased binary semimajor axis decay agrees with the theoretical expectation of @xcite who estimate that @xmath83 .
the enhanced binary eccentricity growth follows as a consequence of equation [ eqn : deda ] . in simulation 11 ,
the binary accretion rate was enhanced by about a factor of @xmath84 relative to the fiducial simulation 4 in agreement with the general findings of both @xcite and @xcite .
we found that accretion had little impact on the binary orbital element evolution . as expected from arguments presented in section [ discstructevolution ] ,
the disc and binary were strongly coupled , producing a prominent @xmath2 spiral wave similar to the one seen in fig .
[ fig : figure5 ] .
the disc also displayed larger eccentricity initially but it did not grow appreciably over the course of the simulation .
the coevolution of a binary with a gaseous circumbinary disc , primarily driven by resonant interactions at the 1:3 eolr , has several important consequences for the subsequent dynamical evolution of the system .
as shown above , an eccentric binary system tends to gain eccentricity and experience a secular decay in semimajor axis due to viscous and resonant interactions with the disc .
this evolution not only changes due to additional feedback with the disc , but also impacts regions in the disc where planets form and migrate . n - body simulations of an unequal mass binary embedded in a protoplanetary disc by @xcite found rapid semimajor axis decay leading the authors to suggest that the binary separation may become small enough that tidal effects or even stellar coalescence may occur for such systems
. tides between stellar companions tend to circularize the orbit over long timescales once the stellar separation becomes sufficiently small .
detailed studies of companions to sun - like stars by @xcite and measurements of solar - type spectroscopic binaries in m35 by @xcite both found that binaries from these populations with periods less than about 10 days tend to be circularized .
theoretical work on the premain - sequence evolution of @xmath85 @xmath86 binaries by @xcite demonstrate that binaries with orbital periods of about 8 days or less are tidally circularized with effectively all of the circularization occurring before the stars reach the main - sequence . for binaries with an initial period slightly greater than the @xmath87 day tidal circularization boundary embedded in a circumbinary disc , binary - disc interactions could potentially decrease @xmath40 enough to make tidal effects important for subsequent evolution given that the lifetimes of protoplanetary discs are of order 1 myr @xcite .
in addition , we would expect some longer period binaries to develop appreciable eccentricity through this mechanism .
observations of spectroscopic binaries discussed by @xcite show that a large number of such binaries have large eccentricities , some up to @xmath88 , suggesting that disc - binary interactions may in fact be an important mechanism in pumping binary eccentricity
. ideally , additional observations of binaries with circumbinary planets , systems guaranteed to have had protoplanetary discs , will allow us to better constrain and model this effect .
the extent to which disc - binary interactions impact astrophysical systems over the disc s lifetime is difficult to measure . over the course of the disc s lifetime
, what may occur is some process that removes the 1:3 eolr from the disc . as @xmath40 decays through disc - binary interactions , the location of the 1:3 eolr shifts inwards .
also as @xmath15 grows , the central gap size increases @xcite .
the combined @xmath40 and @xmath15 evolution could result in the 1:3 eolr moving into the evacuated disc gap , removing its influence from the system , leaving higher order resonances to influence the binary .
for binaries with large @xmath15 , @xcite speculates that the combination of higher order inner and outer lindblad resonances and corotation resonances should combine to reduce the magnitude of @xmath6 and @xmath7 , potentially preventing subsequent evolution .
this picture is not so simple , however , as simulations of binary smbhs embedded in gaseous discs by @xcite and @xcite both find that binary eccentricity growth continues to @xmath89 where this growth did not slow until @xmath90 .
we note that the simulations of @xcite assumed a fixed @xmath40 which neglects the inward motion of the resonances as @xmath40 decays , potentially leaving them in the evacuated region , removing their effects from the system .
the impact of higher order resonances on binary evolution is a complicated matter that requires proper treatment in which both the binary and disc are allowed to coevolve together and likely depends on disc structure and artificial viscosity implementation .
additionally , findings by @xcite show that in principle , there is no limit to the amount of angular momentum that can be lost by a central binary to an external disc suggesting that binary coalescence is not as unrealistic as it sounds .
we caution that when performing simulations of binaries embedded in a gaseous disc that explore the role of semimajor axis decay , one should ensure that their observed semimajor axis decay has converged as both resolution and non - trivial effects such as accretion ( e.g. * ? ? ?
* ) can have a substantial impact .
one effect not explored in this work is the possibility of kozai - lidov ( kl ) oscillations for the general case of a misalign discs in binary systems . for an inclined test particle orbiting one component of a binary ,
periodic kl oscillations allow for the particle s eccentricity to grow at the expense of its inclination @xcite .
for the case of an inclined circumstellar disc about one component of the binary , @xcite found that the disc can exhibit kl cycles with the periodic disc eccentricity maxima approaching @xmath91 .
a later study of similar systems by @xcite demonstrated that misaligned discs can become much more extended than coplanar discs and potentially could overflow the roche lobe of the star .
simulations of misaligned circumbinary discs by @xcite showed that discs of almost all inclinations can tear leading to massive accretion and potentially a merger of the central binary .
given these results in the general case of systems with misaligned circumbinary discs , the binary eccentricity evolution is likely significantly impacted by the disc evolution and depart from the results presented here for thin , coplanar discs .
the disc , if misaligned , could reach large eccentricities due to kl oscillations and via interactions with the binary if it does not tear .
if the disc does in fact tear , the binary would likely not couple with the disc at all but could in fact coalesce as demonstrated by @xcite .
the general case of a binary coupling with an inclined circumbinary disc is greatly complicated by kl oscillations , torque scaling with inclination and the potential for tearing and warrants a more robust future study .
the observed orbital elements of binary stars that host a circumbinary planet are the product of a complex evolutionary history . from fig .
[ fig : figure4 ] and fig .
[ fig : figure7 ] , we see that for systems similar to the ones considered in this work , appreciable changes can occur on order @xmath1 binary orbits .
as shown in section [ varyingdiscmass ] , the mass of the disc strongly influences the binary evolution .
more massive discs , for example , drive much faster @xmath15 growth and @xmath40 decay .
faster dynamical binary evolution due to massive discs could be particularly relevant for _ kepler _ circumbinary planets as the work of @xcite suggests that these circumbinary planets formed and migrated in massive discs .
additionally , disc - binary interactions can make planet formation more difficult .
simulations by @xcite identified an @xmath2 spiral wave in the circumbinary disc that corresponds to an alignment of planetesimal longitudes of periastron .
this wave , whose origin was explained in this work , caused an increase in erosive planetesimal collisions making in - situ formation difficult in circumbinary protoplanetary discs .
the decay of @xmath40 via disc - binary interactions also causes the inward shift of mean motion and lindblad resonances .
these resonances can significantly impact the orbital stability of local objects in the disc in several important ways .
for the restricted three body problem , resonance overlapping can lead to stochastic orbital evolution as shown from the criterion derived by @xcite . for the case of binary orbital evolution driven by tides ,
@xcite point out that evolving binary eccentricity and semimajor axis changes the location of critical resonances and hence where they overlap , potentially making stable systems unstable over time .
the location of mean motion resonances also dictate where circumbinary planets may reside .
the numerical integrations of both @xcite and @xcite show that many circumbinary planets lie in a stable region shepherded by unstable mean motion resonances . if @xmath40 evolves significantly on short enough timescales , so too do the locations of the resonances , sweeping inward and potentially destabilizing orbits .
we note , however , that @xmath40 evolution appears to be a resolution dependent effect which future work should address .
this behavior is of particular importance for studies of planetary migration in circumbinary discs .
studies of circumbinary planetary migration in a viscous , eccentric disc find that planets tend to migrate inwards until they are trapped in or near the 4:1 mean motion resonance in the region of stability identified by @xcite @xcite .
since the resonances and the region of stability move as binary eccentricity and semimajor axis evolve , the final location and stability of migrating planets in circumbinary discs is sensitive to binary evolution . simulations of circumbinary systems , especially those using n - body sph methods like the ones presented in this work , must ensure that they properly account for the disc - binary interactions .
in this work , we showed that unequal mass binary stars embedded in a circumbinary gaseous disc carved out a gap in the disc and caused structural changes within the disc .
resonant interactions with the binary at the 1:3 eolr excited disc eccentricity .
sufficiently eccentric binaries excited a @xmath2 spiral wave within the disc .
this wave corresponded to an alignment of gas particle longitude of periastrons that varied with radius . the spiral wave formed within 50 years for discs about sufficiently eccentric binaries but took longer to strengthen for less eccentric binaries ( see fig .
[ fig : figure5 ] ) .
eccentric binary stars became more eccentric and experienced a secular decrease in semimajor axis while initially nearly circular binaries underwent no eccentricity growth over the timescales considered .
eccentricity growth within the system was understood in the context of the theory of @xcite in which nonlinear coupling between non - zero disc eccentricity and the binary s tidal potential excites an @xmath5 spiral density wave from the 1:3 eolr that mediates angular momentum transfer in the system .
nearly circular binaries weakly couple to the external disc and drive the inner disc edge to become very eccentric .
eccentric binaries , however , strongly couple to the disc leading to eccentricity growth for both the disc and binary .
the origin of the @xmath2 wave within the disc is understood as a recoupling of the @xmath92 spiral density wave with the binary tidal potential .
this model does have limited applicability as disc gap size scales with @xmath15 , so the 1:3 eolr could fall within the evacuated region removing its effect from the system , potentially slowing down binary evolution . for simulations of gaseous circumbinary discs , we caution that the disc - binary interaction must be sufficiently accounted for to properly model the system .
we leave the characterization of the long - term impact of disc - binary interactions to future work
. limitations of this work include the difficulty in integrating the binary orbit .
since the binary feels the force of every other sph particle in our simulations and is integrated using changa s native leapfrog integrator , very conservative timestepping was employed to ensure that the binary orbit was well - resolved and physically accurate .
the conservative timestepping scheme significantly slowed our simulations . in light of this limitation ,
potential future work could include running a long - term higher resolution simulation over at least @xmath93 binary orbits for small yet non - zero @xmath15 in order to better characterize how the disc and binary coevolve .
additional future work could involve examining equal mass binaries or binaries with larger eccentricities than those explored in this work . since binaries with large eccentricities excite higher order resonances within the disc ( e.g. * ? ? ?
* ; * ? ? ?
* ) and carve out gaps that could remove the 1:3 eolr from the disc @xcite , it would be interesting to examine how these other resonances can impact binary evolution .
a study on how different numerical viscosity implementations impact binary evolution would also prove fruitful to examine its influence on disc - binary coevolution , specifically binary semimajor axis decay .
we thank the anonymous referee for helpful comments and suggestions that improved the quality of the manuscript .
we would also like to thank isaac backus for a careful reading of the manuscript and jacob lustig - yaeger for helpful feedback .
this work was facilitated through the use of advanced computational , storage , and networking infrastructure provided by the hyak supercomputer system at the university of washington .
we made use of _ pynbody _
( https://github.com/pynbody/pynbody ) in our analysis for this paper .
this work was performed as part of the nasa astrobiology institute s virtual planetary laboratory , supported by the national aeronautics and space administration through the nasa astrobiology institute under solicitation nnh12zda002c and cooperative agreement number nna13aa93a .
david fleming is supported by an nsf igert dge-1258485 fellowship .
thomas quinn is supported by nasa grant nnx15ae18 g . | the recent discoveries of circumbinary planets by @xmath0 raise questions for contemporary planet formation models . understanding how these planets form requires characterizing their formation environment , the circumbinary protoplanetary disc , and how the disc and binary interact and change as a result .
the central binary excites resonances in the surrounding protoplanetary disc that drive evolution in both the binary orbital elements and in the disc . to probe how these interactions impact binary eccentricity and disk structure evolution , n - body smooth particle hydrodynamics ( sph ) simulations of gaseous protoplanetary discs surrounding binaries based on kepler 38 were run for @xmath1 binary periods for several initial binary eccentricities .
we find that nearly circular binaries weakly couple to the disc via a parametric instability and excite disc eccentricity growth .
eccentric binaries strongly couple to the disc causing eccentricity growth for both the disc and binary .
discs around sufficiently eccentric binaries that strongly couple to the disc develop an @xmath2 spiral wave launched from the 1:3 eccentric outer lindblad resonance ( eolr ) that corresponds to an alignment of gas particle longitude of periastrons .
all systems display binary semimajor axis decay due to dissipation from the viscous disc .
= 1 [ firstpage ] binaries : general hydrodynamics protoplanetary discs |
the idea of a macroscopic state demonstrating quantum mechanical behaviour was introduced by schrdinger in 1935 @xcite .
his famous thought experiment considered how a macroscopic entity ( in this case a domestic cat ) could evolve into a superposition of two distinct physical states ( alive and dead ) when entangled with a microscopic system that obeyed the laws of quantum mechanics .
this possibility of a macroscopic system being simultaneously in two distinct physical states was initially considered to be a flaw in quantum mechanical theory @xcite .
however , experiments have shown the predictions of quantum mechanics to be correct as superpositions of macroscopically distinct physical states have been produced in a variety of systems .
these are often referred to as ` schrdinger cat states ' .
for example , a group of six beryllium ions has been put into a superposition of two hyperfine states @xcite , and persistent currents ( of a few @xmath0a ) of opposing circulation in squids have been detected @xcite .
bose - einstein condensates ( becs ) are attractive candidates for generating macroscopic superpositions due to the large number of atoms that share a single quantum state .
this may provide a useful system in which to further test the validity or boundaries of the assumption of macroscopic realism .
a macroscopic superposition has yet to be demonstrated in a bose - einstein condensate , although there exist numerous proposals for generating either a superposition of relative phase or number states @xcite . in the following we consider a particular kind of superposition state of a single component bec in a double well , that is , a superposition of the two states where the entire condensate is localised in one of the wells ( sometimes referred to as a ` noon ' state ) .
if realisable , this kind of superposition promises to be useful in quantum information applications and precision interferometry , due to the measurement uncertainty scaling inversely with the number of particles ( the so - called heisenberg limit ) .
a major difficulty in realising macroscopic quantum superpositions is decoherence , which occurs when interactions with the surrounding environment cause the pure superposition state to decay into a statistical mixture . however , this paper is not concerned with avoiding decoherence in the realisation of macroscopic superpositions .
instead , we concentrate on measurements aimed at distinguishing a coherent superposition from a statistical mixture .
we find even in a decoherence - free environment , demonstrating a superposition presents several practical challenges .
once realised , measurements of the purity of the state could be useful in studying , for example , rates of decoherence .
this paper begins with an introduction of the two - mode description of a double - well condensate in section [ sec2 ] . in section [ sec3 ]
we consider measurements aimed at distinguishing between a coherent superposition and a statistical mixture .
we focus on quadrature - based measurements , analogous to those used in quantum optics , that can be realized with a ramsey - type interference experiment .
the coherence of the noon state is evident in parity measurements of the number distribution after the two modes are interfered a difficult measurement with standard atom counting techniques .
we analyse the effects of atomic interactions and discuss atom loss during the interference procedure .
finally , we show in section [ sec4 ] that a mesoscopic superposition of 20 atoms could be generated in a reasonable time frame if a feshbach resonance can be used to tune the atomic interactions , before concluding in section [ sec5 ] .
the hamiltonian for a condensate in an external trapping potential , @xmath1 , is @xmath2 } , \label{hamiltonian}\ ] ] where @xmath3 is the field operator for the condensate , and the non - linear interaction parameter is @xmath4 , ( @xmath5 is the s - wave scattering length describing two - body collisions within the condensate , @xmath6 is the atomic mass ) .
we consider the case where the external potential provides a double well confinement for the condensate .
double well potentials can be generated by an optical lattice with an additional harmonic confinement to reduce the number of occupied lattice sites to two @xcite .
they can also be realised on chips , where suitably arranged current carrying wires create a magnetic confinement for the condensate atoms @xcite .
when a double well potential is considered , the above hamiltonian can be simplified by making use of the two - mode approximation .
this means we consider each atom to be in some linear superposition of being in the left well and being in the right well .
we consider the zero temperature case , where all atoms in the system are condensed .
if the ground state energies of the condensate in the two single ( and separate ) wells are sufficiently separated from the energies of the condensate in all other excited single particle states , transitions to or from the two modes of interest and these higher lying states can be neglected .
this is required for the two - mode description to be valid .
in the two - mode approximation , the field operator is expanded as @xmath7 where @xmath8 and @xmath9 are discrete bose annihilation operators for the left and right well respectively , and @xmath10 are the ground state spatial wave functions of the condensate in the left and right wells . substituting this into equation ( [ hamiltonian ] ) ,
we find an effective hamiltonian @xmath11 where we have neglected the spatial overlap of the left and right well densities .
the single well bound state energies , @xmath12 , are @xmath13 @xmath14 , the tunnel coupling , is @xmath15 and the effective non - linear interaction terms are @xmath16 for the remainder of this paper we assume a symmetric potential , where @xmath17 and @xmath18 . by ignoring all possibility of atom loss and decoherence , we can efficiently represent the @xmath19-body wave function using the basis @xmath20 , representing states with @xmath21 atoms in the left well , and @xmath22 atoms in the right well .
any wave function can be written as a superposition of these number states , i.e. @xmath23 where @xmath24 . in this representation ,
the expectation value of the number of atoms in the left well is @xmath25 and the variance in the number difference is @xmath26 ^ 2 .\ ] ] for any initial state , @xmath27 , these coefficients have a time - dependence given by @xmath28 taking the derivative with respect to time , we find @xcite @xmath29 inserting the hamiltonian , ( [ hamiltonian ] ) , into this expression gives the equations of motion for the number state coefficients @xmath30 c_n(t ) + \hbar u_r \left[n^2-n \right ] c_n(t ) \nonumber \\ & - & \hbar\kappa \sqrt{\left(n - n+1 \right ) n } ~c_{n+1}(t ) -\hbar\kappa \sqrt{\left(n+1 \right ) \left(n - n \right ) } ~c_{n-1}(t ) .
\label{cnbydt}\end{aligned}\ ] ] the ground states and dynamics of a condensate in the double well can then be found by solving these equations numerically .
[ fig : gndstates ] shows the ground state probability distributions for the atom number in the left well ( found using imaginary time propagation @xcite ) for three different regimes - negligible , intermediate , and strongly attractive interactions .
the variance in the number difference for these are 12 , 293 , and 396 respectively ( a perfect noon state would have a variance of 400 ) . when non - linear interactions are weak compared to the tunneling , the ground state number distribution is essentially binomial ( fig .
[ fig : gndstates ] ( a ) ) . for large
n this can be approximated as a poissonian distribution and the ground state would be a coherent state ( which has equal mean and variance of atom number in each well ) . for intermediate attractive interactions ,
ground states exist consisting of two well separated peaks in the number state distribution .
these can be considered as superpositions of distinct physical states , and we refer to them as ` mesoscopic ' superpositions @xcite .
the ground state in the limit of infinitely strong attractive interactions is a macroscopic superposition of the entire condensate localized in each well .
this can be understood by realizing that it is energetically preferable for the condensate to be localized in one well , but that given the symmetry of the double well potential these two localized states are degenerate .
thus in the ground state it is equally likely that the condensate will be found entirely in one well as in the other ( hence the two peaks in the probability distribution ) .
this state is the macroscopic superposition that we are interested in .
[ fig : gndstates ] ( c ) shows that with a ratio of non - linear interaction strength to tunnelling rate @xmath31 , the ground state of the double well condensate is close to , but not quite , an ideal superposition .
we return to the problem of creating such a state in sec .
[ sec4 ] . is the atom number in the left well .
interaction strengths are ( a ) @xmath32 , ( b ) @xmath33 , and ( c ) @xmath34.,title="fig : " ] is the atom number in the left well .
interaction strengths are ( a ) @xmath32 , ( b ) @xmath33 , and ( c ) @xmath34.,title="fig : " ] is the atom number in the left well .
interaction strengths are ( a ) @xmath32 , ( b ) @xmath33 , and ( c ) @xmath34.,title="fig : " ]
we now consider the problem of demonstrating that a state is a macroscopic quantum superposition , as opposed to a statistical mixture .
this will allow tests of macroscopic realism , may be an important tool for studying decoherence , and is a prerequisite to building a practical interferometer using noon states .
the obvious measurement to make on such a system is the atom number in each well . for the noon state
, we have equal probability of finding @xmath19 atoms in the left well and none in the right , or vice - versa .
however , such a measurement can not distinguish the coherent noon state @xmath35 from the statistic mixture with density operator @xmath36 there exist complimentary measurements that distinguish between a coherent superposition and a statistical mixture .
one method of finding such an measurement is motivated by the expansion of the density operator of the statistical mixture as the average of all coherent phases @xmath37 , @xmath38 we see that the determination of @xmath37 constitutes proof of coherence as @xmath37 is undefined for the mixed state .
consider an arbitrary operator @xmath39 ; for the mixed state in ( [ mixed_state ] ) , the expectation value of @xmath39 is the average of the two pure , separable states , @xmath40 on the other hand , for the pure noon state @xmath41 , the expectation value is @xmath42 the additional interference terms clearly display a dependence on the value of @xmath37 .
therefore , an appropriate observable @xmath39 has non - zero @xmath43 .
an example of an operator that achieves this is @xmath44 , which coherently transfers @xmath19 atoms from one well to the other or vice - versa .
unfortunately , there is no clear way of directly measuring this observable in an experimental setting .
we now consider a quadrature - based method for distinguishing the entangled noon state from the statistical mixture . in this paper , we define the quadrature operator as @xmath45 the measurement of this observable can be achieved using simple linear interference and number measurements .
such a procedure is analogous to quantum optics experiments using a 50 - 50 beam splitter to interfere two photonic modes before intensity measurement , and allows access to phase information . unlike common quantum optics experiments
, both modes contain a similar number of atoms and neither mode can be interpreted as a local oscillator .
this explains the difference between the above definition and the standard quadrature arising from homodyne measurement , proportional to @xmath46 .
both definitions have been employed in theoretical discussions of becs in the past @xcite . to realize these quadrature measurements , we propose a ramsey - type experiment ( see ref .
@xcite for a description of the ramsey technique ) .
similar experiments have been proposed for double well condensates ( for example , to detect a weak force @xcite ) .
the first step , after creating the superposition state , is to set the tunneling rate between the wells @xmath14 and interaction strength @xmath47 to zero and let the system evolve for some time , @xmath48 , during which an energy imbalance exists between the wells ( i.e. @xmath49 is non - zero ) .
the quadrature angle is set by @xmath50 .
the second stage is to restore the symmetry of the wells ( i.e. set @xmath51 to zero ) , and switch on tunneling for a time of @xmath52 ( this is analogous to a beam splitting operation ) . after this we make a measurement of the atom number in each well .
the difference in atom number is exactly proportional to @xmath53 . we simulate the entire procedure using the equations of motion of the number state coefficients ( [ cnbydt ] ) .
it is then straightforward to numerically calculate the distribution of measurement outcomes , as given by @xmath54 , and thus any moment ( e.g. mean , variance , etc ) of the the quadrature @xmath53 .
the next step is to extract information about the off - diagonal terms of the density matrix by measuring the interference terms in ( [ expectation_pure ] ) .
it is straightforward to see that @xmath55 for @xmath56 . in the special case of @xmath57 ,
interference is observed by a sinusoidal dependence on the value of @xmath58 , as seen in fig .
[ ramsey1atomuzero ] ( a ) ( where @xmath37 is the phase angle used in ( 13 ) ) . in fig
. [ ramsey1atomuzero ] ( b ) we plot the same quantity for @xmath59 , and it is unsurprising that inteference is lacking for this case .
however , higher moments of the quadrature measurements do contain information that can distinguish a pure noon state from a classical mixture .
for @xmath19 atoms , the @xmath19th moment @xmath60 contains exactly one subterm equal to @xmath44 , while all lower moments lack such a term . in fig .
[ ramsey1atomuzero ] ( c ) , we see interference fringes in the quadrature variance for the case @xmath59 . the frequency of these fringes is doubled compared to the case @xmath57 . as a function of the accumulated phase shift for an ideal superposition containing 1 atom .
( b ) the same quantity for an ideal superposition containing 2 atoms displays no interference fringes .
( c ) for the two atom superposition state , the variance ( @xmath61 ) is sensitive to the phase @xmath62 accumulated during the ramsey simulation.,title="fig : " ] as a function of the accumulated phase shift for an ideal superposition containing 1 atom .
( b ) the same quantity for an ideal superposition containing 2 atoms displays no interference fringes .
( c ) for the two atom superposition state , the variance ( @xmath61 ) is sensitive to the phase @xmath62 accumulated during the ramsey simulation.,title="fig : " ] as a function of the accumulated phase shift for an ideal superposition containing 1 atom .
( b ) the same quantity for an ideal superposition containing 2 atoms displays no interference fringes .
( c ) for the two atom superposition state , the variance ( @xmath61 ) is sensitive to the phase @xmath62 accumulated during the ramsey simulation.,title="fig : " ] in general the @xmath19 atom noon state will display fringes with frequency @xmath19 times greater than the first - order coherence measured by @xmath63 allows . specifically
, the @xmath19th quadrature moment contains terms proportional to @xmath64 .
it is this scaling that makes noon states of interest for precision interferometry a noon state with known @xmath37 could potentially be used to measure @xmath65 with accuracy proportional to @xmath66 ( the so - called heisenberg limit ) , compared with the @xmath67 scaling typical when using ` classical ' inteferemetric techniques .
such scaling has been observed in single - photon experiments @xcite , but to - date neither with atoms nor noon states .
the quadrature moments are intrinsically linked with the number distribution @xmath54 after the ramsey interference procedure , and so it follows that the off - diagonal terms in eq .
( [ expectation_pure ] ) are directly visible in this distribution .
we have plotted the output in fig .
[ ramseycninterference ] ( a ) for @xmath68 , and we observe a pattern where each second @xmath69 is zero .
this interference pattern is sensitive to the accumulated relative phase , so that for certain values of @xmath58 the interference pattern is absent and the number state coefficients are given by a binomial distribution .
on the other hand , if the initial state were a statistical mixture a binomial distribution would be expected for all values of the relative phase . after the ramsey procedure is simulated , with accumulated phase @xmath70 .
initial state has 20 atoms and @xmath71 .
non - linear strength @xmath72 is ( a ) zero , ( b ) @xmath73 , and ( c ) @xmath74.,title="fig : " ] after the ramsey procedure is simulated , with accumulated phase @xmath70 .
initial state has 20 atoms and @xmath71 .
non - linear strength @xmath72 is ( a ) zero , ( b ) @xmath73 , and ( c ) @xmath74.,title="fig : " ] after the ramsey procedure is simulated , with accumulated phase @xmath70 .
initial state has 20 atoms and @xmath71 .
non - linear strength @xmath72 is ( a ) zero , ( b ) @xmath73 , and ( c ) @xmath74.,title="fig : " ] the interference can be seen most clearly in the parity , defined as @xmath75 which is the difference in probability that an even or odd number of atoms in measured in one of the modes at the end of the ramsey experiment .
the expectation value of the parity is an oscillatory function of the relative phase between the wells , again with a frequency proportional to the total atom number @xmath19 , as seen in fig .
[ parity ] ( a , b ) .
measurements of the parity were considered in detail in ref .
@xcite , where the authors discuss methods for obtaining heisenberg - limited phase resolution .
line lies under the @xmath32 line.),title="fig : " ] line lies under the @xmath32 line.),title="fig : " ] accurate measurement of the @xmath19th quadrature moment , or equivalently the atom parity after the ramsey experiment , is experimentally very difficult .
measuring the interference pattern in the @xmath76 coefficients would require very accurate atom counting , as the relative phase value giving maximum fringe visibility for an even total atom number gives minimum fringe visibility if the total atom number is odd , and vice versa .
indeed , it is apparent that the loss of a single atom would completely destroy the coherence of the noon state .
therefore , the required counting efficiency @xmath77 is such that @xmath78 , which would be challenging for large values of @xmath19 .
an alternative method of determining the parity has been suggested in ref .
this method does not require extremely accurate atom counting ; however it does involve a third condensate mode and precise control of the nonlinearity .
there are additional complications that may make this procedure difficult , even for moderate values of @xmath19 .
we now consider the effect of finite non - linear interactions during the ramsey interference procedure . with an ideal superposition as the initial state ,
the non - linear interactions degrade the visibility of the interference pattern in the @xmath76 coefficients , as shown in fig .
[ ramseycninterference ] ( b , c ) .
the oscillations in the expectation value of parity are also reduced by the presence of non - linear interactions ( see fig . [ parity ] ) . for a given value of @xmath47 , the visibility of the parity oscillations decreases as the number of atoms in the superposition increase .
however , we note that the phase of the interference fringes is unaffected by the non - linearity , which is important for possible interferometric applications .
finally , when the initial state is an ideal noon state , we observed that including non - linear interactions can result in small fringes in the mean quadrature value even for @xmath56 .
the maximum amplitude of these induced fringes is much less than the total atom number .
these fringes are most noticeable for small atom numbers , as they have amplitudes of only one or two atoms regardless of the total atom number .
there have been a number of proposals for generating superpositions of various kinds involving bose - einstein condensates .
many of these consider superpositions of two - component condensates . in this case ,
methods involving adiabatic manipulation @xcite , and dynamical evolution ( making use of the interplay between nonlinear interactions and tunneling , with a specific initial relative phase ) @xcite , have been suggested to generate superpositions .
superpositions of phase states have also been considered @xcite .
we are interested in superpositions of number states in a single component condensate in a double well potential .
one proposal for generating superpositions of this kind involves using a feshbach resonance to produce a sudden change in the interaction strength , initiating a dynamical evolution where a macroscopic superposition emerges periodically @xcite . in ref .
@xcite , it was claimed that the adiabatic method was not feasible due to the infinitely long evolution time required because of the near degeneracy of the ground and first excited states in the strongly attractive regime .
we find that with realistic parameters a superposition of two well - separated wave packets can be generated on a time scale of seconds using a smooth change in interaction strength .
although this evolution is not necessarily adiabatic it can generate a ` mesoscopic ' superposition state . to estimate some suitable parameters for our effective hamiltonian
we have considered a small condensate of rubidium atoms . around a magnetic field strength of 155 g
there is a feshbach resonance for @xmath79rb atoms , allowing the s - wave scattering length to be tuned from at least 2000 @xmath80 to @xmath81 ( @xmath80 being the bohr radius ) . assuming a cigar - shaped condensate confined by trapping frequencies of 1 khz in both tight directions and 100 hz in the longitudinal direction , and using a gaussian approximation for the condensate wavefunction
, we estimate that the interaction parameter , @xmath47 , given by equation ( [ interactioneqn ] ) could range from approximately 30 s@xmath82 to @xmath83 s@xmath82 .
we look at simulations which are a few seconds in length ( note that condensate lifetimes of greater than 10 seconds are experimentally achievable @xcite ) .
[ linchangeu ] shows the wave function as the interaction strength is changed linearly from 1 s@xmath82 to @xmath83 s@xmath82 over a timespan of 0.5 and 4 seconds .
the initial state is the ground state at u=1 s@xmath82 , and @xmath84 s@xmath82 .
clearly the slower change in interaction strength results , as per the adiabatic theorem , in a final state that is closer to the ideal macroscopic superposition .
this can be seen in fig .
[ fidelity ] , which plots the fidelity of the evolving wave function , i.e. the overlap , @xmath85 , of the ideal superposition state with the evolving wave function .
the variance in the number difference between the wells of the final states in fig .
[ linchangeu ] are approximately 283 and 371 , compared with 400 for an ideal superposition state containing 20 atoms .
once the interaction strength is held constant at @xmath83 s@xmath82 the probability distributions @xmath54 do not change significantly with time .
the parameters could be optimized further to obtain better superposition states . in general , better states are obtained when the final ratio @xmath72 is large and negative .
on the other hand , the timescale for the evolution to be adiabatic is inversely proportional to both @xmath47 and @xmath14 .
the requirement of physically separated wells implies a relatively small value for @xmath14 , while the timespan of the experiment is limited by atomic loss .
we have not attempted to optimize the process of generating a superposition state within these bounds , but rather concentrate on measurements aimed at distinguishing even an imperfect superposition from a statistical mixture .
to @xmath83 s@xmath82 in ( a ) 0.5 , and ( b ) 4 seconds .
initial wavefunction is the ground state at u=1 s@xmath82 and @xmath84 s@xmath82.,title="fig : " ] to @xmath83 s@xmath82 in ( a ) 0.5 , and ( b ) 4 seconds .
initial wavefunction is the ground state at u=1 s@xmath82 and @xmath84 s@xmath82.,title="fig : " ] s@xmath82 and @xmath84 compared to the ideal noon state .
a slower ramp time results in a final wave function closer to that of an ideal superposition state . ]
we now perform the ramsey procedure followed by parity measurements in order to detect coherence in the non - ideal superposition states generated above .
[ ramseynonideal ] and [ paritynonideal ] show typical fringes in the number difference and parity for an initial non - ideal superposition state of a 20 atom condensate . in the parity
, we see clear high frequency components in the interference fringes that correspond to the off - diagonal coherence between states of large atom number difference . however , as the state is not perfect , other frequency components are present , resulting in beating and a low frequency envelope . in the appendix ,
we formalise the relationship between the parity frequency components and coherence and show that @xmath86 for some real numbers @xmath87 .
the component with angular frequency @xmath6 corresponds to coherence between elements separated by @xmath6 atoms in fock space .
in fact , the highest frequency component with angular frequency @xmath19 is proportional only to @xmath88 , and is the only component observed in the results for an ideal superposition ( fig .
[ parity ] ) .
the presence of the same high frequency component in fig .
[ paritynonideal ] ( a ) is an unambiguous demonstration that the generated state contains coherence and is _ not _ a statistical mixture . with the introduction of non - linear interactions during the ramsey interference ,
the amplitude and frequency of the parity fringes become less regular and are no longer periodic over @xmath58 modulo @xmath89
. it would not be easy to use the results when @xmath47 is large as an indicator of the presence of a superposition due to their irregularity , and the fact that high frequency components can not be said to indicate noon - type coherence .
s@xmath82 , and ( c ) @xmath90 s@xmath82 .
initial state was generated from the ground state at u=1 s@xmath82 , @xmath84 s@xmath82 , and linearly changing the interaction strength to @xmath91 s@xmath82 over 4 seconds.,title="fig : " ] s@xmath82 , and ( c ) @xmath90 s@xmath82 .
initial state was generated from the ground state at u=1 s@xmath82 , @xmath84 s@xmath82 , and linearly changing the interaction strength to @xmath91 s@xmath82 over 4 seconds.,title="fig : " ] s@xmath82 , and ( c ) @xmath90 s@xmath82 .
initial state was generated from the ground state at u=1 s@xmath82 , @xmath84 s@xmath82 , and linearly changing the interaction strength to @xmath91 s@xmath82 over 4 seconds.,title="fig : " ] s@xmath92 , and ( c ) @xmath90 s@xmath92 .
initial state was generated from the ground state at u=1 s@xmath92 , @xmath84 s@xmath92 , and linearly changing the interaction strength to @xmath91 s@xmath92 over 4 seconds .
, title="fig : " ] s@xmath92 , and ( c ) @xmath90 s@xmath92 .
initial state was generated from the ground state at u=1 s@xmath92 , @xmath84 s@xmath92 , and linearly changing the interaction strength to @xmath91 s@xmath92 over 4 seconds .
, title="fig : " ] s@xmath92 , and ( c ) @xmath90 s@xmath92 .
initial state was generated from the ground state at u=1 s@xmath92 , @xmath84 s@xmath92 , and linearly changing the interaction strength to @xmath91 s@xmath92 over 4 seconds .
, title="fig : " ]
our simulations indicate that a quantum superposition of a small bose - einstein condensate may be generated in a reasonably short time by a smooth change in the atomic interaction strength from repulsive to attractive .
however , we find that verifying that a superposition has been generated may not be straightforward , even in the ideal case of a perfect ` noon ' state superposition and no decoherence . we have considered a ramsey - type interference experiment as a method to measure the quadrature operator and parity thereof of a double well wave function . for a noon superposition ,
only the highest - order quadrature moment ( and not the mean ) is sensitive to the accumulated relative phase and is therefore useful in distinguishing between a statistical mixture and a coherent superposition .
parity measurements after a ramsey - type experiment on an ideal superposition state show a dependence on the relative phase that could in principle be used to verify the presence of the superposition .
a statistical mixture would not be expected to show any dependence on the relative phase if parity measurements were performed .
the presence of non - linear interactions during the ramsey experiment degrades the amplitude of the parity oscillations .
measurements of a non - ideal superposition state also show a high - frequency dependence on the relative phase a smoking gun indication of noon - type coherence .
however , we observed that the imperfect state displays additional frequencies resulting in a beating pattern .
for some choices of phase angle , the visibility is close to one while the period decreases to @xmath93 , allowing it to remain useful for heisenberg - limited phase measurements .
similarly as for the ideal state , non - linear interactions degrade the amplitude and regularity of the parity oscillations , however for small non - linearity , the high frequency component could still be used to verify that a mesoscopic superposition has been generated .
there are many factors that may make these measurements difficult in practice .
for example , phase diffusion due to non - linear interactions dramatically reduces the visibility of the parity oscillations .
other effects not considered here would also reduce the fringe visibility , such as 3-body loss , and other sources of decoherence .
highly accurate atom counting is required to observe the interference patterns in the @xmath76 coefficients , and the associated parity oscillations as a function of accumulated phase , unless an alternative measurement scheme could be realised .
the authors wish to acknowledge support from the australian research council .
for a state with precisely @xmath19 atoms , the ( discrete ) quadrature operator can take @xmath94 values .
therefore , it is possible to express any linear combination of the measurement probabilities ( i.e. the @xmath95 after the ramsey interference ) as a linear sum of the first @xmath19 quadrature moments .
specifically , the parity can be expressed this way , i.e. @xmath96 where @xmath97 are real numbers .
expanding the power results in @xmath98 where @xmath99 represents the symmetric ordering of @xmath39 .
the operator @xmath100 transfers @xmath101 atoms from the right well to the left , and therefore we can say @xmath102 where @xmath103 are appropriately chosen constants . putting everything together ,
we finally arrive at eq .
( [ frequencies ] ) , where the the constants @xmath87 can be obtained from the above . | the availability of bose - einstein condensates as mesoscopic or macroscopic quantum objects has aroused new interest in the possiblity of making and detecting coherent superpositions involving many atoms . in this article
we show that it may be possible to generate such a superposition state in a reasonably short time using feshbach resonances to tune the inter - atomic interactions in a double - well condensate .
we also consider the important problem of distinguishing whether a coherent superposition or a statistical mixture is generated by a given experimental procedure .
we find that unambiguously distinguishing even a perfect ` noon ' state from a statistical mixture using standard detection methods will present experimental difficulties . |
photoionization of atoms and molecules is one of the most fundamental quantum processes .
it played a key role in the early days of quantum mechanics and has ever since been paving the way towards an improved understanding of the structure and dynamics of matter on a microscopic scale . today
, kinematically complete photoionization experiments allow for accurate tests of the most sophisticated ab - initio calculations . besides , photoionization studies in a new frequency domain are currently becoming feasible by the availability of novel xuv and x - ray radiation sources @xcite , giving rise to corresponding theoretical developments ( see , e.g. , @xcite ) .
various photoionization mechanisms rely crucially on electron - electron correlations .
prominent examples are single - photon double ionization as well as resonant photoionization .
the latter proceeds through resonant photoexcitation of an autoionizing state with subsequent auger decay . in recent years
, a similar kind of ionization process has been studied in systems consisting of two ( or more ) atoms .
here , a resonantly excited atom transfers its excitation energy radiationlessly via interatomic electron - electron correlations to a neighbouring atom leading to its ionization .
this auger - like decay involving two atomic centers is commonly known as interatomic coulombic decay ( icd ) @xcite .
it has been observed , for instance , in noble gas dimers and water molecules @xcite . in metal oxides ,
the closely related process of multi - atom resonant photoemission ( marpe ) was also observed @xcite .
we have recently studied resonant two - center photoionization in heteroatomic systems and shown that this ionization channel can be remarkably strong @xcite . in particular
, it can dominate over the usual single - center photoionization by orders of magnitude .
besides , characteristic effects resulting from a strong coupling of the ground and autoionizing states by a relatively intense photon field were identified .
also resonant two - photon ionization in a system of two identical atoms was investigated @xcite .
we note that photoionization in two - atomic systems was also studied in @xcite and @xcite .
the inverse of two - center photoionization ( in weak external fields ) is two - center dielectronic recombination @xcite . in the presence of an external laser field and two neighbouring atoms @xmath1 and @xmath2 .
apart from the direct photoionization of @xmath0 there are interatomic channels via resonant photoexcitation of the `` molecular '' system @xmath1-@xmath2 and subsequent icd .
, scaledwidth=45.0% ] in the present contribution , we extend our investigations of electron correlation - driven interatomic processes by considering photoionization of an atom @xmath0 in the presence of _ two _ neighbouring atoms @xmath1 ( see figure 1 ) .
all atoms are assumed to interact with each other and with an external radiation field .
we show that the photoionization of atom @xmath0 via photoexcitation of the system of two neighbouring atoms @xmath1 and subsequent icd can be by far the dominant ionization channel .
moreover , we reveal the characteristic properties of the process with regard to its temporal dependence and photoelectron spectra . in particular , by comparing our results with those for photoionization in a system of two atoms @xmath0 and @xmath1 , we demonstrate the influence which the presence of the second atom @xmath1 may have .
atomic units ( a.u . ) are used throughout unless otherwise stated .
let us consider a system consisting of three atoms , @xmath0 , @xmath1 and @xmath2 , where @xmath1 and @xmath2 are atoms of the same element and @xmath0 is different .
we shall assume that all these atoms are separated by sufficiently large distances such that free atomic states represent a reasonable initial basis set to start with .
let the ionization potential @xmath3 of atom @xmath0 be smaller than the excitation energy @xmath4 of a dipole - allowed transition in atoms @xmath1 and @xmath2 . under such conditions ,
if our system is irradiated by an electromagnetic field with frequency @xmath5 , the ionization process of this system ( i.e. , essentially of the atom @xmath0 ) can be qualitatively different compared to the case when a single , isolated atom @xmath0 is ionized .
indeed , in such a case @xmath0 can be ionized not only directly but also via resonant photoexcitation of the subsystem of @xmath1 and @xmath2 , with its consequent deexcitation through energy transfer to @xmath0 resulting in ionization of the latter . in the following ,
we consider photoionization in the system of atoms @xmath0 , @xmath1 and @xmath2 in more detail . for simplicity
, we suppose that the nuclei of all atoms are at rest during photoionization . denoting the origin of our coordinate system by @xmath6
, we assume that the nuclei of the atoms @xmath1 and @xmath2 are located on the @xmath7-axis : @xmath8 and @xmath9 .
the coordinates of the nucleus of the atom @xmath0 are given by @xmath10 .
the coordinates of the ( active ) electron of atom @xmath11 with respect to its nucleus are denoted by @xmath12 , where @xmath13 .
the total hamiltonian describing the three atoms embedded in an external electromagnetic field reads @xmath14 where @xmath15 is the sum of the hamiltonians for the noninteracting atoms @xmath0 , @xmath1 and @xmath2 .
we shall assume that the ( typical ) distances @xmath16 between the atoms are not too large , @xmath17 , where @xmath18 is the speed of light , such that retardation effects in the electromagnetic interactions can be ignored .
if transitions of electrons between bound states in atoms @xmath1 and @xmath2 are of dipole character , then the interaction between each pair of atoms @xmath19 ( with @xmath20 ) can be written as @xmath21 where @xmath22 and @xmath23 is the kronecker symbol .
note that in ( [ inter_atomic ] ) a summation over the repeated indices @xmath24 and @xmath25 is implied . in ( [ inter_atomic ] ) , @xmath26 denotes the interaction of the atom @xmath11 with the laser electromagnetic field .
the latter will be treated as a classical , linearly polarized field , described by the vector potential @xmath27 , where @xmath28 , @xmath29 is the angular frequency and @xmath30 is the field strength .
the interaction @xmath26 then reads @xmath31 where @xmath32 is the momentum operator for the electron in atom @xmath33 .
our treatment of photoionization will be based on the following points : oscillator strengths for dipole - allowed bound - bound transitions can be very strong .
this means that , provided that the distances between all the atoms in our system are of the same order of magnitude , the interaction between atoms @xmath1 and @xmath2 is much more effective than the interaction between atoms @xmath0 and @xmath1 ( or @xmath0 and @xmath2 ) .
besides , atoms @xmath1 and @xmath2 will , in general , couple much more strongly to a resonant laser field than atom @xmath0 . in what follows , we shall assume that the intensity of the laser field is relatively low such that the interaction between atoms @xmath1 and @xmath2 changes the states of the system more substantially than the coupling of these atoms to the laser field .
therefore , we shall begin with building states of the @xmath1-@xmath2 subsystem in the absence of the field . the second step of our treatment will be to include the interaction of the @xmath1-@xmath2 subsystem with the laser field and , in the third step , we complete the treatment of ionization by considering the interaction of atom @xmath0 with both the laser field and the field - dressed subsystem of atoms @xmath1 and @xmath2 .
i. we denote the ground and excited states of the undistorted atoms @xmath1 and @xmath2 by @xmath34 , @xmath35 and @xmath36 , @xmath37 , respectively .
let the corresponding energies of these states be @xmath38 and @xmath39 .
the state @xmath40 of the @xmath1-@xmath2 subsystem can be expanded into the `` complete '' set of undistorted atomic states represented by the configurations ( i ) @xmath41 , ( ii ) @xmath42 , ( iii ) @xmath43 and ( iv ) @xmath44 . in the approximation , which neglects the interatomic interaction , the configurations @xmath42 and @xmath43
are characterized by exactly the same value of the ( undistorted ) energy @xmath45 .
the latter , in turn , strongly differs from the energies @xmath46 and @xmath47 which are characteristic for the configurations @xmath41 and @xmath44 , respectively .
therefore , provided that the distance between the atoms is not too small , the interaction @xmath48 will strongly mix the configurations ( ii ) and ( iii ) only , while the other configurations ( i ) and ( iv ) will be affected only very weakly . taking this into account , it is not difficult to find the states of the subsystem of interacting atoms @xmath1 and @xmath2 which read @xmath49 these two - atomic states are normalized and mutually orthogonal .
they posses energies given by @xmath50 , @xmath51 , @xmath52 and @xmath53 , respectively , where @xmath54 .
note that , for definiteness , @xmath55 has been assumed to be real and negative here , as will always be the case in our examples below ( see section 3 ) .
let us now consider two interacting atoms @xmath1 and @xmath2 embedded in a resonant laser field .
one can look for a state of such a system by expanding it into the new set of states given by eq .
( [ states_of_bb ] ) , @xmath56 inserting the expansion ( [ bb_in_laser_1 ] ) into the corresponding wave equation , we obtain a set of coupled equations for the unknown time - dependent coefficients @xmath57 , @xmath58 , @xmath59 and @xmath60 : @xmath61 the system of equations ( [ bb_in_laser_2 ] ) can be greatly simplified by noting the following .
first , all transition matrix elements of the interaction with the laser field , which involve the asymmetric state @xmath62 , are equal to zero and , thus , only the remaining three states can be coupled by the field .
second , if we suppose that the frequency of the laser field is resonant to the transitions @xmath63 and that the field is relatively weak such that the non - resonant transitions @xmath64 are much less effective than the above resonant ones , the system ( [ bb_in_laser_2 ] ) effectively reduces to @xmath65 which can be readily solved by using the rotating wave approximation . assuming that the field is switched on suddenly at @xmath66
, we obtain two solutions @xmath67 \ ,
\varphi_{g } \nonumber \\ & + & \frac{w_{+,g}}{z_{2 } - z_{1 } } \left ( { \rm e}^{-i z_{2 } t } - { \rm e}^{-i z_{1 } t } \right ) { \rm e}^{- i \omega_0 t } \ , \varphi_{+ } \label{bb'_in_laser_4}\end{aligned}\ ] ] and @xmath68 { \rm e}^{- i \omega_0 t } \ , \varphi_{+}. \label{bb'_in_laser_5}\end{aligned}\ ] ] in the above equations , we have introduced @xmath69 where @xmath70 is the rabi frequency , @xmath71 and @xmath72 .
the two solutions in ( [ bb_in_laser_5 ] ) correspond to two different initial conditions : at @xmath66 the system is either in the state @xmath73 or in @xmath74 .
they are orthogonal to each other and form a `` complete '' set of field - dressed states of the subsystem @xmath1-@xmath2 .
note also that we have neglected the spontaneous radiative decay of the excited state @xmath74 which , in our case , is justified as long as @xmath75 , where @xmath76 is the radiative width of @xmath74 .
now , as the last step , we shall add atom @xmath0 to our consideration .
let @xmath77 and @xmath78 , where @xmath79 is the electron momentum , be the ground and a continuum state of a single , isolated atom @xmath0 .
the wavefunction of the total system @xmath0@xmath1-@xmath2 can be expanded into the following `` complete '' set of states @xmath80 here , the initial conditions are given by @xmath81 , @xmath82 and @xmath83 .
the coupling of atom @xmath0 to both the subsystem @xmath1-@xmath2 and the laser field involves bound - continuum transitions which are normally much less effective than the bound - bound ones . for this reason
, we may assume that the interactions of @xmath0 with the laser field and the @xmath1-@xmath2-subsystem is weak and consider ionization of atom @xmath0 in the lowest order of perturbation theory in these two interactions .
as a result , by inserting the expansion ( [ ioniz_of_a_1 ] ) into the corresponding schrdinger equation we obtain @xmath84 where @xmath85 is the energy of the electron in the initial state @xmath77 of atom @xmath0 and @xmath86 is the electron energy after the emission .
the probability for ionization of the three - atomic system , as a function of time , then reads @xmath87 note that equations ( [ ioniz_of_a_2 ] ) are readily solved analytically .
however , the resulting expressions are somewhat lengthy and will not be given here .
based on the results obtained in the previous section , let us now turn to the discussion of some aspects of photoionization in a system consisting of one lithium and two helium atoms .
we suppose that in our three - atomic system the positions of the lithium and helium atoms are given by the vectors @xmath88 , @xmath89 and @xmath90 , respectively .
our system is initially ( at time @xmath66 ) in its ground configuration and is irradiated by a monochromatic laser field .
the field is linearly polarized along the @xmath7-axis and its frequency is resonant to the @xmath91 - @xmath92 transition in the he - he subsystem , i.e. , @xmath93 .
choosing @xmath94 a.u .
we obtain that the energy spitting @xmath95 between the states @xmath74 and @xmath62 of the he - he subsystem is @xmath96 ev . assuming a field strength of @xmath97 a.u . , the corresponding rabi frequency amounts to @xmath98 ev which is much less than @xmath99 . in figure [ time - develop ] , we present the probability for ionization of our system as a function of time .
the probability shows a non - monotonous behaviour in which time intervals , when the ionization probability rapidly increases , are separated by intervals , when the probability remains practically constant , reflecting oscillations of the electron populations with the rabi frequency @xmath100 between the ground and excited states of the he - he subsystem in a resonant electromagnetic field . for comparison ,
we also show in figure [ time - develop ] results for ionization of a single ( separated ) li atom and for ionization in a two - atomic li - he system . in the latter case ,
the lithium atom is located at the origin ( @xmath88 ) and the coordinates of the helium atom are @xmath101 . the frequency of the laser field is assumed to be resonant to the @xmath102@xmath103 transition frequency of the corresponding bound states of a single he atom .
in contrast to the single - atom ionization , in both the two- and three - atomic cases the ionization probability demonstrates a step - wise temporal development in which time intervals of rapid probability growth are followed by intervals of almost constant probability .
we point out that in the three - atomic case , however , the size of these time intervals is shorter by a factor of @xmath104 . compared to ionization of a single li atom , ionization in the two - atomic system is very strongly enhanced @xcite .
when the three - atomic system is irradiated , the enhancement increases even further . in the range of small values of @xmath105 , where all ionization probabilities still increase monotonously ,
this additional enhancement is equal to a factor of @xmath106 . at larger @xmath105 , however , when the two ionization probabilities exhibit step - wise behaviours , this additional enhancement due to the presence of the second he atom is reduced to a factor close to @xmath107 on average , as can also be seen in figure [ time - develop ] .
all the above features can be understood by noting the following : \i ) for the chosen set of parameters of our two- and three - center systems , the indirect channels of ionization , which involve two- or three - atomic correlations , are substantially stronger than the direct one .
therefore , these correlations have a dominating effect on the ionization .
\ii ) at small @xmath105 , ionization in the two- and three - atomic systems is basically a two - step process : the first step is photoexcitation in the he or he - he subsystem and the second step is a consequent energy transfer to li . in each case , both these steps are described by basically the same dipole transition matrix element of the subsystem . since , compared to a single he atom , this dipole element in he - he is by @xmath104 larger than in he , one obtains a factor of 2 for the enhancement in the ionization amplitude , leading to a factor of @xmath106 in the ionization probability ( see also @xcite ) .
\iii ) at larger @xmath105 , when rabi oscillations show up , the second step `` saturates '' in the sense that the averaged probability to find the corresponding subsystem in the excited state becomes equal to 50% . therefore , the ionization probability in the three - atomic system is now larger ( on average ) by a factor of @xmath107 only .
\iv ) the origin of the step - wise behaviours of the ionization probabilities for the two- and three- atomic systems lies in the oscillations of the population between the ground and excited states in the he atom ( for the two - atomic case ) or in the he - he subsystem ( for the three - atomic case ) .
the scale of these oscillation is set by the rabi frequency and , because in the he - he subsystem the latter is larger by a factor of @xmath104 , the corresponding time intervals are shorter by the same factor .
additional information about the ionization process can be obtained by considering the energy spectrum of emitted electrons .
such a spectrum is shown in figure [ spectra ] for the same systems and parameters as in figure [ time - develop ] and for a pulse duration of @xmath108 ps . in panel ( a ) , we compare the energy spectra of electrons emitted in the process of photoionization of li - he - he and li - he systems . in both cases ,
the main feature is the presence of three pronounced maxima .
the origin of these peaks is similar to the splitting into three lines of the energy spectrum of photons emitted during atomic fluorescence in a resonant electromagnetic field @xcite .
in such a field , the ground and excited levels of the he and he - he subsystems split into two sub - levels , which differ by the corresponding rabi frequency @xmath109 . as a result ,
the resonant electronic correlations between these subsystems and the li atom lead to an energy transfer to the li which peaks at @xmath110 and @xmath111 .
since , as was already mentioned , the rabi frequencies of these subsystems differ by a factor of @xmath104 , the magnitude of the separation between the corresponding maxima in panel ( a ) of figure [ spectra ] also differs by this factor .
note also that the widths of these main maxima as well as the appearance of additional multiple maxima , seen in the figure , are related to the finiteness of the pulse duration ; the distance between the latter is roughly given by @xmath112 .
the distinct influence , which the interatomic electron - electron correlations exert on the shape of the photoelectron spectra , is further highlighted in panel ( b ) of figure [ spectra ] .
it compares the energy spectra of photoelectrons emitted from our li - he - he system and an isolated li atom . in the latter case ,
there is only one main maximum , while the two main side peaks are missing , as one would expect ( the additional multiple maxima are related again to the finiteness of the pulse duration ) .
we have studied resonant photoionization in a system @xmath0-@xmath1-@xmath2 consisting of three atoms , with two atoms @xmath1 of the same element and one different atom @xmath0 .
we have shown that the mutual correlations among the atoms can largely enhance the ionization probability and distinctly modify also other properties of the process in a characteristic manner .
in particular , as compared to the case of resonant photoionization in a two - atom system @xmath0-@xmath1 , it has been demonstrated that the presence of a second atom @xmath1 can ( i ) further enhance the photoionization process , ( ii ) change the time dependence of the ionization probability and ( iii ) move the side peaks in the photoelectron spectrum further apart .
a.b.v . acknowledges the support from the extreme matter institute emmi .
10 rudenko a _ et al _ 2008 _ phys . rev .
lett . _ * 101 *
073003 cederbaum l s , zobeley j and tarantelli f 1997 _ phys .
lett . _ * 79 * 4778 for recent reviews on icd , see averbukh v _
et al _ 2011 _ j. electron spectrosc .
relat . phenom .
_ * 183 * 36 ; hergenhahn u 2011 _ j. electron spectrosc .
relat . phenom . _ | modifications of photoionization arising from resonant electron - electron correlations between neighbouring atoms in an atomic sample are studied .
the sample contains atomic species @xmath0 and @xmath1 , with the ionization potential of @xmath0 being smaller than the energy of a dipole - allowed transition in @xmath1 .
the atoms are subject to an external radiation field which is near - resonant with the dipole transition in @xmath1 .
photoionization of an atom @xmath0 may thus proceed via a two - step mechanism : photoexcitation in the subsystem of species @xmath1 , followed by interatomic coulombic decay . as a basic atomic configuration , we investigate resonant photoionization in a three - atomic system @xmath0-@xmath1-@xmath1 , consisting of an atom @xmath0 and two neighbouring atoms @xmath1 .
it is found that , under suitable conditions , the influence of the neighbouring atoms can strongly affect the photoionization process , including its total probabilty , time development and photoelectron spectra . in particular , by comparing our results with those for photoionization of an isolated atom @xmath0 and a two - atomic system @xmath0-@xmath1 , respectively , we reveal the characteristic impact exerted by the third atom . |
non - abelian stringlike solutions have a long history which starts already in 1973 with the nielsen - olesen seminal paper @xcite .
several general discussions were published @xcite and explicit solutions ( numerical of course ) were obtained @xcite for an @xmath3 gauge theory with higgs fields in the adjoint representation which completely break the symmetry .
however , the major part of the activity in the field of cosmic strings @xcite was concentrated on their abelian counterparts .
one reason for this is that these non - abelian string solutions have their flux directed in a fixed direction in the corresponding algebra so they are essentially abelian . in recent years
, new kinds of non - abelian strings were discovered during attempts to understand the phenomenon of confinement in qcd @xcite , and their properties were studied extensively @xcite .
these new solutions appear in models with a global ( flavor ) @xmath5 symmetry in addition to the @xmath3 local ( color ) symmetry based on scalar fields in the fundamental representation .
they allow rotation of the non - abelian flux in the lie algebra , which makes them genuinely non - abelian .
when @xmath6 these consist a generalization @xcite of the _ semilocal _ strings introduced originally within the extended abelian higgs model @xcite which is the higgs system with a global @xmath7 symmetry in addition to the local @xmath4 .
we therefore term the model discussed here the `` extended non - abelian higgs model '' .
most of the studies of these non - abelian string solutions up to now have been limited to the self - dual ( bps ) case .
however , it is natural to go further and look for more general solutions as has just been done very recently @xcite .
this is the direction which we will take in this work , namely going beyond the bps limit and it will be done together with allowing also for the possibility of rotation ( i.e. spinning solutions ) and of currents along the string axis .
spinning and superconducting cosmic strings have been found recently @xcite in the extended abelian higgs model which gives rise also to the ( embedded ) nielsen - olesen solutions .
these new semilocal solutions , known as _ twisted _ , occur mainly outside the very peculiar ( self - dual ) limit of the coupling constants where the equations of the theory admit bogomolnyi conditions .
another outstanding feature of twisted semilocal strings is that , when they exist , there is a continuous family of them , labelled by the `` twist '' , a parameter entering through a space - time dependent relative phase of the higgs field components .
note however the existence of the electro - weak superconducting strings @xcite which exist without a `` twist '' . in the abelian case ,
the local string is characterized by a magnetic field concentrated in a tube along the symmetry axis . outside the core
the magnetic field strength decays exponentially .
the higgs field vanishes on the axis and reaches asymptotically its symmetry - breaking value . for the twisted semilocal string the geometry is more involved .
the magnetic and higgs fields behave roughly as for local string but the configuration supports in addition an azimuthal ( `` tangential '' ) component of the magnetic field .
the source of this current is the additional higgs component ; its modulus is non - zero on the axis ( forming a condensate ) and vanishes outside the core and its phase is twisted .
the effect of the non trivial phase can be appreciated once computing the gauge invariant noether currents and the magnetic field .
figure [ magneticfield ] gives a pictorial representation of the effect of the twist .
= 4.0 cm + in this work we will present the non - abelian analogues of the abelian semilocal twisted strings and study their properties like current and angular momentum .
untwisted purely magnetic local non - abelian strings will be discussed briefly as a special case .
the field equations corresponding to the models presented here are non linear and coupled and do not admit explicit solutions .
we therefore rely on numerical techniques to construct the solutions and calculate the physical quantities .
several figures are necessary to illustrate the extreme richness of the solutions .
the existence of these new kinds of strings raises the question of the nature of the gravitational fields of these strings and the possibility of new features in this respect .
some initial work has been already done in this area @xcite still for the bps case only and we will turn to that question in a future publication .
another issue which is beyond the scope of this work is the effect of spin on the reconnection probability @xcite of non - abelian strings .
this paper has the following plan . in section 2
we present the extended non - abelian higgs model and discuss its relation with its abelian counterpart . in section 3 we derive the field equations and obtain the physical quantities which are used to characterize the solutions : energy , angular momentum , currents and charges . in section 4
we present the various string solutions and discuss their properties across their parameter space .
section 5 contains our conclusions .
our general framework is based on a lagrangian describing a multiplet of @xmath0 scalar fields with local invariance under @xmath1 and a global invariance under @xmath5 .
the local symmetry further requires @xmath8 non - abelian gauge fields and one abelian field .
the scalar fields can be written in term of a matrix with elements @xmath9 where @xmath10 and @xmath11 transforming according to @xmath12 where @xmath13 and @xmath14 are matrices in the fundamental representations of @xmath3 and @xmath5 respectively .
the lagrangian is : @xmath15 the standard definitions are used for the covariant derivative and gauge field strengths : @xmath16 @xmath17 where @xmath18 are the structure constants of the gauge group and @xmath19 are the ( hermitian ) generators in the fundamental representation .
we use the normalization @xmath20 .
the most general renormalizable symmetry breaking potential is @xmath21 where @xmath22 , @xmath23 and @xmath24 are positive constants .
it can be written also as : @xmath25 the first term forces the scalar fields to develop a non - trivial minimum , while the second term forces the minimal configurations @xmath26 to be such that @xmath27 .
this general system contains some well - known special cases , or seen from the other direction , may be considered as a generalization of several systems . from the point of view of the present discussion our model generalizes the extended abelian higgs model which corresponds to @xmath28 ( and @xmath29 , while @xmath0 is arbitrary ) in our terminology . in the abelian case ,
the second term of the potential vanishes identically .
this model allows for two kinds of stable string solutions as was first discovered @xcite for the case @xmath30 .
the first is just the embedded abelian ( nielsen - olesen ) flux tube which in these circumstances is stable @xcite for @xmath31 .
the second is sometimes called `` skyrmion '' because of the relation with the @xmath32-model lumps @xcite and exists only in the self - dual limit where the masses of the higgs and the gauge particles are equal ( or @xmath33 ) .
its stability is guaranteed by a bogomolnyi - type argument .
both solutions are termed semilocal strings and their discovery was a surprise at the time , since the vacuum manifold of the model is a @xmath34- dimensional sphere which does not give rise to non - contractible loops .
this discovery gave an explicit example for a system where the non - triviality of the first homotopy group of the vacuum manifold is not a necessary condition for the existence of stable string - like solutions .
in this system , the kinetic ( gradient ) term plays a crucial role since it has a lower symmetry than the potential term , and it is this `` mismatch '' between the different symmetries which enables the existence of these solutions @xcite .
the two kinds of solutions mentioned above are static , but stationary spinning semilocal strings also exist in this system , as well as solutions which carry a persistent current .
these solutions @xcite belong to a new family of solutions named _ twisted _ semilocal strings .
the `` twist '' is realized by a relative phase between the two higgs field components , which changes along the string axis and may be also time dependent .
these twisted strings exist for @xmath35 and form a continuous family parametrized by this `` twist '' parameter .
the main physical property of these twisted strings is the current along the string axis which exists in both the static and stationary cases and gives rise to azimuthal component of the magnetic field .
their energy in their rest frame decreases with growing current which implies that they are energetically more favorable to be produced during ( cosmological ) phase transitions than the embedded abelian strings .
these new solutions bifurcate with the embedded abelian flux tubes in the limit of vanishing current , thus clarifying the nature of the `` magnetic spreading '' instability @xcite of the embedded abelian flux tubes for @xmath35 .
we look for cylindrically symmetric non - abelian stringlike solutions and for definiteness , we limit ourselves to the case @xmath36 .
but , the ansatz presented below and the corresponding equations can be generalized easily . the electromagnetic potential and its non - abelian counterpart will be : @xmath37 where @xmath38 belongs to the cartan subalgebra of the gauge group . for @xmath7 there is only one possibility , i.e. @xmath39 but already for @xmath40 which is a rank 2 group there is more freedom and @xmath38 is a linear combination of the two ( commuting ) diagonal gell - mann matrices .
we will denote the diagonal elements of @xmath38 by @xmath41 with @xmath42 .
for the scalar field we take : @xmath43 which follows ( and actually generalizes ) the form of @xcite or @xcite .
the phases @xmath44 and @xmath45 will have linear dependence on @xmath46 , @xmath47 and @xmath48 : @xmath49 and @xmath50 . to write the field equations , it is convenient to use a dimensionless coordinate , to rescale the fields and `` phase parameters '' appropriately and to define a relative gauge coupling constant : @xmath51 with these dimensionless quantities , the field equations of the system are written below .
for the scalar fields we obtain ( recall : @xmath41 are the diagonal elements of @xmath38 ) : @xmath52 \varphi_a \;\;\ ; , a < n\end{aligned}\ ] ] @xmath53 \varphi_n \label{eqfnsemilocal}\end{aligned}\ ] ] @xmath54 \rho_n \label{eqrhosemilocal}\end{aligned}\ ] ] where there is no summation on repeated indices and all sums are for @xmath55 unless indicated otherwise . for the abelian - gauge fields
, we have @xmath56 finally , for the non - abelian fields : @xmath57 in order to obtain localized solutions , appropriate boundary conditions should be imposed .
the boundary conditions will guarantee that the solutions will be regular on the axis @xmath58 and approach a vacuum for @xmath59 .
regularity at @xmath58 yields @xmath60 while the requirement for asymptotic approach to the vacuum is somewhat more involved .
for the scalar field we simply have @xmath61 which makes a clear distinction between the @xmath62 fields @xmath63 and the @xmath64 field .
this last one will either trivially vanish , or have a unique behavior of starting at a non - vanishing central value and decreasing monotonically to zero .
therefore , to avoid singularity at @xmath58 all the solutions discussed in the following will have @xmath65 .
for the asymptotic behavior of the gauge fields we start by the observation that under a lorentz boost in the @xmath48 direction , the quantities @xmath66 transform as lorentz two - vectors . since we will consider in this paper only the case when these two vectors are spacelike ( corresponding to the magnetic case in the classification of @xcite ) , we can choose _ one _ of these two - vectors to be of the form @xmath67 without loosing generality .
independently , one can perform residual gauge transformations of the form @xmath68 and assume in the following @xmath69 as for the azimuthal components of the gauge fields we have to impose @xmath70 these are @xmath62 equations for the two unknowns @xmath71 and @xmath72 , so the three magnetic numbers @xmath73 can not be independent . moreover , since they must be integers , not every @xmath38 in the cartan subalgebra is possible .
a closed formula for these relations is rather complicated so we demonstrate this by two examples for the case @xmath74 where we can parametrize @xmath38 by : @xmath75 where @xmath76 are gell - mann matrices .
for @xmath77 we have the relation @xmath78 and @xmath79 , while for @xmath80 we have @xmath81 and accordingly @xmath82 .
@xmath71 however is given always by @xmath83 .
the concrete solutions which we will present will be mainly with @xmath74 ( other values of @xmath62 can be treated similarly ) . in this section ,
we express several quantities which characterize the different types of solutions .
the quantities are presented for string solutions of the extended non - abelian higgs model .
the corresponding formulas for the `` non extended '' model ( i.e. @xmath84 ) are obtained by setting @xmath85 and @xmath86 in eq .
( [ semilocalansatz ] ) and those that follow .
the energy and angular momentum ( per unit length ) of the solutions are obtained by computing the integrals @xmath87 the energy densities @xmath88 , correspond respectively to the scalar part , the gauge field part and the potential .
they are given by @xmath89 \\
\nonumber + \rho_n'^2 + \left ( \frac{(n - a_\theta - \tilde{a}_\theta g_{n})^2}{x^2}+ ( \varpi - a_0 - \tilde{a}_0 g_{n})^2 + ( q - a_z - \tilde{a}_z g_{n})^2 \right)\rho_n^2 \nonumber \label{energydensity0}\end{aligned}\ ] ] @xmath90 @xmath91\end{aligned}\ ] ] the angular momentum density @xmath92 has the form @xmath93 the solutions are also characterized by the noether charges and currents associated with the global @xmath94 symmetry of the lagrangian . the charges ( per unit length ) are the integrals of the time components of the current densities @xmath95 where @xmath96 denotes any of the generators of the global symmetry and the summation convention is used for both the local and global indices . for the `` diagonal '' ansatz , eq .
( [ semilocalansatz ] ) , we get the following expressions for the global charge densities @xmath97 where we denote by @xmath98 the diagonal elements of the commuting flavor generators as we do for local symmetry . similarly for the current densities : @xmath99 accordingly , the charges and currents will be @xmath100 where the `` single field contributions '' @xmath101 and @xmath102 are defined in an obvious way . in terms of these
, the angular momentum @xmath103 of the string can be written as @xmath104 .
since the non - abelian charges all vanish due to the boundary conditions , these `` single field contributions '' are not independent but satisfy the following identities among the charges and currents @xmath105 these identities allow one to express two of the currents ( or charges ) in terms of the others . as a consequence , for an @xmath94 global symmetry
, the @xmath62 diagonal global currents can be expressed in terms of @xmath106 independent ones .
for example , for local @xmath7 and global @xmath40 , these relations lead to @xmath107 ; therefore @xmath108 by the left equation of ( [ sumchargesandcurrents ] ) and the two diagonal global currents are proportional to @xmath109 . for local @xmath40 and the choice @xmath77
, the global @xmath110 currents depend on two of the `` single field '' currents @xmath102 say , @xmath111 and @xmath112 .
actually two of them are just proportional to @xmath111 and @xmath112 and the third is a combination : @xmath113 these identities provide useful crosschecks of the numerical solutions discussed next .
we solved the system of non - linear equations ( [ eqscalarsemilocal])-([ymeqssemilocal ] ) by using a numerical routine based on the collocation method @xcite .
the solutions were constructed with grids involving typically 1000 points and with accuracy of the order of @xmath114 .
the solutions we found are described in the following . before discussing the semilocal strings
, we pause to describe the simplest kind of non - abelian ones , namely the _ local _ non - abelian strings .
they are obtained by truncating the general system to the case with @xmath115 so @xmath85 and @xmath86 .
this requires the further substitutions @xmath116 in the field equations above .
= 10.0 cm + inspection of the boundary conditions for the gauge fields as @xmath117 , reveals that in the gauge choice ( [ gaugeres ] ) all the parameters @xmath118 should vanish , leading to @xmath119 .
so we solve equations ( [ eqscalarsemilocal])-([ymeqssemilocal ] ) with the above substitutions and with the following boundary conditions : @xmath120 the only stringlike solutions in this system are of the magnetic type , labelled by the integers @xmath73 which are consistent with the value we pick for the angle @xmath121 in the 3 - 8 plane of the cartan subalgebra . in fig .
[ fig1 ] we show the field profiles of a typical @xmath40 local string solutions with @xmath77 . the abelian flux is still quantized but is not an integer as can be easily seen from the figure .
its value is 3/2 .
similarly , the non - abelian flux is @xmath122 .
now we return to the general case of the full system with @xmath123 , that is equations ( [ eqscalarsemilocal])-([ymeqssemilocal ] ) with non - vanishing @xmath124 and @xmath125 .
the solutions have in general @xmath126 while we can still exploit the remaining symmetry to require again @xmath127 for @xmath128 .
in contrast with the local case , the equations for the fields @xmath129 , @xmath130 , @xmath131 and @xmath132 are not trivially satisfied if @xmath133 and we have now non - trivial boundary conditions for these gauge components . the boundary conditions are contained in equations ( [ bczero ] ) , ( [ bcscalarinf ] ) , ( [ gaugeres ] ) and ( [ bcthetainf ] ) and we will not repeat them here . we just point out that the scalar and gauge fields reach their asymptotic values with exponential corrections .
in particular , the field @xmath64 decays according to @xmath134}$ ] , which explains why this field can decay slower than the others for @xmath135 .
we have therefore to solve for @xmath136 , @xmath137 , @xmath138 , @xmath139 and @xmath140 in addition to the @xmath62 scalar components and the two gauge components @xmath141 and @xmath142 . along with @xcite , we call solutions of this type `` twisted '' solutions . the parameter @xmath143 will be referred to as the `` twist '' parameter .
the solutions will be generally `` twisted '' , but special `` untwisted '' solutions ( with @xmath144 ) also occur .
we also distinguish between static solutions with @xmath145 and stationary for @xmath146 .
their properties will be studied below .
the simplest kind of semilocal solutions is the _ embedded _ ( non - abelian ) local flux tubes .
these are purely magnetic solutions of the full system with @xmath85 .
it is therefore evident that @xmath147 and the non - vanishing fields , the @xmath62 scalars @xmath63 , and the two gauge components @xmath141 and @xmath142 , are the same as for the local strings of sec .
[ seclocalstrings ] .
another kind of purely magnetic strings is the _ untwisted _ semilocal strings which generalize the `` skyrmion '' solutions of the abelian model mentioned in sec .
[ secextnah ] .
these solutions are only self - dual as in the abelian case , and are contained already among the results of earlier studies @xcite . in our parametrization
they exist for @xmath148 .
the more general kinds of solutions which contain `` twists '' @xcite will be discussed in the next sections .
the space of twisted semilocal strings decomposes into three regions which are labelled according to the sign of the norm of the two - vector @xmath149 , that is according to the sign of @xmath150 . in the `` timelike '' region ( @xmath151 )
we were unable to find localized solutions as happens also in the abelian case @xcite ( recall also the asymptotic behavior mentioned above ) , so we do not discuss them further .
solutions with @xmath152 are known as `` chiral '' and will be discussed below among the stationary solutions .
if we assume the two - vector @xmath149 to be spacelike , we can set @xmath153 by an appropriate lorentz boost .
the field equations for @xmath129 and @xmath131 are satisfied trivially by @xmath154 and all the charges @xmath155 vanish with this choice . the components @xmath130 and @xmath132 as well as the currents @xmath156 do not vanish . [ !
b ] , title="fig : " ] , title="fig : " ] + = 10.0 cm + in the abelian case ( @xmath28 and @xmath157 with irrelevant @xmath23 ) @xcite twisted semilocal strings appear for large enough values of the potential coupling constant @xmath22 ( precisely for @xmath35 in our notations ) .
they form continuous family of static solutions and are labelled by the `` twist parameter '' @xmath143 appearing in the phase of the scalar function @xmath158 ; this parameter takes values in a finite interval , that is @xmath159 $ ] where @xmath160 is the maximal twist where bifurcation with the embedded flux tube occurs .
this value of @xmath160 depends on the coupling constant @xmath22 . in the limit
@xmath161 the function @xmath158 converges uniformly to the null function and the branch of twisted solutions bifurcates into the local string solutions . here
, we consider the extended non - abelian higgs model with our potential ( [ pot1 ] ) and find that analogous string solutions exists when the potential parameters , @xmath162 , are chosen large enough .
a similar critical phenomenon occurs , and in addition to the critical twist there exists now a critical @xmath163 .
first , we study the case of an @xmath7 gauge group and a global @xmath40 .
we find twisted semilocal strings corresponding to different choices of the magnetic parameters @xmath164 .
[ figsstatictwsu2 ] shows the field profiles of two typical static twisted semilocal strings with different windings and therefore different fluxes .
note the additional scalar field @xmath158 which decreases slowly , yet still exponentially .
the relations among the currents , eq .
( [ sumchargesandcurrents ] ) yield in this case @xmath107 ( see explanation below ( [ sumchargesandcurrents ] ) ) and we observe that it is consistent with the numerical results giving @xmath165 for this case as is indeed seen in fig .
[ figsstatictwsu2 ] .
= 10.0 cm + analogous twisted solutions were obtained for @xmath166 and @xmath167 .
[ fig6 ] depicts a typical solution corresponding to the specific cartan generator fixed by @xmath77 .
the magnetic fields along the string axis ( abelian and non abelian ) and the energy density of a twisted solution ( with @xmath168 or @xmath169 ) are compared in fig .
[ fig12 ] with those of an embedded local string .
note that the twisted strings have a non - vanishing tangential magnetic field ( @xmath170see figure [ magneticfield ] ) whose contribution to the flux vanishes .
t ] = 10.0 cm , title="fig : " ] , title="fig : " ] + the solutions constructed above are static but they can be made stationary by applying a boost along the string axis ( @xmath48 ) .
alternatively , stationary solutions can be obtained directly by setting the parameter @xmath171 non zero in the field equations .
interestingly , allowing the parameter @xmath172 confers the strings with angular momentum as was first demonstrated in the abelian case @xcite .
the resulting angular momentum is therefore more like the `` orbital '' kind rather than an intrinsic spin - like .
the profiles of twisted solutions corresponding to @xmath173 and @xmath174 are presented in fig .
[ fig7 ] for @xmath153 and @xmath175 ( i.e. very close to the chiral limit @xmath176 ) .
we note that the chiral limit @xmath177 leads to a different kind of solutions with @xmath178 decaying as a power instead of exponentially .
this tendency is apparent in fig .
[ fig7 ] namely , through the decay of the function @xmath178 .
= 10.0 cm + since we know from the abelian case ( @xmath28 ) that twisted semilocal strings exist only is a restricted domain of the parameter space , it is natural to try to figure out this domain in the present non - abelian model .
the lagrangian ( [ lag ] ) is characterized by four parameters : @xmath22 , @xmath23 of the scalar potential and two gauge coupling constants @xmath179 and @xmath180 . however , only the relative strength @xmath181 appears in the field equations so the parameter space of the model is three - dimensional .
= 10.0 cm + = 10.0 cm + the parameter space of the _ solutions _ is far more complicated since it contains also the twist @xmath143 , the other phase parameter @xmath171 and the flux numbers .
we investigated this problem in the case of an @xmath40 gauge group and we present here the main results in various planes in this higher dimensional parameter space .
we took representative values of @xmath182 for which it turns out that twisted semilocal strings exist .
the effect of changing the parameter @xmath163 on the twisted solutions is illustrated in fig .
[ fig9 ] where the critical @xmath160-value is plotted as a function of @xmath183 for two different choices of @xmath184 .
the choice @xmath185 corresponds to @xmath77 and the choice @xmath186 to @xmath80 as explained at the end of sec .
[ secfeqs ] .
the twisted semilocal strings exist only for @xmath187 and for values of @xmath183 below a maximal value of @xmath188 .
this value depends on @xmath189 and determines a domain of existence of twisted solutions in the @xmath190 plane . in fig .
[ fig10 ] , a few physical quantities characterizing the solutions corresponding to @xmath191 are presented .
we find @xmath192 which is consistent with the value that is obtained from fig .
we further studied the critical phenomenon appearing when the maximal value of @xmath163 ( with @xmath143 fixed ) is reached as seen in fig .
[ fig10 ] .
our numerical results suggest in particular that the twisted solution bifurcates into the embedded local string at @xmath193 .
note especially the vanishing of @xmath194 , @xmath195 and @xmath196 as @xmath197 .
the evolution of the physical parameters characterizing the twisted solutions for @xmath198 is further presented in fig . [ fig11 ] as a function of the twist parameter @xmath143 .
we see from the plot that the solution bifurcates into the embedded local string for @xmath199 which is of course consistent with @xmath200 of fig .
we notice that the central value of the non abelian magnetic field , @xmath201 changes sign between the bifurcation point and the low twist region ( i.e. @xmath202 ) .
= 10.0 cm + the effect of a boost along the string axis , which is reflected by a non - vanishing @xmath171 , on some physical quantities is presented in fig .
[ fig8 ] . in this figure @xmath203 and @xmath201
are plotted versus @xmath171 for a fixed value of @xmath143 ( for instance @xmath174 ) and for the two values @xmath204 which require different sets of @xmath73 as explained above .
we stress that the plots are made for fixed @xmath143 , so the solutions involved are not related to each other by boosts along the @xmath48 direction .
in particular , the magnetic fields on the @xmath48-axis vary with @xmath171 .
note also that the chiral limit has nontrivial consequences on the mass and spin of the solutions ( per unit length ) which diverge in this limit as indicated in fig .
this work presents string solutions in the extended non - abelian higgs model which may be seen as an expansion of recent discussions of abelian twisted semilocal strings @xcite and of non - abelian untwisted ones @xcite .
the solutions are classified according to the twist parameter @xmath143 as twisted or untwisted ( for @xmath144 ) .
the twisted solutions are characterized by a persistent current along the string axis .
the twisted solutions are further classified as static ( @xmath153 ) or stationary , where the latter are also characterized by a non - zero angular momentum .
the extended non - abelian higgs model allows also two kinds of special solutions both untwisted and both purely magnetic : the embedded local non - abelian flux tubes and the self - dual `` skyrmions '' .
the occurence of twisted semilocal string and their corresponding conserved currents is closely connected to the existence of a scalar field which condensates on the axis and vanishes asymptotically .
for @xmath205 all scalars are higgs - like fields which acquire asymptotically a non vanishing expectation value due to the potential . as a consequence
, there is no place for a condensate and no twisted semilocal string are possible in this case .
all those solutions were constructed by numerically solving the field equations with the appropriate boundary conditions .
the solutions exist only on a non trivial domain of the parameter space of the possible solutions .
we have tried to determine this domain qualitatively and obtained a lot of information about the pattern of the solutions .
we demonstrated the bifurcation phenomenon which occurs in this case at a critical value of the twist @xmath143 which now depends on the various fluxes and on @xmath206 in addition to the parameters in the potential . a crucial step for deciding the physical relevance of non - abelian twisted semilocal strings would be to study their stability by performing a systematic normal mode analysis , as was done recently @xcite for the extended abelian higgs model .
another important issue is the understanding of the low - energy dynamics of the non - abelian twisted semilocal strings .
this can be done e.g. by using the geodesic approximation of @xcite which is based on a parametrization of the zero modes in terms of collective coordinates . for untwisted local and semilocal strings ,
this analysis is reported in details by aldrovandi ( see sec .
vi of @xcite ) .
since the phases of twisted semilocal strings depend on @xmath46 and @xmath48 , the parametrization of the orientational moduli in terms of slowly varying functions of these coordinates does not seem to be straightforward .
the parametrization of the fluctuations used in @xcite might turn out to be very useful .
we plan to address these questions in a further publication , as well as the gravitating counterparts of twisted strings .
another direction is motivated by the non - abelian superconducting strings , recently obtained by volkov @xcite within the @xmath207 electroweak theory .
these `` untwisted '' strings can be used as a starting point towards constructing twisted solutions which will be related to a non - abelian subgroup of the gauge group . *
acknowledgments : * we gratefully acknowledge prof .
m. nitta for useful comments and information about some relevant papers .
one of us ( y. v. ) gratefully acknowledges the belgian f.n.r.s . for a grant supporting a visit at the university of mons - hainaut where most of this work was carried out .
8 h. b. nielsen and p. olesen , nucl . phys . * b 61 * ( 1973 ) 45 .
h. j. de vega , phys .
d * 18 * , 2932 ( 1978 ) .
hasenfratz , phys . lett .
b * 85 * , 338 ( 1979 ) .
a. s. schwarz and yu .
s. tyupkin , phys .
b * 90 * ( 1980 ) 135 .
h. j. de vega and f. a. schaposnik , phys .
lett . * 56 * , 2564 ( 1986 ) .
h. j. de vega and f. a. schaposnik , phys .
d * 34 * , 3206 ( 1986 ) .
j. heo and t. vachaspati , phys .
d * 58 * , 065011 ( 1998 ) .
p. suranyi , phys .
b * 481 * , 136 ( 2000 ) .
f. a. schaposnik and p. suranyi , phys .
d * 62 * , 125002 ( 2000 ) .
m. a. c. kneipp and p. brockill , phys .
d * 64 * , 125012 ( 2001 ) .
k. konishi and l. spanu , int .
j. mod .
a * 18 * , 249 ( 2003 ) .
a. vilenkin and e.p.s .
shellard , _ cosmic strings and other topological defects _ , ( cambridge university press , cambridge , england , 1994 ) .
a. hanany and d. tong , jhep * 0307 * , 037 ( 2003 ) .
r. auzzi , s. bolognesi , j. evslin , k. konishi and a. yung , nucl . phys .
b * 673 * , 187 ( 2003 ) .
m. shifman and a. yung , phys .
d * 70 * , 045004 ( 2004 ) .
a. hanany and d. tong , jhep
* 0404 * , 066 ( 2004 ) . v. markov , a. marshakov and a. yung , nucl .
b * 709 * , 267 ( 2005 ) .
m. eto , y. isozumi , m. nitta , k. ohashi and n. sakai , phys . rev .
* 96 * , 161601 ( 2006 ) .
m. eto , y. isozumi , m. nitta , k. ohashi and n. sakai , j. phys .
a * 39 * , r315 ( 2006 ) .
m. eto , k. konishi , g. marmorini , m. nitta , k. ohashi , w. vinci and n. yokoi , phys .
d * 74 * , 065021 ( 2006 ) .
y. isozumi , m. nitta , k. ohashi and n. sakai , phys .
d * 71 * , 065018 ( 2005 ) .
m. shifman and a. yung , phys .
d * 73 * , 125012 ( 2006 ) .
p. forgacs , s. reuillon and m.s .
volkov , phys .
* 96 * ( 2006 ) 041601 .
p. forgacs , s. reuillon and m.s .
volkov , nucl .
* b 751 * ( 2006 ) 390 .
m. s. volkov , phys .
b * 644 * , 203 ( 2007 ) .
l. g. aldrovandi , phys .
d * 76 * , 085015 ( 2007 ) .
k. hashimoto and d. tong , jcap * 0509 * , 004 ( 2005 ) .
m. eto , k. hashimoto , g. marmorini , m. nitta , k. ohashi and w. vinci , phys .
* 98 * , 091602 ( 2007 ) .
m. hindmarsh , phys .
lett . * 68 * , 1263 ( 1992 ) .
j. preskill , phys .
d * 46 * , 4218 ( 1992 ) .
u. ascher , j. christiansen , r. d. russell , mathematics of computation * 33 * ( 1979 ) 659 ; acm transactions * 7 * ( 1981 ) 209 .
j. garaud and m.s .
volkov arxiv:0712.3589 , to appear in nucl .
n. manton , phys .
b * 110 * , 54 ( 1982 ) .
n. manton and p. sutcliffe , _ topological solitons _ , ( cambridge university press , cambridge , england , 2004 ) . | we find new non - abelian flux tube solutions in a model of @xmath0 scalar fields in the fundamental representation of @xmath1 with @xmath2 ( the `` extended non - abelian higgs model '' ) , and study their main properties . among the solutions there are spinning strings as well as superconducting ones .
the solutions exist only in a non trivial domain of the parameter space defined by the ratio between the @xmath3 and @xmath4 coupling constants , the scalar self - interaction coupling constants , the magnetic fluxes ( abelian as well as non - abelian ) and the `` twist parameter '' which is a non - trivial relative phase of the higgs fields . |
in the present paper we study the curvature of a free boundary which comes up in some obstacle type problems , more specifically laplacian growth ( hele - shaw flow moving boundary problem ) , quadrature domains and partial balayage .
the final aim of the investigations is to show that the free boundary in question satisfies a certain inner ball condition ( a specific upper bound on the curvature ) .
this goal has previously been achieved in the case of two dimensions ( see @xcite and @xcite ; compare also @xcite ) . here
we shall give some partial results ( but no complete solution ) in higher dimensions , and in passing also obtain a new proof for the two dimensional case .
the geometric property we aim at proving can most easily be stated in terms of quadrature domains for subharmonic functions @xcite .
let @xmath0 a positive borel measure with compact support in @xmath1 .
a bounded open set @xmath2 is called a _
quadrature open set _ for subharmonic functions with respect to @xmath0 if @xmath3 and @xmath4 for all integrable ( with respect to lebesgue measure @xmath5 ) subharmonic functions @xmath6 in @xmath7 . a quadrature open set which is connected is a _
quadrature domain_. there is a natural process of balayage of measures to a prescribed density ( partial balayage ) by which @xmath7 can be constructed from @xmath0 when it exists , see e.g. @xcite , @xcite , @xcite .
this balayage process is also equivalent to solving a certain obstacle problem @xcite . in terms of the difference @xmath8 between the newtonian potentials of @xmath0 and @xmath7 , the latter considered as a body of density one ,
the quadrature property spells out to @xmath9 the function @xmath10 appearing here is sometimes called the modified schwarz potential of @xmath11 , see @xcite for example .
what has been proved in two dimensions , and what we like to extend to higher dimension , is that @xmath7 in the above situation can be written as the union of open balls centered in the closed convex hull @xmath12 of @xmath13 : @xmath14 here @xmath15 denotes the radius of the ball at @xmath16 , allowing the possibility @xmath17 , i.e. , that the ball is empty .
we refer to ( [ innerball ] ) as @xmath7 satisfying the inner ball property with respect to @xmath12 . for further discussion and motivations in the present context , see @xcite , @xcite .
the inner ball property concerns the geometry of @xmath11 outside any closed half - space @xmath18 containing @xmath13 . by a certain `` localization '' procedure the part @xmath19 of @xmath7 which is outside
@xmath18 can be shown to be identical with a quadrature open set for some positive measure with support on @xmath20 , see @xcite , @xcite .
for this reason it is enough to prove ( [ innerball ] ) in the case that @xmath0 has support in a hyperplane , and by a further localization one may even assume @xmath0 to have a continuous density on it .
for convenience we shall take the hyperplane in question to be @xmath21 .
it is known ( see @xcite , @xcite ) that given any positive measure with compact support in this hyperplane , which we identify with @xmath22 , there is a uniquely determined quadrature open set , which moreover is symmetric about the hyperplane and is convex in the @xmath23-direction ( i.e. , the intersection with any straight line perpendicular to the hyperplane is connected ) .
thus the quadrature open set can be described in terms of a graph of a function @xmath24 defined in an open subset @xmath25 of the hyperplane .
it is known from the regularity theory of free boundaries ( see @xcite , @xcite , and in the present context @xcite ) that this function @xmath24 is real analytic .
it will be enough to consider the case that @xmath25 is connected , because in the disconnected case the discussions will apply to each component separately .
in the paper we shall therefore study the geometry of quadrature domains for a measure with support in the hyperplane @xmath22 , the corresponding modified schwarz potential @xmath10 , and various differential geometric objects derived from it .
the paper consists of two main parts .
the first part starts with some differential geometric preliminaries ( section [ sec : diffgeom ] ) and ends up with a proof of equivalence of several different formulations of the inner ball condition ( section [ sec : innerball ] ) .
this part is analogous to a corresponding part in @xcite , but new difficulties appear in the higher dimensional case . in the second part of the paper , section [ sec : boundaries ] ,
we develop tools for studying the geometry by means of vector fields and differential forms defined in terms of @xmath10 . to briefly explain , in terms the two dimensional situation , what we do in the second part of the paper , let @xmath26 denote the part of the quadrature domain which lies in the upper half space ( half plane ) and let @xmath27 be the schwarz function ( see @xcite , @xcite ) of @xmath28 .
this is in our case given by @xmath29 , it is analytic in @xmath26 , and equals @xmath30 on @xmath28 .
what we study in the second part of the paper is the higher dimensional counterpart of the mapping @xmath31 defined by @xmath32 the last member referring to polar coordinates .
we also study in higher dimensions that curve @xmath33 which in two dimensions is defined by @xmath34 . in the two dimensional case
a proof of the inner ball property can be based on the topological property of @xmath33 that it separates the two domains @xmath35 and @xmath36 from each other , see @xcite .
there seems to be no direct counterpart of this kind of proof in higher dimensions .
a slightly different proof in two dimensions uses the argument principle for @xmath37 , see corollary [ cor : main ] in the present paper .
we have not been able to generalize this proof either to higher dimensions , even though we think that such a proof may not be completely out of reach .
at least we have computed the relevant mapping degree of @xmath37 in higher dimensions , and this result , theorem [ thm : main ] , may be considered to be the main result in the second part of the paper . another avenue which we have pursued to some extent is the investigation of the general behaviour of the curve @xmath33 .
this curve starts out at the origin and reaches @xmath28 only at points where the largest inner ball centered at the origin touches @xmath28 , see proposition [ prop : gamma ] .
if we could establish for example that @xmath33 were a smooth curve ( no branchings ) a proof of the inner ball property would not be far away .
thus we hope that the partial results we obtain in this paper will turn out to be useful in forthcoming attempts to prove the inner ball property for quadrature domains , explicitly formulated in conjecture [ conj : main ] .
in this section we decompose the coordinates in @xmath1 typically as @xmath38 where @xmath39 , @xmath40 .
we denote by @xmath41 the upper half space .
let @xmath24 be a positive , twice continuously differentiable function defined in a bounded domain @xmath42 and assume that @xmath43 as @xmath44 ( the boundary as a subset of @xmath22 ) .
it is convenient to set @xmath45 for @xmath46 , and then @xmath24 is defined and continuous on all @xmath22 .
set @xmath47 and for any subset @xmath48 , @xmath49 .
the normal line of @xmath28 at a point @xmath50 is given in parametrized form as @xmath51 and it intersects the hyperplane @xmath22 for @xmath52 , i.e. , at the _ foot point _ @xmath53 let @xmath54 be the part of the normal line which is between the base point on @xmath28 and the foot point .
the length of @xmath55 is @xmath56 for any point @xmath57 there is at least one closest point @xmath50 on @xmath28 .
then @xmath58 and , in particular , @xmath59 .
note that @xmath55 is one of the radii in @xmath60 .
it follows that we always have inclusions @xmath61 also , @xmath62 we shall discuss @xmath28 from a differential geometric point of view .
we consider it as a riemannian manifold of dimension @xmath63 and with coordinates @xmath64 .
the riemannian metric is that inherited from @xmath1 .
we shall write certain quantities considered as vectors in @xmath1 in bold .
these involve the `` moving point '' on @xmath28 ( or lift map @xmath65 ) @xmath66 its differential ( a vector - valued differential form ) @xmath67 the unit normal vector @xmath68 and its differential @xmath69 ( which becomes somewhat complicated when written out in components ) . for a point @xmath57
we sometimes write it in bold if we think of it as the vector from the origin to @xmath16 . a tangent vector @xmath70 on @xmath28
may be thought of in an abstract way as a derivation : @xmath71 letting this @xmath70 act on the moving point @xmath72 or , equivalently , letting the differential @xmath73 act on @xmath70 gives the same tangent vector regarded as a vector embedded in @xmath1 : @xmath74 in classical differential geometry ( see for example @xcite ) one associates to any hypersurface in euclidean space two fundamental forms .
the first fundamental form is the metric tensor , which gives the inner product on each tangent space .
it is in our case @xmath75 where the dot denotes the scalar product in @xmath1 .
the second fundamental form is ( up to a sign ) @xmath76 it contains information on how @xmath77 rotates in different directions , i.e. on how @xmath28 is curved within @xmath1 .
the above expression for @xmath78 can be derived as follows : since @xmath77 is orthogonal to @xmath28 we have @xmath79 for all @xmath80 .
therefore @xmath81 proving the formula .
both fundamental forms are symmetric covariant 2-tensors , i.e. , symmetric bilinear forms on each tangent space .
when considered as bilinear forms we shall call them @xmath82 and @xmath83 respectively ( namely @xmath84 , @xmath85 ) .
the above formulas then mean that for ( abstract ) tangent vectors @xmath70 and @xmath86 we have @xmath87 note that @xmath82 is positive definite .
the principal curvatures of @xmath28 and the corresponding principal directions are the eigenvalues and eigendirections of the derivative of the gauss normal map , taking any point @xmath88 onto the normal vector at the point : @xmath89 ; the derivative is the induced linear map between the tangent spaces , intuitively @xmath90 . in terms of abstract tangent vectors this becomes @xmath91 , where @xmath92 , and one can also say that it is the map obtained when expressing the second fundamental form in terms of the first : @xmath93 it is well - known ( and easy to prove ) that these eigenvalues and eigendirections coincide with the stationary points and corresponding stationary directions for the quotient of the fundamental forms , namely for the map @xmath94 ( @xmath95 , @xmath96 ) . for example , the smallest eigenvalue ( the smallest principal curvature ) coincides with the minimum value of @xmath97 .
next we introduce the function @xmath99 namely half of the squared distance from points on @xmath28 to the origin .
more generally , for any @xmath100 with @xmath101 , @xmath102 we set @xmath103 considered to be defined for those @xmath104 for which @xmath105 .
then @xmath106 is twice continuously differentiable with @xmath107 @xmath108 for @xmath109 we can write @xmath110 thus considering the hessian matrix @xmath111 as a tensor , or bilinear form , we see that it is related to the two fundamental forms by @xmath112 we finally notice that @xmath98 is related to the foot point map @xmath113 by @xmath114 more generally , for any @xmath101 we have @xmath115 the poincar metric in @xmath116 is given by @xmath117 the geodesics with respect to the poincar metric are the vertical straight lines @xmath118 together with all vertical semicircles with centers on @xmath22 , namely all curves of the form @xmath119 where @xmath120 is a straight line in @xmath22 , @xmath121 and @xmath122 .
we consider now the variable transformation @xmath123 where @xmath124 , @xmath125 , defined by @xmath126 and , conversely , @xmath127 thus @xmath128 is one - to - one and takes @xmath116 in the @xmath38-space onto the epiparabola @xmath129 in the @xmath130-space . by the definition of @xmath98 , @xmath128 maps @xmath26 onto the set @xmath131 and it maps the graph of @xmath24 onto the graph of @xmath98 .
if @xmath98 is extended to all @xmath22 by setting @xmath132 outside @xmath25 ( this corresponds to extending @xmath24 by zero outside @xmath25 ) then @xmath133 is the epigraph of @xmath98 .
[ poincare ] the above map @xmath134 gives a one - to - one correspondence between the set of geodesics with respect to the poincar metric in @xmath116 and the set of straight lines in @xmath135 .
let @xmath33 be a geodesic in @xmath116 .
if @xmath33 is a vertical line , then also @xmath136 is a vertical line in @xmath135 , hence a straight line . if @xmath33 is of the form ( [ circle ] ) , then just performing the substitution ( [ subst ] ) we see that @xmath136 becomes @xmath137 hence it is a straight line . and
the arguments can easily be run in the other direction : every straight line in the @xmath130-space which has a nonempty intersection with @xmath135 is of the form @xmath136 for some geodesic @xmath33 in @xmath116 . in the new
coordinates @xmath138 the poincar metric takes the form @xmath139 @xmath140 as an easy calculation shows .
hence this is a metric in @xmath135 which is complete and has constant negative curvature , and whose geodesics are euclidean straight lines .
below we state equivalent criteria for the domain @xmath7 defined by ( [ omega ] ) to satisfy the inner ball condition for balls centred on the symmetry plane .
the definition of the inner ball condition can be taken to be statement ( v ) in the theorem .
[ theorem1 ] with assumptions and notations as in section [ sec : diffgeom ] the following statements are equivalent . 1 .
the restriction of @xmath98 to any convex subdomain of @xmath25 is convex .
( note that @xmath25 is not required to be convex itself . )
equivalent formulations : the matrix of second derivatives is positive semidefinite : @xmath141 in @xmath25 ; the extension of @xmath98 to @xmath22 is convex ; @xmath133 is convex as a set .
2 . for every @xmath100 with @xmath142 , @xmath143 ,
the function @xmath106 is convex ( alternatively : strictly convex ) when restricted to any convex subdomain of the set of @xmath104 for which @xmath105 .
the restriction of the foot point map @xmath113 to any convex subdomain of @xmath25 is monotone , i.e. @xmath144 for all @xmath145 in the subdomain .
( here the dot denotes the scalar product in @xmath22 . )
equivalently , the matrix @xmath146 is positive semidefinite .
@xmath147 5 . there exist radii @xmath148 such that @xmath149 6 .
@xmath150 7 .
the principal curvatures at any point @xmath50 of @xmath28 are @xmath151 .
every point in @xmath26 has a unique closest neighbor on @xmath28 .
every point on @xmath28 is a closest point on @xmath11 for some point in @xmath25 .
@xmath152 is convex with respect to the poincar metric in @xmath116 .
( i)@xmath153(ii ) : fix @xmath143 and @xmath104 .
for any @xmath154 with @xmath155 , let @xmath156 if @xmath157 then it follows from that @xmath158 while if @xmath159 equation shows that @xmath160 from these two inequalities the strict convexity of @xmath106 follows .
( we even get a uniform lower bound of @xmath161 in @xmath162 . )
( ii)@xmath153(i ) : just let @xmath163 in .
( i)@xmath164(iii ) : this follows from , which also shows that the matrix @xmath165 is symmetric itself ( @xmath166 ) .
( i)@xmath164(vii ) : as remarked in subsection [ sec : fundamentalforms ] the smallest principal curvature of @xmath28 coincides with the minimum value of the quotient @xmath97 between the two fundamental forms .
therefore it is immediate from that ( vii ) is equivalent to @xmath167 , i.e. , to ( i ) .
( i)@xmath153 ( iv ) : by we only need to show that @xmath168 for all @xmath104 . fix any @xmath169 . then taking @xmath170 in we see that the map @xmath171 has a stationary point at @xmath172 . in view of the interpretation of @xmath173 as half of the squared distance from @xmath174
to @xmath50 it follows that this stationary point is a ( global ) minimum if and only if holds for @xmath172 .
but when @xmath98 ( equivalently @xmath173 ) is convex then every stationary point of @xmath173 is a global minimum . hence holds for @xmath172 .
( iv)@xmath153(v ) : if the representation in ( iv ) holds then we get a representation as in ( v ) by adding small balls @xmath175 for those @xmath104 which are not in the range of @xmath113 .
( clearly @xmath113 maps @xmath25 into @xmath25 when ( iv ) holds , but it need not be onto . ) ( v)@xmath153(iv ) : let @xmath176 .
then , if ( v ) holds , there exist points @xmath177 and @xmath178 such that @xmath179 as @xmath180 .
the smoothness of @xmath28 and the inclusions @xmath181 imply that @xmath182 and @xmath183 and we conclude that @xmath184 . now
( iv ) follows from .
( iv)@xmath164(ix ) : note that @xmath176 is a closest point of @xmath185 if and only if @xmath186 and @xmath187 .
thus @xmath188 is determined by @xmath189 and it follows immediately ( in view also of ) that all points on @xmath28 are such closest points if and only if the inner ball condition in ( iv ) holds .
( iv)@xmath153(vi ) : assume that ( vi ) fails , so that there exists a point @xmath190 for some @xmath191 , @xmath192 .
without loss of generality @xmath193 . then @xmath194 .
since @xmath195 we have @xmath196 .
it follows that @xmath197 . but
@xmath198 , hence ( iv ) does not hold .
( vi)@xmath153(viii ) : if @xmath199 has two closest neighbours @xmath200 then @xmath190 .
( viii)@xmath153(ii ) : this conclusion is somewhat related to @xcite , theorem 2.1.30 ( attributed to motzkin ) , but we give an independent proof .
assume that ( ii ) fails and we shall produce a point @xmath201 with at least two closest neighbors on @xmath28 .
we may assume that @xmath202 is not convex for some @xmath143 in a convex subdomain of @xmath203 .
we extend @xmath202 to all @xmath22 by setting @xmath204 in the latter expression the plus subscript denotes positive part and @xmath24 is assumed to be extended by zero outside @xmath25 .
the function @xmath205 is easily seen to be continuously differentiable and by assumption it is not convex .
let @xmath206 be the convex envelope of @xmath205 , which can be defined as @xmath207 equivalently , @xmath206 is the function whose epigraph , @xmath208 , is the closed convex hull of the epigraph of @xmath205 . in the present case , because of the strict convexity of @xmath205 outside a compact set , the convex hull will be closed right away ( before taking the closure ) .
thus @xmath209 @xmath210 denoting `` convex hull '' . since @xmath205 is not convex there exists @xmath211 such that @xmath212 it is easy to see that the supremum in is attained for each fixed @xmath10 .
hence @xmath213 for some affine @xmath120 satisfying @xmath214 by the point @xmath215 can be written as a finite convex combination of points in the epigraph of @xmath205 . since this convex combination must take place within the graph of @xmath120
, @xmath216 is a convex combination of finitely many points @xmath10 ( at least two are needed ) for which @xmath217 .
it follows that equality is attained in for at least two different @xmath10 .
now write @xmath218 where @xmath142 , @xmath219 and the dot denotes scalar product in @xmath22
. then can be written @xmath220 where equality is attained for at least two different @xmath221 . since the left member in is @xmath222 with equality in at most one point ( namely @xmath223 )
it follows that @xmath224 this positive minimum value can not be attained when @xmath225 because for those @xmath10 the derivative of the left member in is @xmath226 .
note that @xmath223 does not satisfy when @xmath225 because of .
thus @xmath105 whenever equality holds in
. clearly also @xmath227 by and , showing that @xmath228 .
it now follows that @xmath229 for all @xmath104 with equality for at least two different @xmath10 satisfying @xmath105 . in conclusion
we have produced a point @xmath230 having at least two closest neighbors on @xmath28 .
( i)@xmath164(x ) : that @xmath231 is convex with respect to the poincar
metric means by definition that every poincar geodesic @xmath33 in @xmath116 meets @xmath7 at most once ( i.e. , @xmath232 has at most one component ) . by lemma [ poincare ]
this occurs if and only if the straight line @xmath136 in @xmath135 meets @xmath233 at most once .
this is the same as saying that @xmath133 is convex as a set , which is easily seen to be equivalent to the convexity of @xmath98 ( @xmath133 is the epigraph of the extension of @xmath98 obtained by setting @xmath234 outside @xmath25 ) .
this proves ( i)@xmath164(x ) .
let @xmath0 be a finite positive ( and not identically zero ) borel measure with compact support in the hyperplane @xmath238 . then there exists a unique quadrature open set @xmath2 for subharmonic functions with respect to @xmath0 , see @xcite , @xcite . thus @xmath3 and @xmath4 for all integrable ( with respect to lebesgue measure @xmath239 ) subharmonic functions @xmath6 in @xmath7 . by uniqueness
@xmath7 is symmetric with respect to the hyperplane @xmath22 .
we shall assume that @xmath7 is connected ( i.e. , a quadrature domain ) and discuss the curvature of @xmath28 .
in the nonconnected case the discussions will simply apply to each component of @xmath7 .
let @xmath240 be the difference between the newtonian potentials of @xmath0 and @xmath7 , the latter considered as a body of density one .
then @xmath241 from it follows that @xmath242 , hence @xmath243 on @xmath231 ( since @xmath10 attains its minimum value there ) . for proving the inner ball property
it is enough to consider the case that @xmath0 has a continuous density on @xmath22 : @xmath244 with @xmath245 on @xmath246 , @xmath247 on @xmath248 and @xmath249 continuous on @xmath22 ( see subsection [ sec : gammasigma ] for a motivation of this ) . in that case
, @xmath10 is continuous across @xmath22 while @xmath250 makes a jump of size @xmath249 across @xmath22 ( this is the meaning of the distributional laplacian in ) . in summary , assuming @xmath10 satisfies in @xmath26 the overdetermined system @xmath251 where in the last equation @xmath250 should be understood as boundary values from the upper half space .
if @xmath0 is not of the form ( [ dmufdx ] ) , the system ( [ overdetermined ] ) still holds with the last equation appropriately reformulated . returning to the general case ,
the components of @xmath252 are harmonic functions in @xmath26 . by applying the maximum principle
to @xmath250 it follows that @xmath253 @xmath254 thus @xmath10 is strictly decreasing with respect to @xmath23 , showing that @xmath28 is the graph of a function , i.e. , is of the form @xmath255 for some positive function @xmath24 defined in @xmath246 . with a more detailed
analysis one can show @xcite , @xcite , @xcite that the function @xmath24 is actually real analytic and tends to zero at @xmath256 .
it follows that we are in the setting of sections [ sec : diffgeom ] and [ sec : innerball ] .
as have already been remarked , conjecture [ conj : main ] has been settled in the case @xmath257 .
two completely different proofs appear in @xcite and @xcite . a third proof , perhaps somewhat related to the one in @xcite , will be given in the present paper ( corollary [ cor : main ] ) .
let @xmath261 be spherical coordinates in @xmath1 in the sense that @xmath262 , @xmath263 ( if @xmath264 )
. then @xmath265 .
in order to analyse the curvature of @xmath28 we introduce a function @xmath258 and a 2-form @xmath260 according to thus , @xmath267 ( @xmath268 ) with @xmath269 note that the restriction of @xmath258 to @xmath270 coincides with the function @xmath98 in subsection [ phi ] when the latter is regarded as a function of @xmath271 : with @xmath28 given by @xmath272 we have @xmath273 recall that the hodge star operator @xcite is defined on basic forms @xmath274 as @xmath275 where the sign is chosen so that @xmath276 and @xmath277 denotes any permutation of @xmath278 .
interior multiplication of a @xmath113-form @xmath279 with a vector field @xmath280 is the @xmath281-form obtained by letting @xmath280 occupy the first position when @xmath279 is viewed as an antisymmetric operator acting on @xmath113 vector fields : @xmath282 we shall also need the lie derivative @xmath283 with respect to a vector field @xmath70 , and recall that its action on any @xmath113-form @xmath279 is @xmath284 and that its action on a vector field @xmath86 is the commutator @xmath285,\ ] ] when all vector fields are considered as differential operators ( @xmath70 corresponds to the directional derivative @xmath286 etc . ) . in our euclidean
setting of differential geometry there are natural one - to - one correspondences between vector fields , @xmath287-forms and @xmath288-forms .
the basis elements correspond to each other according to @xmath289 and the forms / vectors simply keep their coefficients with respect to the above bases under the identifications .
this means for example that if @xmath290 are vectors , @xmath279 a @xmath287-form and @xmath291 an @xmath288-form , then @xmath292 @xmath293 the definition of @xmath260 can be expressed in terms of interior multiplication and the hodge star operator as @xmath294 it is easy to check that , when spelled out , simply becomes , so we just verify : differentiating we get @xmath297 when @xmath298 . for @xmath299
there is an additional term @xmath300 , which cancels the term @xmath301 .
summing over @xmath302 therefore gives @xmath303 on the other hand , differentiating gives @xmath304 by a change of order of differentiating , and in view of @xmath305 in @xmath26 , now follows . introducing the coexterior derivative @xmath306 , defined on @xmath113-forms by @xmath307 the system can be written @xmath308 trivially @xmath309 , and since @xmath260 by definition is exact , @xmath310 .
recall that the hodge laplacian is @xmath311 and that it in our euclidean setting simply equals the ordinary laplacian acting on the coefficients of the forms it is applied to . from ( [ drhodeltaomega ] )
we find ( using @xmath312 etc . ) next we introduce the vector field @xmath315 for any function @xmath316 , @xmath317 hence the cauchy - riemann system can be written in terms of @xmath70 and @xmath260 as @xmath318 immediate consequences , in view of @xmath319 , are [ projection ] at any point @xmath325 , @xmath326 equals the orthogonal projection of the vector @xmath327 onto the tangent space of @xmath28 at @xmath328 .
in particular , @xmath70 is tangent to @xmath28 , i.e. , @xmath329 on @xmath28 .
since all components of @xmath330 vanish on @xmath28 , the @xmath288-form @xmath331 vanishes along @xmath28 ( i.e. , its integral over any piece of @xmath28 is zero ) .
this means exactly that @xmath70 is tangent to @xmath28 . but
@xmath332 and @xmath333 is orthogonal to @xmath28 because @xmath334 on @xmath28 .
it follows that @xmath70 is the projection of @xmath327 onto the tangent space of @xmath28 . in two dimensions , with the identification @xmath339 and using @xmath340 as coordinates ( and with @xmath341 ) we can define the schwarz function in @xmath26 by @xmath342 it is holomorphic in @xmath26 and extends continuously to @xmath343 with @xmath344 in addition , by , @xmath345 in the two dimensional case @xmath330 is a @xmath346-form , i.e. , a function , and it is exactly the conjugate harmonic function of @xmath258 .
more precisely , @xmath258 , @xmath330 are the real and imaginary parts of the analytic function @xmath347 : @xmath348 in the case of three dimensions we can identify the @xmath259-form @xmath260 with the vector field @xmath349 the cauchy - riemann system @xmath350 then takes the form @xmath351 in terms of @xmath326 and @xmath352 we thus have the equations @xmath353 in particular @xmath354 . note the triple curl identity @xmath355 .
notice also that the vector field @xmath352 is everywhere tangent to the family of spheres @xmath356 .
thus , in spherical coordinates @xmath357 , where @xmath358 ( now ) is the angle between the vector @xmath359 and the @xmath360-axis and @xmath316 is the polar angle for the projection of @xmath359 onto the @xmath361-plane , on writing @xmath362 we have @xmath363 . on @xmath364 ,
i.e. , for @xmath365 we have @xmath366 because @xmath367 on @xmath25 . for the interior multiplication , @xmath368 as to the lie derivatives of @xmath352 ( vector ) , @xmath260 ( @xmath287-form ) , @xmath330 ( @xmath259-form ) we have , when the results are identified with vector fields , @xmath369 = ( \boldsymbol{\xi } \cdot\nabla ) \boldsymbol{\omega } - ( \boldsymbol{\omega}\cdot\nabla ) \boldsymbol{\xi } , \\ l_{\boldsymbol{\xi } } ( \omega ) & = \text { as a vector } = \operatorname*{\mathrm{curl\,}}(\boldsymbol{\omega } \times \boldsymbol{\xi } ) = ( \boldsymbol{\xi } \cdot\nabla ) \boldsymbol{\omega } - ( \boldsymbol{\omega}\cdot\nabla ) \boldsymbol{\xi } ,
\\ l_{\boldsymbol{\xi } } ( \ast\omega ) & = \text { as a vector } = { \nabla}(\boldsymbol{\xi}\cdot\boldsymbol{\omega } ) = ( \boldsymbol{\xi } \cdot\nabla ) \boldsymbol{\omega } + ( \boldsymbol{\omega}\cdot\nabla ) \boldsymbol{\xi}.\end{aligned}\ ] ] we now return to the general case . since , by , @xmath370 in @xmath26 , the integral curves of @xmath252 constitute a smooth ( even real analytic ) family of curves which fill up @xmath26 .
the curves are directed downwards ( @xmath371 ) , start immediately inside @xmath28 ( recall that @xmath372 on @xmath28 ) and end up at points of @xmath373 . thus a subdomain @xmath374 which is bounded by integral curves of @xmath252 may be thought of as a tube going from @xmath28 to @xmath375 .
let @xmath376 be a domain such that @xmath377 is smooth and consists of integral curves of @xmath252 . then , assuming @xmath378 with @xmath249 continuous on @xmath25 , @xmath379 @xmath239 denoting lebesgue measure .
the vanishing of @xmath260 at a point @xmath16 means by definition that all components of @xmath260 vanish .
note that , since @xmath385 and @xmath386 in @xmath26 , @xmath260 vanishes if and only if @xmath387 and @xmath388 are proportional , i.e. , if and only if @xmath252 is parallel to the vector from @xmath16 to the origin , and hence points towards the origin .
this is also the same as saying that the projection @xmath389 of @xmath252 onto the tangent plane of the sphere @xmath390 vanishes .
hence @xmath391 recall that @xmath392 denotes the vector from the origin to the point @xmath16 .
a possible proof of conjecture [ conj : main ] may be based on establishing that @xmath33 is a smooth curve , when nonempty .
the strategy would then be as follows .
suppose @xmath7 does not satisfy the inner ball condition , i.e. , that the equivalent conditions in theorem [ theorem1 ] do not hold .
then there exists by condition ( viii ) a point @xmath393 which has at least two closest neighbors on the boundary . here
@xmath101 , @xmath143 . by `` localization '' ( see @xcite , @xcite ) the set @xmath394 will still be the upper half of a quadrature domain , namely for a measure with support on the hyperplane @xmath395 .
moreover , this measure will have a continuous density function @xmath249 on @xmath395 , namely given by @xmath396 ( cf .
( [ overdetermined ] ) ) .
this is because the localization measure on @xmath395 can be obtained from @xmath10 by keeping @xmath10 unchanged in @xmath397 and defining a new continuation of @xmath10 to @xmath398 by reflection in @xmath395 .
note by ( [ dudxnneg ] ) that the above @xmath249 is strictly positive at points belonging to @xmath7 .
now we translate our system of coordinates so that @xmath399 becomes the origin in the new coordinate system , and we also adapt the notations in general to fit the new coordinates . in this new situation
@xmath7 is a subharmonic quadrature domain for a measure @xmath0 in the hyperplane @xmath400 with @xmath401 @xmath249 continuous on @xmath22 , @xmath402 on @xmath403 , and the largest ball @xmath404 centered at the origin and contained in @xmath7 touches @xmath28 in at least two points .
the set @xmath33 is expected to meet @xmath28 at these points ( cf .
proposition [ prop : gamma ] below ) , and in addition reach the origin , thus it need to branch . thus if we can prove that @xmath33 is a smooth curve we reach a contradiction .
the conclusion then is that the original @xmath7 will have to satisfy the inner ball condition . if @xmath406 , then @xmath33 is empty .
if @xmath407 then @xmath33 intersects every hemisphere @xmath408 with @xmath409 . for @xmath122 sufficiently small
there is exactly one intersection point . at the origin
@xmath33 starts out in the direction @xmath410 .
the closure of @xmath33 reaches @xmath411 only at the origin . to prove the first statement , assume @xmath412 , for example , and consider the component @xmath413 of @xmath260
we have @xmath414 in @xmath26 , @xmath415 on @xmath28 . on @xmath22 , @xmath416 , which by assumption ( and ) vanishes for @xmath417 and
is nonnegative ( and not identically zero ) for @xmath418 .
hence @xmath419 , and so @xmath420 , in @xmath26 . to prove the second statement , consider any hemisphere @xmath421 , @xmath422 , and the tangent vector field @xmath423 on it .
on the boundary , @xmath424 , @xmath425 points into the lower hemisphere , because @xmath426 .
thus it follows from well - know index theorems for vector fields ( see @xcite , for example ) that @xmath425 has to vanish somewhere in @xmath421 .
next , assume @xmath427 ( @xmath428 )
. then @xmath429 , @xmath430 for some @xmath431 ( @xmath392 , @xmath432 denote @xmath16 and @xmath433 considered as vectors ) .
an easy calculation , using @xmath434 , shows that @xmath435 by assumption , @xmath252 is smooth in @xmath26 up to @xmath22 and @xmath436 for small @xmath122 ( for some @xmath437 ) .
it follows that @xmath438 , hence @xmath439 if we let @xmath440 .
thus the difference quotients @xmath441 on @xmath421 tend to infinity as @xmath440 , which contradicts the smoothness of @xmath252 up to @xmath22 .
the conclusion is that there can not be two different points in @xmath442 for @xmath122 sufficiently small . from the definition of @xmath448 we see that @xmath449 hence @xmath260 is algebraically determined by the @xmath63 components @xmath450 .
it follows that all information of @xmath258 and @xmath260 is contained in the map @xmath451 in the case of two dimensions this @xmath452 is , with a switch of coordinates , the same as the function @xmath453 in : @xmath454 .
in particular we conclude from the above that @xmath455 let @xmath456 denote the jacobian matrix of @xmath37 : @xmath457 equation with @xmath458 says that @xmath459 equivalently , if @xmath37 considered as a vector field , @xmath460 note also , by corollary [ rhoharmonic ] , that @xmath37 is a harmonic map ( each component of @xmath37 is a harmonic function ) .
clearly @xmath37 extends continuously to @xmath461 . on @xmath28 , @xmath462 hence @xmath28 , of dimension @xmath63 , is mapped by @xmath37 onto a set of dimension at most one .
it follows that @xmath463 at all points of @xmath28 ( since @xmath37 loses at least @xmath464 dimensions ) . to be more precise , for any vector @xmath86 , @xmath465 and writing , at a given point on @xmath28 , @xmath86 as @xmath466 with @xmath467 tangent to @xmath28 and @xmath468 gives ( since @xmath469 on @xmath28 ) , @xmath470 where @xmath471 is the @xmath472:th unit vector ( for the last member proposition [ projection ] was used ) .
thus the range of @xmath456 is contained in the ( at most ) two dimensional vector space spanned by @xmath473 and @xmath474 . at points on @xmath28 where @xmath336 is stationary , i.e.
, @xmath338 , we have @xmath475 , hence @xmath476 .
since the trace of @xmath456 is zero it follows that all eigenvalues of @xmath456 are zero . in two dimensions it even follows that @xmath477 , because @xmath456 is symmetric when @xmath257 .
returning to the mapping properties of @xmath37 , we see from that @xmath37 maps @xmath28 onto the segment @xmath478 of the @xmath479-axis , possibly except for endpoints , where @xmath480 on @xmath246 we have @xmath481 thus , since we have assumed that @xmath405 on @xmath25 , @xmath482 meets the @xmath479-axis only if @xmath407 .
the meeting point is @xmath483 , hence has @xmath484 . in summary
, @xmath37 maps the boundary @xmath485 of @xmath26 as follows : @xmath486 is mapped onto a point @xmath487 with @xmath488 , @xmath489 is mapped to points on the @xmath479-axis , with @xmath490 , and for the remaining points @xmath491 of @xmath485 , @xmath492 does not meet the @xmath479-axis .
more precisely , @xmath16 and @xmath492 are by located on opposite sides of the @xmath479-axis .
it follows from all this that @xmath485 encloses the interval @xmath493 $ ] on the @xmath479-axis @xmath494 times in the sense of degree theory .
thus [ thm : main ] assume @xmath407 and that @xmath0 is of the form with @xmath405 on @xmath25 .
then the map @xmath37 has mapping degree ( brouwer degree ) @xmath494 with respect to each point @xmath495 with @xmath496 . in the case of two dimensions
a stronger conclusion is possible because the jacobi determinant has constant sign : @xmath498 .
switching the order between @xmath260 and @xmath258 one gets the analytic function @xmath499 ( which we still denote @xmath500 ) , with nonnegative jacobi determinant , and all arguments become more familiar , e.g. , the mapping degree can be identified with the argument variation divided by @xmath501 .
we repeat everything in this case , with stronger assertions .
[ cor : main ] with assumptions as in the theorem we have , in the case @xmath257 , that if @xmath407 then there is exactly one point on @xmath28 of shortest distance to the origin and @xmath33 is an analytic smooth arc from origin to that point . in particular , the equivalent conditions in theorem [ theorem1 ] hold .
in addition , @xmath33 is an integral curve of @xmath70 .
let @xmath502 be the distance from the origin to @xmath28 .
we apply the argument principle to the analytic function @xmath503 in @xmath26 .
we have @xmath504 , @xmath505 for @xmath506 , @xmath507 for @xmath508 . for @xmath509 , @xmath510 , hence @xmath511 .
it follows that as @xmath512 runs through @xmath485 , then @xmath513 encircles each point on the open interval @xmath514 exactly once .
in addition , no other point on @xmath515 is encircled , although some points lie on the image curve @xmath516 .
but now @xmath33 is exactly the inverse image of @xmath515 under @xmath517 , hence it follows that @xmath37 is univalent on @xmath33 .
this means that @xmath518 , for @xmath519 , is a bijective analytic parametrization of @xmath33 .
since @xmath28 is known to be analytic without singular points in the present geometry @xcite , @xcite , @xcite the schwarz function @xmath27 , and hence @xmath500 , extends to be analytic in a full neighborhood of @xmath28 .
in particular , @xmath37 is an open mapping in such a neighborhood .
let @xmath520 be a point on distance @xmath521 from the origin . then
, with the above extended @xmath37 , @xmath522 , and any neighborhood @xmath523 of @xmath524 in @xmath525 is mapped onto a full neighborhood @xmath526 of @xmath527 .
the latter neighborhood contains the interval @xmath528 on the real axis for some @xmath529 , hence @xmath523 contains the final part of @xmath33 .
since this is true for any neighborhood @xmath523 of @xmath524 it follows that @xmath530 consists of just the point @xmath524 .
in particular , there is only one point on @xmath28 on distance @xmath521 from the origin .
the last statement of the corollary follows from the fact that @xmath531 ( see ) in two dimensions , or simply from the analyticity of @xmath499 : @xmath33 is a level line of @xmath330 and @xmath326 is perpendicular to level lines of @xmath258 , which are orthogonal to those of @xmath330 .
now we return to @xmath532 dimensions .
we still expect that there is only one point on @xmath28 of shortest distance to the origin , that @xmath33 reaches @xmath28 only at that point , and that there are no other stationary points of @xmath336 .
the following proposition gives some partial results in this direction .
[ prop : gamma ] the closure of @xmath33 in @xmath1 intersects @xmath28 only at points of @xmath28 where @xmath338 , i.e. , where @xmath336 is stationary . if there is a strict local maximum of @xmath336 at some point @xmath533 , then @xmath16 is in the closure of @xmath33 .
it is easy to see that , at points in @xmath26 close to @xmath28 , @xmath252 is essentially directed in the inward normal direction . at any point @xmath534
where @xmath96 , the vector @xmath359 from the origin to @xmath16 has a nonzero angle to the normal vector .
it follows from ( [ gamma ] ) that @xmath252 can not be parallel to @xmath359 close to such points , hence that @xmath33 does not reach points of @xmath28 where @xmath96 .
next , let @xmath533 be a point at which @xmath336 attains a strict local maximum .
then for some neighborhood @xmath535 of @xmath16 @xmath536 ( @xmath537 ) . for a slightly smaller @xmath538 , say @xmath539 , @xmath540 is a piece of a sphere cut off by @xmath28 . in this piece @xmath541 , while on the boundary @xmath542 .
it follows that there is a point on @xmath543 where @xmath544 attains a local maximum .
but at such a point the gradient @xmath252 is perpendicular to @xmath543 , which means that the point belongs to @xmath33 .
since @xmath545 was arbitrarily close to @xmath538 , this means that we can produce points on @xmath33 arbitrarily close to @xmath16 .
let @xmath546 be the hessian of @xmath10 .
it is a symmetric matrix with @xmath547 in @xmath26 .
the rank of @xmath18 is at least one ( since the sum of eigenvalues is different from zero ) . on @xmath28
the rank is exactly one since @xmath243 on @xmath28 and hence all derivatives of @xmath252 in tangential directions of @xmath28 are zero . from @xmath548 , where the last product is matrix multiplication , it follows that the definition of @xmath70 can be written in terms of @xmath18 as @xmath549 thus , on @xmath28 , @xmath550 and in particular @xmath551 at stationary points ( for @xmath336 ) on @xmath28 .
more generally we can write , for any @xmath552 , @xmath553 on @xmath28 , @xmath554 , hence @xmath555 is perpendicular to @xmath28 . therefore , on @xmath28 , @xmath556 for some @xmath557 .
it follows that @xmath558 defines the only eigendirection for @xmath18 corresponding to a nonzero eigenvalue , and since the sum of eigenvalues equals one , that @xmath559 moreover , it follows that @xmath18 annihilates every vector tangent to @xmath28 .
thus , on @xmath28 , @xmath18 simply is the orthogonal projection onto the normal of @xmath28 : @xmath560 this follows from a straight - forward computation of the second derivatives of @xmath571 .
assuming for simplicity that @xmath572 and computing , e.g. , the derivatives in the @xmath573-direction in the case @xmath574 we have , with @xmath575 , @xmath576 where @xmath577 denotes the spherical area of @xmath578 .
we may assume that @xmath572 .
set @xmath582 .
classical balayage of a point mass at any point @xmath583 onto @xmath584 gives the mass distribution on @xmath584 with density equal to the poisson kernel ( or normal derivative of green function ) , i.e. , the function of @xmath585 @xmath586 it follows that the density @xmath587 of classical balayage of @xmath0 is obtained by integrating the above with respect to @xmath588 : @xmath589 note that the last expression is a function only of @xmath337 .
one sees immediately that as such a function the second order derivative in any direction is nonnegative ( compare previous proof ) .
in fact , any derivative of even order is nonnegative ( a similar remark applies to the previous proposition ) . | it is known that any subharmonic quadrature domain in two dimensions satisfies a natural inner ball condition , in other words there is a specific upper bound on the curvature of the boundary .
this result directly applies to free boundaries appearing in obstacle type problems and in hele - shaw flow . in the present paper
we make partial progress on the corresponding question in higher dimensions . specifically , we prove the equivalence between several different ways to formulate the inner ball condition , and we compute the brouwer degree for a geometrically important mapping related to the schwarz potential of the boundary .
the latter gives in particular a new proof in the two dimensional case . |
a fact usually assumed in astrophysics is that the main part of the mass of a typical spiral galaxy is concentrated in a thin disk ( @xcite ) .
accordingly , the obtention of the gravitational potential generated by an idealized thin disk is a problem of great astrophysical relevance and so , through the years , different approaches has been used to obtain such kind of thin disk models .
wyse and mayall ( @xcite ) studied thin disks by superposing an infinite family of elementary disks of different radii .
brandt ( @xcite ) and brandt and belton ( @xcite ) constructed flat galaxy disks by the flattening of a distribution of matter whose surface of equal density were similar spheroids .
a simple potential - density pair for a thin disk model was introduced by kuzmin ( @xcite ) and then rederived by toomre ( @xcite ) as the first member of a generalized family of models .
the toomre models are obtained by solving the laplace equation in cylindrical coordinates subject to appropriated boundary conditions on the disk and at infinity .
the kuzmin and toomre models of thin disks , although they have surface densities and rotation curves with remarkable properties , represent disks of infinite extension and thus they are rather poor flat galaxy models .
accordingly , in order to obtain more realistic models of flat galaxies , is better to consider methods that permit the obtention of finite thin disk models . a simple method to obtain the surface density , the gravitational potential and the rotation curve of thin disks of finite radius was developed by @xcite . the hunter method is based in the obtention of solutions of laplace equation in terms of oblate spheroidal coordinates , which are ideally suited to the study of flat disks of finite extension . by superposition of solutions of laplace equation
, expressions for the surface density of the disks , the gravitational potential and its rotational velocity can be obtained as series of elementary functions .
the simplest example of a thin disk obtained by means of the hunter method is the well known kalnajs disk ( @xcite ) , which can also be obtained by flattening a uniformly rotating spheroid ( @xcite ) .
the kalnajs disk have a well behaved surface density and represents a uniformly rotating disk , so that its circular velocity is proportional to the radius , and its stability properties have been extensively studied ( see , for instance , hunter ( @xcite ) , @xcite and @xcite ) . in this paper
we use the hunter method in order to obtain an infinite family of thin disks of finite radius .
we particularize the hunter general model by considering a family of thin disks with a well behaved surface mass density .
we will require that the surface density be a monotonically decreasing function of the radius , with a maximum at the center of the disk and vanishing at the edge , in such a way that the mass distribution of the higher members of the family be more concentrated at the center .
the paper is organized as follows . in sec .
2 we present a summary of the hunter method used to obtain the thin disk models of finite radius and also we obtain the general expressions for the gravitational potential , the surface density and the circular velocity . in the next section , sec .
3 , we present the particular family of models obtained by imposing the required behavior of the surface densities and then , in sec . 4 , we analyze its physical behavior . finally , in sec . 5
, we summarize our main results .
in order to obtain finite axially symmetric thin disk models , we need to find solutions of the laplace equation that represents the outer potential of a thin disklike source .
according with this , we need to solve the laplace equation for an axially symmetric potential , @xmath0 where @xmath1 are the usual cylindrical coordinates .
we will suppose that , besides the axial symmetry , the gravitational potential has symmetry of reflection with respect to the plane @xmath2 , @xmath3 so that the normal derivative of the potential , @xmath4 , satisfies the relation @xmath5 in agreement with the attractive character of the gravitational field .
we also assume that @xmath6 do not vanishes on the plane @xmath7 , in order to have a thin distribution of matter that represents the disk .
given a potential @xmath8 with the above properties , the density @xmath9 of the surface distribution of matter can be obtained using the gauss law ( @xcite ) .
so , using the equation ( [ eq : con2 ] ) , we obtain @xmath10_{z = 0^{+}}.\label{eq : sigma}\ ] ] now , in order to have a surface density corresponding to a finite disklike distribution of matter , we impose the boundary conditions @xmath11 so that the matter distribution is restricted to the disk @xmath7 , @xmath12 .
we introduce now the oblate spheroidal coordinates , whose symmetry adapts in a natural way to the geometry of the model .
this coordinates are related to the usual cylindrical coordinates by the relation ( @xcite ) , @xmath13 where @xmath14 and @xmath15 .
the disk has the coordinates @xmath16 , @xmath17 .
on crossing the disk , @xmath18 changes sign but does not change in absolute value .
this singular behavior of the coordinate @xmath18 implies that an even function of @xmath18 is a continuous function everywhere but has a discontinuous @xmath18 derivative at the disk
. in terms of the oblate spheroidal coordinates , the laplace equation can be written as @xmath19_{,\xi } + [ ( 1 - \eta^2 ) \phi_{,\eta}]_{,\eta},\ ] ] and we need to find solutions that be even functions of @xmath18 and with the boundary conditions @xmath20 where @xmath21 is an even function which can be expanded in a series of legendre polynomials in the interval @xmath22 ( @xcite ) . according with this , the newtonian gravitational potential for the exterior of a finite thin disk with an axially symmetric matter density
can be written as ( @xcite ) , @xmath23 where @xmath24 are arbitrary constants , @xmath25 and @xmath26 are the usual legendre polynomials and the legendre functions of second kind , respectively . with this general solution for the gravitational potential ,
the surface matter density is given by @xmath27 and , as we will shown later , the arbitrary constants @xmath24 must be chosen properly so that the surface density presents a physically reasonable behavior .
besides the matter density , another quantity commonly used to characterize galactic matter distributions is the circular velocity @xmath28 , also called the rotation curve , defined as the tangential velocity of the stars in circular orbits around the center .
now , given @xmath29 , we can easily evaluate @xmath30 through the relation @xmath31_{z=0 } , \label{eq:1.22}\ ] ] in such a way that , by using ( [ eq : potencial ] ) , we obtain @xmath32 for the circular velocity .
we will now particularize the above general model by considering a family of finite thin disk with a well behaved surface mass density .
we will require that the surface density will be a monotonically decreasing function of the radius , with a maximum at the center of the disk and vanishing at the edge . in order to do this ,
we impose the conditions @xmath33 and we also require that @xmath34 where @xmath35 is the total mass of the disk . now , by using the boundary condition ( [ eq : bc1 ] ) , the surface density can be written in the form @xmath36 where @xmath37 is an even function of @xmath18 , monotonically increasing at the interval @xmath38 , and such that @xmath39 furthermore , we must to impose the condition @xmath40 in agreement with ( [ eq:4.12 ] ) .
a simple function @xmath21 that agrees with all the above requirements was given by @xcite and can be written as @xmath41 where , in order to fulfill the condition ( [ eq : limit ] ) , we must take @xmath42 . with this particular choice of @xmath21
we obtain an infinite family of finite disks with surface mass densities given by @xmath43^{m-1/2}. \label{eq:4.20}\ ] ] as we can easily see , the disk with @xmath44 corresponds to the well known kalnajs disk ( @xcite ) .
accordingly , this family of finite thin disks can then be considered as a generalization of the kalnajs disk now , from the equation ( [ eq:4.10 ] ) , the function @xmath21 can be written as @xmath45 with @xmath46 the coefficients @xmath47 are founded , by using the orthogonality property of the legendre polynomials , through the expression @xmath48 the above equation can be expressed as ( @xcite ) @xmath49,\ ] ] so that , using the gamma function properties , we obtain : @xmath50,\ ] ] for @xmath51 and @xmath52 for @xmath53 .
with the above values of the @xmath24 we can compute the different physical quantities that characterize the behavior of the models .
so , for instance , the gravitational potential of the first three members of the family are given by @xmath54 , \label{eq:4.22 } \\
\phi_{2}(\xi,\eta ) & = - \frac{mg}{a } [ \cot^{-1 } \xi + \frac{10 a}{7 } ( 3\eta^{2 } - 1 ) \nonumber \\ & \quad \quad + \ b ( 35 \eta^{4 } - 30 \eta^{2 } + 3 ) ] , \label{eq:4.23 } \\
\phi_{3}(\xi,\eta ) & = - \frac{mg}{a } [ \cot^{-1 }
\xi + \frac{10 a}{6 } ( 3 \eta^{2 } - 1 ) \nonumber \\ & \quad \quad + \
\frac{21 b}{11 } ( 35 \eta^{4 } - 30 \eta^{2 } + 3 ) \nonumber \\ & \quad \quad + \ c ( 231 \eta^{6 } - 315 \eta^{4 } + 105 \eta^{2 } - 5 ) ] , \label{eq:4.24 } \end{aligned}\ ] ] where @xmath55 , \\ b & = \frac{3}{448 } [ ( 35 \xi^{4 } + 30 \xi^{2 } + 3 ) \cot^{-1 } \xi - 35 \xi^{3 } - \frac{55}{3 } \xi ] , \\ c & = \frac{5}{8448 } [ ( 231 \xi^{6 } + 315 \xi^{4 } + 105 \xi^{2 } + 5 ) \cot^{-1 } \xi \nonumber \\ & \quad \quad - 231 \xi^{5 } - 238 \xi^{3 } - \frac{231}{5 } \xi ] , \end{aligned}\ ] ] with similar , but more involved , expressions for greater values of @xmath56 . by taking @xmath16 at the above expressions , we can obtain the value of the potential at the plane of the disk .
now , is easy to see that , for all the members of the family , we will obtain finite values for these quantities .
in particular , for the first member of the family , the potential at the disk is given by @xmath57 for @xmath58 , and this expression is completely equivalent to the corresponding expression in @xcite . in the same way ,
we obtain for the surface mass densities of the first three members of the family the expressions @xmath59 and for the corresponding circular velocities the expressions @xmath60 where has been introduced the dimensionless radial variable @xmath61 . in order to graphically illustrate the behavior of the different particular models
, we first introduce the dimensionless surface density of the disks , defined as @xmath62 for @xmath63 .
so , in fig .
[ fig : dens ] are depicted the dimensionless surface mass densities @xmath64 for the models corresponding to @xmath65 .
as we can see , the disks with higher values of @xmath56 present a mass distribution that is more concentrated at the center and less at the edge .
accordingly , these disks can then be considered as appropriated models of galaxies with a central bulb .
now , in order to graphically illustrate the behavior of the circular velocities or rotation curves , we introduce the dimensionless quantity @xmath66 we plot , in fig .
[ fig : vel1 ] the dimensionless rotation curves for the models corresponding to @xmath67 .
the circular velcotiy corresponding to @xmath44 is proportional to the radius , representing thus a uniformly rotating disk . on the other hand , for @xmath68 , the circular velocity increases from a value of zero at the center of the disks , until it attain a maximum at a critical radius and then decreases to a finite value at the edge of the disk .
also we can see that the value of the critical radius decreases as the value of @xmath56 increases .
we presented an infinite family of axially symmetric thin disks of finite radius obtained by means of a particularization of the hunter method .
the disk models so obtained are generalizations of the well known kalnajs disk , which corresponds to the first member of the family .
the particularization of the hunter model was obtained by requiring that the surface density was a monotonically decreasing function of the radius , with a maximum at the center of the disk and vanishing at the edge , in such a way that the mass distribution of the higher members of the family were more concentrated at the center .
we also analyzed the rotation curves of the models and we find for the first member of the family , the kalnajs disk , a circular velocity proportional to the radius , representing thus a uniformly rotating disk , whereas for the other members of the family the circular velocity increases from a value of zero at the center of the disks until reach a maximum at a critical radius and then decreases to a finite value at the edge of the disk .
also we find that the value of the critical radius decreases as the value of @xmath56 increases .
we believe that the obtained thin disk models have some remarkable properties and so they can be considered as appropriated realistic flat galaxy models , in particular if the superposition of these thin disks with appropriated halo distributions ( @xcite ) is considered .
we are now considering some research in this direction .
we are now also working in the non axially symmetric generalization of the here presented disks models and also in the obtention of the relativistic generalization of them for the axially symmetric case .
the authors want to thank the financial support from colciencias , colombia . | an infinite family of axially symmetric thin disks of finite radius is presented .
the family of disks is obtained by means of a method developed by hunter and contains , as its first member , the kalnajs disk .
the surface densities of the disks present a maximum at the center of the disk and then decrease smoothly to zero at the edge , in such a way that the mass distribution of the higher members of the family is more concentrated at the center .
the first member of the family have a circular velocity proportional to the radius , representing thus a uniformly rotating disk . on the other hand , the circular velocities of the other members of the family increases from a value of zero at the center of the disks until a maximum and then decreases smoothly until a finite value at the edge of the disks , in such a way that for the higher members of the family the maximum value of the circular velocity is attained nearest the center of the disks .
stellar dynamics galaxies : kinematics and dynamics . |
in @xcite , c. bishop sets the following problem : _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ `` suppose we stand in a forest with tree trunks of radius @xmath6 and no two trees centered closer than unit distance apart . can the trees be arranged so that we can never see further than some distance @xmath7 , no matter where we stand and what direction we look in ? what is the size of @xmath8 in terms of @xmath9 ? '' _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ this is an example of a visibility problem , a topic which has attracted substantial interest over the past decades see @xcite for a survey . among the problems which gave impetus to research in this field , one can mention the art gallery problem ( see @xcite for details ) or , closer to the spirit of the question set by c. bishop , plya s orchard problem . in ( * ? ? ? * chap . 5 ,
problem 239 ) , g. plya asks `` how thick must [ be ] the trunks of the trees in a regularly spaced circular orchard grow if they are to block completely the view from the center '' .
he then provides a solution in the case that the observer stands at the origin in the plane and that the centres of the trees are the elements of @xmath10 lying in a disk of integer radius @xmath11 .
allen @xcite extended this result to the case when the disk has a non integer radius and kruskal @xcite dealt with the situation where the trees are centred at non
zero points of any lattice .
chamizo @xcite also studied an analogue of this problem in hyperbolic spaces and cusick @xcite considered the case when the trees have the shape of any given convex body ( cusick relates this case with the lonely runner conjecture see @xcite for further developments ) .
g. plya @xcite also took an interest in the visibility in a random and periodic forest , a topic related to the distribution of free path lengths in the lorentz gas which is still an active domain of research see @xcite and the references therein . on another front
, problems of visibility appear in the context of quasi
crystals @xcite and of probabilistic billiards in relation with the study of the behaviour of particles @xcite .
they are also much studied from an algorithmic point of view and one can therefore find a wealth of literature dealing with them in computer science see , e.g. , @xcite and the references therein .
lastly , one should mention that bishop s question finds its origin in a problem of rectifiability of curves .
[ [ section ] ] let @xmath12 be a fixed integer . define formally a forest @xmath1 in @xmath13 as a collection of points in @xmath13 . given @xmath14 , an @xmath15tree in this forest shall refer to a closed ball centred at an element in @xmath1 .
a set @xmath16 is a _ dense forest _ if there exists a function @xmath17 defined in a neighbourhood of the origin such that the following holds for all @xmath2 small enough : @xmath18 , \;\;\ ; \exists \bm{f}\in{\mathfrak{f } } , \;\;\ ; \left\|\bm{x}+t\bm{v}-\bm{f}\right\|_2\le { \epsilon},\ ] ] where @xmath19 stands for the euclidean norm in @xmath13 and @xmath20 for the euclidean sphere in dimension @xmath21 .
the function @xmath3 is then referred to as a _ visibility function _ for @xmath1 .
thus , in a forest with visibility function @xmath3 , given @xmath2 , _ any _ line segment of length @xmath22 intersect an @xmath9tree ( this is the main difference with a plya s orchard type problem , where one only takes into account those line segments with one of the end points at the origin ) .
it is clear that @xmath23 is a dense forest whenever the set @xmath1 is itself dense . to avoid this pathological case
, one may consider at least two types of restrictions for the set @xmath1 . on the one hand
, one may ask for there to exist a strictly positive real number @xmath24 such that the gap between any two elements in @xmath1 is at least @xmath24 .
the forest @xmath1 is then said to be _
uniformly discrete_. this is essentially the condition required by c. bishop in the statement of his problem .
y. solomon and b. weiss @xcite proved the existence of a uniformly discrete dense forest in any dimension .
however , their construction is not fully explicit as the forest they obtain is defined as a set of `` visit times '' for the action of a group on a suitable compact metric space .
furthermore , no bound is given for the corresponding visibility function .
on the other hand , one may consider a concept weaker than uniform discreteness , namely that of _ finite density_. more precisely , a forest @xmath1 is said to be of finite density if @xmath25 where @xmath26 stands for the cardinality of a subset @xmath27 and where , given @xmath28 , @xmath29 denotes the euclidean ball with radius @xmath30 centred at @xmath31 ( being of finite density is that of @xmath1 being discrete . ] ) . clearly , a uniformly discrete dense forest has a finite density . using dirichlet s theorem in diophantine approximation ,
y. peres @xcite proved the existence of a dense forest with finite density in the plane ( i.e. , when @xmath32 ) whose visibility function @xmath3 satisfies the estimate @xmath33 .
it is a question left open at end of @xcite to determine whether this bound can be improved .
the following , which is the main result of this paper , shows in particular that it is indeed the case .
[ mainthmintro ] let @xmath12 .
then there exists a dense forest @xmath23 with finite density whose visibility function @xmath3 satisfies the estimate @xmath34 for any @xmath5 .
here , @xmath35 means that the implicit constant implied by this notation depends at most on @xmath36 and @xmath37 .
the construction of the dense forest in theorem [ mainthmintro ] is inspired by the work of y. peres and has the merit to be explicit .
it will actually be defined as the union of @xmath21 uniformly discrete forests of a discrete set @xmath23 is defined by @xmath38 the condition @xmath39 is then equivalent to @xmath1 being the finite union of uniformly discrete sets . ] . as a consequence ,
the argument in @xcite can easily be modified to show that , given @xmath2 , there exists a uniformly discrete set @xmath40 depending on @xmath9 satisfying equation for a particular function @xmath3 .
it should be noted that the visibility problem under consideration is closely related to the danzer problem , which asks for the existence in @xmath13 of a set of finite density intersecting any convex body of volume one .
clearly , such a set is in particular a dense forest with visibility function @xmath41 . as the danzer problem is still open
, one may consider to weaken the conditions therein in two directions .
on the one hand , one may relax the density constraint .
thus , y. solomon and b. weiss @xcite showed that there exists a set @xmath42 intersecting any convex body with volume one such that @xmath43 . on the other hand ,
one may require less restrictive conditions for the convex bodies under consideration . while john s theorem ( @xcite , see also ( * ? ? ?
* proposition 5.4 ) ) shows that one obtains essentially an equivalent problem when limiting oneself to the subset of boxes in @xmath13 , one may relax the volume constraint for these boxes is the image under an orthogonal transformation of an aligned hyperrectangle of the form @xmath44 $ ] , where @xmath45 ( @xmath46 ) and @xmath47 . ] . for obvious reasons ,
the constructions of dense forests are in this vein . in this respect
, the interest of theorem [ mainthmintro ] lies in the fact that it provides the best known bound for the visibility function of a dense forest .
this bound is essentially the best one can hope for with the fourier analytic proof presented in sections [ sec3 ] and [ sec4 ] ( see also the introduction of section [ sec4 ] for further details ) .
the construction of the dense forest in section [ sec1 ] has an independent interest .
[ [ notation . ] ] * notation . * + + + + + + + + + + + in addition to those already introduced , the following pieces of notation shall be used throughout : for any @xmath48 , let @xmath49 let @xmath50 denote the fourier transform of an integrable function @xmath51 defined in @xmath52 . for any @xmath53 , where @xmath54 , @xmath55 and @xmath56 , the denominator @xmath57 of @xmath58 shall refer to the well defined integer @xmath59 . given @xmath60 $ ]
, @xmath61 shall stand for the usual @xmath62norm in @xmath52 .
for any two subsets @xmath63 , set @xmath64 and let @xmath65 ( resp .
@xmath66 ) denote the ball with radius @xmath67 centred at @xmath68 ( resp . the unit sphere in @xmath69 ) defined with respect to the @xmath62norm .
notation such as @xmath70 , @xmath71 and @xmath72 shall mean that the implicit constants implied by these relations depend at most on the parameters @xmath73 and @xmath74 .
also , @xmath75 shall refer to the lebesgue measure of a measurable set @xmath76 and @xmath77 to the usual dot product between two vectors @xmath78 . given @xmath79 with @xmath80 ,
let @xmath81 denote the interval of integers with endpoints @xmath82 and @xmath83 .
lastly , @xmath84 shall stand for the @xmath37dimensional torus @xmath85 and @xmath86 for the corresponding sup norm ; that is , for any @xmath68 , @xmath87
in view of the equivalence of norms in finite dimension and in view of the nature of the result to prove , it should be clear that one may choose to work with any norm instead of the usual euclidean one in definition .
this choice will indeed only affect the value of the implicit multiplicative constant in .
it will be more convenient for us to work with the sup norm .
thus , definition shall be modified in the following way : @xmath88 , \;\;\ ; \exists \bm{f}\in{\mathfrak{f } } , \;\;\ ; \left\|\bm{x}+t\bm{v}-\bm{f}\right\|_\infty\le { \epsilon}\ ] ] for any given @xmath14 small enough .
note that @xmath89 is the boundary of the hypercube @xmath90^{d+1}$ ] ; that is , @xmath91^{d+1}\ : : \ : \exists i_0\in \llbracket 1 , d+1\rrbracket , \ , \left|x_{i_0}\right|=1 \right\}.\ ] ] given @xmath92 , define @xmath93 fix @xmath2 and @xmath94 . for @xmath95 and @xmath96 , let @xmath97 be a line segment .
assume first that there exists a forest @xmath98 such that for any @xmath99 and any @xmath95 , @xmath100 this means that any line segment of euclidean length at most a constant multiple of @xmath8 intersects an @xmath9tree in @xmath101 provided it be supported on a line spanned by a vector @xmath102 .
see figure [ foretasyfigure1 ] below for an illustration in the plane of the type of forests @xmath101 that shall be considered later . [ foretasyfigure1 ] consider now _ any _ @xmath103 .
then , there exists an index @xmath104 such that @xmath105 . if @xmath106 , let @xmath107 denote the rotation in the @xmath108plane bringing @xmath109 onto @xmath110 and leaving the orthogonal of this plane invariant .
thus , @xmath111 , which implies from that @xmath112 for any @xmath113 . in other words , if one denotes by @xmath114 the forest @xmath115 obtained by applying the rotation @xmath116 to the forest @xmath101 ( @xmath117 ) , then the visibility @xmath22 in the forest @xmath118 satisfies the estimate @xmath119 and is in particular finite .
if the forest @xmath101 is furthermore uniformly discrete , then the forest @xmath1 has a finite density as the union of a finite number of uniformly discrete forests . for the forest @xmath101 in the plane represented in figure [ foretasyfigure1 ] , the resulting forests @xmath120 and @xmath121 are respectively depicted in figures [ foretasyfigure ] and [ foretuniffigure ] [ foretasyfigure ] [ foretuniffigure ] it remains to define a uniformly discrete forest @xmath101 satisfying condition .
to this end , consider a sequence @xmath122 of real numbers to be specified later .
let @xmath123 clearly , this forest is symmetric with respect to the hyperplane @xmath124 .
more explicitly , on each affine hyperplane @xmath125 where @xmath126 , the forest is given by @xmath127 assume first that one is able to prove for all @xmath102 and @xmath113 such that @xmath128 .
consider then any line segment @xmath129 determined by the two vectors @xmath130 and @xmath131 .
if @xmath132 , then the symmetrical of @xmath129 with respect to the hyperplane @xmath124 lies in the half
space @xmath133 and it is therefore plain from the symmetry of the forest @xmath101 that then holds for @xmath129 . if , however , the latter line segment lies on both sides of the hyperplane @xmath134 , consider the extended line segment @xmath135 . even if it means increasing the value of the implicit constant appearing in the statement of theorem [ mainthmintro ] , it is cleary enough to prove that holds for @xmath135 .
this is straightforward : the middle point of @xmath135 determines two line segments , one of which lies either in the half
space @xmath136 or in the half space @xmath137 . in the former case
, holds by assumption and in the latter case this follows from the preceding argument .
all this shows that , when proving , it is enough to restrict ourselves to the case that @xmath128 , where @xmath102 and @xmath113 . even if it means defining the line segment @xmath138 from its second endpoint ( other than @xmath139 ) , it may further be assumed without loss of generality that the vector @xmath102 is given by @xmath140 . letting @xmath141 denote the components of the vector @xmath139 ,
these assumptions imply that @xmath142 where @xmath143 , @xmath144^d$ ] and @xmath145 .
if one can prove that the distance in the hyperplane @xmath146 between the point and an element of the set is less than @xmath9 for some @xmath147 , this would plainly imply that holds .
the problem has therefore been reduced to find @xmath94 such that the following holds for a given @xmath2 : @xmath148^d , \;\ ; \;\forall { \bm{z}}\in{\mathbb{r}}^d , \;\;\ ; \exists k\in \llbracket 0 , v\rrbracket , \;\;\ ; \left\|k{\bm{v } } ' + { \bm{z}}-a_{k}{\mathds{1}_\textrm{d}}\right\|_{\infty}^{{\mathbb{t}^d}}\ , \le \ , { \epsilon}.\ ] ] the following lemma essentially shows that it is enough to check for a finite number of @xmath149^d$ ] provided that @xmath150 which , in view of the statement of theorem [ mainthmintro ] , will be assumed throughout without loss of generality .
[ lemreduc ] assume that holds for a given @xmath2 .
define @xmath151^d\cap { \mathbb{q}}^d\ ; : \ ; \textrm{den}\left({\bm{q}}\right)\le v^2 \right\}.\end{aligned}\ ] ] assume furthermore that @xmath152 then , for all @xmath153^d$ ] and all @xmath154 , there exists @xmath147 such that @xmath155 let @xmath156^d$ ] and @xmath154 .
clearly , one can find @xmath157 such that @xmath158 . under assumption , there exist @xmath159 and @xmath160 satisfying the inequality @xmath161 .
it then follows that @xmath162
let @xmath163 . set @xmath164^{d}$ ] and denote by @xmath165 the characteristic function of the set @xmath166 ; that is , @xmath167 given @xmath168 , @xmath157 and @xmath154 , define @xmath169 thus , @xmath170 if , and only if , there exists @xmath171 such that @xmath172 . given @xmath5 , we shall prove that one can ensure that @xmath173 uniformly in @xmath157 and @xmath174 . in view of the reduction operated in and in view of lemma [ lemreduc ] , this means that any line segment with length at most a constant multiple of @xmath175 , contained in the half
space @xmath133 and supported on a line spanned by a vector in @xmath176 hits a @xmath177tree in the forest @xmath101 . from the discussion held in section [ sec1 ] , this is clearly enough to establish theorem [ mainthmintro ] . in order to establish
, the characteristic function @xmath165 shall be approximated by a finite trigonometric sum . in dimension one (
i.e. , when @xmath32 the set @xmath178 then reduces to the interval @xmath179 $ ] ) , this can be achieved thanks to the well known beurling
selberg extremal functions which approximate the characteristic function of @xmath179 $ ] by integrable functions with compactly supported fourier transform . in higher dimensions ( i.e. , when @xmath180 )
, similar constructions can be found in @xcite and ( * ? ? ?
* proof of theorem 5.25 ) to handle the case of the characteristic function of the hypercube @xmath178 . here
, we adopt the formalism of @xcite and give a simplified version of theorem 5.1 within which deals with the general case of aligned parallelotopes .
this is achieved in the following proposition , where the majorant function @xmath181 is introduced for the sake of completeness but will not be used thereafter .
( ( * ? ? ?
* theorem 5.1))[thmhaynes ] given an integer @xmath182 , there are trigonometric polynomials @xmath183 and @xmath181 whose fourier coefficients are supported in @xmath184 such that for all @xmath185 , @xmath186 moreover , there exist two strictly positive constants @xmath187 and @xmath188 for which @xmath189 and @xmath190 for all @xmath191 , where @xmath192 by choosing @xmath193 as @xmath194 we find that @xmath195 with the notation of proposition [ thmhaynes ] . in view of and , one thus has : @xmath196 thus , @xmath170 as soon as the right hand side of is strictly positive . the following theorem , which is the core of the proof of theorem [ mainthmintro ] , shall be established in section [ sec4 ] .
the sequence @xmath122 is in particular specified therein .
[ mainthm ] let @xmath197 be a constant depending on @xmath12 .
then , there exists a real number @xmath198 such that , setting @xmath199 the following estimate holds for any @xmath5 uniformly in @xmath200 and @xmath157 : @xmath201 in view of and , @xmath202 .
thus , from theorem [ mainthm ] , inequality implies that @xmath170 uniformly in @xmath157 as soon as @xmath203 that is , as soon as @xmath204 since this holds for all @xmath5 , this establishes theorem [ mainthmintro ] modulo the proof of theorem [ mainthm ] .
a classical heuristic argument ( cf .
@xcite ) shows that the upper bound in is , up to the factor @xmath36 in the exponent , the best one can hope for .
theorem [ mainthm ] actually holds for any sequence @xmath122 of the form @xmath205 for all @xmath206 , where @xmath207 is a sequence of integers such that @xmath208 indeed , such sequences satisfy the following property that shall play a crucial role in the proof of theorem [ mainthm ] : [ lemcompatge ] let @xmath209 and @xmath210 be integers .
assume that the sequence of integers @xmath211 satisfies .
then @xmath212 for all @xmath8 large enough ( depending on @xmath24 ) .
this implies in particular that @xmath213 assume that @xmath214 for @xmath215 .
comparing the leading terms in this equation shows that , under , they must be equal and must appear on both sides with the same multiplicity for @xmath8 large enough .
an easy induction then proves that @xmath216 .
let @xmath210 and @xmath217 be integers .
fix @xmath157 and @xmath200 .
then @xmath218 because @xmath219 , by orthogonality , the integral above does not vanish if , and only if , @xmath220 . from lemma [ lemcompatge ] , this implies that , for @xmath8 large enough , @xmath221 given @xmath5 , define now the set @xmath222\ ; : \ ; v^{1/2+\eta}\,\le\ , \left|\sum_{k = v+1}^{2v } { \textrm{e}}\left({\bm{m}}\bm{\cdot}\left(k{\bm{q}}-\alpha_k \theta { \mathds{1}_\textrm{d}}\right)\right ) \right|\right\}.\ ] ] it then follows from this definition that @xmath223 whence @xmath224 note that @xmath225 defining @xmath226 one thus obtains @xmath227 if the integer @xmath24 is chosen such that @xmath228 , then this inequality implies that the series @xmath229 converges , hence , from the borel
cantelli lemma , @xmath230 this means that for almost all @xmath231 $ ] , there exists a constant @xmath232 depending at most on @xmath233 and @xmath37 such that for all @xmath209 , for all @xmath200 and all @xmath157 , @xmath234 this completes the proof of theorem [ mainthm ] .
the construction of the forest @xmath1 rests on the choice of the sequence @xmath122 which in theorem [ mainthm ] depends on a non explicit parameter @xmath198 . in view of the applications of visibility problems in computer science ( see , e.g. , @xcite and the references therein ) , it is desirable to exhibit an explicit sequence with a good estimate for the visibility function of the corresponding forest
. this is the task undertaken in this section , where the result established in theorem [ thmdernier ] below is as good as theorem [ mainthmintro ] under the so called exponent pair conjecture but is weaker unconditionally . in order to estimate the exponential sum appearing in ( see also ) , given @xmath235 and @xmath157 , set @xmath236 a pair @xmath237 ^ 2 $ ] will be said to be an _
admissible pair of exponents _ if the inequality @xmath238 holds uniformly in @xmath239 .
clearly , @xmath240 is an admissible pair of exponents . as in section [ sec3 ]
, we are looking for a value of @xmath209 depending only on the sequence @xmath122 such that the quantity @xmath243 defined in is strictly positive . from inequality
, this happens as soon as @xmath244 under assumption , this happens whenever @xmath245 in order to find `` good '' exponent pairs , one must first define the sequence @xmath122 . inspired by van der corput s method of exponential sums which will be outlined below , it is natural to ask that this sequence should be the leading term in the argument of the exponential appearing in .
however , one should then expect that the faster the sequence @xmath122 grows , the less accurate the estimate of with this method .
hence , it is natural to set for instance @xmath251 and @xmath252 .
let @xmath256 for all @xmath257 .
thus , we are reduced to estimate the exponential sum @xmath258 . in view of the type of estimate we are looking for ( cf . ) , we may without loss of generality work with the modified exponential sum @xmath259 ( @xmath209 ) , which will slightly simplify the reasoning below .
we first apply to this sum a classical inversion step which is a consequence of the poisson formula .
( ( * ? ? ?
* lemma 3.6))[invstep ] assume that @xmath51 is a four times continuously differentiable function on an interval @xmath260\subset [ v , \ , 2v]$ ] such that @xmath261 .
set @xmath262 and @xmath263 and assume furthermore that there exists a constant @xmath264 such that the following estimates hold on @xmath260 $ ] : @xmath265 for any integer @xmath266 $ ] , define the real number @xmath267 $ ] by @xmath268 . from , it should be clear that for any @xmath282 , @xmath283 and @xmath210 , @xmath284 uniformly in @xmath285 $ ] , where the implicit constant is absolute .
one may therefore apply the theory of exponent pairs in order to estimate the exponential sum appearing on the right hand side of ( see , e.g. , @xcite for a justification of this claim ) . for a detailed exposition on this topic ,
the reader is referred to @xcite .
we recall hereafter the facts which will be useful to our purpose .
the inversion step ( i.e. lemma [ invstep ] ) has reversed the role played by the integers @xmath286 and @xmath8 in the initial sum : in , the integer @xmath286 now determines the range of the sum and @xmath8 is considered as a parameter . as a consequence , an admissible exponent pair for the latter sum takes the form of a pair @xmath287 ^ 2 $ ] satisfying the inequality @xmath288 thus , if @xmath289 is an admissible exponent pair in with @xmath290 , in view of and , @xmath291 is an admissible pair for ( in , the second term of the sum on the right hand side is bounded above by a constant multiple of @xmath292 which dominates the term @xmath293 because @xmath294 and @xmath295 ) .
let @xmath5 . according to the exponent pair conjecture
, @xmath296 is an admissible exponent pair in , in which case @xmath297 is also an admissible exponent pair in . from lemma [ lemexppairvisibilite ] , this implies that the visibility function @xmath3 in the forest constructed from the sequence defined in is such that @xmath298 this is essentially the result stated in theorem [ mainthmintro ] . in order to obtain an unconditional result ,
one may successively repeat , following van der corput s idea , a so called @xmath299process and a so
called @xmath300process when treating the exponential sum in .
the @xmath299process consists of an application of the weyl
van der corput lemma ( cf .
* lemma 2.5 ) ) and transforms an exponent pair @xmath301 into a pair @xmath302 .
the @xmath300process is essentially the inversion step presented in lemma [ invstep ] above and transforms an exponent pair @xmath301 into a pair @xmath303 .
as the latter step is involutive ( i.e. @xmath304 gives back the initial sum ) , any application of van der corput s method takes the form @xmath305 where @xmath306 means that the @xmath299process is applied @xmath67 times .
here , one must start with any given pair of exponents on the right hand side of ( e.g. , the trivial one @xmath307 ) to obtain a new pair of exponents after the successive applications of the @xmath299 and @xmath300 processes as indicated in .
the choice of the parameters @xmath308 represents a difficult problem of optimisation which was rigorously formalised by van der corput in the twenties and then simplified by philips ( 1933 ) .
more recently , graham and kolesnik ( * ? ? ?
* chap . 5 ) proposed an algorithm to this end . in our case , in view of lemma [ lemexppairvisibilite ] and of
, the goal is to find optimal exponent pairs @xmath289 so as to minimise the quantity @xmath309 the algorithm provided in ( * ? ? ?
5 ) yields that the pair of exponents @xmath310 resulting from the operations @xmath311 is an admissible pair for which approaches the infimum of . as the algorithm is infinite and as further iterations do not seem to provide substantial improvement ( the value of decreases by less than 0.001 when @xmath32 for the next 20 iterations )
, we prefer to state below the unconditional result obtained with the pair .
the interested reader may try to improve on this bound by delving further into the algorithm detailed in ( * ? ? ?
* chap.5 ) .
thus , from lemma [ lemexppairvisibilite ] , the visibility function @xmath3 in the forest under consideration satisfies the estimate @xmath312 for any @xmath5 .
note that in the plane ( i.e. when @xmath32 ) , this gives a visibility function such that @xmath313 instead of @xmath314 ( @xmath5 ) under the exponent pair conjecture .
this is still much better than the bound @xmath315 obtained by y. peres in @xcite .
dumir and r.j .
hans - gill . in _
number theory and discrete geometry .
proceedings of the international conference in honour of professor r. p. bambah , chandigarh , india , november 30december 3 , 2005 _
, p.119127 , mysore : ramanujan mathematical society , 2008 . | we address a visibility problem posed by solomon & weiss @xcite .
more precisely , in any dimension @xmath0 , we construct a forest @xmath1 with finite density satisfying the following condition : if @xmath2 denotes the radius common to all the trees in @xmath1 , then the visibility @xmath3 therein satisfies the estimate @xmath4 for any @xmath5 , no matter where we stand and what direction we look in .
the proof involves fourier analysis and sharp estimates of exponential sums . |
the production and weak decay properties of strangeness - rich systems is of fundamental importance for our understanding of relativistic heavy - ion collisions and certain astrophysical phenomena , such as neutron stars .
the simplest systems with strangeness , hypernuclei with one or two bound @xmath1 s , have been used to study both the strong and the weak baryon - baryon ( bb ) interaction in the su(3 ) sector . until now , hypernuclear weak decay represents the only source of information on the @xmath13 four fermion interaction where , in contrast to the weak @xmath14 nn case , both the weak parity - conserving ( pc ) and parity - violating ( pv ) amplitudes can be studied . in the absence of exact solutions to low - energy qcd ,
effective field - theory techniques based on chiral expansions have been fairly successful in the description of hadronic observables in the ( non - strange ) su(2 ) sector .
the stability of the chiral expansion is less clear for the su(3 ) sector , due to the significant amount of su(3 ) symmetry breaking . a well - known failure of su(3 ) chiral perturbation theory has been the prediction@xcite of the four pc @xmath15-wave amplitudes in the weak nonleptonic decays of octet baryons , @xmath16 , with @xmath17 , @xmath18 or @xmath19 . since large cancellations among tree - level amplitudes are held responsible for the problem with the weak @xmath15-wave octet amplitudes , it is imperative to assess whether this situation is universal within su(3 ) @xmath7pt or limited to a few exceptional cases .
however , other weak octet baryon - baryon - meson ( bbm ) vertices can only be accessed through reactions that allow for the virtual exchange of mesons , such as the reactions @xmath2 and @xmath20 . the process @xmath2 has been extensively studied in an approach where the long - range part of this interaction is based on the exchange of the su(3 ) pseudoscalar meson octet ( @xmath21 , @xmath22 , and @xmath23 ) .
the pseudoscalar baryon - baryon - meson vertices are considered fixed by experiment in the case of the pion , and by su(3 ) chiral algebra for the @xmath22 and the @xmath23 . since the large momentum transfer in the reaction ( typically 400 mev / c ) leads to a mechanism where
short - range effects must be included , they have been modelled either through the exchange of the vector meson octet@xcite ( @xmath24,@xmath25 and @xmath26 ) or quark exchange@xcite .
the vector baryon - baryon - meson vertices are constrained by much weaker su(6 ) considerations .
the @xmath2 process is then embedded in nuclear many - body matrix elements using either correlated faddeev amplitudes in the case of few - body systems , hypernuclear shell model wave functions or nuclear matter solutions within the local density approximation , depending on the mass number of the hypernucleus under investigation . while this description of hypernuclear weak decay is not as rigorous as effective field theory would require , it nevertheless has been reasonably successful in describing the available experimental data . since the late 1960 s ,
the production and decay of single-@xmath1 hypernuclei has been studied experimentally in great detail , but only very few events involving doubly - strange objects have been reported @xcite .
double @xmath1 hypernuclei are produced via the @xmath27 reaction at kek ( japan ) and bnl ( usa ) , where a @xmath5 hypernuclear fragment can be formed by @xmath28 capture on a nucleus .
the finuda experiment at da@xmath29ne ( frascati , italy ) can produce double-@xmath1 hypernuclei by stopping slow @xmath30 ( coming from the @xmath29 decay ) into thin targets to obtain data with higher energy resolution .
studying the weak decay of those objects opens the door to a number of new exotic @xmath1-induced decay modes : @xmath31 and @xmath32 .
both of these decays would involve hyperons in the final state and should be distinguishable from the ordinary @xmath33 mode .
the @xmath31 channel is especially intriguing since the dominant pion exchange is forbidden , thus forcing this reaction to occur mostly through kaon exchange .
one would therefore gain access to the weak @xmath5k and @xmath6k vertices . in this paper
, we extend previous weak decay calculations of single-@xmath1 hypernuclei into the @xmath34 sector , thus exploring the power of the @xmath35 process to shed light on the novel weak vertices . in order to take into account the effects of the strong interaction between the baryons ,
correlated wave functions are obtained from a @xmath36-matrix calculation for the initial @xmath1n and @xmath5 states , while a @xmath37-matrix equation is solved for the final nn and yn states using the nijmegen interaction models@xcite , in particular the nsc97f one .
in the weak nonmesonic decay of double-@xmath1 hypernuclei , new hyperon - induced mechanisms , the @xmath3 and the @xmath38 transitions ( denoted as @xmath39 throughout the text ) become possible in addition to the dominant @xmath40 decay mode . assuming the initial hypernucleus to be at rest , the nmd rate is given by : @xmath41 where the quantities @xmath42 , @xmath43 , @xmath44 and @xmath45 are the mass of the hypernucleus , the energy of the residual -particle system , and the total asymptotic energies of the emitted baryons , respectively . the integration variables @xmath46 and @xmath47 stand for the momenta of the two baryons in the final state . in the expression above ,
the momentum - conserving delta function has been used to integrate over the momentum of the residual nucleus .
the sum , together with the factor @xmath48 , indicates an average over the initial hypernucleus spin projections , @xmath49 , and a sum over all quantum numbers of the residual -particle system , @xmath50 , as well as the spin and isospin projections of the emitted particles , @xmath51 and @xmath52 .
the total nonmesonic decay rate can be written as : @xmath53 where the rate corresponding to @xmath40 transitions @xmath54 is divided into a neutron - induced rate , @xmath55 and a proton - induced one , @xmath56 , while the @xmath1-induced transitions @xmath57 give rise to @xmath1n , @xmath58n and @xmath59p final states , hence : @xmath60 note that while the individual rates have been written above as exclusive observables , in principle , charge - exchange final state interactions ( fsi ) with the residual nucleus obscurs a clean experimental discrimination of these channels on the basis of the charge of the emitted particles .
monte carlo intranuclear cascade models are necessary @xcite to extract the partial weak decay rates of hypernuclei from experiment @xcite . the impossibility of measuring the partial rates directly from the charge of the emitted particles was also shown in the case of the weak decay of the hypertriton @xcite , a system where the strong interaction effects can be treated exactly using fadeev equations with a realistic baryon - baryon potential . in order to draw conclusions regarding the weak dynamics , one has to write the hypernuclear transition amplitude , @xmath61 , in terms of the elementary two - body transitions , @xmath62 , and nuclear structure details have to be treated with as few approximations and ambiguities as possible .
we work in a shell - model framework , hence the @xmath1 hyperons and the nucleons are described by single - particle orbitals . in addition , we assume a weak coupling scheme by virtue of which the @xmath1 hyperons couple only to the ground state of the nuclear core . therefore , in the case of the @xmath12he hypernucleus studied here , with quantum numbers , , the state will be given by @xmath63 where antisymmetry forces the two @xmath1 hyperons to be in a @xmath64 state , since they are assumed to be in the lowest s - shell before the decay occurs .
this is so because , in general , hypernuclei with @xmath1 s in excited orbitals will rapidly decay into the ground state through electromagnetic or strong de - excitation processes , which are orders of magnitude faster than those mediated by the weak interaction .
the single - particle orbitals for nucleons and @xmath1 s are taken as harmonic oscillator states with parameters @xmath65 fm and @xmath66 fm , respectively .
the nucleon parameter is chosen to account for the @xmath67he form factor .
the parameter for the @xmath1 wave function reproduces the hartree - fock ( hf ) probability of finding the two @xmath1 particles at relative distance @xmath68 , obtained in ref .
@xcite by adjusting their model parameters to the binding energies of the three observed double @xmath1-hypernuclei , @xmath12he , @xmath69b and @xmath70b @xcite . to obtain the rate corresponding to the @xmath40 transition , with the initial @xmath71 being @xmath72 or @xmath73
, we have to write the non - strange nuclear core as one nucleon coupled to a 3-particle system ( with quantum numbers @xmath74 and @xmath75 ) and decouple one of the two @xmath1 s in the cluster .
therefore , the initial @xmath1n pair will convert into a final nn pair with quantum numbers @xmath76 , leaving a residual 4-particle system with quantum numbers @xmath77 . performing all the necessary decoupling and recoupling operations ,
we finally arrive at : @xmath78 where @xmath79 , @xmath80 , @xmath81 , @xmath82 , @xmath83 , @xmath84 , @xmath85 and @xmath86 .
we have assumed the @xmath1 at the weak vertex to behave as a @xmath87 isospurion in order to naturally incorporate the experimentally observed @xmath88 rule
. the momentum states @xmath89 and @xmath90 have been transformed to total ( @xmath91 ) and relative ( @xmath92 ) momenta and the final two - nucleon state , @xmath93 , must be properly antisymmetrized , which means that the factor @xmath94 will appear when a decomposition in partial waves is performed for the outgoing plane wave . for the @xmath57 transition we do nt need to decouple one of the hyperons from the cluster , neither a nucleon from the core . the residual 4-particle system , which coincides with the @xmath95he core in eq .
( [ eq : decomposition ] ) , contains no strangeness , while the final two - particle state contains one hyperon that can be either a @xmath1 ( @xmath96 ) , a @xmath59 ( @xmath97 ) or a @xmath58 ( @xmath98 ) . the @xmath57 hypernuclear amplitude is given by : @xmath99 where @xmath100 , @xmath101 and @xmath102 .
in a meson - exchange description , the @xmath2 and @xmath57 transitions are assumed to proceed via the exchange of virtual mesons belonging to the ground - state pseudoscalar and vector meson octets .
the amplitude , which is schematically represented by the feynman diagram depicted in fig .
[ [ fig : feydia ] ] , reads @xmath103 where @xmath104 is the free baryon field of positive energy , @xmath105 the dirac operator characteristic of the baryon - baryon - meson ( bbm ) vertex and @xmath106 the meson propagator . in table
[ tab : vert ] we show the strong and weak hamiltonians for pseudoscalar ( ps ) and vector ( v ) mesons .
the constants @xmath107 , @xmath108 , @xmath109 , @xmath110 and @xmath111 correspond to the weak couplings , while @xmath112 ( @xmath113 ) represents the strong ( vector , tensor ) one .
the nonrelativistic reduction of this amplitude gives us the potential in momentum space @xcite , which for pseudoscalar mesons reads : @xmath114 where @xmath115 is the fermi weak constant , @xmath116 is the momentum carried by the meson directed towards the strong vertex , @xmath117 the meson mass and @xmath118 ( @xmath119 ) is the average of the baryon masses at the strong ( weak ) vertex . for vector mesons the potential reads : @xmath120 in eqs .
( [ eq : pion ] ) , ( [ eq : rhopot ] ) the operators @xmath121 , @xmath122 , @xmath123 , @xmath124 and @xmath125 contain , apart from the weak coupling constants , the specific isospin dependence of the potential , which is @xmath126 for the isovector @xmath21 and @xmath24 mesons , @xmath127 for the isoscalar @xmath23 and @xmath25 mesons and a combination of both operators for the isodoublet @xmath22 and @xmath26 .
@xcite introduces a compact way to write the transition potential in @xmath128-space : @xmath129 where the index @xmath130 runs over the different mesons exchanged ( @xmath131 represents @xmath132,k,@xmath133,k@xmath134 ) and @xmath109 over the different spin operators : central spin independent ( c ) , central spin dependent ( ss ) , tensor ( t ) and parity violating ( pv ) @xmath135 { \hat r } & \mbox{pv ( vector mesons ) } \ , { \rm .}\\ \end{array } \right.\ ] ] in order to account for the finite size and structure of the particles involved in the process , form factors ( ff ) are included . in our previous calculations of the decay of single-@xmath1 hypernuclei @xcite we used a monopole ff , @xmath136 , at both the weak and strong vertices , with different cut - offs depending on the exchanged meson , @xmath137 .
those cut - offs were taken from the jlich yn interaction model @xcite .
the latest version of the nijmegen potentials@xcite , which is used here , also gives meson - dependent cut - offs but uses an exponential ff at each vertex of the type @xmath138 .
therefore , the results presented here are also obtained with the exponential - type ff , although , for technical reasons , it is matched to a function of the type @xmath139 at @xmath140 mev / c , the most relevant momentum transfer in the weak decay transition . by construction , this type of monopole ff is equivalent to the exponential one at @xmath141 and we have verified that , up to a momentum transfer of 600 mev / c above which the transition amplitude is negligible , the differences between both functions are less than 2% .
the modified _ cut - off_s , @xmath142 , to be used in the monopole - type ff are listed in table [ tab : ff ] . to incorporate the effects of the strong baryon - baryon ( bb ) interaction
we solve a t - matrix scattering equation in momentum space for the outgoing two - particle system ( nn or yn ) . for the initial two - particle system ( @xmath1n or @xmath5 )
we take into account medium effects , thus intermediate states only propagate into states allowed by the pauli operator ( @xmath36-matrix ) .
for the initial interacting @xmath1n pair we previously found @xcite that multiplying the independent two - particle wave function by a spin - independent correlation function of the type : @xmath143 with @xmath144 fm , @xmath145 @xmath146 , @xmath147 fm , @xmath148 , produced a correlated wave function which averaged the ones obtained from a microscopic finite - nucleus @xmath36-matrix calculation@xcite using the soft - core and hard - core nijmegen models @xcite . in the case of the interacting @xmath149 system , we again assume a correlation function of the type of eq .
( [ eq : fsief ] ) .
the parameters , @xmath150 fm , @xmath151 @xmath146 , @xmath152 fm and @xmath153 , are determined to reproduce the ratio between the correlated and uncorrelated @xmath64 @xmath149 wave functions in nuclear matter at saturation density , @xmath154 , taking @xmath155 mev / c as a representative momentum of the two @xmath1 s in a finite nucleus .
the nuclear - matter correlated wave function , @xmath156 , has been obtained from a coupled - channel g - matrix calculation using the nsc97f model of the new nijmegen potentials . in fig .
[ fig : prob ] , we show the probability per unit length of finding two @xmath1 particles at a distance @xmath68 , given by latexmath:[$ 4 \pi r^2 the uncorrelated two-@xmath1 wave function , which in this work is given by the 1s harmonic oscillator wave function with the parameter @xmath158 . the dashed line incorporates the calculated correlation function , @xmath154 , a result that is well parametrized by a phenomenological correlation function of the form of eq .
( [ eq : fsief ] ) , as shown by the dotted line . as mentioned before
, the strong bbm couplings are taken from the nsc97f@xcite potential . in the weak sector
only the couplings corresponding to decays involving pions are experimentally known .
the weak couplings for heavy mesons are obtained following refs.@xcite .
the pv amplitudes for the nonleptonic decays @xmath159 involving pseudoscalar mesons are derived using the soft - meson reduction theorem and su(3 ) symmetry .
this allows us to relate the physical amplitudes of the nonleptonic hyperon decays into a pion plus a nucleon , @xmath160 , with the unphysical amplitudes involving other members of the meson and baryon octets . on the other hand , su(6@xmath161 permits relating the amplitudes involving pseudoscalar mesons with those for vector mesons .
the weak pc coupling constants are obtained using a pole model , where the pole terms due to the @xmath162 ground state ( singular in the su(3 ) soft meson limit ) are assumed to be the dominant contribution .
no meson pole terms are included .
the values of these couplings for the weak vertices are listed in table [ tab : wcc ] .
the results for the decay rates of @xmath12he into the different channels are summarized in tables [ tab : resfree.nsc97f],[tab : res.nsc97f ] and [ tab : diffsi.nsc97f ] , where the rate is presented in units of the free @xmath1 decay , @xmath163 . apart from giving the various partial rates ,
the ratio between the neutron - induced and proton - induced decay rates , @xmath164 , is also quoted .
only @xmath165 transitions have been considered .
we begin by discussing the results of table [ tab : resfree.nsc97f ] , which list calculations absent of either strong correlations ( initial and final ) and ff .
although the results are clearly unrealistic , they serve to illustrate , in comparison with the following table , the effect of the strong interaction on the weak decay process .
the @xmath40 transitions receive contributions mainly from the @xmath21 , @xmath24 , @xmath22 and @xmath26 mesons ; however , once short - range strong correlations are included ( see table [ tab : res.nsc97f ] ) , the role of the heavier mesons is strongly reduced . the results in table [ tab : resfree.nsc97f ]
illustrate that isospin symmetry forbids isovector meson contributions , @xmath166 , for the uncorrelated @xmath167 decay rate and isoscalar meson contributions , @xmath168 , for the uncorrelated @xmath38 rates . finally , we note that the nucleon - induced rate is about 20 times larger than the total hyperon - induced contribution . when ff and strong correlations in the initial and final states are included , the picture changes considerably , as can be seen from the results shown in table [ tab : res.nsc97f ] .
most of the @xmath169 rate comes from the exchange of the @xmath21 and @xmath22 mesons .
note that the small @xmath164 ratio for the one - pion exchange mechanism substantially increases when the @xmath22-meson is incorporated due to a destructive ( constructive ) interference in the @xmath170-induced ( @xmath171-induced ) rate .
this aspect is discussed recently in ref .
@xcite , which updates the meson - exchange model of ref .
@xcite and finds agreements with results obtained by groups combining the @xmath21 and @xmath22 exchange with either a direct quark mechanism@xcite or @xmath172 exchange @xcite .
since there are now two @xmath1 particles contributing to the @xmath40 decay , one would expect the result to be roughly twice the rate in @xmath173he , which for this potential turns out to be @xmath174 , as shown in ref .
we obtain instead a significantly larger value of @xmath175 .
this finding can be traced to the @xmath1 which is more strongly bound in @xmath12he than it is in @xmath173he by about 80% ( i.e. , the @xmath1 single - particle separation energy changes from 3.12 mev to around 5 mev ) .
this increased binding is reflected in the @xmath1 single - particle wave function , with a harmonic oscillator parameter @xmath66 fm instead of the value @xmath176 fm used in the decay of single-@xmath1 hypernuclei @xcite . as mentioned in sec .
ii , the parameter @xmath66 fm simulates the uncorrelated @xmath5 probability of ref .
@xcite , which reproduces the binding energy of the three observed double-@xmath1 hypernuclei .
there is also some influence from the slightly different kinematics involved in the @xmath169 decay of @xmath177he with respect to that in @xmath178he . from the results presented in table [ tab : res.nsc97f ] , we observe that the @xmath167 rate receives a tiny contribution from the isovector @xmath21 and @xmath24 mesons .
this is due to the combined transition @xmath179 , which is possible through the coupling between @xmath1n and @xmath18n states induced by the strong interaction .
analogously , the strong @xmath180n coupling also induces very small contributions from the isoscalar mesons to the @xmath18n rate , @xmath181 .
we find the rate @xmath182 to be dominated by @xmath22-exchange .
we note that the @xmath183 contribution is very similar to the value for @xmath22-exchange alone ; adding the @xmath23 meson decreases the rate by 35% . however , there is strong evidence that we are overestimating the role of the @xmath23 meson by using strong @xmath23nn and @xmath184 couplings that rely on su(3 ) symmetry .
all indications from a careful analysis of reactions like @xmath23 photoproduction@xcite point towards a @xmath23nn coupling much smaller than the one provided by su(3 ) symmetry . in this case , the weak decay of double-@xmath1 hypernuclei going to final @xmath1n states would be almost solely determined by @xmath22 exchange and the measurement of this decay channel would provide valuable information on the weak @xmath149k couplings . in particular , since this partial rate is dominated by pc transitions , as can be inferred from the values of the @xmath185 weak couplings listed in table [ tab : wcc ] , one would gain access to the pc @xmath185 amplitude .
the rate to final @xmath18n states , @xmath181 , on the other hand is very small and , as expected , dominated by the @xmath21-exchange mechanism .
the ratios between the rates that include correlations are : @xmath186 correlations have a stronger influence on the yn channels , which are a factor @xmath187 smaller compared to the nn rate , even though the @xmath18n transition allows a final state with a lower relative momentum and , therefore , is less sensitive to the strongly repulsive core of the yn interaction . however , the nn rate is dominated by tensor transitions that are absent in the yn channels because the interacting @xmath149 pair is in a @xmath64 state . with the tensor strength distributed towards larger distances compared to those for central and parity - violating transitions ,
the nn transition becomes comparatively less affected by strong correlations than the yn ones .
note that correlations tend to increase the @xmath1n rate while decreasing the @xmath18n rate .
the reason is that the uncorrelated @xmath18n rate is dominated by @xmath21 and @xmath26 exchanges , the @xmath26 contribution being about twice that of the @xmath21 meson , whereas the @xmath1n rate comes mainly from strange meson exchange , with the @xmath22 contribution twice as large as that of the @xmath26 .
therefore , the supression of short - range physics induced by correlations affects the @xmath18n rate more than the @xmath1n one .
in addition , it is known@xcite that the @xmath18n interaction for the new nijmegen potentials is quite repulsive in the @xmath188 channel , affecting the @xmath18n wave functions significantly , which are pushed out from the weak potential interaction range .
the results for various approaches to the treatment of fsi are compared in table [ tab : diffsi.nsc97f ] . those labelled with
fsi eff . " have been obtained by multiplying the free nn , @xmath1n and @xmath18n final states by the momentum and channel independent function , with @xmath189 @xmath190 . as recently discussed in ref .
@xcite , treating fsi in this simplified phenomenological fashion overestimates the rate @xmath191 calculated rigorously with a @xmath37-matrix by about a factor of two .
the rate @xmath182 obtained with the phenomenological fsi treatment is almost one order of magnitude smaller compared with the result obtained within a @xmath37-matrix approach , while on the other hand , the rate @xmath181 is overpredicted by almost an order of magnitude . as in ref .
@xcite , we again find that in order to be able to extract information on the weak decay amplitudes a careful treatment of fsi is essential in the study of the weak decay of double-@xmath1 hypernuclei .
we finally compare our results with the ones presented recently by itonaga et al .
@xcite , where the weak decay mechanism contains @xmath21 , @xmath24 and correlated two - pion exchange with scalar ( @xmath192 ) and vector ( @xmath24 ) quantum numbers .
the @xmath1 wave function in @xmath12he was taken the same as in their @xmath173he calculation @xcite , thus neglecting the possible effect of a stronger binding of the @xmath1 in a double-@xmath1 hypernucleus .
hence , the nucleon - induced rate obtained in @xcite is @xmath193 , equal to twice their decay rate in @xmath173he .
this rate is about 20% smaller than the one obtained here .
the rates for @xmath1n and @xmath194 final states , @xmath195 and @xmath196 , are about a factor two smaller than the ones obtained in this work .
we note that the @xmath5 correlation function used in ref .
@xcite is in fact the @xmath64 @xmath1n one used in their study of the decay of @xmath173he .
we also note that the @xmath22-exchange mechanism , dominant in the @xmath197 transition amplitude , is absent in the work of ref .
@xcite . in ref .
@xcite , we explored su(3 ) symmetry breaking effects in the decay of single-@xmath1 hypernuclei by using weak nnk couplings derived in the framework of heavy baryon chiral perturbation theory@xcite , where one - loop su(3 ) corrections to leading order were evaluated . here
we explore similar effects not only for the nnk but also for the @xmath5k and @xmath6k couplings , which were calculated in refs .
@xcite . in we list the values for the weak pv ( @xmath198-wave ) and pc ( @xmath15-wave ) couplings involving the kaon .
compared to the ones quoted in ref .
@xcite , the numbers in table [ chpt ] differ since they were obtained using a common mass splitting of @xmath199 mev between the octet and decuplet baryons @xcite . comparing the values listed in table [ chpt ] with those in table [ tab : wcc ]
we observe that the one - loop corrected nnk @xmath15-wave constants are approximately a factor two smaller than the tree level ones , while the @xmath198-wave are very similar , except for the @xmath200 coupling , which , again , is half the tree - level value .
this explains the reduced @xmath22-exchange contribution to the nucleon - induced rates with respect to the tree - level result .
note that the magnitude of the ( constructive ) interference between @xmath21 and @xmath22 in the @xmath201 channel has not changed much , while it has become less destructive in the @xmath202 channel , thereby producing a smaller @xmath164 ratio .
once all the mesons are included , the nn rate obtained with the one - loop corrected nnk couplings is a factor 1.4 larger than that obtained with the tree - level values , while the ratio @xmath164 is smaller by a factor of two .
although the one - loop corrected @xmath5k couplings of table [ chpt ] are not much different from the corresponding tree - level values , they lead to a reduction of the @xmath22-meson exchange contribution to the @xmath1n rate by about a factor two .
hence , the contribution of the @xmath23 meson and the vector mesons becomes more relevant , making the extraction of the @xmath149k coupling from this @xmath203 decay mode more difficult .
the decay rate into final @xmath18n states is the only one that increases when the one - loop corrected @xmath6k couplings are used , despite the magnitude of the @xmath15-wave amplitudes being ten times smaller than the corresponding tree - level values , whereas the @xmath198-wave amplitudes are only moderately larger .
in fact , table [ tab : res.nsc97f ] reveals that the @xmath22-exchange contribution to the @xmath18n rate is two times smaller than that obtained with the tree - level @xmath6k couplings , as expected from the reduced magnitude of the couplings .
however , due to the opposite sign of the loop - corrected @xmath15-wave coupling with respect to the corresponding tree level value , the interference between the @xmath21 and @xmath22 contributions is now constructive instead of destructive , which explains the final increased loop - corrected rate . finally , we have investigated how our results would change if we allowed the one - loop corrected bbk couplings to vary within their estimated error bands .
the corresponding range of variation on the rates is quoted in table [ tab : summarize ] , where the results obtained with the tree - level bbk couplings have also been included to facilitate the comparison between both calculations .
the results summarized in table [ tab : summarize ] show clearly that the inclusion of weak bbk couplings obtained with one - loop corrections to the leading order in @xmath7pt produces a @xmath40 rate 40 % larger than the tree - level value , a ratio @xmath164 twice smaller , a @xmath31 rate roughly twice smaller , and a @xmath38 rate almost twice as large .
in this study we investigated the nonmesonic weak decay modes of double-@xmath1 hypernuclei within a one - meson - exchange framework .
our results are given for the specific double-@xmath1 hypernucleus @xmath12he , but can easily be extended to heavier systems .
the standard nucleon - induced nonmesonic @xmath40 rate is found to be more than twice as large as in @xmath178he due to the increased binding of the second @xmath1 hyperon .
two new hyperon - induced modes become possible , @xmath167 and @xmath204 , the latter coming in two charge states .
we find the total hyperon - induced rate to be as large as 4% of the total nonmesonic rate .
this new rate is dominated by the @xmath167 mode .
in fact , this transition turns out to be the more interesting one ( rather than the @xmath204 ) since it allows direct access to exotic vertices like @xmath5k , unencumbered by the usually dominant pion exchange .
indeed , one - loop log corrected @xmath7pt results modify the @xmath205 by 50% while changing the @xmath40 only at the 15% level , demonstrating the power of this weak mechanism to test @xmath7pt in the weak su(3 ) sector . with a free @xmath1 in the final state
this new mode should be distinguishable from the usual nucleon - induced decay channels .
given the potential benefits to hadronic physics , the experimental program of investigating the production and decay of double-@xmath1 hypernuclei should be intensified .
the authors would like to thank christoph hanhart , barry r. holstein , roxanne p. springer and isaac vidaa for their respective inputs and comments to different aspects of the present paper , regarding both the strong and weak baryon - baryon interaction .
one of the authors , a.p . , would like to thank the institute for nuclear theory and the university of washington ( seattle , usa ) where an important part of this work was developped .
this work has been partially supported by the u.s .
dept . of energy under grant no .
de - fg03 - 00-er41132 , by the dgicyt ( spain ) under contract pb98 - 1247 and by the generalitat de catalunya project sgr2000 - 24 .
su(3 ) allows us to obtain the weak pv couplings involving two baryons and one pseudoscalar ( @xmath213 ) meson . in order to obtain the corresponding constants for vector mesons we use su(6)@xmath214
here , we quote the results for all the couplings that appear in the @xmath40 and @xmath215 channels .
j. golak , k. miyagawa , h. kamada , h. witala , w. glckle , a. parreo , a. ramos and c. bennhold , phys .
c * 55 * , 2196 ( 1997 ) ; j. golak , h. kamada , k. miyagawa , h. witala , and w. glckle , phys . rev . lett . * 83 * , 3142 ( 1999 ) .
.possible @xmath229 channels present in the weak decay of @xmath230he .
the notation used is the following : @xmath231 stands for the relative orbital momentum for the initial @xmath5 or @xmath1n pair , @xmath232 for the relative angular momentum of the two - particle state right after the weak transition occurs , and @xmath233 for the relative angular momentum of the final two - particle state after including the effects of the strong interaction ( fsi ) .
the symbols c , t and pv stand for central , tensor and parity violating .
[ cols="<,^,^,^,^,^,^,^,^,^ " , ] | the strangeness - changing ( @xmath0 ) weak baryon - baryon interaction is studied through the nonmesonic weak decay of double-@xmath1 hypernuclei . besides the usual nucleon - induced decay @xmath2
we discuss novel hyperon - induced decay modes @xmath3 and @xmath4 .
these reactions provide unique access to the exotic @xmath5k and @xmath6k vertices which place new constraints on chiral pertubation theory ( @xmath7pt ) in the weak su(3 ) sector . within a meson - exchange framework
, we use the pseudoscalar @xmath8 octet for the long - range part while parametrizing the short - range part through the vector mesons @xmath9 .
realistic baryon - baryon forces for the @xmath10 and @xmath11 sectors account for the strong interaction in the initial and final states . for @xmath12he
the new hyperon - induced decay modes account for up to 4% of the total nonmesonic decay rate .
predictions are made for all possible nonmesonic decay modes . |
the problem of describing all finite - dimensional lie algebras of vector fields is a starting point for the symmetry analysis of ordinary differential equations , because , having solved this problem , one finds all possible algebras of contact symmetries for ordinary differential equations . over the complex numbers this classification
was done at the end of the last century by sophus lie @xcite .
he showed that , with three exceptions , all lie algebras of contact vector fields , viewed up to equivalence , are lifts of lie algebras of vector fields on the plane .
the largest algebra of the three exceptions ( so - called _ irreducible algebras of contact vector fields _ ) is the algebra of contact symmetries of the equation @xmath0 and is isomorphic to @xmath1 , while the other two are its subalgebras of dimension 6 and 7 . in this paper
we show that the problem of describing algebras of vector fields can be formulated in a natural way in terms of filtered and graded lie algebras .
this allows not only to give a new up - to - date proof of sophus lie s classification , which is as yet missing in the literature , but also to solve this problem over the field of real numbers .
it turns out that in the real case there are 8 irreducible contact lie algebras of vector fields on the plane , and one of them involves an arbitrary parameter .
lie algebras of vector fields on the plane were also classified ( both in real and complex case ) by sophus lie @xcite , so that the description of irreducible lie algebras of vector fields on the plane which is given in the present paper , basically concludes the description of all finite - dimensional contact lie algebras of vector fields over the field of real numbers .
it should be noted that the problem of finding all irreducible contact lie algebras over the real numbers was also considered by f. engel in @xcite , which is mentioned in sophus lie s three - volume treatise @xcite ( volume 3 , chapter 29 , pages 760761 ) .
p olver , in his recently published book @xcite , cites this problem as unsolved .
let @xmath2 be the set of 1jets of mappings from @xmath3 to @xmath3 , and let @xmath4 denote the natural projection @xmath5 .
we fix a coordinate system @xmath6 on @xmath7 in which the 1jet of the mapping @xmath8 at the point @xmath9 has the coordinates @xmath10 .
the projection @xmath4 has in these coordinates the form @xmath11 .
we can introduce a natural _ contact structure _
indeed , there is a two - dimensional distribution @xmath12 on @xmath7 which is not completely integrable and has the property that all its integral curves whose projection onto the plane is diffeomorphic , are precisely the curves of the form @xmath13 with @xmath14 , @xmath15 . in terms of coordinates ,
this distribution is given by the vector fields @xmath16 and @xmath17 or , alternatively , by the differential 1form @xmath18 .
a ( local ) diffeomorphism @xmath19 of the manifold @xmath7 is said to be _ contact _
if @xmath19 preserves the contact distribution @xmath12 , ie , if @xmath20 for all @xmath21 . a vector field on @xmath7
is called contact if it generates a local one - parameter transformation group that consists solely of contact diffeomorphisms .
it is easy to show that a vector field @xmath22 is contact if and only if @xmath23 for some smooth function @xmath24 .
if @xmath22 is a contact vector field , then the function @xmath25 is called the _ characteristic function _ of @xmath22 .
it completely determines the field @xmath22 , which in this case is denoted by @xmath26 and has the form @xmath27 the mapping @xmath28 establishes an isomorphism between the space of all smooth functions and that of contact vector fields on @xmath7 .
this allows to make the space @xmath29 into a lie algebra by letting @xmath30)$ ] .
if @xmath19 is a ( local ) diffeomorphism of the plane , then there exists a unique local contact transformation @xmath31 such that the following diagram is commutative : @xmath32 the transformation @xmath33 is then called the ( _ first _ ) _ prolongation of the diffeomorphism _ @xmath19 and , in terms of coordinates , has the from @xmath34 similarly , for any vector field @xmath22 on the plane there exists a unique contact vector field @xmath35 on @xmath36 such that @xmath37 .
this vector field @xmath35 is called the ( _ first _ ) _ prolongation of the vector field _
@xmath22 and has the form @xmath38 its characteristic function is @xmath39 .
the mapping @xmath40 is an embedding of the lie algebra of vector field on the plane into the lie algebra of contact vector fields on @xmath36 .
the contact vector fields that lie in the image of this mapping are called _ point _ contact vector fields .
point vector fields @xmath41 are characterized by the following two equivalent properties : 1 .
any point vector field @xmath41 is an infinitesimal symmetry of the _ vertical distribution _
@xmath42 on @xmath36 ( @xmath43 ) ; 2 .
the characteristic function of @xmath41 is linear in @xmath44 .
a lie algebra @xmath45 of contact vector fields is called _ reducible _ if there is a local contact diffeomorphism @xmath19 such that the lie algebra @xmath46 consists only of point vector fields .
otherwise , @xmath45 is said to be _
irreducible_. [ th1 ]
a lie algebra @xmath45 of contact vector fields is irreducible if and only if it preserves no one - dimensional subdistribution of the contact distribution .
every lie algebra that consists of point vector fields preserves the vertical distribution @xmath42 , which is a one - dimensional subdistribution of the contact distribution @xmath12 .
consequently , any reducible lie algebra of contact vector fields also preserves a one - dimensional subdistribution of @xmath12 .
conversely , let @xmath45 be a lie algebra of vector fields that preserves some one - dimensional subdistribution @xmath47 of the contact distribution .
if @xmath48 and @xmath49 are two functionally independent first integrals of @xmath47 , then , as one can easily verify , the local diffeomorphism @xmath50 is contact and transforms the vertical distribution @xmath42 to @xmath47 .
it follows that the lie algebra @xmath51 preserves the vertical distribution and hence consists of point vector fields .
any irreducible lie algebra of contact vector fields is transitive at a point in general position .
let @xmath45 be an irreducible lie algebra of contact vector fields .
for an arbitrary point @xmath52 , we let @xmath53 and define @xmath54 and @xmath55 . then @xmath56 is obviously an open subset in @xmath36 .
the lie algebra @xmath45 is transitive at point in general position if and only if @xmath57 .
assume the contrary .
then the subspaces @xmath58 form a completely integrable distribution @xmath47 in @xmath56 which is invariant under @xmath45 .
consider the following two possibilities : 1@xmath59 : @xmath60 then the intersection @xmath61 is one - dimensional at the points in general position , and this family of subspaces forms a one - dimensional subdistribution of the contact distribution which is invariant under @xmath45 . 2@xmath59 : @xmath62 in this case @xmath47 can be locally embedded into a two - dimensional completely integrable distribution which , as follows from its construction , is also invariant under @xmath45 .
then , arguing as in the previous case , we conclude that the lie algebra @xmath45 preserves a one - dimensional subdistribution of the contact distribution . in this paper
we restrict ourselves to a local description of finite - dimensional lie algebras of contact vector fields at a point in general position .
in particular , from now on we shall assume that all irreducible algebras of contact vector fields are transitive .
let @xmath45 be a transitive lie algebra of contact vector fields on @xmath2 , let @xmath63 be an arbitrary point in @xmath36 , and let @xmath64 be the subalgebra of @xmath45 that consists of all vector fields in @xmath45 vanishing at the point @xmath63 .
it is easy to show that the subalgebra @xmath65 is _ effective _ , ie , contains no nonzero ideals of @xmath45 ( see , for example , ( * ? ? ?
* theorem 10.1 ) ) .
we can identify @xmath66 with @xmath67 in the obvious way .
then @xmath68 is identified with a certain submodule @xmath69 of the @xmath65module @xmath67 . since the distribution @xmath12 is not completely integrable , the subspace @xmath70
will not be closed with respect to the multiplication in @xmath45 .
we define a decreasing chain of subspace in @xmath45 as follows : @xmath71 for all @xmath72 , @xmath73\subset { \mathfrak g}_p\}\text { for all } p\ge0.\ ] ] it is easily shown by induction that @xmath74\subset{\mathfrak g}_{p+q}$ ] for all @xmath75 , so that the family of subspaces @xmath76 defines a filtration of the lie algebra @xmath45 .
a filtered lie algebra @xmath45 is called a _ contact lie algebra _ if a. @xmath71 for all @xmath72 ; b. @xmath77 , @xmath78 , and @xmath79+{\mathfrak g}_{-1}={\mathfrak g}$ ] ; c. @xmath80\subset { \mathfrak g}_p\}$ ] for all @xmath81 ; d. @xmath82 . two contact lie algebras are said to be isomorphic if they are isomorphic as filtered lie algebras .
show that any transitive lie algebra @xmath45 of contact vector fields is a contact lie algebra with respect to the above filtration . the properties
a ) and c ) follow immediately from the way that the filtration in @xmath45 is introduced .
let us prove b ) .
since the contact distribution has codimension 1 , we get @xmath77 . from transitivity of @xmath45 on @xmath7
we get @xmath83 .
next , since @xmath12 is not completely integrable , the subspace @xmath79+{\mathfrak g}_{-1}$ ] is strictly greater than @xmath84 and , hence , is equal to @xmath45 .
finally , let @xmath85 .
then , obviously , @xmath86 is an ideal in @xmath45 contained in @xmath65 .
now since @xmath65 is an effective subalgebra , it follows that @xmath87 .
this proves d ) .
conversely , let @xmath45 be an arbitrary finite - dimensional contact lie algebra
. then the pair @xmath88 determines a unique ( up to local equivalence ) realization of @xmath45 as a transitive lie algebra of vector fields on @xmath89 . and the subspace @xmath84 allows us to define a @xmath45invariant two - dimensional distribution on @xmath89 which is not completely integrable .
therefore , the lie algebra @xmath45 admits a unique ( up to local equivalence ) realization as a transitive lie algebra of contact vector fields .
thus , the local classification of finite - dimensional transitive lie algebras of contact vector fields on @xmath36 is equivalent to the classification ( up to isomorphism ) of the corresponding contact lie algebras .
observe that the latter problem is algebraic and , as we shall see later , can be solved by purely algebraic means .
all @xmath45invariant distributions on @xmath36 are in one - to - one correspondence with the submodules of the @xmath65module @xmath67 .
in particular , the contact distribution corresponds to the submodule @xmath90 . from theorem [ th1 ] now easily follows the next algebraic criterion for the irreducibility of @xmath45 . [ lem1 ]
a transitive lie algebra @xmath45 of contact vector fields is irreducible if and only if the @xmath65module @xmath90 is irreducible .
the major tool in the study of filtered lie algebras is to consider graded lie algebras associated with them .
as we shall see later on , with a few exceptions , irreducible contact lie algebras can be completely restored from their associated graded lie algebras . a @xmath91graded lie algebra @xmath92
is called a _ graded contact lie algebra _ if a. @xmath93 for all @xmath94 ; b. @xmath95 , @xmath96 , and @xmath97={\mathfrak h}_{-2}$ ] ; c. @xmath98=0\}=\{0\}$ ] for all @xmath81 .
if @xmath45 is a contact lie algebra , then it is clear that the associated graded lie algebra @xmath99 satisfies all three conditions in the above definition and is therefore a graded contact lie algebra .
moreover , if @xmath45 is a filtered lie algebra such that the associated graded lie algebra @xmath100 is contact and @xmath101 , then it is easy to show that @xmath45 itself is a contact lie algebra .
the concept of irreducibility for contact lie algebras can be carried over to graded contact lie algebras . from lemma [ lem1 ]
it immediately follows that a contact lie algebra @xmath45 is irreducible if and only if so is the @xmath102module @xmath103 in the corresponding graded contact lie algebra @xmath100 .
the graded contact lie algebras that satisfy this condition will be called _
irreducible_. the classification of all irreducible graded contact lie algebras can be carried out using the methods developed in the works of tanaka @xcite . slightly modifying the terminology of those papers
( see also @xcite ) , we introduce the concept of transitive graded lie algebra , which generalizes the concept of graded contact lie algebra .
a graded lie algebra @xmath104 is said to be _ transitive _ if it satisfies the following conditions : a. there exists a natural number @xmath105 such that @xmath106 for all @xmath107 ; b. @xmath108={\mathfrak g}_{-p-1}$ ] for all @xmath109 ; c. if @xmath110 for @xmath111 and @xmath112=\{0\}$ ] , then @xmath113
. it immediately follows from this definition that @xmath114 is the graded nilpotent lie algebra generated by @xmath84 . following tanaka @xcite ,
we shall call graded nilpotent lie algebras of this kind _
fundamental_. in particular , the fundamental nilpotent lie algebra corresponding to a graded contact lie algebra is none other than the three - dimensional heisenberg algebra .
let @xmath115 be an arbitrary fundamental graded lie algebra . then ,
as was shown by tanaka @xcite , there exists a unique transitive graded lie algebra @xmath116 that satisfies the following conditions : 1 .
@xmath117 for @xmath118 ; 2 .
@xmath119 is the largest among all transitive graded lie algebras satisfying condition ( 1 ) .
this lie algebra @xmath119 is called the ( _ algebraic _ ) _ extension _ of @xmath45 . in particular ,
any transitive graded lie algebra @xmath45 may be identified with a graded subalgebra of @xmath119 , where @xmath115 .
the lie algebra @xmath119 has a clear geometrical meaning .
namely , let @xmath7 be a connected lie group with lie algebra @xmath120 , and let @xmath121 be a left - invariant distribution on @xmath7 such that @xmath122
. denote by @xmath123 the lie algebra of all germs of infinitesimal symmetries of @xmath121 at the identity element @xmath124 of @xmath7 .
consider the following two subspaces in @xmath123 : @xmath125 where @xmath126 denotes the germ of the vector field @xmath22 at the point @xmath124 .
now let @xmath127\text { for all } p\geq1\\ a_p=&\{\xi\in{\mathcal a}_{p-1}\mid[\xi,{\mathcal a}_{-1}]\subset{\mathcal a}_{p-1}\}\text { for all } p\geq1.\end{aligned}\ ] ] then the family of subspaces @xmath128 forms a decreasing filtration of the lie algebra @xmath123 , and @xmath119 can be identified with the associated graded algebra , ie , @xmath129 for all @xmath130 .
this geometrical interpretation allows to describe , without difficulty , the structure of @xmath119 in the case that we are interested in , namely in the case of graded contact lie algebras .
let @xmath131 be the three - dimensional heisenberg algebra and @xmath132 $ ] , while @xmath133 is a two - dimensional subspace complementary to @xmath134 $ ] . in this case we may assume without loss of generality that @xmath121 is precisely the contact distribution on @xmath36 . using the description of all infinitesimal symmetries of the contact distribution , it is not hard to determine the structure of the lie algebra @xmath135 .
it can be identified with the space of polynomials in @xmath136 with the bracket operation given by @xmath137 $ ] .
the space @xmath138 consists of all homogeneous polynomials of degree @xmath139 , assuming that the variables @xmath136 are of degree 1 , 2 and 1 respectively .
for example , @xmath140 we shall now fix some fundamental graded lie algebra @xmath120 and describe how one can classify all finite - dimensional graded subalgebras @xmath100 of the lie algebra @xmath119 such that @xmath141 for all @xmath111 . in what follows
we shall always assume that @xmath142 for all @xmath118 .
suppose that for some @xmath143 we have a collection of subspaces @xmath144 , @xmath145 , such that @xmath146\subset{\mathfrak h}_{p+q}$ ] @xmath147 , @xmath148 .
using induction , we define a sequence of subspaces @xmath149 as follows : @xmath150\subset{\mathfrak h}_p\}\ ] ] for all @xmath151 . it can be easily shown that @xmath152 is a graded subalgebra of @xmath119 .
this subalgebra is called the _ extension _ of the collection @xmath153 .
note that @xmath154 is the largest of all graded subalgebras whose @xmath155th grading space coincides with @xmath156 for all @xmath157 .
one the other hand , we can associate @xmath153 with the graded subalgebra @xmath158 generated by these subspaces .
now let @xmath100 be an arbitrary graded subalgebra of @xmath119 such that @xmath159 for all @xmath160 .
then , obviously , for any @xmath161 we have @xmath162 finally , notice that @xmath163 is precisely the algebra of all derivations of @xmath120 that preserve the grading ( see @xcite ) , and all the subspaces @xmath164 are invariant under the natural action of @xmath102 on @xmath165 .
based on these remarks , the following algorithm for the classification of the desired kind of subalgebras in @xmath119 suggests itself . * * step i**describe , up to conjugation , all subalgebras @xmath166 .
go to step iii . *
* step ii**suppose that for some @xmath143 , a collection of subspaces @xmath144 , @xmath145 , is already constructed such that @xmath167\subset{\mathfrak h}_{p+q}\quad\forall p , q\leq k,\ p+q\leq k\\ { \rm(ii)}&\quad\dim\tilde{\mathfrak g}({\mathfrak m},{\mathfrak h}_0,\dots,{\mathfrak h}_k)<\infty.\end{aligned}\ ] ] let @xmath168,\\ & { \mathfrak g}_{k+1}({\mathfrak m},{\mathfrak h}_0,\dots,{\mathfrak h}_k)=\{x\in{\mathfrak g}_{k+1}({\mathfrak m})\mid[x,{\mathfrak h}_{-1}]\subset{\mathfrak h}_k\}.\end{aligned}\ ] ] at this point we describe all @xmath102-invariant subspaces @xmath169 in @xmath170 such that @xmath171 * * step iii**find the subalgebras @xmath172 and @xmath173 .
if these subalgebras are not the same , go to step ii .
if , however , they coincide , then @xmath174 is one of the desired subalgebras .
now we shall use this algorithm to classify all irreducible graded contact lie algebras over the field of real numbers .
[ thm1 ] let @xmath131 denote the three - dimensional real heisenberg algebra , considered as a graded lie algebra , and let @xmath135 be the universal extension of @xmath131
. then any finite - dimensional irreducible graded contact lie algebra @xmath100 is isomorphic to one and only one of the following subalgebras of @xmath135 :
@xmath175 2 .
@xmath176 3 .
@xmath177 4 .
@xmath178 5 .
@xmath179 6 .
@xmath180 , @xmath181 fix a basis @xmath182 in the space @xmath183 .
then the action of the elements of @xmath184 on @xmath183 is given by the following matrices : @xmath185 therefore , the lie algebra @xmath184 may be identified with @xmath186 , and the @xmath184module @xmath183 with the natural @xmath186module .
[ lem2 ] any irreducible subalgebra of @xmath186 is conjugate to one and only one of the following subalgebras : @xmath187 { \rm(iii)}&\ { \mathfrak{sl}}(2,{\mathbb r})&{\rm(iv)}&\ { \mathfrak{gl}}(2,{\mathbb r } ) \end{aligned}\ ] ] if a subalgebra of @xmath186 is nonsolvable , then it is either three - dimensional and coincides with @xmath188 , or four - dimensional and is equal to the whole of @xmath186 .
any solvable irreducible subalgebra is commutative .
if it is one - dimensional , then , as follows from the classification of real jordan normal forms of @xmath189 matrices , it is conjugate to the subalgebra ( i ) .
if @xmath45 is two - dimensional , it coincides with the centralizer of one of the jordan normal forms , which implies that it is conjugate to the subalgebra ( ii ) .
if we identify @xmath186 and @xmath184 , the subalgebras listed in lemma [ lem2 ] are identified with the following subspaces @xmath190 : @xmath191 { \rm(iii)\ } & \langle x^2 , z^2 , 2y - xz\rangle & { \rm(iv)\ } & \langle x^2 , xz , z^2 , y\rangle\end{aligned}\ ] ] consider separately each one of these cases : ( i)it is easily verified that in this case we have @xmath192
. therefore @xmath193 , and we arrive at the algebra which is listed in the theorem under the number @xmath194 .
( ii)here we have @xmath195 and the action of the subalgebra @xmath102 on this space is irreducible . therefore , the space @xmath196 is either zero or coincides with the whole of @xmath197 . in the former case
we immediately find that @xmath193 ( subalgebra @xmath198 ) . in the second case the subalgebras @xmath199 and @xmath200 coincide and are equal to the subalgebra @xmath201 of the theorem .
( iii)here @xmath202 , and the @xmath102module @xmath203 is irreducible .
hence either we have @xmath204 and then @xmath193 ( subalgebra @xmath205 ) , or @xmath206 . in the latter case , however , the space @xmath207 generates a finite - dimensional subalgebra .
( iv)here @xmath208 , and the @xmath102module @xmath197 is a sum of two irreducible submodules @xmath209 and @xmath210 of the form @xmath211 the submodule @xmath209 generates a finite - dimensional subalgebra , so that either @xmath204 or @xmath212 . in the former case @xmath193 ( subalgebra @xmath213 ) , while in the latter the subalgebras @xmath199 and @xmath200 coincide and are equal to the subalgebra @xmath214 of the theorem .
in order to classify all finite - dimensional irreducible contact lie algebras , it will suffice to describe all filtered lie algebras whose associated graded lie algebras are listed in theorem [ thm1 ] . to solve this latter problem
, we shall need the following result .
[ pas : l3 ] let @xmath45 be a finite - dimensional filtered lie algebra , and @xmath100 the associated graded lie algebra . if there is an element @xmath215 such that @xmath216=px_p\quad\forall x_p\in{\mathfrak h}_p\ ] ] then @xmath100 , viewed as a filtered lie algebra , is isomorphic to @xmath45 .
suppose @xmath217 for some @xmath218 . for every @xmath130 ,
consider the subspace @xmath219=px\}.\ ] ] it is easy to show that , @xmath220 for all @xmath130 .
thus , the subspace @xmath221 may be identified with @xmath222 , and since @xmath223\subset{\mathfrak g}^{i+j}(\bar e)$ ] , this identification is in agreement with the structure of the lie algebras @xmath45 and @xmath100 .
hence , we have found an isomorphism of the lie algebras @xmath45 and @xmath100 which is compatible with their filtrations . for the graded lie algebras listed in theorem [ thm1 ] under the numbers @xmath214 , @xmath213 , @xmath201 , @xmath198 , we can choose @xmath124 to be equal to @xmath224 , as this element is contained in all of these algebras . then
, in view of lemma [ pas : l3 ] , the description of the corresponding filtered lie algebras in these four cases is trivial .
consider the remaining two cases @xmath205 and @xmath194 .
@xmath205let @xmath100 be the graded lie algebra that appears under the number @xmath205 in theorem [ thm1 ] , and let @xmath45 be a contact lie algebra whose associated graded lie algebra is isomorphic to @xmath100 . since @xmath225 , the subalgebra @xmath65 can be identified with the subalgebra @xmath102 , which is isomorphic to @xmath188 . consider the @xmath65module @xmath45 .
it is completely reducible , and its decomposition into a sum of irreducible submodules has the form : @xmath226 , where the submodule @xmath227 is one - dimensional and is a complement of @xmath84 , while the submodule @xmath228 is two - dimensional and complements @xmath65 in @xmath84 .
therefore , the submodules @xmath229 , @xmath230 can be identified with the subspaces @xmath231 of the graded lie algebra @xmath100 , which allows to identify @xmath45 and @xmath100 as vector spaces .
the structure of the lie algebra @xmath45 is completely determined by the mappings @xmath232 and @xmath233 defined as restrictions of the bracket operation in @xmath45 to the corresponding subspaces .
the jacobi identity shows that these mappings are both @xmath65invariant .
since the @xmath65module @xmath234 is one - dimensional and trivial , we have @xmath235 .
similarly , @xmath236 .
now , computing the jacobi identity for the basis vectors of @xmath227 and @xmath228 , we find that the mapping @xmath237 is zero .
thus , the identification of the spaces @xmath45 and @xmath100 is in agreement with the lie algebra structures of these spaces , so that the lie algebra @xmath45 is isomorphic to @xmath100 , viewed as a filtered lie algebra .
@xmath194as in the above case , we can identify @xmath65 and @xmath102 .
now , since the @xmath102modules @xmath102 and @xmath103 are not isomorphic for any value of @xmath237 , we conclude that @xmath84 contains a @xmath65invariant subspace @xmath228 which is a complement of @xmath65 . choose a basis @xmath238 for @xmath65 and a basis @xmath239 for @xmath228 in such a way that @xmath240&=\alpha u_1-u_2\\ [ e , u_2]&=u_1+\alpha u_2.\end{aligned}\ ] ] then the elements @xmath241 , together with the element @xmath242 $ ] , will , obviously , form a basis of @xmath45 , and @xmath243=2\alpha u_3 $ ] . furthermore , checking the jacobi identity , we find that in case @xmath244 we have @xmath245=[u_1,u_3]=0 $ ] , and the lie algebra @xmath45 is isomorphic to @xmath100 , viewed as a filtered lie algebra .
if @xmath246 , we have @xmath245= \beta u_3 $ ] and @xmath247=-\beta u_2 $ ] for some @xmath248 .
note that the parameters @xmath249 and @xmath250 with @xmath251 give here isomorphic lie algebras , whatever the value of @xmath252 may be .
therefore , up to isomorphism of contact lie algebras we may assume that @xmath253 .
if @xmath254 , we find that @xmath45 is again isomorphic to @xmath100 , viewed as a filtered lie algebra .
if @xmath255 or @xmath256 , the lie algebra @xmath45 is isomorphic to @xmath186 or @xmath257 respectively , while the subalgebras @xmath45 can be written , under this identification , in matrix form as follows : @xmath258 summing up what has been said , we obtain the following result : any finite - dimensional irreducible contact lie algebra is isomorphic to one and only one of the following : * any of the graded contact lie algebras listed in theorem [ thm1 ] , if they are viewed as filtered lie algebras ; * @xmath259 , where @xmath260 for @xmath109 , @xmath261 * @xmath262 , where @xmath260 for @xmath109 , @xmath263 from now on , to refer to irreducible contact algebras of type i , we shall employ the notation i.@xmath264 , where @xmath264 is the number of the corresponding graded contact lie algebra in theorem [ thm1 ] .
[ [ section ] ] now we shall find explicit representations in contact vector fields for the lie algebras of vector fields described above . note that the mapping @xmath28 that maps an arbitrary function @xmath265 of the variables @xmath136 into the vector field
whose characteristic function is @xmath265 , defines an embedding of the lie algebra @xmath135 into the algebra of all contact vector fields . in this way
we immediately obtain the explicit representations in vector fields for those contact algebras @xmath45 which are isomorphic to their corresponding graded algebras .
in particular , from the representations ( a ) and ( b ) it follows that these two algebras of contact vector fields can both be embedded into the 10dimensional algebra i.1 and into the 8dimensional algebra i.4 .
the representations ( c ) are notable for the fact that the characteristic functions here are independent of @xmath272 .
consider the set of all contact vector fields of the form @xmath26 , where the function @xmath265 has the form @xmath273 with @xmath274 and @xmath275 being an arbitrary function of @xmath276 .
it is easy to show that this condition is equivalent to the requirement that @xmath26 be an infinitesimal symmetry of the one - dimensional distribution @xmath47 generated by the vector field @xmath277 .
thus we see that this space of vector fields forms an infinite - dimensional subalgebra @xmath278 of the lie algebra of all contact vector fields .
consider the projection @xmath279 given by @xmath280 it is easily verified that this mapping is a homomorphism of lie algebras whose kernel is one - dimensional and is generated by @xmath281 , while its image coincides with the set of all vector fields on the plane that preserve , up to a constant factor , the volume form @xmath282 on the plane : @xmath283 thus , with every lie algebra of contact vector fields that preserves a one - dimensional distribution complementary to the contact one , we can associate a lie algebra of vector fields on the plane .
conversely , the inverse image of any subalgebra of the lie algebra ( [ eq : pr ] ) of vector fields on the plane is some lie algebra of contact vector fields in the jet space .
note that all irreducible lie algebras of contact vector fields , except i.1 and i.4 , preserve a one - dimensional distribution complementary to the contact distribution , and hence can be embedded into @xmath278 .
the corresponding lie algebras of vector fields on the plane are as follows : * the lie algebra corresponding to the group of affine transformations of the plane ; * the lie algebra corresponding to the group of equi - affine transformations of the plane ( ie , area - preserving affine transformations ) ; * the lie algebra corresponding to the group of similitude transformations ; * @xmath284 , @xmath285 ( if @xmath246 , this lie algebra corresponds to the group of euclidean transformations ) ; * the lie algebra corresponding to the group of all transformations of the hyperbolic plane ; * the lie algebra corresponding to the group of rotations of the sphere .
the above correspondence allows to describe without any difficulty all differential and integral invariants for all lie algebras @xmath45 of contact vector fields that satisfy the following conditions : indeed , let @xmath287 be the standard coordinate system in the space @xmath288 of @xmath264th jets of curves on the plane .
( see @xcite for definition of jet spaces and notions of differential and integral invariants . )
denote by @xmath289 the @xmath264th prolongation of the lie algebra @xmath45 .
it then follows from the condition ( b ) that the differential and integral invariants of @xmath45 that have the order @xmath264 are independent of @xmath272 and may be considered on the manifold of the trajectories of the vector field @xmath290 .
these trajectories are given by the equations @xmath291 and can be parametrized by the coordinates @xmath292 . furthermore
, it turns out that if @xmath293 , the action of the algebra @xmath289 on that quotient manifold is equivalent to the action of the lie algebra @xmath294 on the space of @xmath295th jets , and the mapping @xmath296 that establishes this equivalence has the form : @xmath297 therefore , all differential and integral invariants of @xmath45 may be derived from the invariants of @xmath298 by substituting @xmath299 instead of @xmath300 for @xmath301 .
we remark that sophus lie @xcite found all invariants for those lie algebras of vector fields on the plane that correspond to the cases i.2 and i.3 .
the invariants of the 10dimensional irreducible lie algebra i.1 were computed in @xcite over the complex numbers , and they remain unchanged on passing to the real case .
now we shall specify nontrivial integral and differential invariants of the least order for the rest of irreducible contact lie algebras of vector fields ; all other invariants can be derived from these by means of differentiation ( see @xcite ) .
@xmath302 \text{i.5 } & \frac{(1+y_2 ^ 2)y_4 - 3y_2y_3 ^ 2}{y_2 ^ 2 } & \frac{y_3dx}{1+y_2 ^ 2 }
\\[2 mm ] \text{i.6 } & \frac{y_3e^{-\beta{\operatorname{arctg}}y_2}}{(1+y_2 ^ 2)^{3/2 } } & e^{\beta{\operatorname{arctg}}y_2}(1+y_2)^{1/2}dx \\[2 mm ] \text{ii.1 } & \frac{\sqrt{1+y_1 ^ 2}((1+y_1 ^ 2)y_3 - 3y_1y_2 ^ 2-y_1(1+y_1 ^ 2)^2)}{((1+y_1 ^ 2)^2+y_2 ^ 2)^{3/2 } } & \left(\frac{(1+y_1 ^ 2)^2+y_2 ^ 2}{1+y_1 ^
2}\right)^{1/2}dx \\[2 mm ] \text{ii.2 } & \frac{\sqrt{1-y_1 ^ 2}((1-y_1 ^ 2)y_3 + 3y_1y_2 ^ 2+y_1(1-y_1 ^ 2)^2)}{((1-y_1 ^ 2)^2+y_2 ^ 2)^{3/2 } } & \left(\frac{(1-y_1 ^ 2)^2+y_2 ^ 2}{1-y_1 ^ 2}\right)^{1/2}dx\\[2 mm ] \hline \end{array}\ ] ] where @xmath303 for the algebras ii.1 and ii.2 , we have chosen here their representations in contact vector fields that appear earlier under the letter ( c ) .
notice that all contact lie algebras listed in the table above are reducible over the field of complex numbers .
hence , for each of these algebras there exists a certain complex analytic contact transformation which takes it to one of the known canonical forms for contact lie algebras over @xmath304 .
thus , the inverse thatsformation ( prolonged as many times as needed ) brings known invariants to the invariants of the initial lie algebra .
for example , the contact transformation @xmath305 takes the contact lie algebra i.5 to the algebra with the following space ( over @xmath304 ) of characteristic functions : @xmath306 this contact lie algebra is reducible and is the first prolongation of the following lie algebra of vector fields on the plain : @xmath307 the differential invariants of the least order for this lie algebra were computed already ( see , for example , @xcite ) and have the form : the third prolongation of the inverse transformation @xmath308 takes these invariants to those given in the table above . in the similar way we can compute invariants for other contact lie algebras of vector fields given in the table .
consider the problem of classifying those infinite - dimensional subalgebras in the lie algebra of contact vector fields that correspond to lie pseudo - groups of contact transformations ( ie , those that can be defined with the help of a finite number of differential equations ; see @xcite ) .
as in the finite - dimensional case , all these algebras can be naturally divided into two classes : reducible ones , which are actually extensions of infinite - dimensional lie algebras of vector fields on the plane , and irreducible ones . over the field of complex numbers
all irreducible infinite - dimensional lie algebras of contact vector fields were described by sophus lie @xcite , who showed that , apart from the lie algebra of all contact vector fields , there exist exactly two infinite - dimensional irreducible subalgebras , namely , the lie algebra @xmath309 , which we already mentioned earlier , and its commutant @xmath310=\{x_{g(x , z)}\}$ ] .
the methods for the description of contact lie algebras that have been developed in this paper , can be easily generalized to the infinite - dimensional case .
in particular , with these lie algebras we can again associate graded contact lie algebras that can be embedded into the universal extension @xmath135 of the three - dimensional heisenberg algebra .
let @xmath100 be an infinite - dimensional graded subalgebra in @xmath135 such that @xmath311 for @xmath118 and such that the @xmath102module @xmath103 is irreducible . then
all possible types of subalgebras @xmath102 over the real numbers are listed in lemma [ lem2 ] . as follows from the proof of theorem [ thm1 ] , in the cases ( i ) and
( ii ) the lie algebra @xmath100 is finite - dimensional .
the consideration of the remaining cases ( iii ) and ( iv ) is the same over the complex and real numbers , and gives the infinite - dimensional lie algebras of contact vector fields described above .
thus the classification of irreducible infinite - dimensional lie algebras of contact vector fields remains unchanged on passing from the complex to the real case .
, _ klassifikation und integration von gewhnlichen differentialgleichungen zwischen @xmath312 , die eine gruppe von transformationen gestatten _ , i
vi , from : `` gesamelte abhandlungen '' , volume 5 , teubner , leipzig ( 1924 ) 240310 , 362427 , 432448 | the paper is devoted to the complete classification of all real lie algebras of contact vector fields on the first jet space of one - dimensional submanifolds in the plane .
this completes sophus lie s classification of all possible lie algebras of contact symmetries for ordinary differential equations . as a main tool
we use the abstract theory of filtered and graded lie algebras .
we also describe all differential and integral invariants of new lie algebras found in the paper and discuss the infinite - dimensional case . |
telomeres are tandem repeated noncoding sequences of nucleotides at both ends of the dna in eukaryotic chromosomes stabilizing the chromosome ends and preventing them from end - to - end fusion or degradation @xcite .
polymerase can not completely replicate the 3 end of linear dna , so telomeres are shortened at each dna replication @xcite .
this end replication problem leads to a finite replicative capacity for normal somatic cells @xcite .
they can only divide up to a certain threshold , the hayflick limit @xcite .
the enzyme telomerase , repressed in most normal somatic cells , synthesizes and elongates telomere repeat sequences at the end of dna strands so that certain cells like germline cells are immortal and indefinite in growth @xcite .
most forms of cancer follow from the accumulation of somatic mutations @xcite .
cancer - derived cell lines and 85 - 90% of primary human cancers are able to synthesize high levels of telomerase and thus are able to prevent further shortening of their telomeres and proliferate indefinitely @xcite .
but if cells are premalignant or already cancerous and telomerase is not yet activated , the proliferative capacity of these cells and therefore the accumulation of mutations is determined by the remaining telomere length @xcite .
so the frequency of malignant cancer should be higher for longer telomeres in normal somatic cells .
recently published data show that longer telomeric dna increased the life span of nematode worms @xcite .
experiments are running @xcite whether there is a positive effect on the longevity also for organisms with renewing tissues if the telomere length in fetal cells is increased @xcite .
as the probability for the incidence of cancer is correlated to the replicative potential of the mutated cells @xcite , one can ask the following question : is an extension of life span possible if telomeres in embryonic cells are elongated and cancer is considered ?
an answer to this question could be given by computer simulations as the presented model focuses on organismal ageing due to the loss of telomeres in dna and neglects other effects which lead to a decreasing survival probability with age .
as shortening of telomeres is one of the supposed mechanisms of ageing on cellular level , most stochastical and analytical studies investigate this relationship @xcite . a theoretical model which directly relates telomere attrition to human ageing
was first suggested by aviv et al .
@xcite . here
we present a different telomere dynamics model providing requirements to study the effects of ( a ) different mean telomere lengths in constituting cells , as well as of ( b ) somatic mutations leading to cancer progression and indefinite proliferation due to telomerase activity on the ageing of organisms .
every organism is developed from a single progenitor cell , the zygote ( figure [ bild01 ] ) .
the initial telomere length of zygote cells is assumed to be normally distributed with mean @xmath2 and standard deviation @xmath3 @xcite .
telomere repeats lost per division ( trlpd ) are randomly chosen at each division of every cell from a normal distribution with mean @xmath4 and standard deviation @xmath5 .
a dividing cell produces a clone who inherits the replicative capacity of the progenitor cell at this age .
cells can divide until nearly all their original telomeres are lost . for every organism
the dynamics of the model is as follows : divisions of the zygote and the stem cells derived from it occur 6 times in the early embryo .
each of these cells is the progenitor of one tissue .
this is followed by a period of population doublings where all cells divide once in every timestep until @xmath6 cells are present in each tissue . in the following maturation stage ,
cells are chosen randomly for division until each tissue reaches the adult size of @xmath7 cells .
it takes about 26 timesteps until an organism is mature .
ageing starts now . in every
timestep first cells die with 10% probability due to events like necrosis or apoptosis .
10% of the cells of the corresponding tissue are then randomly chosen for division to fill this gap .
the replacement does not have to be complete as the chosen cell could probably not divide anymore due to telomere attrition .
after some time the tissue will start shrinking .
the random choice of dying and dividing cells in differentiated tissues is in accordance with nature as for example in epithelium the choice of cells to be exported from the basal layer is random @xcite .
the organism dies , if its total cell population size shrinks to 50% of the mature size .
the results presented in the following do not depend qualitatively on the choice of this threshold .
in the implementation of this model , linear congruential generators are used to produce the required random numbers and normally distributed variables are generated with the box - muller algorithm @xcite . resulting age distributions for different mean telomere lengths in the zygote cells are shown in ( figure [ bild02 ] ) .
the shape of these distributions is analogous to empirical data of many human and animal populations .
we obtain a positive effect on the longevity of the organisms if the mean telomere length in the precursor cell is increased .
the chosen mean doubling potentials of the zygote cells are 30 , 40 and 50 with the choice of @xmath8 and @xmath9 .
the number of mitotic divisions observed in human fibroblasts is higher @xcite , but the choice of this parameter is reasonable as the number of considered cells per mature organism ( 640000 ) in this model is also much lower than in human organisms where the total number of cells is of the order of @xmath10 @xcite .
non - dividing cells are not included in the model as we focus on ageing due to the progressive shrinking of tissues driven by telomere shortening .
our results will be given below as part of fig.5 .
clonal cancer is now introduced in the model . in accordance to the model of moolgavkar
et al . , one of our assumptions is that malignant tumors arise from independently mutated progenitor cells @xcite . for most forms of carcinoma , transformation of a susceptible stem cell into a cancer cell
is suggested to be a multistage process of successive mutations with a relatively low probability for the sequential stages @xcite .
two independent and irreversible hereditary mutation stages are considered here , which can occur at every level of development of the organism during cell division .
the first premalignant stage to be considered is a promotion stage : a dividing cell can mutate with small probability @xmath11 @xcite .
all descendant cells inherit this mutation .
this mutation leads to a partial escape from homeostatic control of growth by the local cellular environment @xcite .
cells on the promotion stage have a selective advantage over unaffected cells @xcite . in our model
they are chosen first for division during maturation and for filling up the gap in the ageing period .
the subsequent transition can occur again with probability @xmath11 during division .
if a cell reaches this second stage of mutation it is a progenitor of a carcinoma .
an explosive clonal expansion to a fully malignant compartment happens @xcite .
this cell and the clonal progeny doubles in the current timestep until it is no more possible due to telomere attrition .
this expansion leads only to an increase of the malignant cell population size . as a certain fraction of cells
is killed per unit time and clonal expansions only occur with a very small probability , the tumor environments may not continue growing , eventually shrink , or even die @xcite .
we assume that it is necessary for advanced cancer progression and therefore for the development of a deadly tumor that fully mutated cells are able to activate telomerase @xcite . in our model ,
telomerase activation is possible at every age of the organism in normal and mutated cells during division with a very low probability @xmath12 .
the irreversible loss of replicative potential is stopped in these cells . as the contribution of telomerase to tumorigenicity is not yet completely understood @xcite , we assume that death of an organism due to cancer occurs if telomerase is reactived in at least one fully mutated cell @xcite .
we treat the time interval between the occurence of the deadly tumor and death as constant , so we set this interval to zero .
figure [ bild03 ] shows simulation results for @xmath13 and @xmath14 .
as we consider a lower complexity by chosing a lower number of tissues and cells per organism , we assume higher mutation rates for the incidence of cancer than observed in nature @xcite .
the age distribution for shorter initial telomere lengths considering cancer is shifted to the left but still very old organisms exist . for longer telomeres the age distribution
is again shifted to the left but even behind the distributions for shorter telomeres with and without considering carcinogenesis .
thus without considering cancer , organisms with longer zygote telomeres live longer , as the life expectancy of the organisms increases linear for longer telomeres .
but if cancer is considered this effect is reversed for longer initial mean telomere lengths ( fig .
[ bild04 ] ) .
the force of mortality @xcite resulting from this model is shown for @xmath13 with and without considering cancer in comparison to empirical human mortality data ( figure [ bild05 ] ) .
it agrees to some extent with human mortality functions provided cancer is incorporated into the model and decelerates at advanced ages , as claimed for human and animal populations @xcite .
the hump in the curve at younger ages , occurring also for other parameter sets , fits to data of many human mortality tables .
the expected simulation result of the basic model without cancer is an increase of life span of most organisms with longer initial telomeres . after introducing somatic mutations promoting cancer and telomerase activation in this model ,
the survival probability is lower for each considered initial telomere length in certain time intervals in adult ages .
but even low probabilities for the two mutation stages and for the activation of telomerase lead to a strong reduction of life span for longer telomeres .
so the implication of two - stage carcinogenesis for the incidence of cancer in this simple model of cell proliferation in organisms is that life expectancy and life span of complex organisms can not be increased by artificially elongating telomeres in primary cells , for example during a cloning procedure .
further improvements , extentions and applications of this model are possible . with respect to the role of telomeres and telomerase in carcinogenesis ,
maybe this computational approach can contribute to the development of a comprehensive theoretical model in oncology uniting mutagenesis and cell proliferation @xcite .
+ + * acknowledgements * + we wish to thank the european project cost - p10 for supporting visits of mm and ds to the cebrat group at wrocaw university and the julich supercomputer center for computing time on their crayt3e .
cs was supported by foundation for polish science .
blackburn , nature 350 ( 1991 ) 569 .
harley , a.b .
futcher , c.w .
greider , nature 345 ( 1990 ) 458 .
a.m. olovnikov , j theor biol 41 ( 1973 ) 181 .
l. hayflick , p.s .
moorhead , exp cell res . 25 ( 1961 ) 585 .
l. hayflick , exp gerontology 38 ( 2003 ) 1231 .
greider , e.h .
blackburn , cell 43 ( 1985 ) 405 .
shay , w.e .
wright , nature rev 1 ( 2000 ) 72 .
nordling , br j cancer 7 ( 1953)68 .
fearon , b. vogelstein , cell 61 ( 1990 ) 759 .
kim , m.a .
piatyszek , k.r .
prowse , c.b .
harley , m.d .
west , p.l . ho , g.m .
coviello , w.e .
wright , s.l .
weinrich , j.w .
shay , science 266 ( 1994 ) 2011 .
moolgavkar , e.g. luebeck , cancer 38 ( 2003 ) 302 .
joeng , e.j .
song , k.j .
lee , j. lee , genet 36 ( 2004 ) 6071 .
lanza , j.b .
cibelli , c. blackwell , v.j .
cristofalo , m.k .
francis , g.m .
baerlocher , j. mak , m. schertzer , e.a .
chavez , n. sawyer , p.m. lansdorp , m.d .
west , science 288 ( 2000 ) 665 .
shay , w.e .
wright , radiat res 155 ( 2001 ) 188 .
levy , r.c .
allsopp , a.b .
futcher , c.w .
greider , c.b .
harley , j mol biol 225 ( 1992 ) 951 .
o. arino , m. kimmel , g.f .
webb , j theor biol .
177 ( 1995 ) 45 .
p. olofsson , m. kimmel , math biosci 158 ( 1999 ) 75 .
i. rubelj , z. vondracek , j theor biol . 197
( 1999 ) 425 .
sozou , t.b .
kirkwood , j theor biol 213 ( 2001 ) 573 .
a. aviv , d. levy , m. mangel , mech ageing dev 124 ( 2003 ) 829 .
j. op den buijs , p.p .
van den bosch , m.w .
musters , n.a .
van riel , mech ageing dev 125 ( 2004 ) 437 .
tu , e. fischbach , int j mod phys c 16 ( 2005 ) 281 .
box , m.e .
muller , ann math stat 29 ( 1958 ) 610 .
l. cairns , nature 255 ( 1975 ) 197 .
frank , y. iwasa , m.a .
nowak , genetics 163 ( 2003 ) 1527 . r.c .
allsopp , h. vaziri , c. patterson , s. goldstein , e.v .
younglai , a.b .
futcher , c.w .
greider , c.b .
harley , proc natl acad sci usa 89 ( 1992 ) 10114 .
moolgavkar , a. dewanji , d.j .
venzon , risk anal 8 ( 1998 ) 383 .
p. armitage , r. doll , br j cancer 8 ( 1954 ) 1 .
a. sarasin , mutat res 544 ( 2003 ) 99 .
p. armitage , r. doll , br j cancer 11 ( 1957 ) 161 .
e. hiyama , k. hiyama , t. yokoyama , y. matsuura , m.a .
piatyszek , j.w .
shay nat med 1 ( 1995 ) 249 .
shay , j natl cancer inst 91 ( 1999 ) 4 .
t. hiyama , h. yokozaki , y. kitadai , k. haruma , w. yasui , g. kajiyama , e. tahara , virchows arch 434 ( 1999 ) 483 .
v. kanjuh , m. kneevi , e. ivka , m. ostoji , b. beleslin , archive of oncology 9 ( 2001 ) 3 .
blasco , w.c .
hahn , trends cell biol 13 ( 2003 ) 289 .
moolgavkar , a.g .
knudson jr .
, j natl cancer inst 66 ( 1998 ) 1037 .
e.g. luebeck , s.h .
moolgavkar , proc natl acad sci usa 99 ( 2002 ) 15095 .
drake , b. charlesworth , d. charlesworth , j.f .
crow , genetics 148 ( 1998 ) 1667 .
thatcher , v. kannisto , j.w .
vaupel , _ the force of mortality at ages 80 to 120 .
_ monographs on population aging , vol . 5 , odense university press , 1998 .
robine , j.w .
vaupel , exp gerontology 36 ( 2001 ) 915 .
vaupel , j.r .
carey , k. christensen , t.e .
johnson , a.i .
yashin , n.v .
holm , i.a .
iachine , a.a .
khazaeli , p. liedo , v.d .
longo , yi zeng , k.g .
manton , j.w .
curtsinger , science 280 ( 1998 ) 855 .
gatenby , p.k .
maini , nature 421 ( 2003 ) 321 . | as cell proliferation is limited due to the loss of telomere repeats in dna of normal somatic cells during division , telomere attrition can possibly play an important role in determining the maximum life span of organisms as well as contribute to the process of biological ageing . with computer simulations of cell culture development in organisms ,
which consist of tissues of normal somatic cells with finite growth , we otain an increase of life span and life expectancy for longer telomeric dna in the zygote . by additionally considering a two - mutation model for carcinogenesis and indefinite proliferation by the activation of telomerase
, we demonstrate that the risk of dying due to cancer can outweigh the positive effect of longer telomeres on the longevity .
* does telomere elongation lead to a longer lifespan if cancer is considered ? *
michael masa@xmath0 , stanisaw cebrat@xmath1 and dietrich stauffer@xmath0 @xmath0 institute for theoretical physics , cologne university , d-50923 kln , euroland @xmath1 institute of genetics and microbiology , university of wrocaw , + ul .
przybyszewskiego 63/77 , pl-54148 wrocaw , poland _ keywords _ : biological ageing ; computer simulations ; telomeres ; telomerase ; cancer |
modern power networks are increasingly dependent on information technology in order to achieve higher efficiency , flexibility and adaptability @xcite .
the development of more advanced sensing , communications and control capabilities for power grids enables better situational awareness and smarter control .
however , security issues also arise as more complex information systems become prominent targets of cyber - physical attacks : not only can there be data attacks on measurements that disrupt situation awareness @xcite , but also control signals of power grid components including generation and loads can be hijacked , leading to immediate physical misbehavior of power systems @xcite . furthermore , in addition to hacking control messages , a powerful attacker can also implement physical attacks by directly intruding upon power grid components .
therefore , to achieve reliable and secure operation of a smart power grid , it is essential for the system operator to minimize ( if not eliminate ) the feasibility and impact of physical attacks .
there are many closely related techniques that can help achieve secure power systems .
firstly , coding and encryption can better secure control messages and communication links @xcite , and hence raise the level of difficulty of cyber attacks . to prevent physical attacks ,
grid hardening is another design choice @xcite .
however , grid hardening can be very costly , and hence may only apply to a small fraction of the components in large power systems .
secondly , power systems are subject to many kinds of faults and outages @xcite , which are in a sense _ unintentional _ physical attacks . as such outages
are not inflicted by attackers , they are typically modeled as random events , and detecting outages is often modeled as a hypothesis testing problem @xcite . however , this event and detection model is not necessarily accurate for _ intentional _ physical attacks , which are the focus of this paper .
indeed , an intelligent attacker would often like to strategically _ optimize _ its attack , such that it is not only hard to detect , but also the most viable to implement ( e.g. , with low execution complexity as well as high impact ) .
recently , there has been considerable research concerning data injection attacks on sensor measurements from supervisory control and data acquisition ( scada ) systems .
a common and important goal among these works is to pursue the integrity of network _ state estimation _ , that is , to successfully detect the injected data attack and recover the correct system states .
the feasibility of constructing data injection attacks to pass bad data detection schemes and alter estimated system states was first shown in @xcite . there
, a natural question arises as to how to find the _
sparsest unobservable _ data injection attack , as sparsity is used to model the complexity of an attack , as well as the resources needed for an attacker to implement it . however , finding such an _ optimal attack _ requires solving an np - hard @xmath2 minimization problem . while efficiently finding the sparsest unobservable attacks in general remains an open problem , interesting and exact solutions under some special problem settings have been developed in @xcite @xcite @xcite .
another important aspect of a data injection attack is its impact on the power system .
as state estimates are used to guide system and market operation of the grid , several interesting studies have investigated the impact of data attacks on optimal power flow recommendation @xcite and location marginal prices in a deregulated power market @xcite @xcite .
furthermore , as phasor measurement units ( pmus ) become increasingly deployed in power systems , network situational awareness for grid operators is significantly improved compared to using legacy scada systems only . however , while pmus provide accurate and secure sampling of the system states , their high installation costs prohibit ubiquitous deployment .
thus , the problem of how to economically deploy pmus such that the state estimator can best detect data injection attacks is an interesting problem that many studies have addressed ( see , e.g. @xcite among others . ) compared to data attacks that target state estimators , physical attacks that directly disrupt power network physical processes can have a much faster impact on power grids .
in addition to physical attacks by hacking control signals or directly intruding upon grid components , several types of load altering attacks have been shown to be practically implementable via internet - based message attacks @xcite .
topological attacks are another type of physical attack which have been considered in @xcite .
dynamic power injection attacks have also been analyzed in several studies .
for example , in @xcite , conditions for the existence of undetectable and unidentifiable attacks were provided , and the sizes of the sets of such attacks were shown to be bounded by graph - theoretic quantities . alternatively , in @xcite and @xcite , state estimation is considered in the presence of both power injection attacks and data attacks .
specifically , in these works , the maximum number of attacked nodes that still results in correct estimation was characterized , and effective heuristics for state recovery under sparse attacks were provided . in this paper
, we investigate a specific type of physical attack in power systems called _ power injection attacks _ , that alter generation and loads in the network .
a linearized power network model - the dc power flow model - is employed for simplifying the analysis of the problem and obtaining a simple solution that yields considerable insight .
we consider a grid operator that employs pmus to ( partially ) monitor the network for detecting power injection attacks .
since power injection attacks disrupt the power system states immediately , the timeliness of pmu measurement feedback is essential .
furthermore , our model allows for the power injections at some buses to be `` unalterable '' .
this captures the cases of `` zero injection buses '' with no generation and load , and buses that are protected by the system operator . under this model
we study the open @xmath2 minimization problem of finding the sparsest unobservable attacks given any set of pmu locations .
we start with a feasibility problem for unobservable attacks .
we prove that the existence of an unobservable power injection attack restricted to any given set of buses can be determined with probability one by computing a quantity called the structural rank .
next , we prove that the np - hard problem of finding the sparsest unobservable attacks has a simple solution with probability one .
specifically , the sparsity of the optimal solution is @xmath0 , where @xmath1 is the `` vulnerable vertex connectivity '' that we define for an augmented graph of the original power network .
meanwhile , the entire set of globally optimal solutions ( there can be many of them ) is found in polynomial time .
we further introduce a notion of potential impacts of unobservable attacks .
accordingly , among all the sparsest unobservable attacks , an attacker can easily find the one with the greatest potential impact .
finally , given optimized pmu placement , we evaluate the sparsest unobservable attacks in terms of their sparsity and potential impact in the ieee 30 , 57 , 118 and 300-bus , and the polish 2383 , 2737 and 3012-bus systems .
the remainder of the paper is organized as follows . in section [ secform ] , models of the power network , power injection attacks , pmus and unalterable buses
are established .
in addition , the minimum sparsity problem of unobservable attacks is formulated . in section [ secfeas ]
we provide the feasibility condition for unobservable attacks restricted to any subset of the buses . in section [ secmin ]
we prove that the minimum sparsity of unobservable attacks can be found in polynomial time with probability one . in section [ secnum ] , a pmu placement algorithm for countering power injection attacks
is developed , and numerical evaluation of the sparsest unobservable attacks in ieee benchmark test cases and large - scale polish power systems are provided .
conclusions are drawn in section [ secconc ] .
we consider a power network with @xmath3 buses , and denote the set of buses and the set of transmission lines by @xmath4 and @xmath5 respectively . for a line @xmath6 that connects buses @xmath7 and @xmath8 , denote its reactance by @xmath9 as well as @xmath10 , and define its _ incidence vector _ @xmath11 as follows : @xmath12 based on the power network topology and line reactances
, we construct a weighted graph @xmath13 where the edge weight @xmath14 .
the power system is generally modeled by nonlinear ac power flow equations @xcite . in this paper , a linearized model - the dc power flow model - is employed as an approximation of the ac model , which allows us to find a simple closed - form solution to the problem from which we glean significant insights . under the dc model , the real power injections @xmath15 and
the voltage phase angles @xmath16 satisfy @xmath17 , where @xmath18 is the _
laplacian _ of the weighted graph @xmath19 .
we assume that @xmath9 is positive which is typically true for transmission lines ( cf .
chapter 4 of @xcite ) .
furthermore , the power flow on line @xmath20 from bus @xmath7 to bus @xmath8 equals @xmath21 .
we consider attackers inflicting power injection attacks that alter the generation and loads in the power network .
we denote the power injections in normal conditions by @xmath22 , and denote a power injection attack by @xmath23 .
thus the post - attack power injections are @xmath24 .
we consider the use of pmus by the system operator for monitoring the power network in order to detect power injection attacks .
with pmus installed at the buses , we consider the following two different sensor models : 1 .
a pmu securely measures the voltage phasor of the bus at which it is installed .
a pmu securely measures the voltage phasor of the bus at which it is installed , as well as the current phasors on all the lines connected to this bus .
we denote the set of buses with pmus by @xmath25 , and let @xmath26 be the total number of pmus , where @xmath27 denotes the cardinality of a set . without loss of generality ( wlog ) , we choose one of the buses in @xmath28 to be the angle reference bus .
we say that a power injection attack @xmath29 is _ unobservable _ if it leads to _ zero _ changes in all the quantities measured by the pmus . with the first pmu model described above
, we have the following definition : an attack @xmath30 is unobservable if and only if @xmath31 where @xmath32 denotes the @xmath33 sub - vector of @xmath34 obtained by keeping its @xmath35 entries whose indices are in @xmath28 .
is a weighted laplacian matrix , the elements of @xmath36 sum to @xmath37 . ] with the second pmu model described above , for any bus @xmath38 , it is immediate to verify that the following three conditions are equivalent : 1 .
there are no changes of the voltage phasor at @xmath7 and of the current phasors on all the lines connected to @xmath7 .
there are no changes of the voltage phasor at @xmath7 and of the power flows on all the lines connected to @xmath7 .
3 . @xmath39 $ ] , there is no change of the voltage phasor at @xmath40 , where @xmath41 $ ] is the closed neighborhood of @xmath7 which includes @xmath7 and its neighboring buses @xmath42 .
thus , for forming unobservable attacks , the following two situations are equivalent to the attacker : * the system operator monitors the set of buses @xmath28 with the second pmu model ;
* the system operator monitors the set of buses @xmath43 $ ] with the first pmu model , where @xmath43 $ ] is the closed neighborhood of @xmath28 which includes all the buses in @xmath28 and their neighboring buses @xmath44 .
thus , the unobservability condition with the second pmu model is obtained by replacing @xmath28 with @xmath43 $ ] in .
wlog , we employ the first pmu model in the following analysis , and based on the discussion above all the results can be directly translated to the second pmu model . in forming an unobservable attack ,
an attacker generally has two objectives : minimize execution complexity and maximize its impact on the grid . note that these two objectives can be competing interests that are not simultaneously achievable .
we will first focus on finding the minimum execution complexity for an attack to be unobservable , which constitutes the main part of this work . among attacks with the minimum complexity
, we then find the one with the maximum impact . for an attack vector @xmath29
, we use its zero norm @xmath45 to model its execution complexity .
this is because attackers are typically resource - constrained , and can choose only a limited number of buses to implement attacks . for minimizing attack complexity ,
an attacker is interested in finding the sparsest attacks that satisfy the unobservability condition : @xmath46 since @xmath47 , a more compact form of is as follows : @xmath48 where @xmath49 denotes the complement of @xmath35 , and @xmath50 is the submatrix of @xmath51 formed by choosing all its rows and a set of columns @xmath52 .
we now note that problem is np - hard : specifically , as a special case of the cospark problem of a matrix @xcite problem resembles a security index problem discussed in @xcite , which has been proven to be np - hard . under some special problem settings for data injection attacks , problems of this type
have been shown to be solvable exactly in polynomial time @xcite @xcite @xcite . in general ,
low complexity heuristics have been developed for solving @xmath2 minimization problems ( e.g. , @xmath53 relaxation ) .
we now generalize our model to allow a subset of buses to be `` unalterable buses '' , meaning that their nodal power injection can not be changed by attackers .
this allows us to model the following scenarios : * a `` zero injection '' bus that simply connects multiple lines without nodal generation or load , and hence its power injection is always zero and can not be changed .
* a `` protected '' bus by the system operator , and its power injection is not accessible by the attacker .
we denote the set of unalterable buses by @xmath54 .
the other buses @xmath55 are termed `` alterable '' buses .
generalizing , the sparsest unobservable attack problem is established as follows : @xmath56 when @xmath57 , reduces to .
generalizing , eq . has the following equivalent form : @xmath58 _ given the locations of the sensors _
@xmath28 , we now introduce a variation of the graph @xmath19 that will prove key to developing the main results later .
[ gmdef ] given a set of buses @xmath59 , @xmath60 is defined to be the following augmented graph based on @xmath19 : 1 .
@xmath60 includes all the buses in @xmath19 , and has one additional unalterable dummy bus .
2 . define an augmented set @xmath61 that contains @xmath28 and the unalterable dummy bus .
@xmath60 includes all the edges of @xmath19 , and an edge is added between every pair of buses in @xmath61 , and its weight can be chosen arbitrarily as any positive number .
we note that the dummy bus is only connected to the set of sensors @xmath28 .
we observe the following key facts .
first , an unobservable attack in the original graph @xmath19 leads to zero changes in all the voltage phase angles in @xmath28 .
thus , any line between a pair of buses in @xmath28 would see a zero change of the power flow on it .
it is then clear that the added dummy bus and lines in @xmath60 do not lead to any power flow changes in the network under any unobservable attack .
we thus have the following lemma : [ lemaug ] an attack is unobservable by @xmath28 in @xmath19 if and only if it is unobservable by @xmath28 in @xmath60 .
this allows us to work with the augmented graph @xmath60 instead of @xmath19 .
it is clear that the weights of the added edges in @xmath60 do not matter for lemma [ lemaug ] to hold .
in this section , we address the following question whose solutions will be useful in solving the minimum sparsity problem : assuming that the attacker can only alter the power injections at a subset of the buses , denoted by @xmath62 , does there _ exist _ an attack that is unobservable by a set of pmus @xmath28 ? for any given @xmath63 ,
a feasible non - zero attack @xmath64 must satisfy @xmath65 .
in other words , it must not alter the power injections at the buses in @xmath66 . from , there exists an unobservable non - zero attack if and only if @xmath67 since @xmath68 , we have that is equivalent to @xmath69 where @xmath70 is the submatrix of @xmath51 formed by its rows @xmath66 and columns @xmath52 .
an illustration of is depicted in figure [ feasfig ] , where the submatrix formed by the shaded blocks represents @xmath70 . from
, we have the following lemma on the feasibility condition of unobservable attacks .
[ lemfeas1 ] given @xmath63 and @xmath28 , there exists an unobservable non - zero attack if and only if @xmath70 is column rank deficient . to analyze when this column rank deficiency condition , @xmath71 , is satisfied , we start with the following observations based on the fact that @xmath51 is the laplacian of the weighted graph @xmath19 . 1 . the _ signs _ ( @xmath72 , @xmath73 , or @xmath37 ) of the entries of @xmath51 are fully determined by the _ network topology _ : @xmath74 2 .
the _ values _ of the non - zero entries of @xmath51 are determined by the _ line reactances _
when all the line reactances in the power network are known , so are the entries of the submatrix @xmath70 , and it is immediate to compute whether @xmath77 . _ without _ knowing the exact values of any line reactances , we will show that whether @xmath77 can be determined almost surely by computing the _ structural rank _ of @xmath70 , defined as follows @xcite .
[ indent ] a set of independent entries of a matrix @xmath78 is a set of nonzero entries , no two of which lie on the same line ( row or column ) .
the structural rank of a matrix @xmath78 , denoted by @xmath79 , is the maximum number of elements contained in at least one set of independent entries .
a basic relation between the structural rank and the rank of a matrix is the following @xcite , @xmath80 in the literature , structural rank is also termed `` generic rank '' @xcite .
specifically , we consider _ generic _ power grid parameters , i.e. , we assume that the line reactances @xmath81 are independent , but not necessarily identical random variables drawn from continuous probability distributions .
we assume that the reactances are bounded away from zero from below ( as lines do not have zero reactances in practice ) .
as such , the analysis in this work is along the line of _ structural properties _ as in @xcite and @xcite , and we will develop results that hold _ with probability one_. we believe the independence ( but not identically distributed ) assumption is sufficiently general in practice . in particular , there are uncertainties in factors that influence the reactance of a line ( e.g. the distance that a line travels , the degradation of a line over time ) .
these uncertainties can be modeled as independent ( but not identically distributed ) random variables , leading to the model employed in this paper . clearly
, @xmath70 is always column rank deficient when @xmath82 .
next , we discuss the case of @xmath83 .
we begin with the special case @xmath84 , for which we have the following lemma whose proof is relegated to appendix [ secprfleml ] : [ leml ] let @xmath85 be the laplacian of a connected graph @xmath19 with strictly positive edge weights . for any set of node indices @xmath86 , denote by @xmath87 the submatrix of @xmath51 formed by its rows @xmath88 and columns @xmath88 .
then @xmath89 , @xmath87 is of full rank .
note that lemma [ leml ] holds deterministically without assuming generic edge weights of the graph .
for the case of @xmath84 , we let @xmath90 , and lemma [ leml ] proves that @xmath91 .
this implies the intuitive fact that there exists no attack restricted to @xmath63 that is unobservable by a set of pmus @xmath92 .
now , we address the general case of arbitrary @xmath63 and @xmath28 . we have the following theorem demonstrating that having @xmath93 _ almost surely guarantees _ @xmath94 .
the proof is relegated to appendix [ secprffeasthm ] .
[ feasthm ] for a connected weighted graph @xmath95 , assume that the edge weights are independent continuous random variables strictly bounded away from zero from below , and denote the laplacian of @xmath19 by @xmath85 .
then , any @xmath96 submatrix of @xmath51 , with @xmath97 , has a rank of @xmath98 with probability one if it has a structural rank of @xmath98 . from theorem [ feasthm ] , with @xmath83 , if @xmath99 , we have with probability one that @xmath100 , and there exists no attack restricted to @xmath63 that is unobservable by a set of pmus @xmath28 .
it has been known in the literature that ( see e.g. , @xcite ) , a full structural rank of a matrix leads to a full rank matrix with probability one , as long as the nonzero entries in the matrix are drawn independently from continuous probability distributions . however , it is worth noting that this is not sufficient for proving theorem [ feasthm ] .
this is because , as in theorem [ feasthm ] , we are interested in matrices that are _ submatrices of a graph laplacian _ : even with the edge weights of the graph drawn independently , the entries in these submatrices are _ correlated _ due to the special structure of a graph laplacian .
such correlation leads to technical difficulties for the proof , which can be overcome as shown in appendix [ secprffeasthm ] .
we note that the structural rank of a matrix can be computed in polynomial time by finding the maximum bipartite matching in a graph @xcite .
since whether an entry of @xmath51 is non - zero is solely determined by the topology of the network , we have the following corollary . given @xmath63 and @xmath28 , whether a non - zero unobservable attack exists can be determined with probability one based solely on the knowledge of the grid topology .
in this section , we study the problem of finding the sparsest unobservable attacks given any set of pmus @xmath28 ( cf . ) .
as remarked in section [ secmm ] , @xmath2 minimization such as is np - hard .
we will show that the sparsest unobservable attack can in fact be found in _
polynomial time with probability one_. we first introduce a key concept a _ vulnerable vertex cut_. we then state our main theorem that yields an explicit solution for the sparsest unobservable attack problem .
we prove that this solution both upper and lower bounds the optimum of , hence proving the theorem .
we start with the following basic definitions : a vertex cut of a connected graph @xmath19 is a set of vertices whose removal renders @xmath19 disconnected .
the vertex connectivity of a graph @xmath19 , denoted by @xmath101 , is the size of the minimum vertex cut of @xmath19 , i.e. , it is the minimum number of vertices that need to be removed to make the remaining graph disconnected . from the definition of the augmented graph @xmath60 in section [ sec : aug ] , since all the buses in @xmath61 ( containing @xmath28 and the dummy bus ) are pair - wise connected , we have the following lemma : [ auglem ] for any vertex cut of the augmented graph @xmath60 , there is no pair of the buses in @xmath61 that are disconnected by this cut .
accordingly , we introduce the following notations which will be used later on : [ ntn1 ] given a vertex cut of @xmath60 , we denote the set of buses disconnected from @xmath61 after removing the cut set by @xmath102 . the cut set itself is denoted by @xmath103 . with the vertex cut @xmath103
, @xmath60 is partitioned into three subgraphs : 1 .
@xmath102 , which does not contain any bus in @xmath61 , i.e. , @xmath104 .
2 . @xmath103 , which is the vertex cut set itself , and may contain buses in @xmath61 .
3 . @xmath105 $ ] , which contains ( not necessarily exclusively ) all the remaining buses in @xmath61 after removing the cut set .
an illustrative example with a cut @xmath103 of size 2 is depicted in figure [ 3spgen ] in section [ upsec ] .
we note that there is a slight abuse of notation in @xmath106 : in general , a cut set does not necessarily consist of exactly all the neighboring nodes of @xmath107 .
nonetheless , as will be shown in the remainder of the paper , we need only care about the _ minimum _
cut set , which indeed consists of exactly all the neighboring nodes of @xmath107 , namely , @xmath106 .
leveraging the above notation , we now introduce a key type of vertex cut on @xmath60 .
a vulnerable vertex cut of a connected augmented graph @xmath60 is a vertex cut @xmath103 for which @xmath108\vert \ge \vert n({\mathcal}{s})\vert+1 $ ] . in other words ,
the number of _ alterable _ buses in @xmath109 $ ] is no less than the cut size plus one .
the reason for calling such a vertex cut `` vulnerable '' will be made exact later in section [ upsec ] .
the basic intuition is the following . in order to have @xmath110 ( unobservability ) ,
the key is to have the phase angle changes on the cut @xmath103 be zero , with power injection changes ( which can only happen on the alterable buses ) restricted in @xmath109 $ ] .
as will be shown later , this can be achieved if a cut @xmath103 is `` vulnerable '' as defined above .
we note that it is possible that no vulnerable vertex cut exists ( e.g. , in the extreme case that all buses are unalterable ) .
accordingly , we define the following variation on the vertex connectivity .
the vulnerable vertex connectivity of an augmented graph @xmath60 , denoted by @xmath111 , is the size of the minimum vulnerable vertex cut of @xmath60 .
if no vulnerable vertex cut exists , @xmath111 is defined to be @xmath112 .
we note that the concepts of vulnerable vertex cut and vulnerable vertex connectivity do not apply to the original graph @xmath19 .
we immediately have the following lemma : [ kplem ] if a vulnerable vertex cut exists , then @xmath113 .
suppose a vulnerable vertex cut exists , and @xmath114 .
denote the minimum vulnerable vertex cut by @xmath103 ( cf .
notation [ ntn1 ] ) . now consider the set @xmath28 : it is a vertex cut of @xmath60 that separates the dummy bus and @xmath115 .
because there are at least @xmath116 alterable buses in @xmath109\subseteq n[\bar{m}^c]$ ] , @xmath28 is also a _ vulnerable vertex cut_. this contradicts the minimum vulnerable vertex cut having size at least @xmath117 .
we now state the following theorem that gives an explicit solution of the sparsest unobservable attack problem in terms of the vulnerable vertex connectivity @xmath111 .
[ mainthm ] for a connected grid @xmath118 , assume that the line reactances @xmath119 are independent continuous random variables strictly bounded away from zero from below .
given any @xmath28 and @xmath54 , the minimum sparsity of unobservable attacks , i.e. , the global optimum of , equals @xmath120 with probability one .
we note that finding the minimum vulnerable vertex connectivity of a graph is computationally efficient . for polynomial time
algorithms we refer the readers to @xcite and @xcite . in particular , vertex cuts are enumerated @xcite starting from the minimum and with increasing sizes , until a minimum vulnerable vertex cut is identified .
we now prove theorem [ mainthm ] by upper and lower bounding the minimum sparsity of unobservable attacks in the following two subsections .
we show that _ any _ vulnerable vertex cut @xmath103 provides an upper bound on the optimum of as follows .
[ upthm ] for a connected grid @xmath19 and a set of pmus @xmath28 , for any vulnerable vertex cut of @xmath60 denoted by @xmath103 ( cf .
notation [ ntn1 ] ) , there exists an unobservable attack of sparsity no higher than @xmath121 . a vulnerable vertex cut @xmath103 partitions @xmath60 into @xmath102 , @xmath103 and @xmath122}$ ] , with @xmath123 .
similarly to the range space interpretation of the sparsest unobservable attack , it is sufficient to show that there exists a non - zero vector in the range space of @xmath124 such that i ) it has a sparsity no higher than @xmath121 , and ii ) non - zero power injections occur only at the alterable buses . by re - indexing the buses , wlog , i ) let @xmath125 , and ii ) let @xmath124 have the following partition as depicted in figure [ multicol ] : 1 .
the top submatrix @xmath126 is an @xmath127 matrix .
the middle submatrix ( which will be shown to be @xmath128 ) consists of all the remaining rows , each of which has at least one _ non - zero _ entry .
3 . the bottom submatrix is an _ all - zero _ matrix . in particular , from the definition of the laplacian ,
the middle submatrix of @xmath124 , as described above , is exactly @xmath128 because its row indices correspond to those buses not in @xmath102 but connected to at least one bus in @xmath102 . from the definition of the vulnerable vertex cut , @xmath108\vert \ge \vert n({\mathcal}{s})\vert+1 $ ]
now , pick any set of @xmath129 alterable buses in @xmath130 $ ] , denote this set by @xmath63 , and denote the other buses in @xmath109 $ ] by @xmath131\backslash { \mathcal}{a}$ ] . clearly , @xmath132 .
therefore , @xmath133 ( which is a submatrix of @xmath134{\mathcal}{s}}$ ] ) has @xmath135 columns but only @xmath136 rows , and is hence column rank deficient .
now , we let @xmath137 be a non - zero vector in the null space of @xmath133 : @xmath138 then , we construct an attack vector @xmath139 : it has some possibly non - zero values at the indices that correspond to @xmath63 , and has _
zero values at all other indices . _
thus , @xmath140 theorem [ upthm ] explains our terminology of a `` vulnerable vertex cut '' , since if a vertex cut is vulnerable , it leads to an unobservable attack . if a vulnerable vertex cut of @xmath60 exists , applying theorem [ upthm ] to the _ minimum _ one , we have that the optimum of is upper bounded by @xmath120 .
if no vulnerable vertex cut exists , @xmath141 is a trivial upper bound .
we now provide a graph - theoretic interpretation of theorem [ upthm ] .
as shown in figure [ multicol ] and [ 3spgen ] , all the buses can be partitioned into three subsets @xmath142 and @xmath105 $ ] , corresponding to the row indices of the top , middle and bottom submatrices of @xmath124 , respectively .
@xmath103 is a vulnerable vertex cut of @xmath60 that separates @xmath102 from @xmath105 $ ] .
the sparse attack @xmath29 ( cf . )
is formed by injecting / extracting power at @xmath121 alterable buses in @xmath109 $ ] , such that the phase angle changes at @xmath143 are all zero .
note that @xmath144 .
the example with @xmath145 in figure [ 3spgen ] illustrates a @xmath146-sparse attack with power injection / extractions at ( assumed alterable ) buses @xmath147 and @xmath148 , such that the phase angle changes at @xmath143 are all zero .
we end this subsection by introducing a notion of `` potential impact '' of unobservable attacks .
we make the following observation : as long as an attacker takes control of all the power injections in a vulnerable vertex cut @xmath103 ( assuming they are alterable ) , it can always _ cancel out the effects of anything that happens within @xmath109 $ ] _ on the measurements taken in @xmath149 .
thus , by taking control of all the buses in @xmath103 , an attacker can successfully _ hide _ from the system operator a power injection attack with a zero norm as large as @xmath150 the potential impact of unobservable attacks associated with a vulnerable vertex cut @xmath103 is defined as @xmath151|$ ] .
definition [ impactdef ] is one characterization of attack impact based solely on graph theoretic properties . in practice , there are many different notions of attack impact depending on , e.g. , the interpretation of the attacks and the operating objective of the system . employing definition [ impactdef ] , we can _ differentiate _ the potential impacts of multiple sparsest unobservable attacks _ with the same sparsity_. an illustration is depicted in figure [ impfig ] . in this example , two vulnerable vertex cuts both of size two , @xmath152 and @xmath153 , are enclosed by solid ovals .
accordingly , both cuts enable 3-sparse unobservable attacks . however , their potential impacts are significantly different .
cut @xmath154 only disconnects one other bus , namely @xmath155 from the set of pmus @xmath28 , and hence its potential impact equals @xmath156| = 3 $ ] . in comparison ,
cut @xmath157 disconnects all the vertices above @xmath157 from @xmath28 , and hence its potential impact equals @xmath158| \gg 3 $ ] . with this definition of potential impact ,
it is then natural for an attacker to _ seek the sparsest unobservable attack with the greatest potential impact_. as an immediate byproduct of the analysis of potential impact , by letting @xmath159 , we obtain the _ maximum _ potential impact of all unobservable attacks in a power network : [ maxcor ] for a connected power grid @xmath118 , given any @xmath28 denoting the pmu locations , the maximum potential impact among all the unobservable attacks equals @xmath160|$ ] .
we first define the following property of a matrix @xmath161 , which will be shown to be equivalent to having @xmath162 .
[ cond1def ] @xmath163 we have the following lemma whose proof is relegated to appendix [ secprfcond1 ] : [ cond1equ ] property [ cond1def ] is equivalent to having @xmath162 .
we now prove the lower bounding part of theorem [ mainthm ] , namely , with probability one , all unobservable power injection attacks @xmath36 must have @xmath164 .
the key idea is in showing that the equivalence between property [ cond1def ] and a full structural rank ( cf .
lemma [ cond1equ ] ) implies a connection between the vulnerable vertex connectivity and the feasibility condition of unobservable attacks ( cf .
lemma [ lemfeas1 ] ) .
we focus on @xmath60 and consider its corresponding laplacian @xmath51 .
suppose there exists a power injection attack @xmath165 such that @xmath166 denote the buses with non - zero power injection changes by @xmath62 , and hence @xmath65 . from , @xmath167 , and @xmath168 , implying that @xmath70 is column rank deficient .
we first consider the case that a vulnerable vertex cut exists , i.e. , @xmath169 .
the proof for the case of @xmath170 follows similarly . for notational simplicity
, we will use @xmath171 instead of @xmath111 in the remainder of the proof .
[ [ if - a - vulnerable - vertex - cut - exists - i.e .- barkappa - infty ] ] if a vulnerable vertex cut exists , i.e. , @xmath172 + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + we will prove that , for all @xmath173 with @xmath174 , _
@xmath70 is of full column rank with probability one _ , i.e. , can only happen with probability zero . from lemma [ kplem ] , @xmath175 .
it is then sufficient to prove for the `` worst cases '' with @xmath176 , i.e. , @xmath177 and @xmath70 is a square matrix . from theorem [ feasthm ] and lemma [ cond1equ ] , it is sufficient to show that _
@xmath70 satisfies property [ cond1def ] _ , and hence is of full rank with probability one .
recall from the definition of the laplacian @xmath51 that , for any column ( or row ) of @xmath51 , @xmath178 , its non - zero entries correspond to bus @xmath179 and those buses that are connected to bus @xmath179 . with this
, we now prove that @xmath70 satisfies property [ cond1def ] .
consider any set of @xmath7 ( @xmath180 ) buses in @xmath66 , denoted by @xmath181 .
\i ) if @xmath182 : based on the definition of the laplacian @xmath51 , the @xmath7 columns of @xmath183 _ that correspond to the buses @xmath181 themselves _ each has at least one non - zero entry .
ii ) if @xmath184 : we prove that @xmath185 must contain at least @xmath171 buses .
this is because , otherwise , @xmath186 , contradicting that @xmath171 is the minimum size of vulnerable vertex cuts for the following reasons : 1 .
@xmath187 , and thus @xmath188 has at least @xmath189 alterable buses . 2 .
@xmath186 implies that @xmath190 , and thus @xmath185 is a vertex cut that separates @xmath181 and @xmath191 .
3 . because @xmath184 and @xmath28 are pairwise connected in @xmath60 , @xmath192 $ ] . thus ,
@xmath191 and @xmath28 are disjoint . from 1 ) ,
3 ) , and the fact that @xmath186 , we observe that @xmath185 is a _ vulnerable vertex cut _ of size @xmath193 , contradicting @xmath171 being the vulnerable vertex connectivity . now ,
based on the definition of the laplacian @xmath51 , the @xmath194 submatrix _
@xmath195 must have at least @xmath196 columns each of which has at least one non - zero entry for the following reasons : _ * the @xmath7 columns of @xmath195 that correspond to the buses @xmath181 themselves each has at least one non - zero entry . * as @xmath181 are connected to at least @xmath171 other buses , each one of these @xmath171 neighbors of @xmath181 corresponds to one column of @xmath195 that has at least one non - zero entry .
accordingly , the @xmath197 submatrix @xmath183 has at least @xmath7 columns each of which has at least one non - zero entry . summarizing i ) and
ii ) , @xmath70 satisfies property [ cond1def ] , and is thus of full column rank with probability one .
therefore , can only happen with probability zero .
[ [ if - no - vulnerable - vertex - cut - exists - i.e .- barkappa - infty ] ] if no vulnerable vertex cut exists , i.e. , @xmath198 + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + if @xmath199 , i.e. , all buses have pmus , then clearly no unobservable attack exists .
we now focus on @xmath200 .
suppose @xmath201 .
consider the set @xmath202 containing @xmath28 and the dummy bus .
@xmath203 ( cf . )
implies that @xmath204 $ ] , and thus @xmath205 $ ] has at least @xmath206 alterable buses . since @xmath207 separates the dummy node and @xmath208
, @xmath28 is a _ vulnerable vertex cut_. this contradicts the nonexistence of a vulnerable vertex cut .
therefore , @xmath209 . in this case , the same proof as in the above case i ) when a vulnerable vertex cut exists applies , and can only happen with probability zero . with the proofs of upper and lower bounds ,
we have now proved theorem [ mainthm ] . in addition , from the proof of theorem [ upthm ] , we have a _ constructive solution _ of the sparsest unobservable attack in polynomial time .
we conclude this section by noting the following fact similar to that in section [ secfeas ] : the minimum sparsity of unobservable attacks is fully determined with probability one by the _ network topology , the locations of the alterable buses , and the locations of the pmus_. @ l @ l @ p8.0 cm @ + + + & if no unobservable attack exists given the current set of pmus @xmath28 , stop .
+ & step 1 : find all the minimum vulnerable vertex cuts of @xmath210}$ ] ; + & among them , find the cut with the greatest potential impact , + & denoted by @xmath211})$ ] .
+ & step 2 : among all the buses disconnected from @xmath43 $ ] by @xmath211})$ ] + & as well as those in the cut set @xmath211})$ ] , place the next pmu + & at the one such that the resulting maximum potential impact + & among all the remaining unobservable attacks is minimized . +
in this section , we evaluate the sparsest unobservable attacks and their potential impacts when the system operator deploys pmus at optimized locations .
we first provide an efficient algorithm for optimizing pmu placement by the system operator .
next , we provide comprehensive evaluation of our analysis and algorithms in multiple ieee power system test cases as well as large - scale polish power systems .
our matlab codes are openly available for download .
we have seen in section [ secmin ] that the minimum sparsity and potential impacts of unobservable attacks are determined fully by the network topology , the locations of the alterable buses , and the pmu placement .
note that , unlike network states and parameters which can vary over short and medium time scales , the transmission network topology and the alterable buses typically stay the same over relatively long time scales .
this motivates the system operator to optimize the pmu placement according to this information . for the best performance in countering power injection attacks
, the system operator wants to _ raise the minimum sparsity of unobservable attacks , as well as mitigate the maximum potential impact of unobservable attacks_. algorithm 1 ( cf . table [
alg1table ] ) is developed for the system operator to greedily place pmus to pursue both objectives . in this algorithm
, we have assumed that the _ second pmu model _ in section [ secsensemod ] is employed , and the algorithm can be adapted to the first pmu model by replacing @xmath43 $ ] with @xmath28 .
algorithm 1 is essentially a successive cut / attack elimination procedure .
the purpose of step 1 is to identify the sparsest unobservable attack with the greatest potential impact .
specifically , step 1 can be performed as follows : 1 .
assign arbitrarily one of the buses in @xmath28 as the _ source _ node ; 2 .
for each of the buses in @xmath212 $ ] , assign it as the _ destination _ node , and compute all the minimum vulnerable vertex cuts that separate such a source - destination pair .
3 . among all the computed source - destination vertex cuts that have the same minimum size ,
compute their corresponding potential impacts , and select the minimum vertex cut with the greatest potential impact , denoted by @xmath211})$ ] .
we note that all the minimum vulnerable vertex cuts can be enumerated in polynomial time ( c.f .
@xcite ) . in our numerical evaluation
using matlab on a laptop with intel core i7 3.1-ghz cpu and 8 gb of ram , it takes less than @xmath213 seconds on average for every pmu placed for the ieee 300 bus systems .
this per - pmu time increases to about 50 seconds for the polish 3012 bus system . in step 2 ,
our primary goal is to ensure that the cut set @xmath211})$ ] found in step 1 _ does not remain a legitimate vertex cut _ after placing the next pmu .
this can be achieved by placing the next pmu among the buses disconnected from @xmath43 $ ] by @xmath211})$ ] as well as those in @xmath211})$ ] . among such candidate buses
, we choose the one that renders the _ minimum _ maximum potential impact among all the remaining unobservable attacks ( cf .
corollary [ maxcor ] ) had the next pmu been placed at it .
we evaluate our results in the ieee 30-bus , ieee 57-bus , ieee 118-bus , ieee 300-bus , polish 2383-bus , polish 2737-bus , and polish 3012-bus systems .
the evaluation is performed based on the software toolbox matpower @xcite . in each of these systems , we apply algorithm 1 to generate a set of pmu locations greedily , with the number of pmus @xmath35 increasing from one until all attacks become observable . moreover , from algorithm 1 , for all @xmath35 , the minimum sparsity of unobservable attacks as well as the maximum potential impact among the sparsest unobservable attacks are found ( cf .
step 1 in algorithm 1 ) .
we assume that all buses are alterable in the test cases . in general ,
for a given set of pmus , one can also search for the maximum potential impact among all @xmath214-sparse unobservable attacks for any given sparsity @xmath214 , ( as opposed to evaluate that among the sparsest attacks only as in algorithm 1 ) .
however , this problem is np - hard in @xmath214 . in light of this , we selectively focused on some level of sparsity of unobservable attacks that is _ not _ minimally sparse , and evaluated their maximum potential impacts .
specifically , the minimum sparsity of unobservable attacks and the maximum potential impact among these sparsest attacks both as functions of the number of pmus @xmath35 are plotted for the ieee 30 and 118-bus power systems and the polish 3012-bus system , in figures [ eval30 ] , [ eval118 ] and [ eval3012 ] respectively . in addition , * for the ieee 30-bus system , the maximum potential impact among all 2-sparse , 3-sparse , 4-sparse and 5-sparse unobservable attacks for the entire range of @xmath35 are plotted .
( note that the minimum sparsity of unobservable attacks does not exceed 3 for all @xmath35 ) .
* for the ieee 118-bus system , the maximum potential impact among all 3-sparse attacks when @xmath215 is plotted .
( note that for @xmath216 the minimum sparsity of unobservable attacks is 2 ) .
we make the following observations which appear in all seven of the evaluated systems : * in all seven systems , all the attacks become observable with _ less than a third _ of the buses installed with pmus ( assuming the second pmu model ) .
the average percentage of the number of pmus needed to have full network observability equals @xmath217 .
while this number resembles a well - known estimate of such percentage to be one third @xcite , it also demonstrates the efficacy of algorithm 1 in pmu placement . *
the topologies of the tested power systems tend to allow sparse power injection attacks . in other words ,
the vertex connectivity of these power networks is often small .
furthermore , there are often many unobservable attacks with the same minimum sparsity : this is why even after adding a lot more pmus into the network , with each addition eliminating the previous sparsest attack , the minimum sparsity of an unobservable attack can still remain the same . *
while there are many unobservable attacks with the same sparsity , the potential impacts among them can vary significantly .
moreover , as more pmus are added , the maximum potential impact among all the sparsest unobservable attacks drops quickly until it reaches the minimum sparsity .
similar behavior is demonstrated for all the @xmath214-sparse unobservable attacks ( @xmath218 ) for the ieee 30-bus system as shown in figure [ eval30 ] .
we have studied physical attacks that alter power generation and loads in power networks while remaining unobservable under the surveillance of system operators using pmus . given a set of pmus
, we have first shown that the existence of an unobservable attack that is restricted to any given subset of the buses can be determined with probability one by computing the structural rank of a submatrix of the network laplacian @xmath51 .
next , we have provided an explicit solution to the open problem of finding the sparsest unobservable attacks : the minimum sparsity among all unobservable attacks equals @xmath0 with probability one .
the constructive solution allows us to find all the sparsest unobservable attacks in polynomial time . as a result
, @xmath0 is a fundamental limit of this minimum sparsity that is not only explicitly attainable , but also unbeatable by all possible unobservable attacks .
we have then introduced a notion of potential impacts of unobservable attacks . for the system operator to raise the minimum sparsity
while simultaneously mitigating the maximum potential impact of all unobservable attacks , we have devised an efficient algorithm of greedily placing the pmus . with optimized pmu deployment , we have evaluated the sparsest unobservable attacks and their potential impacts in the ieee 30 , 57 , 118 , 300-bus systems and the polish 2383 , 2737 , 3012-bus systems . finally , while this work has studied a static system model and power injection attacks , extension to dynamic systems , measurements and power injection attacks
remains an interesting future direction , for which we expect that similar insights will apply .
first , we denote the laplacian of the _ induced subgraph _ @xmath219 $ ] by @xmath220 .
denote the number of connected components of the induced subgraph @xmath219 $ ] by @xmath221 . by properly re - indexing the nodes
, we have @xmath222 where @xmath220 is a block - diagonal matrix whose each block @xmath223 is _ positive semidefinite _ and corresponds to one connected component of @xmath219 $ ] , @xmath224 and @xmath225 is diagonal , which we write in a block diagonal form whose each block @xmath226 is itself a diagonal matrix with _ non - negative _ entries , @xmath227 since the original graph @xmath19 is connected , each connected component of the induced subgraph @xmath219 $ ] must be connected to at least one node in @xmath228 .
this implies the following fact : now , for any non - zero vector @xmath229 , we write it as a concatenation of @xmath221 sub - vectors : @xmath230^t~ [ \bm{x}_{\mathcal}{i}^2]^t \ldots [ \bm{x}_{\mathcal}{i}^c]^t]^t,\end{aligned}\ ] ] where the length of each sub - vector @xmath231 follows the size of the sub - matrix @xmath232 . 1 . if @xmath234 , then immediately @xmath235 .
if @xmath236 , then @xmath237 , which implies @xmath238 namely , @xmath239 is in the null space of @xmath232 . note that as @xmath232 corresponds to a single connected component of @xmath219 $ ] , the dimension of the null space of @xmath232 is one , and is spanned by the all one vector @xmath240^t$ ] with the appropriate length .
thus , @xmath239 must be in the form of @xmath241 , for some @xmath242 .
from fact 1 , @xmath243 has non - negative diagonal entries with at least one of them strictly positive , and we have @xmath244^t \bm{d}_{\mathcal}{i}^j\bm{x}_{\mathcal}{i}^j > 0 $ ] , and hence @xmath245 .
[ lemcond1 ] for a matrix @xmath161 , if the following conditions are satisfied , @xmath315,i } \text { has at least one } { \nonumber}\\ & ~~~~~~~~~~~~~~~~~~\text{non - zero entry , } \label{cond13}\end{aligned}\ ] ] \ii ) assume that the lemma is true for all @xmath317 .
for @xmath318 : first , because the upper left @xmath319 submatrix of @xmath78 satisfies the induction assumption , each of @xmath78 s first left @xmath320 columns must contain at least one non - zero entry .
from , the last column of @xmath78 has at least one non - zero entry .
thus , the case of @xmath321 in property [ cond1def ] holds for @xmath78 . *
if the last @xmath7 rows of @xmath78 are selected to form @xmath323 , ( i.e. @xmath325 ) , from , the columns @xmath326 each has one non - zero entry , namely , @xmath327 .
* otherwise , there exists a row @xmath328 , @xmath329 , which is not selected in @xmath323 ( cf .
figure [ cond1fig ] ) . in this case
, the row indices of @xmath323 can be partitioned into two subsets : @xmath330 and @xmath331 . on the one hand ,
note that the upper left @xmath332 submatrix of @xmath78 satisfies the induction assumption .
thus , _ among the first @xmath333 columns _ of the rows @xmath334 , there exists @xmath179 columns each of which has one non - zero entry . on the other hand , from , @xmath335
are all non - zero , and none of these non - zero entries appears in the first @xmath333 columns .
therefore , there exist @xmath7 columns of @xmath323 such that each of them has at least one non - zero entry . for @xmath343 , from the induction assumption b ) , there exists a non - zero permutated diagonal for the @xmath344 submatrix @xmath345,[1:n]}$ ] : wlog , assume that it corresponds to @xmath346 ( cf .
figure [ cond1fig2 ] . )
now , we use proof by contradiction , and assume that @xmath323 does _ not _ have a non - zero permuted diagonal .
then , the sub - row @xmath347}$ ] must be all zero , because otherwise any non - zero entry within @xmath347}$ ] will form a non - zero permuted diagonal of @xmath323 with @xmath348 . from property [ cond1def ] ,
the @xmath349 row of @xmath323 must have at least one non - zero entry .
wlog , assume that @xmath350 is non - zero .
then , the sub - row @xmath351}$ ] must be all - zero , because otherwise any non - zero entry within @xmath351}$ ] will form a non - zero permuted diagonal of @xmath323 with @xmath350 and @xmath352 . from property [ cond1def ] , the first two rows of @xmath323 must have at least two columns each of which has at least one non - zero entry .
since @xmath353 , there is at least one more column of @xmath354,[1,n]}$ ] that has at least one non - zero entry .
wlog , assume that the sub - column @xmath354,2}$ ] has at least one non - zero entry , ( cf . the shaded area in figure [ cond1fig2 ] ) .
similarly , consider the @xmath355 row of @xmath323 .
note that the submatrix @xmath356 , [ 1:r-1]}$ ] satisfies , and , and hence satisfies property [ cond1def ] by lemma [ lemcond1 ] . from the induction assumption ii )
, @xmath356 , [ 1:r-1]}$ ] has a non - zero permuted diagonal .
then , the sub - row @xmath357}$ ] must be all - zero , because otherwise any non - zero entry within @xmath357}$ ] will form a non - zero permuted diagonal of @xmath323 with the non - zero permuted diagonal of @xmath356 , [ 1:r-1]}$ ] and @xmath358 . therefore , the submatrix @xmath359,[n , n]}$ ] must be all - zero .
this implies that there are only @xmath360 ( instead of @xmath7 ) columns of @xmath323 each of which has at least one non - zero entry , and hence contradicts with property [ cond1def ] .
yue zhao ( s06 , m11 ) is an assistant professor of electrical and computer engineering at stony brook university .
he received the b.e .
degree in electronic engineering from tsinghua university , beijing , china in 2006 , and the m.s . and ph.d .
degrees in electrical engineering from the university of california , los angeles ( ucla ) , los angeles in 2007 and 2011 , respectively .
his current research interests include smart grid , renewable energy integration , optimization theory , stochastic control , and statistical signal processing .
andrea goldsmith ( s90 , m93 , sm99 , f05 ) is the stephen harris professor in the school of engineering and a professor of electrical engineering at stanford university .
she was previously on the faculty of electrical engineering at caltech .
goldsmith co - founded and served as cto for two wireless companies : wildfire.exchange , which develops software - defined wireless network technology for cloud - based management of wifi systems , and quantenna communications , inc . , which develops high - performance wifi chipsets .
she has previously held industry positions at maxim technologies , memorylink corporation , and at&t bell laboratories .
she is a fellow of the ieee and of stanford , and has received several awards for her work , including the ieee comsoc edwin h. armstrong achievement award as well as technical achievement awards in communications theory and in wireless communications , the national academy of engineering gilbreth lecture award , the ieee comsoc and information theory society joint paper award , the ieee comsoc best tutorial paper award , the alfred p. sloan fellowship , the wice outstanding achievement award , and the silicon valley / san jose business journal s women of influence award .
she is author of the book `` wireless communications '' and co - author of the books `` mimo wireless communications '' and `` principles of cognitive radio , '' all published by cambridge university press , as well as an inventor on 28 patents .
she received the b.s .
, m.s . and ph.d .
degrees in electrical engineering from u.c .
berkeley .
goldsmith has served as editor for the ieee transactions on information theory , the journal on foundations and trends in communications and information theory and in networks , the ieee transactions on communications , and the ieee wireless communications magazine as well as on the steering committee for the ieee transactions on wireless communications .
she participates actively in committees and conference organization for the ieee information theory and communications societies and has served on the board of governors for both societies .
she has also been a distinguished lecturer for both societies , served as president of the ieee information theory society in 2009 , founded and chaired the student committee of the ieee information theory society , and chaired the emerging technology committee of the ieee communications society . at stanford
she received the inaugural university postdoc mentoring award , served as chair of stanford s faculty senate in 2009 , and currently serves on its faculty senate , budget group , and task force on women and leadership .
h. vincent poor ( s72 , m77 , sm82 , f87 ) received the ph.d .
degree in electrical engineering and computer science from princeton university in 1977 . from 1977 until 1990
, he was on the faculty of the university of illinois at urbana - champaign . since 1990
he has been on the faculty at princeton , where he is the dean of engineering and applied science , and the michael henry strater university professor of electrical engineering .
he has also held visiting appointments at several other institutions , most recently at imperial college and stanford .
his research interests are in the areas of information theory , stochastic analysis and statistical signal processing , and their applications in wireless networks and related fields such as smart grid . among his publications in these areas
is the recent book mechanisms and games for dynamic spectrum allocation ( cambridge university press , 2014 ) .
poor is a member of the national academy of engineering and the national academy of sciences , and is a foreign member of the royal society .
he is also a fellow of the american academy of arts and sciences and of other national and international academies .
he received a guggenheim fellowship in 2002 and the ieee education medal in 2005 .
recent recognition of his work includes the 2014 ursi booker gold medal , the 2015 eurasip athanasios papoulis award , the 2016 john fritz medal , and honorary doctorates from aalborg university , aalto university , hkust , and the university of edinburgh .
first , for a matrix @xmath246 with a full structural rank , we define an equivalent term , `` a non - zero permuted diagonal '' , for a set of @xmath247 independent entries ( cf .
definition [ indent ] ) .
this term is based on the following intuition : for example , for @xmath161 , a non - zero permuted diagonal ( i.e. , a set of @xmath3 independent entries ) corresponds to a permutation function @xmath248 , such that @xmath249 . for a @xmath253 submatrix of @xmath51 with a non - zero permuted diagonal ,
we denote it by @xmath254 .
we denote the set of row indices of @xmath51 that are selected in forming @xmath254 by @xmath255 , and similarly the set of selected column indices by @xmath256 : @xmath257 clearly , if @xmath258 , @xmath254 is of full rank from lemma [ leml ] . now , consider the case that @xmath259 , where @xmath260 .
in other words , @xmath88 denotes the common indices that appear in both the row indices @xmath261 and the column indices @xmath262 , @xmath263 denotes the indices that appear in @xmath261 but not in @xmath262 , and @xmath264 denotes the indices that appear in @xmath262 but not in @xmath261 .
wlog , @xmath254 has the form as in figure [ ijk11 ] , in which the common row and column indices @xmath88 are located in the upper left part of @xmath254 , and @xmath254 consists of four blocks @xmath265 .
since @xmath254 has a non - zero permuted diagonal , there exists a permutation function @xmath266 , such that @xmath267 .
in other words , the mapping @xmath268 forms a bijection between @xmath269 and @xmath270 , such that @xmath271 we now consider the following two cases : wlog , assume that the entry @xmath274 in @xmath273 is on a non - zero permuted diagonal of @xmath254 , i.e. , @xmath275 . as in figure
[ ijk12 ] , we partition @xmath254 into @xmath276 , [ 1:t ] } , \bm{b}'_{t+1 , [ 1:t ] } , \bm{b}'_{[1:t ] , t+1}$ ] and @xmath274 . because @xmath274 is on a non - zero permuted diagonal of @xmath254 , the @xmath277 submatrix @xmath276 , [ 1:t]}$ ] has a non - zero permuted diagonal . from the induction assumption ,
@xmath276 , [ 1:t]}$ ] is of full rank with probability one .
when @xmath276 , [ 1:t]}$ ] is of full rank , let @xmath278 , [ 1:t]}\bm{b}'_{[1:t ] , t+1}.\end{aligned}\ ] ] thus , @xmath276 , t+1 } = \bm{b}'_{[1:t ] , [ 1:t]}\bm{\alpha}$ ] .
then , @xmath254 is rank - deficient if and only if @xmath279 } \bm{\alpha } \label{rankdef1}\end{aligned}\ ] ] note that , except for @xmath280 itself , there are only three other entries in the laplacian @xmath51 that are correlated with @xmath274 : @xmath281 however , as @xmath282 , _ none _ of the above three entries is selected into the submatrix @xmath254 .
therefore , _
@xmath274 is independent to all other entries in @xmath254 _ , and is hence independent to @xmath283 } \bm{\alpha}$ ] .
because @xmath274 is drawn from a continuous distribution , the probability that is satisfied is zero . as a result
, @xmath254 is of full rank with probability one .
wlog , assume that the entry @xmath287 in @xmath286 is on a non - zero permuted diagonal of @xmath254 , i.e. , @xmath288 .
as in figure [ ijk21 ] , we partition @xmath254 into @xmath289 } , \bm{b}'_{[2:t+1 ] , [ 1:t ] } , \bm{b}'_{1 , t+1}$ ] and @xmath290,t+1}$ ] . because @xmath287 is on a non - zero permuted diagonal of @xmath254 , the @xmath277 submatrix @xmath290 , [ 1:t]}$ ] has a non - zero permuted diagonal . from the induction assumption ,
@xmath290 , [ 1:t]}$ ] is of full rank with probability one .
when @xmath290 , [ 1:t]}$ ] is of full rank , let @xmath291 , [ 1:t]}\bm{b}'_{[2:t+1 ] , t+1}. \label{alphadef}\end{aligned}\ ] ] thus , @xmath290 , t+1 } = \bm{b}'_{[2:t+1 ] , [ 1:t]}\bm{\alpha}$ ] .
then , @xmath254 is rank - deficient if and only if @xmath292 } \bm{\alpha } \label{rankdef2}\end{aligned}\ ] ] note that @xmath293 .
we have @xmath294 where @xmath295 , and @xmath296 is _ independent _ to @xmath287 .
substitute for @xmath297 in , we have @xmath298 note that @xmath287 is _ independent to @xmath299 , and independent to the right hand side of _ . because @xmath287 is drawn from a continuous distribution , if @xmath300 , the probability ( conditioned on @xmath300 ) that is satisfied is zero .
next , we prove that the probability of @xmath301 is zero . from , if @xmath301 , @xmath302 , t+1 } = \sum_{j=1}^t\alpha_j\bm{b}'_{[2:t+1 ] , j } { \nonumber}\\ \leftrightarrow~ & \bm{b}'_{[2:t+1 ] , 1 } + \bm{b}'_{[2:t+1 ] , t+1 } = \sum_{j=2}^t\alpha_j\bm{b}'_{[2:t+1 ] , j}.\end{aligned}\ ] ] thus , @xmath301 implies that @xmath290 , 1 } + \bm{b}'_{[2:t+1 ] , t+1}$ ]
is in the range space of @xmath290 , [ 2:t]}$ ] .
note that , all the entries in the two vectors @xmath290 , 1}$ ] and @xmath290 , t+1}$ ] are mutually independent non - diagonal entries of @xmath51 .
now , consider that we make the following _ change of distributions _ of certain entries in @xmath254 ( and also @xmath51
correspondingly ) : * as in figure [ ijk22 ] , @xmath307 , the _ only _ entry in @xmath290 , [ 2:t]}$ ] that is correlated with @xmath308 and @xmath309 is @xmath310 where @xmath311 .
note that @xmath312 depends on @xmath313 only via the _ sum _ @xmath305 .
* @xmath314 , @xmath308 and @xmath309 are independent to all the entries in @xmath290 , [ 2:t]}$ ] .
the joint distribution of @xmath290 , 1}$ ] and @xmath290 , [ 2:t]}$ ] after the change of distributions is equal to the joint distribution of @xmath290 , 1 } + \bm{b}'_{[2:t+1 ] , t+1}$ ] and @xmath290 , [ 2:t]}$ ] before the change .
we now note that , after the change of distributions , the @xmath277 matrix @xmath290 , [ 1:t ] } = [ \bm{b}'_{[2:t+1 ] , 1 } ~\bm{b}'_{[2:t+1 ] , [ 2:t]}]$ ] still satisfies the induction assumption , and is hence of _ full rank with probability one_. thus , before the change of distributions , @xmath290 , 1 } + \bm{b}'_{[2:t+1 ] , t+1}$ ] falls in the range space of @xmath290 , [ 2:t]}$ ] _ with probability zero_. therefore , @xmath301 with probability zero , hence the probability that is satisfied is zero . as a result
, @xmath254 is of full rank with probability one . | physical security of power networks under power injection attacks that alter generation and loads is studied .
the system operator employs phasor measurement units ( pmus ) for detecting such attacks , while attackers devise attacks that are _ unobservable _ by such pmu networks .
it is shown that , given the pmu locations , the solution to finding the sparsest unobservable attacks has a simple form with probability one , namely , @xmath0 , where @xmath1 is defined as the vulnerable vertex connectivity of an augmented graph .
the constructive proof allows one to find the entire set of the sparsest unobservable attacks in polynomial time .
furthermore , a notion of the potential impact of unobservable attacks is introduced . with optimized pmu deployment
, the sparsest unobservable attacks and their potential impact as functions of the number of pmus are evaluated numerically for the ieee 30 , 57 , 118 and 300-bus systems and the polish 2383 , 2737 and 3012-bus systems .
it is observed that , as more pmus are added , the maximum potential impact among all the sparsest unobservable attacks drops quickly until it reaches the minimum sparsity . |
one of the main limiting factors in determining the masses of the compact objects in low - mass x - ray binaries is uncertainty in the fraction of the infrared light that is produced by the companion star ( e.g. , hynes , robinson , & bitner 2005 ) .
when lmxbs are accreting at low rates , the optical and infrared light from these systems is dominated by the companion stars , so their mass functions can be measured to high accuracy based on the doppler motion of the companion ( e.g. , * ? ? ?
the inclination then must be constrained by modeling the modulations in the optical and infrared light curves that are produced by the varying aspect of the distorted , roche - lobe - filling companion . generally ,
the contribution of the accretion disk to the optical and near - infrared emission is uncertain , and varying the fraction of the light that is assumed to be produced by the accretion disk can lead to differences of a factor of 2 in the derived mass of a compact object ( e.g. , gelino , harrison , & orosz 2001 ) .
fortunately , the broad - band spectrum of a multi - temperature accretion disk is significantly flatter than that of a stellar photosphere , so mid - infrared observations could constrain the relative contributions of the two components .
however , the spectra of lmxbs in the mid - infrared have not been well studied , and several indirect lines of evidence suggest that these systems might contain circumbinary material that could emit in the mid - infrared .
gfirst , some of the white dwarf analogs to lmxbs , cataclysmic variables ( cvs ) , exhibit spectral features that lie at the mean radial velocities of the systems @xcite and excess mid - infrared emission @xcite that could be interpreted as arising in circumbinary material ( see also * ? ? ?
* ; * ? ? ?
second , the supernovae that produced the compact objects could have left fall - back disks around the binaries .
indeed , the first fall - back disk has recently been found around a young , highly - magnetized neutron star ( wang , chakrabarty , & kaplan 2006 ) .
third , the planets around the isolated millisecond pulsar psr 1257@xmath412 @xcite could not have survived the supernova that produced the neutron star , and must have formed afterward ( see , e.g. , * ? ? ?
millisecond pulsars are usually assumed to have been spun up by accretion as lmxbs , in which case planets could form from material present during the binary phase . therefore , to search for evidence of circumbinary material ,
we have observed four nearby , quiescent lmxbs with the _ spitzer _ space telescope .
we chose the lmxbs in our sample to be detectable with _ spitzer _ if they contained optically - thick circumbinary disks passively illuminated by the mass donor stars ( see * ? ? ?
* ; * ? ? ?
* and below ) .
based on a simple model , we chose sources with : ( 1 ) @xmath5 magnitudes brighter than 17 , ( 2 ) locations more than @xmath6 in projection from the galactic center , and ( 3 ) no 2mass sources within 5 that were brighter than our targets . in table [ tab : targets ] , we list the positions , the orbital periods , estimates of the primary masses , the spectral types of the companions , and the quiescent @xmath5 magnitudes of the four systems in our sample ( see the table notes for references ) .
our measurements were taken in the 4.5 and 8.0 @xmath0 m bands with infrared array camera ( irac ) , and in the 24 @xmath0 m band with the multiband imaging photometer ( mips ; tab .
[ tab : fluxes ] ) .
we used the post - basic - calibration data provided by the _ spitzer _ science center ( ssc ) for most of our analysis .
however , the mips image of a 0620@xmath100 contained latent features with a low spatial frequency and a @xmath72% amplitude that were left by a previous observation of a bright , extended source .
we corrected the image by creating a flat field from the median of the individual dithered images , dividing each snapshot by the flat , and re - creating the mosaicked image using the script provided by the ssc .
three - color images centered on each lmxb are displayed in figure [ fig : img ] .
each target is detected at 4.5 and 8.0 @xmath0 m .
only a 0620@xmath100 and gs 2023 + 338 are also detected at 24 @xmath0 m . we computed the irac fluxes of each source using the point - spread - function - fitting routine apex from the ssc , and the mips fluxes and upper limits using aperature photometry ( tab . [ tab : fluxes ] and fig . [ fig : sed ] ) .
llcccccccc[htp ] a 0620@xmath100 & v616 mon & 95.68561 & @xmath8 & [email protected] & 7.8 & 8.712.9 & k4v & 1.2 & 14.55(6 ) + gs 2023 + 338 & v404 cyg & 306.01594 & @xmath10 & 2.23.7 & 155.3 & 10.113.4 & k0iii & 4.0 & 12.50(5 ) + xte j1118 + 480 & kv uma & 169.54498 & @xmath11 & [email protected] & 4.1 & 6.57.2 & k6v & 0.06 & 16.9(2 ) + cen x-4 & v822 cen & 224.59135 & @xmath12 & 1.2 & 15.1 & 1.4 & k5v & 0.3 & 14.66(8 ) lcccccccccc[htb ] a 0620@xmath100 & 2005 mar 25 & 400 & 448(13 ) & 194 & 249(10 ) & 149 & 2005 mar 06 & 180 & 54(18 ) & 43 + gs 2023 + 338 & 2004 oct 09 & 36 & 3020(90 ) & 670 & 1450(40 ) & 500 & 2004 oct 16 & 30 & 153(70 ) & 46 + xte j1118 + 480 & 2004 nov 21 & 400 & 46(1 ) & 17 & 45(7 ) & 34 & 2005 may 13 & 240 & @xmath1316 & @xmath1316 + cen x-4 & 2004 aug 12 & 300 & 199(6 ) & @xmath190 & 95(17 ) & @xmath114 & 2005 aug 28 & 150 & @xmath1330 & @xmath1330 to understand the origin of the mid - infrared emission , in figure [ fig : sed ] we plot for each lmxb the observed and de - reddened fluxes in the infrared and optical bands ( from tab . [ tab : fluxes ] and the references in tab .
[ tab : targets ] ) .
the optical and near - infrared intensities of quiescent lmxbs often vary by several tenths of a magnitude on time scales of years ( not counting outbursts ; * ? ? ?
* ; * ? ? ?
* ) , presumably because of changes in the accretion flow .
none of the fluxes were obtained simultaneously , so when comparing the flux measurements in figure [ fig : sed ] one can only expect them to be self - consistent to within @xmath1430% . to estimate the contributions of the companions to the spectra , we have obtained model stellar spectra computed with the next - generation phoenix code ( kindly provided by t. barman ; * ? ? ?
* ) that correspond to the temperatures and radii of the spectral types in table [ tab : targets ] .
we have estimated the contribution of the accretion disk using a standard @xcite model , assuming the disk extends out to the tidal truncation radius ( @xmath1560% of the binary separation ; frank , king , & raine 1992 ) , is inclined by 60@xmath16 to our line of sight , and has the largest accretion rate that is consistent with the optical photometry ( see the caption of fig . [ fig : sed ] for values ) . we have confirmed that our conclusions are robust against changing the assumed spectral types of the companions within the values reported in the literature , and against varying the parameters of the accretion disk so long as the models are consistent with the optical photometry .
we find that the spectrum of cen x-4 is consistent with that expected for the companion star , with a possible contribution at long wavelengths from the accretion disk .
there are clear excesses of mid - infrared flux above that expected from the companions of a 0620 - 00 , gs 2023 + 338 , and xte j1118 + 480 . in order to conservatively estimate the amounts of the mid - infrared fluxes that are not produced by the companions
, we computed the maximum contributions of the companions photospheres by normalizing the model stellar spectra to match the de - reddened @xmath5 fluxes , and reported the difference between the predicted and observed mid - infrared fluxes as the excesses in table [ tab : fluxes ] ( @xmath17 ) .
several lmxbs have now been detected in outburst in the mid - infrared ( e.g. , smith , beall , & swain 1990 ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) , but these are the first detections of quiescent systems . in outburst , the mid - infrared emission could originate from an accretion disk in which @xmath18 g s@xmath19 @xcite , from free - free emission in a strong wind driven from the accretion disk @xcite , or from radio jets with flat spectra extending to infrared wavelengths @xcite .
we use these hypotheses as a starting point for understanding the mid - infrared emission from our sample of quiescent lmxbs
. the mid - infrared excess from gs 2023 + 338 could originate in an accretion disk , because the 2.224 @xmath0 m fluxes in figure [ fig : sed ] follow the @xmath20 law that one would expect from the rayleigh - jeans tail of a black body .
we find that our model for a viscously - heated accretion disk @xcite from 2 adequately describes the excess mid - infrared flux from gs 2023 + 338 in figure [ fig : sed ] .
in contrast , the 424 @xmath0 m fluxes from a 0620@xmath100 and xte j1118 + 480 do not follow the @xmath21 law expected for the rayleigh - jeans tail of a blackbody , because there is too much flux at 8 @xmath0 m . a blackbody spectrum that peaks near 8 @xmath0 m would have a temperature of only 640 k. for these two sources , we plot blackbody spectra that match the peak of the mid - infrared excesses ( @xmath22 ) in figure [ fig : sed ] .
the sum of the spectra of stellar photospheres and optically - thick blackbodies match the observed mid - infrared fluxes well .
however , the inferred emitting areas are large compared to the binary separations . for a planck spectrum , the solid angles of the emitting regions correspond to circular radii of @xmath23 where @xmath24 is the distance to the source .
in contrast , the disks will be contained within @xmath1560% of the binary separations @xmath25 , which are given by : @xmath26 where @xmath27 is the orbital period in days , @xmath28 is the mass of the accretor in solar masses , and @xmath29 is the mass ratio of the binary ( e.g. , * ? ? ? * ) . given the fluxes in table [ tab : fluxes ] , we find that the excess mid - infrared emission originates from regions @xmath142 times larger than the orbital separations of gs 2023 + 338 and xte j1118 + 480 , and @xmath304 times larger than the accretion disks .
therefore , the mid - infrared emission is produced in a region that extends beyond the binary orbit .
jets could produce an emitting region larger than the binary orbit , but given the spectra of the mid - infrared excesses we suggest that jets are only minor contributors .
two lmxbs in our sample have recently been detected as radio sources in quiescence : gs 2023 + 338 with flat - spectrum radio emission with an intensity of 350@xmath0jy over the frequency range 1.48.4 ghz @xcite , and a 0620@xmath100 with a radio flux of 50 @xmath0jy at 8.4 ghz @xcite .
the 24@xmath0 m excesses from a 0620@xmath100 is equal to the flux in the radio , so it could be produced by a jet with a flat ( @xmath31 , with @xmath32 ) spectrum between the radio and mid - infrared . for gs 2023 + 338 ,
the 24 @xmath0 m flux is slightly lower than the radio flux , which would imply that its jet has a steeper @xmath33 spectrum .
however , the 8 @xmath0 m excesses from a 0620@xmath100 , gs 2023 + 338 , and xte j1118 + 480 are not consistent with a flat - spectrum radio jet , because for xte j1118 + 480 and a 0262@xmath100 they are 4 times larger than the excesses at 24 @xmath0 m , and for gs 2023 + 338 it is 10 times larger .
if we assume the 24 @xmath0 m emission is from a flat - spectrum jet , then it contributes @xmath13 25% to the 8 @xmath0 m flux from xte j1118 + 480 and a 0620@xmath100 , and @xmath1310% from gs 2023 + 338 .
we suggest that the excesses from a 0620@xmath100 and xte j1118 + 480 originate from circumbinary dust that re - processes the light of the companions . to estimate the masses of the dust , we assume it is contained in optically thick disks
. the disks could lie as close to the center of masses of the binaries as 1.7@xmath25 , at which point they would be tidally truncated @xcite .
we can estimate the temperature profile of such disks by assuming they are flat , in which case @xmath34 where @xmath35 and @xmath36 are the temperatures and radii of the disks , and @xmath37 and @xmath38 are the temperatures and radii of the companions ( e.g. , * ? ? ?
* ; * ? ? ?
using equation [ eq : orbit_sep ] and the parameters in table [ tab : targets ] , we would expect the inner edges of the circumbinary disks to have temperatures of @xmath14600 k , which is consistent with our detection of excesses that peak near 8 @xmath0 m .
the lack of excess flux at 24@xmath0 m implies that material does not re - process much stellar light beyond radii @xmath143 times larger than the binary separations .
if we assume the disks are composed of dust with an opacity @xmath39@xmath7300 @xmath40 g@xmath19 at 8 @xmath0 m ( e.g. , * ? ? ?
* ) , then for them to have an optical depth of @xmath41@xmath71 , the disks would only need to contain @xmath7@xmath2 g of dust . if we assume the emitting material is optically thin , we find dust masses that are similar within a factor of a few ( e.g. , eq .
3 in * ? ? ?
if the gas - to - dust ratio is similar to that of the interstellar medium , @xmath7100 , then the total mass of the circumbinary material could be @xmath7@xmath3 g.
the detection of excess mid - infrared flux from a 0620@xmath100 and xte j1118 + 480 ( fig .
[ fig : sed ] ) provides evidence that circumbinary material is present around some lmxbs , but its spectrum suggests that it will only be detectable in the mid - infrared . the excess emission peaks at 8 @xmath0 m , and if we model it as a single - temperature blackbody , we predict that the circumbinary material produced negligible flux shortward of @xmath143 @xmath0 m
. moreover , the lack excess mid - infrared emission from cen x-4 demonstrates that such material is not ubiquitous .
finally , for gs 2023 + 338 , the mid - infrared emission appears to originate from a hot accretion disk , and the contribution of the accretion disk to the mid - infrared light can be predicted based on the optical measurements ( the dotted line in fig .
[ fig : sed ] ) .
these results provide reassurance that efforts to determine the inclinations of lmxbs by modeling the ellipsoidal modulations of their infrared and optical light curves are not compromised by the presence of circumbinary material . for a 0620@xmath100 and xte j1118 + 480
, we suggest that the circumbinary disks are either the remains of fall - back disks produced in the supernovae that formed the compact objects ( e.g. , * ? ? ?
* ; * ? ? ?
* ) , or material injected into circumbinary orbits during the process of mass loss by the roche - lobe filling companions ( e.g. , * ? ? ?
these circumbinary disks contain @xmath7@xmath2 g of dust , which for a standard dust - to - mass ratio of @xmath7100 implies a total mass of @xmath7@xmath3 g. if the disks are produced by matter ejected into circumbinary orbits during the process of mass transfer , then it represents only a tiny fraction of the @xmath7@xmath42 g ( @xmath70.1 @xmath43 ) that will be lost by the companions over the lifetimes of these lmxbs .
alternatively , @xcite have proposed that fall - back disks containing @xmath30@xmath3 g of material could form asteroids and inject them into the magnetospheres of pulsars with a high enough rate to explain the the observed intermittency in their radio pulses .
asteroids might not survive the outbursts of lmxbs @xcite , but the mid - infrared excesses that we have identified could be the remnants of similar fall - back disks
. however , the circumbinary material is not massive enough either to affect the evolution of the orbital angular momentum of lmxbs ( e.g. , @xmath7@xmath44 g in * ? ? ?
* ) , or to form planetesimals ( e.g. , @xmath7@xmath45 g in * ? ? ?
as the formation and evolution of circumbinary matter around lmxbs and cvs are considered further , it may be that the paucity of matter around a 0620@xmath100 and xte j1118 + 480 ( and the apparent lack of circumbinary matter around the one neutron star lmxb in our sample , cen x-4 ) may be the most lasting aspect of this result .
we are grateful to t. barman for providing the model stellar spectra , to e. gallo and s. migliari for alerting us to a mistake in our original photometry , to m. jura for conversations about circumstellar dust , to j. mcclintock for discussions about the contribution of the accretion disk to the mid - infrared light , and to the referee for comments that clarified the text .
support for this work was provided by nasa through an award issued by jpl / caltech .
0 casares , j. , charles , p. a. , naylor , t. , & pavlenko , e. p. 1993
, , 265 , 834 chevalier , c. , ilovaisky , s. a. , van paradijs , j. , pedersen , j. , & van der klis , m. 1989 , , 210 , 114 cordes , j. m. & shannon , r. m. 2006 , astro - ph/0605145 draine , b. t. & lee , h. m. 1984 , , 285 , 89 dubus , g. , campbell , r. , kern , b. , taam , r. e. , & spruit , h. c. 2004 , , 349 , 869 evans , a. , geballe , t. r. , rawlings , j. m. c. , eyres , s. p. s. , & davies , j. k. 1997 , , 292 , 192 fender , r. p. 2001
, , 322 , 31 frank , j. , king , a. , & raine , d. 1992 , `` accretion power in astrophysics '' , cambridge university press gallo , e. , fender , r. p. & hynes , r. i. 2005 , , 356 , 1017 gallo , e. , fender , r. p. , miller - jones , j. c. a. , merloni , a. , jonker , p. g. , heinz , s. , maccarone , t. j. , & van der klis , m. 2006 , submitted to , astro - ph/0605376 gelino , d. m. , harrison , t. e. , & orosz , j. a. 2001 , , 122 , 2668 harrison , t. e. & gehrz , r. d. 1991 , , 101 , 587 harrison , t. e. & gehrz , r. d. 1994 , , 108 , 1899 hauschildt , p. h. , lowenthal , d. k. , & baron , e. 2001 , , 134 , 323 homan , j. , buxton , m. , markoff , s. , bailyn , c. d. , nespoli , e. , & belloni , t. 2005 , , 624 , 295 howell , s. b. 2006 , in press , astro - ph/0606202 hynes , r. i. , robinson , e. l. , & bitner , m. 2005 , , 630 , 405 jura , m. 2003 , , 584 , l91 mcclintock , j. e. , narayan , r. , garcia , m. r. , orosz , j. a. , remillard , r. a. , murray , s. s. 2003 , , 593 , 435 mcclintock , j. e. , & remillard , r. a. 2006 , to appear in `` compact stellar x - ray sources , '' eds . w. h. g. lewin & m. van der klis , cambridge university press , astro - ph/0306213 migliari , s. , tomsick , j. a. , maccarone , t. j. , gallo , e. , fender , r. p. , nelemands , g. , & russell , d. m. 2006 , astro - ph/0605051 mikoajewsa , k. , ruthowski , a. , gonalves , d. r. , & szostek , a. 2005 , , 362 , l13 miller , m. c. & hamilton , d. p. 2001
, , 550 , 863 ruden , s. p. & pollack , j. b. 1991 , , 375 , 740 shahbaz , t. , naylor , t. , & charles , p. a. 1993 , , 265 , 655 shakura , n. i. & sunyaev , r. a. 1973 , , 24 , 337 smith , h. a. , beall , j. h. & swain , m. r. 1990 , , 99 , 273 solheim , s .- e . & sion , e. m. 1994 , , 287 , 503 taam , r. e. , sandquist , e. l. , & dubus , g. 2003 , , 592 , 1124 thorsett , s. e. & chakrabarty , d. 1999 , , 512 , 288 torres , m. a. p. , callanan , p. j. , garcia , m. r. , zhao , p. , laycock , s. , & kong , a. k. h. 2004 , torrres , m. a. p. , casares , j. , martnez - pais , i. g. , & charles , p. a. 2002 , , 334 , 233 van paradijs , j. , telesco , c. m. , kouveliotou , c. , & fishman , g. j. 1994 , , 429 , 19 wang , z. , chakrabarty , d. , & kaplan , d. l. 2006 , , 440 , 772 wolszczan , a. & frail , d. a. 1992 , _ nature _ , 355 , 145 zurita , c. , casares , j , martinez - pais , i. g. , piccioni , a. , bernabei , s. , bartolini , c. , & guarnieri , a. 2002 , iauc 7868 | we report the discovery of excess 4.5 and 8@xmath0 m emission from three quiescent black hole low - mass x - ray binaries , a 0620@xmath100 , gs 2023 + 338 , and xte j1118 + 480 , and the lack of similar excess emission from cen x-4 .
the mid - infrared emission from gs 2023 + 338 probably originates in the accretion disk . however , the excess emission from a 0620@xmath100 and xte j1118 + 480 is brighter and peaks at longer wavelengths , and most likely originates from circumbinary dust that is heated by the light of the secondary star .
for these two sources , we find that the inner edges of the dust distributions lie near 1.7 times the binary separations , which are the minimum radie at which circumbinary disks would be stable against tidal disruption .
the excesses are weak at 24 @xmath0 m , which implies that the dust does not extend beyond about 3 times the binary separations .
the total masses of circumbinary material are between @xmath2 and @xmath3 g. the material could be the remains of fall - back disks produced in supernovae , or material from the companions injected into circumbinary orbits during mass transfer . |
the study of quantum mechanical hamiltonians that are @xmath0-symmetric but not hermitian@xcite-@xcite has recently found an unexpected application in classical optics@xcite-@xcite , due to the fact that in the paraxial approximation the equation of propagation of an electromagnetic wave in a medium is formally identical to the schrdinger equation , but with different interpretations for the symbols appearing therein .
the equation of propagation takes the form [ opteq ] i=-(+v(x ) ) , where @xmath6 represents the envelope function of the amplitude of the electric field , @xmath5 is a scaled propagation distance , and @xmath7 is the optical potential , proportional to the variation in the refractive index of the material through which the wave is passing .
a complex @xmath8 corresponds to a complex refractive index , whose imaginary part represents either loss or gain . in principle the loss and gain regions can be carefully configured so that @xmath8 is @xmath0 symmetric , that is @xmath9 .
propagation through such a medium exhibits many new and interesting properties , such as nonreciprocal diffraction @xcite and birefringence @xcite .
one of the main features of complex optical lattices is the non - conservation of the total power . in the @xmath0-symmetric case
this can lead to effects such as power oscillations @xcite .
it has been argued that one can distinguish three universal dynamics @xcite related to broken or unbroken symmetry . while this is in general true
, the behaviour can be modified considerably for special initial conditions , as we will discuss in the present paper .
many familiar effects such as bloch oscillations and dynamical localisation get drastically modified in the presence of imaginary potentials and @xmath0-symmetry @xcite .
the new features of complex optical lattices provide exciting opportunities for engineering applications . as an example
, the possibility of realizing unidirectional light propagation has been envisaged @xcite . in the case of high intensities
the propagation equation ( [ opteq ] ) gets modified due to the kerr - nonlinearity , leading to an additional term proportional to @xmath10 .
it has been shown in @xcite that the influence of the nonlinearity on the non - reciprocal effects can be advantageous for applications such as unidirectional couplers .
it is interesting to note that the nonlinear propagation equation also has a counterpart in quantum dynamics , as the mean - field description of bose - einstein condensates , where there has also been interest in @xmath0 symmetric models @xcite .
however , for the purposes of this paper , we shall limit ourselves to the linear case . a model system exemplifying some of the novel features of beam propagation in @xmath0-symmetric optical lattices uses the sinusoidal potential [ vsin ] v = v_0 .
this model has been studied numerically and theoretically , e.g. in refs .
the propagation in @xmath5 of the amplitude @xmath6 is governed by the analogue schrdinger equation ( [ opteq ] ) , which for an eigenstate of @xmath11 , with eigenvalue @xmath12 and @xmath5-dependence @xmath13 reduces to the eigenvalue equation [ h ] --v_0= .
these eigenvalues are real for @xmath2 , which corresponds to unbroken @xmath0 symmetry , where the eigenfunctions respect the ( anti - linear ) symmetry of the hamiltonian . above @xmath4 pairs of complex conjugate eigenvalues begin to appear , and indeed above @xmath14 all the eigenvalues are complex@xcite
. clearly one would expect oscillatory behaviour of the amplitude below the threshold at @xmath4 and exponential behaviour above the threshold , but the precise form of the evolution at @xmath4 is less obvious . at first sight
one would expect linear growth ( see , e.g. ref .
@xcite ) because of the appearance of jordan blocks associated with the degenerate eigenvalues that merge at that value of @xmath15 , but , as longhi@xcite has emphasized , this behaviour can be significantly modified depending on the nature of the initial wave packet .
it is this problem that we wish to discuss in the present paper . in section 2
we explicitly construct the bloch wave - functions and the associated jordan functions corresponding to the degenerate eigenvalues and then use the analogue of the method of stationary states to construct the @xmath5-dependence .
we find that the explicit linear dependence arising from the jordan associated functions is indeed cancelled by the combined contributions from the non - degenerate wave - functions ( which individually give an oscillatory behaviour ) . in section 3
we analyze this cancellation in detail , showing how the coefficients of the two contributions are closely related , and obtaining an approximate analytic expression for the @xmath5-derivative of their sum .
our conclusions are given in section 4 .
at the threshold @xmath4 , the potential @xmath8 in eq .
( [ vsin ] ) becomes the complex exponential @xmath16 , for which the schrdinger equation reads [ h1 ] --v_0(2ix / a)= .
this is a form of the bessel equation , as can be seen by the substitution @xmath17 , where @xmath18 , giving [ bessel ] y^2+y-(y^2+q^2)=0 , where @xmath19 .
thus the spectrum is that of a free massive particle , shown in the reduced zone scheme in fig . 1 , and for @xmath20 not an integer the solutions @xmath21 and @xmath22 are linearly independent , and have exactly the correct periodicity , @xmath23 , to be the bloch wave - functions .
it is important to note , however , that because the original potential is @xmath0-symmetric rather than hermitian , these functions are not orthogonal in the usual sense , but rather with respect to the @xmath0 inner product ( see eq .
( [ orthogonality ] ) ) .
however , for @xmath24 , a non - zero integer , @xmath25 and @xmath26 are no longer independent , but are in fact equal , signalling the degeneracy of the eigenvectors at those points , and the formation of spectral singularities and jordan blocks . in that case
the bloch eigenfunctions do not form a complete set , and we must search for other functions , still with the same periodicity , to supplement them .
these are the jordan associated functions ( see appendix and refs .
@xcite ) , which we denote by @xmath27 , defined not by the eigenvalue equation itself , but by [ jordan ] _
n(y)=i_n(y ) , and the periodicity condition @xmath28 , corresponding to the bloch periodicity @xmath29 . a particular solution of this equation , which can be expressed explicitly in terms of generalized hypergeometric functions , is given by [ particular ] _
n^pi(y)=_y_0^y i_n(z ) , as is easily checked by differentiation and use of the wronskian identity @xmath30 .
however , the corresponding @xmath31 has a discontinuity in its imaginary part at @xmath32 and does not have the required periodicity .
this problem can be rectified by recognizing that we may add to @xmath33 any multiple of @xmath25 , or more importantly @xmath34 .
the latter displays exactly the same kind of discontinuity has a branch cut along the negative real axis , and when @xmath35 passes through @xmath36 one should really go to the next riemann sheet . ] as @xmath33 , and by choosing its coefficient judiciously the discontinuities can be made to cancel . specifically ,
we take [ fixed ] _
n(y)=_n^pi(y)-k_n(y ) . an alternative derivation of this relation using the definition of @xmath37 in terms of @xmath38 will be given in the next section .
with the addition of the last term the resulting @xmath37 is not only free of discontinuities , but also obeys the correct bloch periodicity condition .
this is shown in figure 2 , where we plot the real and imaginary parts of @xmath31 and @xmath37 for @xmath39 .
note the @xmath0 symmetry of @xmath37 , namely @xmath40 .
\(a ) ( b ) the complete set of functions , orthogonal with respect to the @xmath0 metric , now consists of the bloch eigenfunctions @xmath41 , supplemented by the jordan associated functions @xmath37 for @xmath42 with @xmath43 , and a general wave - function @xmath44 may be expanded as [ ptexpansion ] f(x)=c_0 _ 0(x)+_kn / a c_k _ k(x)+_n > 0(_n i_n(y)+_n _
n(y ) ) . here
we have discretized the problem by putting the system in a box of length @xmath45 , in which case @xmath46 . the coefficients @xmath47 can be obtained@xcite by using the @xmath0 orthogonality [ orthogonality ] dx _
-k(x)_k(x)=_k kdx _ -k(x)_k(x ) , as [ ptcoeffs ] c_k= . here
the sign of the denominator alternates from band to band , reflecting the indefinite nature of the @xmath0 metric . however , this relation breaks down precisely at the brillouin zone boundaries , where the single bloch eigenfunctions are self - orthogonal , @xmath48 , another indication that we need the supplementary jordan functions .
thus it is @xmath49 , rather than @xmath50 , which is determined by integrating @xmath44 with respect to @xmath25 : [ jordancoeff ] _
n = , while @xmath50 is subsequently determined by integrating @xmath44 with respect to @xmath51 . as an immediate check of the correctness of the functions @xmath52 , the identity [ one ] 1=i_0(y)-2i_2(y)+2i_4(y)- implies that dx _ 2n(y)=2(-1)^n dx i_2n(y)_2n(y ) , for @xmath43 , a relation we have verified numerically . in standard
quantum mechanics one method of solving the time - dependent schrdinger equation i(x , t)=h ( x , t ) is to expand the initial wave - function @xmath53 in terms of the ( complete ) set of orthonormal eigenfunctions @xmath54 as ( x,0)=_m a_m _ m(x ) , with the coefficients @xmath55 given by the overlap a_m = dx ^*_m(x)(x,0 ) . the wave - function at time @xmath56 is then given by ( x , t)=_m a_m e^-ie_m t _
here we have essentially the same problem , with @xmath5 taking the role of @xmath56 , but with the crucial difference that we must include the jordan associated functions in the sum , which now takes the form of eq .
( [ ptexpansion ] ) , with the coefficients determined as in eqs .
( [ ptcoeffs ] ) and ( [ jordancoeff ] ) . as is well known , and has been emphasized again recently in refs .
@xcite and@xcite , the time dependence then takes a different form .
the jordan function @xmath57 satisfies the equation @xmath58 , and hence e^-iht_r & = & e^-ie_r t e^-i(h - e_r)t _ r&= & e^-ie_r t(_r -i t _
thus , in addition to the usual phase factors making up the sum for the time - dependent wave - function , one has an explicit factor of @xmath56 multiplying the degenerate eigenfunctions associated with a jordan block , giving the complete time dependence of the initial wave - function of eq .
( [ ptexpansion ] ) as [ j+rest ] f(x , t)&=&c_0 _ 0(x)+_kn / a c_k _ k(x)e^-ik^2 t + & & + _
n > 0((_n -it_n)i_n(y)+_n
_ n(y))e^-i(n^2 ^ 2/a^2)t .
however , the explicit factor of @xmath56 only appears when the coefficient of the jordan function is nonzero .
the expansion of eq .
( [ one ] ) is a case in point . in the corresponding optical problem
it therefore appears that the @xmath5-dependence should be expected to be oscillatory for @xmath59 , exponential for @xmath3 and linear precisely at the symmetry - breaking threshold @xmath4 for initial states that excite a jordan function .
however , longhi@xcite has shown numerically and in perturbation theory that this linear dependence may not be realized , depending on the nature of the initial wave - function . since we now have explicit expressions for the bloch and jordan functions we are in a position to investigate the origin of this phenomenon in the context of the method of stationary states .
we will take as our input a gaussian profile of the form [ gaussian ] ( x,0)=f(x)e^-(x / w)^2+i k_0 x , with offset @xmath60 and width @xmath61 . because of the periodicity of the bloch eigenfunctions and jordan associated functions the range of the integral in eqs .
( [ ptcoeffs ] ) and ( [ jordancoeff ] ) can be reduced to @xmath62 , provided that @xmath44 is replaced by [ fn ] f_q(x)_m =- n^n-1 e^-im qf(x+ma ) , in which we recall that @xmath63 .
that is , c_k= , and similarly for @xmath50 , @xmath49 .
in particular [ inintegral ] _
n = . for @xmath64 large , the sum in eq .
( [ fn ] ) can be extended to infinity without significant error .
however , we have to be careful about the periodicity in @xmath65 of the discretized version .
that is , @xmath66 as given in eq .
( [ fn ] ) is really a function of @xmath67 .
thus we first replace @xmath65 by @xmath68 , where @xmath68 is the nearest to @xmath69 of the set of equivalent momenta , satisfying @xmath70 , to obtain the result [ fq ] f_q(x)= w e^-^2(|q - q_0)^2w^2/a^2+i|q x / a _ 3 ( + i^2 ( |q - q_0 ) , e^-^2w^2/a^2 ) , where @xmath71 is the jacobi elliptic theta function , which has the expansion@xcite _ 3(z , v)=1 + 2_s=1^v^s^2(2sz ) . for a broad wave - packet , with @xmath72 , the argument @xmath73 is very small , so that @xmath74
. moreover the prefactor in eq .
( [ fq ] ) is also small except for the case @xmath75 .
we are now in a position to identify under what conditions the jordan associated functions are excited , that is , when @xmath49 , as given by eq .
( [ inintegral ] ) , is non - vanishing . in order to obtain an appreciable value of @xmath76 the scaled offset momentum @xmath69 must be an integer @xmath77 , with @xmath78 mod 2
now , however , there is the integral in the numerator to be considered . given that the phase of @xmath79 is approximately @xmath80 and that @xmath25 has the expansion i_n(y)=(y)^n
_ s=0^ , where we recall that @xmath81 , it is easily seen that the integral vanishes unless @xmath77 is negative and @xmath82 . in that case
the jordan functions @xmath83 are excited . however , if @xmath69 is not a negative integer , no jordan blocks are excited , and thus no linear growth is to be expected .
for @xmath84 no jordan function is excited because the ground - state level @xmath85 is non - degenerate .
thus no linear growth is expected in this case . nor is it expected for the case @xmath86 . in the set - up chosen by longhi in ref .
@xcite , on the other hand , the offset @xmath69 was taken to be -1 , in which case @xmath87 is excited .
note the left - right asymmetry here : the hamiltonian is not parity invariant , but only @xmath0 invariant .
figures 3 and 4 show the different propagation behaviour for a wide beam in the two cases @xmath88 and @xmath84 respectively .
the parameters are : @xmath89 , @xmath90 and @xmath91 . in the first case , where one jordan mode @xmath87 , is excited
, the beam spreads out but does not split , as shown in fig .
3(b ) we show the different contributions to the maximum amplitude .
the lower curve shows the contribution of the jordan block sector only , which indeed rises linearly , as expected .
however , the intermediate curve , the contribution of all other modes , mysteriously begins to decrease after an initial rise , and the upper curve , which takes into account all the contributions , exhibits the saturation first noted by longhi@xcite . as a function of @xmath35 , @xmath5 .
( b ) maximum value of @xmath92 as a function of @xmath5 .
the parameters are : @xmath89 , @xmath91 , @xmath90 and @xmath84.,title="fig:",width=226 ] in the second case , where no jordan mode is excited , the behaviour is quite different .
the beam does not significantly spread , but instead splits into two , exhibiting the phenomenon of birefringence@xcite , and the maximum amplitude , after some initial fluctuations , decreases slowly with @xmath5 .
the behaviour for the case when @xmath93 is similar .
in this section we investigate in detail how the contributions from the other ( non - jordan ) eigenfunctions conspire to cancel out the explicit linear growth in @xmath5 coming from the jordan sector .
a priori such a cancellation seems highly unlikely , but since it does occur it must be because the two sets of contributions are in fact closely connected .
mathematically the cancellation can not be complete , but can only happen only over a limited range in @xmath5 . however , that is sufficient for the physical situation , because our lattice is finite in the @xmath35 direction , and at very large values of @xmath5 the beam will have encountered the edges of the lattice . we now show that the contributions of the jordan block and of the nearby eigenfunctions are indeed closely related .
we will consider specifically the first case of the previous section , where the jordan mode @xmath87 is excited . recall that @xmath94 is expanded as ( x,0)=c_0 _ 0(x)+_kn / a c_k _
k(x)+_n > 0(_n i_n(y)+_n _
n(y ) ) , as in eq . ( [ ptexpansion ] ) .
the expression for @xmath47 is given in eq .
( [ ptcoeffs ] ) , in which the integration over @xmath35 can be reduced to the interval @xmath95 $ ] by exploiting the periodicity of the bloch functions , to obtain [ cknum ] c_k= , where f_q(x)_m =- n^n-1 e^-im qf(x+ma ) .
similarly @xmath49 is given by _
the denominator of @xmath47 would vanish at @xmath24 .
in fact it turns out that it is precisely a sinc function : _
( ka)=a ( q ) .
the denominator of @xmath49 is proportional to the derivative of this previous denominator with respect to @xmath96 . thus _
0^a dx _ -k(x)_k(x)&=&_0^a dx i_-q(y)i_q(y)&=&_0^a dx ( i_-q(y)i_q(y)+i_-q(y)i_q(y ) ) .
but by differentiating the general relation @xmath97 and setting @xmath98 we find that @xmath99 .
thus , using the derivative definition of the jordan associated function as as given in eq .
( [ particular ] ) by a multiple of @xmath100 @xmath101 , we see by reference to eq .
( [ fixed ] ) that ._0^a dx _ -k(x)_k(x)|_q=1&=&4 _ 0^a dx i_1(y)(_1^pi(y)-k_1(y))&=&4_0^a dx i_1(y)_1(y ) .
the numerator of @xmath49 is a smooth continuation of the numerator of @xmath47 , which from now on we denote as @xmath102 .
moreover , @xmath102 is highly peaked around @xmath98 ( and also around @xmath103 ) , as is shown in fig .
5 . in fact , near @xmath98 it is given by the gaussian @xmath104 , where @xmath105 , while the denominator is proportional to @xmath106 .
we are now in a position to examine the @xmath5-development of the jordan contribution and the related contributions from nearby values of @xmath65 .
recall that @xmath6 is given by ( x , z)&=&c_0 _ 0(x)+_kn / a c_k _ k(x)e^-ik^2z + & & + _
n > 0[(_n -iz_n)i_n(y)+_n
_ n(y)]e^-i(n^2 ^ 2/a^2)x .
thus the total contribution from the neighbourhood of @xmath98 and @xmath103 is [ flat ] ( x , z)_j i_1(y)e^-iz , where @xmath107 . in principle
the upper limit of the sum is infinity , but in practice it can be taken less than @xmath64 .
note that if @xmath106 were continuous the limit of the second term as @xmath108 would be twice the first term , which comes from the jordan function .
it turns out that although @xmath109 does exhibit the expected linear behaviour in @xmath5 initially , it subsequently@xmath109 has an extremely wide and flat plateau before it eventually rises again as @xmath5 approaches @xmath110 .
this is illustrated in fig .
6(a ) for @xmath111 .
a hint of such a plateau - like behaviour can be seen in the simple function @xmath112 , which is plotted in fig .
however , in this case the plateau is much less pronounced .
\(a ) ( b ) we can understand the extreme flatness of the plateau in fig .
6(a ) in terms of jacobi @xmath113 functions . for simplicity let us set @xmath89 , so that we are analyzing the function j(z)&&z+_r=1^()(2_r z)e^-_r^2/(w^2/4+iz ) + & & z+_r=1^()(2_r z)e^-4_r^2/w^2for the values of @xmath5 we are considering . while this is not itself a @xmath113 function , its derivative with respect to @xmath5 can be so expressed : j(z ) & & 1 + 2_r=1^(2_r z)e^-_r^2w^2/4&=&1 + 2_r=1^(2rz / n)e^-r^2w^2/(4n^2)&=&_3 ( , e^-w^2/(4n^2 ) ) .
the behaviour of this @xmath114 function is not immediately apparent , since the second argument is of @xmath115 for @xmath116 .
however , it can be made clear by using the alternative notation @xmath117 , where @xmath118 , and applying jacobi s imaginary transformation@xcite , whereby _
3(z|)=(-i)^-e^-i z^2/()_3(z| ) , where @xmath119 .
this converts @xmath120 to [ jp ] j(z)=2 e^-4z^2/w^2_3(z , e^-4 ^ 2n^2/w^2 ) . in this form
the second argument of @xmath113 is small , so that for moderate values of @xmath5 we can approximate @xmath113 by 1 , in which case the behaviour is dominated by the preceding gaussian , which rapidly falls from 1 to a very small value , corresponding to the plateau in @xmath121 .
the gaussian is eventually overwhelmed by the hyperbolic cosines occurring in the expansion of @xmath113 , as must be the case , since @xmath113 is periodic in @xmath5 .
the plot of this function is given in fig .
on general grounds one expects linear growth of the amplitude @xmath6 at the symmetry - breaking threshold @xmath4 , due to the degeneracy of a subset of the eigenfunctions at this point and the consequent development of jordan blocks .
however , it has been observed numerically@xcite that this growth becomes saturated at large @xmath5 , at least for wide input beams .
we have been able to explain this saturation phenomenon by analyzing in detail the separate contributions from the jordan blocks and the contributions from the nearby bloch functions in which they are embedded . for the particular potential considered here ,
the bloch eigenfunctions are associated bessel functions of the first kind , @xmath122 , and we were able to explicitly construct the associated jordan functions at the exceptional points @xmath24 .
hence we were in a position to use the analogue of the method of stationary states to generate the @xmath5 dependence . in this method
we were able to isolate the separate contributions of the jordan blocks from the other non - degenerate states with normal @xmath5 dependence generated by multiplying each eigenfunction by its appropriate @xmath5-dependent phase .
we found that , in cases where the associated jordan functions are excited ( fig .
3 ) the explicit linear increase of the jordan - block contributions is precisely compensated by a linear decrease of the contribution of the non - degenerate states , which of course have no _ explicit _ linear @xmath5-dependence . in cases where the jordan associated functions are not excited ( fig .
4 ) , there is no initial linear increase , but rather a rapid oscillatory behaviour followed by a very slow decrease .
the topology of the beams in these two cases is markedly different . in the first case , although the maximum amplitude becomes constant , the beam spreads laterally in a linear fashion , and hence the total power grows linearly , with the beam taking energy from the lattice . in the second case
the total power turns out to be constant for large @xmath5 , with the slow decrease in the maximum amplitude being matched by the slow broadening of the individual beams .
we then examined in detail the mechanism of saturation , which is only possible because of the close relation between the contribution of the jordan block and the contributions of the bloch functions in which it is embedded . with the aid of a certain amount of approximation
we were able to write the respective contributions in the relatively simple form of eq .
( [ flat ] ) and to express its @xmath5-derivative as a @xmath113 function .
this @xmath113 function encodes the initial increase and the subsequent extremely wide and flat plateau . as a mathematical function
it also encodes further steps and plateaux for larger @xmath5 , but these are not physically relevant because they correspond to values of @xmath35 outside the range of the finite lattice .
since the references for jordan associated functions are not readily accessible , we give here a brief outline of their main properties .
they are the analogue of the jordan associated vectors that occur in simple matrix eigenvalue problems of the form ( h-)u=0 where @xmath11 is non - hermitian . for hermitian problems , while two eigenvalues @xmath123 and @xmath124 may become degenerate at a particular critical value of a parameter in the hamiltonian , there remain two distinct eigenvectors . however , in the non - hermitian case it is possible that the eigenvectors also become degenerate : @xmath125 .
the reduction in the number of eigenvectors at such a point means that the set of eigenvectors no longer forms a complete basis .
however , the basis can be completed by inclusion of the jordan associated vector @xmath126 ( generalized eigenfunction ) , defined by the generalized eigenvalue equation ( h-_1)v_1=u_1 .
of course this associated vector is not uniquely defined : to it may be added any multiple of @xmath127 . ) , which has the single eigenvalue @xmath15 and only one eigenvector @xmath128 .
the independent vector @xmath129 needed to complete the basis is indeed a solution ( undetermined up to a multiple of @xmath130 ) of the generalized eigenvalue equation [ v ] ( m-)v = u .
the eigenvectors of a non - hermitian matrix are not orthogonal in the usual sense .
instead one needs to introduce the left eigenvectors @xmath131 , satisfying @xmath132 , which are different from the usual ( right ) column vectors satisfying @xmath133 .
the orthogonality is then between left and right eigenvectors : @xmath134 for distinct eigenvalues @xmath123 and @xmath124
. when the eigenvectors become degenerate , they are self - orthogonal .
this is exemplified by the jordan matrix @xmath135 in eq .
( [ m ] ) , whose left and right eigenvectors are @xmath136 and @xmath137 , respectively , which are indeed orthogonal .
it is easy to show in general that the associated vectors @xmath138 are orthogonal in the same sense to eigenvectors @xmath139 and associated vectors @xmath140 belonging to different eigenvalues . in the event that @xmath11 has a @xmath0 symmetry , for some definition of the reflection operator @xmath141 , all of these overlaps
can instead be expressed in terms of right eigenvectors alone with the aid of the @xmath0 metric .
the continuum analogue of this situation is the eigenvalue problem ( h - e_n)_n=0 where @xmath11 is a non - hermitian differential operator with eigenvalue @xmath142 and ( right ) eigenfunction @xmath143 .
again , it is possible that , at a particular critical value of a parameter in @xmath11 two eigenvalues @xmath144 and @xmath142 , and also the corresponding eigenfunctions @xmath145 and @xmath143 , become degenerate . at that point
the single eigenfunction becomes self - orthogonal with respect to the metric defined by @xmath11 , which , for a @xmath0-symmetric problem , is the @xmath0 metric .
the reduction in the number of eigenfunctions at such a point again means that the set of eigenfunctions no longer forms a complete basis .
however , in complete analogy with eq .
( [ v ] ) , the basis can be completed by inclusion of the jordan associated functions @xmath146 ( generalized eigenfunctions ) , defined by the generalized eigenvalue equation ( h - e_n)_n=_n .
this is precisely how the jordan functions were introduced in eq .
( [ jordan ] ) .
they are defined only up to multiples of solutions of the homogeneous equation and are guaranteed to be orthogonal , using the @xmath0 metric , to each other and to eigenfunctions belonging to other eigenvalues .
we found it necessary to exploit this freedom in eq .
( [ fixed ] ) in order to ensure that the @xmath146 satisfy the appropriate boundary conditions .
an alternative definition of the jordan functions is as derivatives of the eigenfunctions with respect to the energy@xcite .
thus , differentiating the eigenvalue equation @xmath147 with respect to @xmath148 we get precisely ( h - e)-=0 , so that we may identify @xmath149 with @xmath150 , again modulo solutions of the homogeneous equation .
in our present problem this latter definition leads to extremely simple formulae for the functions @xmath52 .
for example , using eq .
( 9.6.44 ) of ref .
@xcite it yields @xmath151 , which is easily seen to be a solution of eq .
( [ jordan ] ) with the correct periodicity .
99 c. m. bender and s. boettcher , phys .
lett . * 80 * , 5243 ( 1998 ) . c. m. bender , contemp
* 46 * , 277 ( 2005 ) ; rep .
* 70 * , 947 ( 2007 ) .
a. mostafazadeh , int .
* 7 * , 1191 ( 2010 ) . c. m. bender ,
d. c. brody and h. f. jones , phys .
lett . * 89 * , 270401 ( 2002 ) ; 92 , 119902(e ) ( 2004 ) . c. m. bender ,
d. c. brody and h. f. jones , phys .
d * 70 * , 025001 ( 2004 ) ; 71 , 049901(e ) ( 2005 ) .
a. mostafazadeh , j. math . phys .
* 43 * , 205 ( 2002 ) ; j. phys .
a * 36 * , 7081 ( 2003 ) .
r. el - ganainy , k. g. makris , d. n. christodoulides and z. h. musslimani , optics letters * 32 * , 2632 ( 2007 ) .
z. musslimani , k. g. makris , r. el - ganainy and d. n. christodoulides , phys .
lett . * 100 * , 030402 ( 2008 ) .
k. makris , r. el - ganainy , d. n. christodoulides and z. musslimani , phys .
. lett . *
100 * , 103904 ( 2008 ) . s. klaiman , u. gnther and n. moiseyev , phys .
* 101 * 080402 ( 2008 ) .
a. guo , g. j. salamo , d. duchesne , r. morandotti , m. volatier - ravat , v. aimez , g. a. siviloglou and d. n. christodoulides , phys .
lett . * 103 * , 093902 ( 2009 ) .
s. longhi , phys .
a * 81 * , 022102 ( 2010 ) .
k. makris , r. el - ganainy , d. n. christodoulides and z. musslimani , phys .
a * 81 * , 063807 ( 2010 ) . c. rter ,
k g. makris , r. el - ganainy , d. n. christodoulides , m. segev and d. kip , nature physics * 6 * , 192 ( 2010 ) . h. ramezani , t. kottos , r. el - ganainy and d. christodoulides , phys .
rev . a * 82
* , 043803 ( 2010 ) .
m. v. berry , j. phys .
a * 31 * , 3493 ( 1998 ) .
m. c. zengh , d. n. christodoulides , r. fleischmann and t. kottos , phys .
a * 82 * , 010103(r ) ( 2010 ) .
s. longhi , phys .
103 * , 123601 ( 2009 ) .
s. longhi , phys .
b * 80 * , 235102 ( 2009 ) . m. kulishov , j. m. laniel , n. blanger , j. azaa and d. plant , optics express , * 13 * , 3068 ( 2005 ) .
e. m. graefe , h. j. korsch and a. e. niederle , phys .
* 101 * , 150408 ( 2008 ) ; e. m. graefe , h. j. korsch and a. e. niederle , phys .
a * 82 * 013629 ( 2010 ) .
b. midya , b. roy and r. roychoudhury , phys .
a * 374 * , 2605 ( 2010 ) .
w. d. heiss , european physical journal d * 60 * , 257 ( 2010 ) .
t. kato , _ perturbation theory for linear operators _ ( springer , berlin,1966 ) . m. v. keldysh , russ .
* 26 * , 15 , 1971 . m. abramowitz and i. a. stegun , _ handbook of mathematical tables _ ,
dover , new york 1970 .
e. t. whittaker and g. n. watson , a course in modern analysis , 4th ed .
, cambridge university press 1990 .
e. hernndez , a. juregui and a. mondragn , phys . rev .
a * 67 * , 022721 ( 2003 ) . | the @xmath0 symmetric potential @xmath1 $ ] has a completely real spectrum for @xmath2 , and begins to develop complex eigenvalues for @xmath3 . at the symmetry - breaking threshold @xmath4
some of the eigenvectors become degenerate , giving rise to a jordan - block structure for each degenerate eigenvector . in general this
is expected to result in a secular growth in the amplitude of the wave .
however , it has been shown in a recent paper by longhi , by numerical simulation and by the use of perturbation theory , that for a broad initial wave packet this growth is suppressed , and instead a saturation leading to a constant maximum amplitude is observed .
we revisit this problem by explicitly constructing the bloch wave - functions and the associated jordan functions and using the method of stationary states to find the dependence on the longitudinal distance @xmath5 for a variety of different initial wave packets .
this allows us to show in detail how the saturation of the linear growth arises from the close connection between the contributions of the jordan functions and those of the neighbouring bloch waves . |
the relativistic heavy ion collider ( rhic ) is under construction at brookhaven national laboratory .
in addition to its primary purpose , the search for quark - gluon plasma , a proposal to explore spin physics at rhic @xcite has been approved .
the major goals of the spin physics program at rhic are : * elucidation of the spin structure of the nucleon , and * precision tests of symmetries .
rhic offers a unique opportunity for those studies because of its capability of accelerating polarized proton beams up to @xmath0= 500 gev at the luminosity of 2@xmath110@xmath2@xmath3sec@xmath4 or more with large polarizations of @xmath570% .
obviously we will reach the high - energy frontier for polarized proton - proton collisions at rhic .
the phenix detector is one of the two large detectors at rhic @xcite .
its basic design concept is to detect photons , leptons , and hadrons with high momentum resolution and strong particle identification .
it consists of two spectrometers covering the central rapidity region ( central arms ) , which include an electromagnetic ( em ) calorimeter with fine segmentation ( @xmath6 ) , and two endcap muon spectrometers ( muon arms ) .
since hadron reactions with photonic or leptonic final states such as prompt photon production and lepton production from weak boson decays play major roles in spin physics program , phenix is very well suited to spin physics at rhic .
the studies are done by measuring the spin asymmetries in the cross sections for various reactions . by use of the spin rotators located upstream and downstream of phenix experimental hall , any combination of beam polarizations is possible .
thus we can measure the _ helicity _ dependent cross sections @xmath7 , @xmath8 , @xmath9 , and @xmath10 separately , where @xmath11 and @xmath12 represent positive and negative helicity states of the beam particles as well as _
dependent cross sections , @xmath13 , @xmath14 , @xmath15 , @xmath16 , where @xmath17 and @xmath18 represent transverse polarization of the beam particles . among these asymmetries
, we will discuss only two asymmetries in this presentation , a double longitudinal - spin asymmetry , @xmath19 , and a single longitudinal - spin asymmetry , @xmath20 ; @xmath21 the quantity @xmath19 is often used to extract helicity dependent structure functions ; @xmath20 extracts parity violation effects in the reaction . in the following section
, the sensitivity of our measurements is calculated assuming integrated luminosity of 320 pb@xmath4 and 800 pb@xmath4 for @xmath22 gev and 500 gev , respectively , which corresponds to 10 weeks of running with 70% machine efficiency .
the results of polarized muon scattering off a polarized proton target reported by the emc collaboration have stimulated both experimental and theoretical works to elucidate the spin structure of the proton .
the fraction of the proton spin carried by quarks , @[email protected]@xmath240.14 , was amazingly small comparing to the canonical expectation [email protected] @xcite .
post - emc experiments , which have provided data with much better statistics including deuterium and @xmath25he targets , have confirmed the emc results .
a recent global analysis gives @xmath26 , and thus still shows a significant deficit .
phenix will measure the gluon polarization @xmath27 and anti - quark polarization @xmath28 with flavor @xmath29 identified not only to search for the origin of the deficit but also to check the validity of assumptions in the analysis to obtain @xmath23 , e.g. su(3)@xmath30 .
these measurements will be described in the following subsections .
one of the most reliable channels to measure the gluon polarization is high @xmath31 prompt photon production .
the production is dominated by the gluon compton process , followed in significance by the annihilation process . by neglecting the contribution from annihilation channel , ( which is justified in several predictions @xcite )
, the asymmetry @xmath19 can be written at the leading order ( lo ) as a function of photon @xmath31 , @xmath32 here @xmath33 , @xmath34 stands for the scattering angle of partons in their cms , and @xmath35 represents the double longitudinal spin asymmetry for the parton cross sections .
it should be noted that the phenix acceptance ( @xmath36 ) strongly selects the samples in symmetric quark - gluon scattering at @xmath37 and this selection allows great simplification of the expression in eq .
( [ e : asymdg ] ) @xcite .
since @xmath38 is calculated in qcd and @xmath39 has been measured in lepton scattering experiments , @xmath27 can be extracted from the measured @xmath19 . to overcome experimental difficulties due to
the huge background from hadron decays , phenix s finely segmented em calorimeter plays a crucial role in avoiding the fake prompt photon signal that results from the merging of two photons from a high-@xmath31 @xmath40 .
since the phenix calorimeter is as fine as @xmath6 , the prompt photon can be identified up to 30 gev/@xmath41 or more without serious background .
the yield for the assumed integrated luminosities has been calculated using pythia for the phenix acceptance and listed in table [ t : dg_sensitiv ] for both @xmath42200 gev and 500 gev . in addition
, the sensitivity of the measurement of @xmath27 has been evaluated using @xmath35 and the measured @xmath39 .
the listed errors are statistical only .
we have identified the origin of the systematic errors and have begun studies to minimize them . in addition , studies of @xmath27 measurements with other channels such as @xmath43 , open charm / beauty , and heavy quarkonium production are in progress .
.sensitivity summary for the measurements of gluon polarization via prompt @xmath44 production .
[ cols="^ , < , < , < , < , < , < " , ] [ t : dg_sensitiv ] the polarized - dis experiments are sensitive to neither differences between anti - quarks and quarks nor their flavors , since the photon couples to the square of their electric charge . therefore the measurement of anti - quark polarization and the flavor decomposition will improve the knowledge on the spin of the nucleon significantly .
the parity violating asymmetry @xmath45 for @xmath46 production is presented at lo as @xmath47 for @xmath48 production , the @xmath49 and @xmath50 should be exchanged in this expression .
the asymmetry is just the linear combination of the quark and anti - quark polarizations .
furthermore , the flavor in the reaction is almost fixed .
thus flavor decomposition is possible .
the asymmetry converges to @xmath51 at the limit of @xmath52 and to @xmath53 at the limit of @xmath54 .
@xmath55 production can be identified by the detection of muon with @xmath5620 gev/@xmath41 . with the assumed @xmath57=800 pb@xmath4
, we expect about 5000 events for each of @xmath58 and @xmath59 .
furthermore the @xmath60-range can be selected using the muon momentum as shown in figure [ f : polpdf](a ) . using the muon sample which is divided into several energy bins ( and thus @xmath60-bins ) ,
we have estimated the sensitivities of our measurements ( with statistical errors only ) .
the results are plotted on two model predictions for polarized structure functions @xcite in figure [ f : polpdf](b ) .
error bars indicate the expected statistical errors
. our measurement will be quite sensitive to both quark and anti - quark helicity distributions , although further studies are needed to minimize the background contributions .
the nature of parity non - conservation itself can be directly probed using the polarized beams at rhic , with @xmath45 the measure of violation .
taxil and virey studied various possibility to find new physics at rhic through the measurements of @xmath20 using polarized proton beams @xcite .
while their predictions are only for single - jet production , which is not detectable with phenix , we expect that such asymmetries , however , should persist with inclusive or leading particle production .
sensitivity studies for phenix are underway .
the phenix physics scope has been extended to include the spin physics program .
it will provide measurements of @xmath27 and @xmath28 that greatly reduce the uncertainties in the knowledge of the spin structure of the nucleon .
precision tests of symmetries in the standard model is foreseen .
the first collision of the polarized proton beams is expected to start in the year of 2000 .
we thank the technical staffs of the participating institutions for their vital contributions . this detector construction project is supported by the department of energy ( u.s.a . ) ,
sta and monbu - sho ( japan ) , ras , rmae , and rms ( russia ) , bmbf ( germany ) , frn and the knut & alice wallenberg foundation ( sweden ) , and mist and nserc ( canada ) . 9 d.p .
morrison for the phenix collaboration , these proceedings .
proposal on spin physics using the rhic ( r5 ) , aug .
1992 ; updated sep . 1993 .
jaffe , nucl .
a547 * ( 1992 ) 17c .
gordon , nucl . phys .
* b501 * ( 1997 ) 197 . c. bourrely and j. soffer nucl
b445 * ( 1995 ) 341 .
p. taxil and j.m .
b364 * ( 1995 ) 181 , * b383 * ( 1996 ) 355 . | the phenix experiment at rhic has extended its scope to cover spin physics using polarized proton beams .
the major goals of the spin physics at rhic are elucidation of the spin structure of the nucleon and precision tests of the symmetries .
sensitivities of the spin physics measurements with the phenix detector system are reviewed . |
the generalized baker s construction @xcite defines a large class of invertible , lebesgue - measure - preserving maps of the unit square @xmath34\times[0,1]$ ] .
specifically , a two - dimensional map @xmath35 on @xmath36 is determined by a measurable _ cut function _
@xmath37 on @xmath38 $ ] satisfying @xmath39 .
the graph @xmath40 partitions the square @xmath36 into upper and lower pieces and the line @xmath41 , where @xmath42 , partitions the square into a ` left half ' @xmath43 \times [ 0,1]$ ] and a ` right half ' @xmath44 \times [ 0,1]$ ] .
the _ generalized baker s transformation _ ( gbt ) @xmath35 maps the left half into the lower piece and the right half into the upper piece in such a way that : * vertical lines in the left ( right ) half are mapped affinely into vertical ` half lines ' under ( over ) the graph of the cut function @xmath37 . * @xmath35 preserves two - dimensional lebesgue measure . * the factor action @xmath1 of @xmath35 restricted to vertical lines is ( conjugate to ) a piecewise monotone increasing , lebesgue - measure - preserving interval map on @xmath38 $ ] with two monotonicity pieces @xmath43 $ ] and @xmath44 $ ] .
the action of a typical gbt is presented in figure [ fig : gbt ] . when @xmath45 the map is the classical baker s transformation where the action on vertical lines is an affine contraction and the map @xmath1 is @xmath46 . on the other hand ,
every measure - preserving transformation @xmath47 on a ( nonatomic , standard , borel ) probability space with entropy satisfying @xmath48 is measurably isomorphic to some generalized baker s transformation on the square @xmath36 ( see @xcite ) . in order to proceed , we establish some notation .
each gbt @xmath35 has a skew - product form @xmath49 where @xmath50 defines @xmath1 implicitly .
furthermore , by construction the vertical lines @xmath51 are mapped into themselves by @xmath35 so that @xmath52 are fixed points of @xmath1 .
if @xmath37 is continuous then @xmath1 is differentiable on both @xmath53 and @xmath54 and @xmath55 since @xmath56 $ ] for each @xmath57 $ ] , @xmath58 , so @xmath1 is expanding , each branch of @xmath1 is increasing , and may have infinite derivative at preimages of places where @xmath59 .
we will call @xmath1 the * expanding factor * of @xmath35 .
note also that if @xmath37 is a _ decreasing _ function then @xmath60 is increasing on @xmath53 and decreasing on @xmath54 .
[ lem.setup ] let @xmath1 be defined by ( [ eqn.map1 ] ) .
each @xmath61 $ ] has two preimages under @xmath1 : @xmath62 and @xmath63 and moreover 1 . for every @xmath64
$ ] , @xmath65 2 .
@xmath66 and @xmath67 lebesgue almost everywhere ; 3 .
lebesgue measure @xmath26 is @xmath1invariant .
see appendix .
let @xmath16 .
maps @xmath19 arise as expanding factors of gbts whose cut functions @xmath37 satisfy * @xmath37 is continuous and decreasing function on @xmath38 $ ] with @xmath39 .
* there is a constant @xmath69 and a @xmath70 function @xmath71 on @xmath72 such that near @xmath73 @xmath74 where @xmath75 ; * either @xmath76 or there are constants @xmath77 , @xmath78 and a @xmath70 function @xmath79 on @xmath72 such that near @xmath80 @xmath81 where @xmath82 .
it follows from these conditions that @xmath37 is @xmath70 on @xmath72 and therefore each @xmath83 has two piecewise increasing @xmath84 branches @xmath85 with respect to the partition into intervals @xmath86 and @xmath54 ( where @xmath87 ) .
the branch @xmath88 has continuous extension to @xmath43 $ ] ( similarly for @xmath89 and @xmath44 $ ] ) and @xmath90 as @xmath91 . near @xmath92 each @xmath93 has the formula @xmath94 giving an _ indifferent fixed point _ ( ifp ) at @xmath17 .
if @xmath95 then @xmath1 also has an ifp at @xmath96 , and the order of tangency of the graph of @xmath1 near @xmath22 is @xmath97 . for maps with ifps at both @xmath17 and @xmath22 where the order of tangency is higher at @xmath22 then @xmath17
, the conjugacy @xmath98 will put the higher order tangency at @xmath17 .
there is thus no loss of generality in assuming that the `` most indifferent '' point is at @xmath92 . in case @xmath76 ,
equation ( [ eqn.mapd ] ) shows that the fixed point at @xmath22 is hyperbolic .
* example 1 [ alves - arajo map @xmath13 ] . *
see @xcite .
. then @xmath101 and @xmath102 .
then @xmath103 in @xcite , rahe established that the map @xmath35 is bernoulli ( using techniques from @xcite ) .
moreover , @xmath104 is conjugate ( by an affine scaling @xmath38\to[-1,1]$ ] ) to a map presented in @xcite which has non - integrable first hyperbolic time . despite this
, @xmath105 exhibits polynomial decay of correlations for hlder observables with rate @xmath106 @xcite.@xmath68 * example 2 [ symmetric case ] .
* let @xmath107 for @xmath108 $ ] and @xmath109 for @xmath110 $ ]
. then @xmath37 is symmetric ( and @xmath111 ) ; let the expanding factor of the corresponding gbt be denoted by @xmath112 . then @xmath112 has indifferent fixed points at @xmath17 and @xmath22 with tangency of order @xmath113 ; moreover @xmath114 as @xmath115 . in @xcite
it is shown that hlder continuous functions have correlation decay rate @xmath116 under @xmath112 .
the paper @xcite obtains similar results for a conjugate class of maps on @xmath14 $ ] .
the results in this paper , theorem [ prop : htimelower ] and theorem [ th : hintclt ] , imply that the _ first hyperbolic time _ is integrable if and only if @xmath117 , showing that the map @xmath105 emerges as an interesting transition point in the class @xmath24 .
the class of examples discussed here actually have a rather long history in the mathematical physics literature .
for other examples see @xcite .
to analyse @xmath19 we make a convenient partition of @xmath38 $ ] .
first , observe that since @xmath118 has four @xmath22@xmath22 and onto branches , @xmath1 admits a period-@xmath119 orbit @xmath120 where @xmath121 .
next , for each @xmath122 let @xmath123 and @xmath124 .
then @xmath125 defining @xmath126 and @xmath127 allows a similar partitioning of @xmath128 and @xmath129 ( note that @xmath130 and @xmath131 ) .
put @xmath132 these intervals partition @xmath38 $ ] from left to right as @xmath133 with @xmath134 ( and similarly for the @xmath135 intervals ) .
[ l.meas ] for @xmath19 , and the notation established above . 1 .
@xmath136 ; 2 .
@xmath137 ; 3 . for @xmath138 , @xmath139 ; 4 .
@xmath140 ; 5 . for @xmath141 . when @xmath95 , similar estimates hold for the @xmath142 intervals with @xmath21 replaced by @xmath143
. for parts ( i)(iv )
see ( * ? ? ?
* lemma 1 ) ; for part ( v ) see appendix .
@xmath68 when @xmath93 and @xmath76 the fixed point at @xmath22 is hyperbolic .
the corresponding decay rates for @xmath144 are exponential , and ( iii ) does not hold .
some of the estimates and statements below can be modified in this latter case is no longer an exceptional point ; see the definition in section [ sec.2 ] . ] .
[ assum:1 ] for the remainder of the paper we assume that @xmath95 so that @xmath1 also has an ifp at @xmath22 with tangency of order @xmath145 where @xmath146 . consequently , the estimates in parts ( i ) , ( ii ) , ( iv ) and ( v ) of lemma [ l.meas ] reveal decay of the sets @xmath147 which is no faster than for @xmath148 .
[ rem : compare ] below we use a young - tower built under the first - return time function to the set @xmath149 to prove upper bounds on the distribution of first @xmath150-hyperbolic times @xmath151 for certain @xmath150 . indeed , for @xmath19 ( @xmath117 ) , and any @xmath152 , @xmath153 .
the reason for this distribution is that points in @xmath154 ( whose lebesgue measure @xmath155 ) require approximately @xmath8 iterates to achieve enough expansion to be a hyperbolic time .
thus , @xmath156 for any @xmath157 .
similar bounds are used in @xcite to prove decay of correlation results by building a young tower @xmath158 whose tail set decays in the same way as the distribution of @xmath151 ; the resulting , a central limit theorem holds ( * ? ? ?
* theorem 2 ) . ]
decay of correlations are @xmath159close to the typical rate for maps with indifferent fixed points with tangencies of @xmath160 .
interestingly , direct calculations in @xcite where a first - return tower is built over @xmath161 give decay of correlations for hlder observables with rate @xmath116 for maps in @xmath15 .
the same computations give a clt for the entire range @xmath25 .
these improved asymptotics are due to the fact that orbits of @xmath1 experience very rapid expansion when they pass near @xmath162 , giving partial compensation for return from the neighbourhoods of @xmath163 , something that is not accounted for by the hyperbolic times analysis .
non - uniformity of expansion is expressed relative to a certain ( finite ) set @xmath164 $ ] of exceptional points .
the notation here is precisely as in @xcite .
each @xmath1 is locally @xmath84 on @xmath38\setminus\mathcal{s}$ ] , and must satisfy a non - degeneracy of the following type : there exist constants @xmath165 such that such that for _ every _
@xmath64 \setminus \mathcal{s}$ ] we have @xmath166 and if @xmath167\setminus\mathcal{s}$ ] and @xmath168 then @xmath169 @xmath170 is used to denote the usual euclidean distance to the set @xmath171 ( since @xmath1 is one - dimensional there is no need to impose a separate lipschitz condition on @xmath172 ) .
[ lem : nondegs1d ] under the conditions of assumption [ assum:1 ] , for @xmath19 set @xmath173 .
then there exists a @xmath174 so that @xmath175 satisfies conditions ( [ def : exceptional1 ] ) and ( [ def : exceptional2 ] ) . hence @xmath171 is a non - degenerate set of exceptional points for @xmath1 . in case @xmath76 then only @xmath92 and @xmath176 are required to be exceptional points .
see appendix .
let @xmath171 be a non - degenerate exceptional set , @xmath177 as in ( [ def : exceptional1 ] ) and ( [ def : exceptional2 ] ) and fix @xmath178 .
let constants @xmath179 and @xmath180 be given and define a truncated distance function @xmath181 as in @xcite , @xmath12 is called a @xmath182-hyperbolic time for @xmath64\setminus\mathcal{s}$ ] if for all @xmath183 @xmath184 although orbits escape subexponentially from @xmath171 and the rate of growth of derivatives along orbits is not uniform , the essential properties of uniformly expanding maps are captured at hyperbolic times . since the invariance ( and ergodicity @xcite ) of lebesgue measure is already at our disposal ,
establishing the long - run positive density ( in time ) of hyperbolic times is relatively straightforward .
[ lem : controlh1 ] let @xmath185 .
then for every @xmath186 and for every @xmath187 , there exists a @xmath180 such that for almost every @xmath64 $ ] @xmath188 apply ( birkhoff s ) ergodic theorem ( for the ergodic system @xmath189 ) to the function @xmath190 to obtain the first estimate . for the second estimate ,
choose @xmath180 such that @xmath191 and apply the ergodic theorem again .
@xmath68 this lemma says that for a given choice of @xmath192 , there is a good choice of @xmath150 for which there are ( many ) hyperbolic times for almost every @xmath64\setminus\mathcal{s}$ ] .
the two conditions established in lemma [ lem : controlh1 ] imply that @xmath1 is a _ non - uniformly expanding map _ in the sense of alves , bonatti and viana @xcite .
it now follows that : * positive density of hyperbolic times ( * ? ? ?
* lemma 5.4 ) * _ for every @xmath193 , for every @xmath187 there exist @xmath194 and @xmath180 ( depending only on @xmath195 and @xmath192 ) so that for almost every @xmath64 $ ] , for all sufficiently large @xmath196 there exist @xmath182-hyperbolic times @xmath197 for x , with @xmath198 .
_ now that we have established existence of hyperbolic times , we define @xmath199 to be the first @xmath200hyperbolic time for @xmath5 .
( if there are no @xmath201hyperbolic times for @xmath5 , set @xmath202 . )
the young tower partition gives simple lower bounds on @xmath199 .
although crude , these lower bounds are still sufficient for our first main result , theorem [ prop : htimelower ] below .
[ lem : htimelower ] let @xmath19 and fix @xmath203 .
let @xmath192 and @xmath204 be as above .
let @xmath205 be minimal such that @xmath206\cup [ y_{k_1},1]}\frac{df}{dx } < 1 $ ] .
then for @xmath207 ( @xmath208 ) , @xmath209 .
since @xmath60 is increasing on @xmath53 and decreasing on @xmath54 and @xmath210 , @xmath205 is well - defined .
now let @xmath208 , @xmath207 and fix @xmath12 with @xmath211 .
for each @xmath212 with @xmath213 we have @xmath214 $ ] , so that @xmath215 . comparing with ( [ eqn : defht ] ) , @xmath12 can not be a @xmath150-hyperbolic time for @xmath5 . since @xmath216 was arbitrary , @xmath217 .
@xmath68 [ prop : htimelower ] let @xmath19 , with @xmath203 .
there is a constant @xmath218 ( depending on @xmath219 ) such that @xmath220 and @xmath221 fails to be integrable with respect to lebesgue measure when @xmath222 .
choose @xmath205 as in lemma [ lem : htimelower ] , so @xmath223 .
hence @xmath224 by lemma [ l.meas ] ( ii ) . if @xmath225 then @xmath226 @xmath68
suitable choices of @xmath150 make @xmath151 integrable when @xmath117 . to prove this we distinguish @xmath227 = \cup_{j=1}^\infty i_j\cup i^\prime_j\qquad\pmod{m}\ ] ] as a `` good '' set , where expansion is very rapid and hyperbolic times are easy to control .
the derivative growth condition in ( [ eqn : defht ] ) is satisfied for @xmath228 on @xmath161 , but for points close to @xmath162 , the condition on @xmath229 fails for @xmath228 . controlling @xmath151 involves trading expansion with proximity to @xmath175 , and it turns out that getting enough expansion is the difficult part .
the idea is to control derivative growth upon successive returns to @xmath161 .
long excursions near @xmath163 lead to `` expansivity deficits '' relative to @xmath230 , with `` expansion recovery '' by passage through @xmath231 .
this is made quantitatively precise using a large deviations result of melbourne and nicol @xcite .
choose @xmath232 such that @xmath233 .
now choose @xmath234 such that @xmath235}\frac{df}{dx}\geq 1 \quad\textnormal{and}\quad \sigma\,\min_{[x_{k_0},y_{k_0}]}\frac{df}{dx } \geq 1.\ ] ] define @xmath236,\\ 0 & \textnormal{otherwise.}\end{array}\right.\ ] ] notice that @xmath237 [ lem : esth1 ] let @xmath19 , @xmath238 and put @xmath239 . if @xmath240 then @xmath241 and for all @xmath242 @xmath243 first , for any @xmath244 , @xmath245 .
thus @xmath246 and hence @xmath247 ( since @xmath248 is the only negative value of @xmath196 ) .
thus,@xmath241 and @xmath249 .
therefore @xmath250 for each @xmath251 . to complete the proof , apply ( [ eqn : setsigma ] ) to notice that for @xmath252 , we have @xmath253 ; if @xmath61\setminus [ x_{k_0},y_{k_0}]$ ] then @xmath58 so @xmath254 ; all other @xmath5 belong to @xmath255 $ ] so that @xmath256 . in particular , for each type of @xmath5 , @xmath257 .
hence @xmath258 [ lem : setdelta ] let @xmath19 , @xmath260 satisfy ( [ eqn : setsigma ] ) and let @xmath261 be fixed .
then there is @xmath204 such that whenever @xmath262 and @xmath263 , @xmath264 choose @xmath265 such that @xmath266 ( note that this is always possible , since there is a constant @xmath218 such that @xmath267 for all @xmath268 ) .
choose @xmath259 small enough that @xmath269\subset\cup_{k\geq k_b}(i_k\cup i^\prime_k)$ ] and @xmath270\cup[1-\delta,1)\subset \cup_{k\geq k_b}(j_k\cup j^\prime_k)$ ] .
for @xmath271 there are three cases to consider :
( i ) @xmath272 for @xmath273 ; ( ii ) @xmath274 for @xmath275 ; ( iii ) otherwise . in case ( i ) , for each @xmath276 , @xmath277 .
since @xmath278 it follows that @xmath279 so that @xmath280 . by the choice of @xmath265 , @xmath281 in case ( ii ) , let @xmath282 be such that @xmath283 . then , since @xmath284 on @xmath161 , @xmath285 in the final case , @xmath286.@xmath68 [ thm : isht ] let @xmath287 be fixed , let @xmath219 satisfy ( [ eqn : setsigma ] ) and choose @xmath259 as in lemma [ lem : setdelta ] .
let @xmath288 be as defined in lemma [ lem : esth1 ] .
then @xmath289 is a @xmath150hyperbolic time for @xmath5 and if @xmath151 is the first @xmath150hyperbolic time then @xmath290 if @xmath291 then @xmath292 and @xmath22 is a hyperbolic time .
otherwise , comparing ( [ eqn : defht ] ) with lemmas [ lem : esth1 ] and [ lem : setdelta ] shows that @xmath293 is a hyperbolic time . for the integral
, @xmath294 @xmath68 let @xmath295 $ ] , and partition ( modulo sets of measure @xmath17 ) according to @xmath296 for @xmath122 . for @xmath297 , @xmath298 so we put @xmath299 and equip @xmath300 with the measure @xmath301 obtained by direct upwards translation of @xmath302 . the tower map @xmath303
is defined in the usual manner : @xmath304 for @xmath305 and @xmath306 .
it is easy to check that @xmath307 so that @xmath308 .
standard arguments show that branches of the map @xmath309 have uniformly bounded distortion on @xmath161 , and since @xmath309 is uniformly expanding on @xmath161 , the tower map @xmath13 satisfies the usual regularity conditions @xcite for maps on a young tower denotes the jacobian @xmath310 then @xmath311 , where @xmath312 is the class of @xmath313hlder functions , defined with respect to the usual @xcite separation time @xmath314 . ] . since lebesgue measure @xmath26 is invariant for @xmath1 , @xmath315 and hence @xmath301 is actually @xmath13-invariant on @xmath300 .
as is usual , @xmath316,m)$ ] arises as a factor of @xmath317 via the semi - conjugacy @xmath318 .
then @xmath319 , and @xmath320 . next , lift @xmath288 to the tower :
put @xmath321 and @xmath322 let @xmath323 note that @xmath324 by ( [ eqn : nve ] ) . put @xmath325 .
then @xmath326 has zero mean , and belongs to every hlder class @xmath312 since it is piecewise constant with respect to the tower partition ( see @xcite ) .
[ lem : ldest ] for every @xmath327 and @xmath157 there is a constant @xmath218 ( not independent of @xmath328 ) such that @xmath329 using a result of melbourne and nicol , apply ( * ? ? ?
* theorem 3.1 ) ( setting @xmath330 and @xmath331 ) with the observation in ( [ eqn.towertailest ] ) that @xmath332.@xmath68 this estimate is enough to prove our main theorem : [ th : hintclt ] let @xmath19 where @xmath117 and let @xmath333 be fixed .
there is a choice of @xmath150 where @xmath334 and @xmath204 such that @xmath335 for any @xmath152 .
in particular , the first @xmath150-hyperbolic time for @xmath1 is integrable .
let @xmath232 be large enough that @xmath336 is negative and choose @xmath219 to satisfy ( [ eqn : setsigma ] ) and @xmath259 as in lemma [ lem : setdelta ] .
the first inequality in the theorem follows because @xmath337 ( compare with theorem [ thm : isht ] ) .
next , note that @xmath338 for @xmath339 , @xmath340 so that @xmath341 for any @xmath342 . for each such @xmath343 , @xmath344 choosing @xmath345 and @xmath346 ( where @xmath347 is given ) , by lemma [ lem : ldest ] there is a constant @xmath348 ( depending on @xmath328 ) such that the set on the rhs of ( [ eqn : htailldset ] ) has @xmath301measure bounded by @xmath349 .
the second inequality in the theorem now follows from ( [ eqn : esth1tower ] ) and ( [ eqn : htailldset ] ) .
the integrability of @xmath151 now follows from theorem [ thm : isht ] .
in this paper we have undertaken a detailed study of the asymptotics of _ hyperbolic times _ for a parameterized family @xmath350 of non - uniformly expanding , lebesgue - measure - preserving maps of the interval .
these one - dimensional maps arise naturally as the expanding factors of a class of two - dimensional _ generalized baker s transformations _ on the unit square .
a central result in the literature , alves , bonatti and viana @xcite , proves that if a non - uniformly hyperbolic map has the property that almost every @xmath5 has a ( uniform ) positive frequency of hyperbolic times , then it admits an absolutely continuous invariant measure .
according to lemma [ lem : controlh1 ] , each of our maps @xmath112 has this property ; of course the resulting invariant measure is already known to be lebesgue .
therefore our main result concerns the statistics of _ first hyperbolic times _
@xmath351 for our maps @xmath112 and in particular how the quantities @xmath352 depend on @xmath12 and @xmath21 .
we show that this is entirely determined by the strength of the ( most indifferent ) fixed point for the map @xmath112 .
in particular , * @xmath351 is integrable if and only if @xmath353 , corresponding to relatively fast polynomial escape from the indifferent fixed points at @xmath354 within the range of our family of maps @xmath355 ( theorems [ prop : htimelower ] and [ th : hintclt ] ) . * for @xmath356 and @xmath357 we establish @xmath358 which implies , by using results from @xcite , correlation decay for hlder observables at rate @xmath359 .
* finally , a clt holds in case @xmath360 , again by applying the results of @xcite to the case @xmath361 .
the conclusions which are obtained in @xcite depend on the construction of a suitable markov or young tower via first hyperbolic times and their statistics and the analysis of the return times on the resulting tower . on the other hand , @xcite details a specific construction of a young tower for the maps @xmath112 and
proves an improved correlation decay rate of @xmath362 for all @xmath363 .
it is shown that this correlation decay rate is sharp for hlder data .
the estimates in @xcite also imply a clt for @xmath353 .
see remark [ rem : compare ] for further discussion .
we interpret this as follows .
the analysis via hyperbolic times provides a relatively general approach to analysis of nonuniformly hyperbolic systems that leads to a particular young tower construction related to the hyperbolic times .
it is not so surprising that this approach does not always yield optimal results such as sharp estimates on decay of correlation rates or the clt . in the case of our maps
@xmath112 , for example , a dedicated tower construction in @xcite can produce optimal results in the form of sharp estimates on correlation decay by bypassing the intermediate construction of hyperbolic times .
_ proof of lemma [ lem.setup ] : _
( i ) the first equation is immediate from equation ( [ eqn.map1 ] ) for the second , again use equation ( [ eqn.map1 ] ) and write @xmath364 and so @xmath365 ( ii ) differentiating the expressions in ( i ) via lebesgue s theorem gives : @xmath366 ( iii ) apply ( i ) and ( ii ) : @xmath367 ) = \int_0^{x_l } { { \bf 1}}dt + \int_a^{x_r } { { \bf 1}}dt = \int_0^x \frac{dx_l}{dx } dx + \int_0^x \frac{dx_r}{dx}\ , dx = \int_0^x { { \bf 1}}dx= m[0,x ]
. \qquad\box\ ] ] _ proof of lemma [ lem : nondegs1d ] : _ first , note that
@xmath374 , and @xmath375 for every @xmath376 , so for the lower bound in ( [ def : exceptional1 ] ) holds whenever @xmath377 .
we will establish the right hand side inequality after proving ( [ def : exceptional2 ] ) for that , it suffices to work on @xmath53 , since the other interval is similar . by ( [ eqn.mapd ] ) , @xmath378 , and hence @xmath379 if @xmath380 ( @xmath381 ) then @xmath382 so @xmath383 and ( since @xmath37 is decreasing ) , @xmath384 .
moreover , @xmath385 . hence @xmath386 .
if @xmath387 then @xmath388 $ ] and @xmath389 ( since @xmath37 is decreasing and @xmath70 ) .
if @xmath390 then @xmath391 so @xmath392 , @xmath393 and @xmath394 .
hence @xmath395 ( since @xmath396 and @xmath397 by lemma [ l.meas ] ) .
for any @xmath398 there is a constant @xmath399 such that @xmath400 to complete the proof of ( [ def : exceptional2 ] ) let @xmath401 .
then the mean value theorem gives a @xmath402 between @xmath403 such that @xmath404 ( using [ e : regest ] ) ) .
but since @xmath405 , @xmath406 so choosing @xmath407 completes the regularity estimate . a similar argument estimating @xmath408 on the interval @xmath409
gives @xmath410 giving the upper bound in ( [ def : exceptional1 ] ) for any @xmath411.@xmath68 | we establish some statistical properties of the _ hyperbolic times _ for a class of nonuniformly expanding dynamical systems .
the maps arise as factors of area preserving maps of the unit square via a geometric baker s map type construction , exhibit intermittent dynamics , and have unbounded derivatives .
the geometric approach captures various examples from the literature over the last thirty years .
the statistics of these maps are controlled by the order of tangency that a certain `` cut function '' makes with the boundary of the square .
using a large deviations result of melbourne and nicol we obtain sharp estimates on the distribution of first hyperbolic times . as shown by alves , viana and others , knowledge of the tail of the distribution of first hyperbolic times
leads to estimates on the rate of decay of correlations and derivation of a clt . for our family of maps
, we compare the estimates on correlation decay rate and clt derived via hyperbolic times with those derived by a direct young tower construction .
the latter estimates are known to be sharp .
let @xmath0 be a dynamical system which is expanding on average , but not necessarily uniformly with every time - step . amongst the important questions to ask about @xmath1
are : is there an invariant srb - probability measure ? how quickly do correlations between observables decay under iteration by @xmath1 ? does a central limit theorem hold ?
are these properties stable to perturbations of @xmath1 ? when @xmath1 is _ uniformly expanding _ the answers to these questions are well understood ( see eg @xcite ) , but the situation for non - uniformly expanding @xmath1 is more delicate .
difficulties arise from the fact that orbits of @xmath1 may experience periods of local contraction as well as expansion ( for example , quadratic maps @xcite ) , rapidly varying derivatives near singularities leading to unbounded distortion ( eg @xcite ) , or indifferent fixed points @xcite .
the theory of _ hyperbolic times _ has proved useful for analysing the statistical properties of non - uniformly expanding maps @xcite .
the idea was introduced in @xcite to handle specific non - uniformly expanding families ( alves - viana maps @xcite and certain quadratic maps @xcite ) , and has since been developed for various non - uniformly expanding and partially hyperbolic classes @xcite .
gouzel @xcite has used hyperbolic times to show that the alves - viana maps exhibit stretched exponential decay of correlations .
alves , luzzatto and pinheiro @xcite prove exponential decay of correlations ( and a clt ) for a class of non - uniformly expanding maps by using hyperbolic times and a young tower construction .
further results are obtained in @xcite for one - dimensional families .
a survey paper discussing many of these ideas is found in @xcite .
roughly speaking , hyperbolic times are defined as follows ) in section [ sec.2 ] . ] : given an orbit @xmath2 of a point @xmath3 , an integer @xmath4 is a _ hyperbolic time _ for @xmath5 if for all @xmath6 the cumulative derivative @xmath7 grows exponentially in @xmath8 .
in addition , if the map has a nonempty set @xmath9 of singular points we require the distance from @xmath10 to @xmath9 to contract at exponential rate in @xmath8 , essentially an exponential escape condition .
these exponential rates are to be chosen uniformly for @xmath11 .
note that for uniformly expanding maps with bounded distortion both conditions automatically hold and every @xmath12 is a hyperbolic time .
therefore , the idea is to choose certain times at which the accumulated expansion and escape from the singular set mimic the uniformly expanding case even though there may have been times along the way where these properties failed . in this way
, many good statistical properties can be recovered .
there are at least two important statistics associated with hyperbolic times : their long - run frequency of occurrence , and the distribution of _ first hyperbolic times_. obtaining precise quantitative control of the distribution of hyperbolic times can be an important step @xcite in further analysis of statistical properties of the map , including the above - mentioned rates of decay of correlation and clt .
in @xcite a map @xmath13 on the interval @xmath14 $ ] is introduced that has positive density of hyperbolic times , but for which the first hyperbolic time fails to be integrable .
@xmath13 has a number of special properties ( symmetry , preservation of lebesgue measure , and a pair of indifferent fixed points with quadratic tangencies ) . in this paper
we present a class of interval maps @xmath15 ( parametrised by$ ] . ]
@xmath16 ) which arise as nonuniformly expanding one - dimensional factors of geometrically derived _ generalized baker s transformations _ ( gbts ) @xcite .
each map in @xmath15 has an indifferent fixed point ( ifp ) at @xmath17 , and in fact the map @xmath13 in @xcite is conjugate to a certain map @xmath18 .
each @xmath19 has a positive long - run frequency of hyperbolic times ( by an argument from @xcite ) , and integrability of first hyperbolic times holds when @xmath20 .
however , as @xmath21 increases through @xmath22 this integrability is lost ( lemma [ prop : htimelower ] ) . in this way , @xmath23 appears as a transition point for our families @xmath24 . as becomes clear in our proof of lemma [ prop : htimelower ] , the non - integrability is entirely due to lower bounds on the first hyperbolic time which are determined by escape statistics from the neighbourhood of the ifp(s ) .
these same escape statistics are then used to provide upper bounds on the first hyperbolic times in theorem [ th : hintclt ] , completing the analysis and providing sharp estimate on tail asymptotics for hyperbolic times for the range @xmath25 . while precise statements are given below , roughly speaking : if @xmath26 denotes lebesgue measure and @xmath27 denotes the first hyperbolic time on an orbit beginning at @xmath5 then @xmath28 . in @xcite
hyperbolic times asymptotics are used to estimate correlation decay rates and to establish clt s for nonuniformly hyperbolic systems . when applied to our family @xmath24
, these results imply upper bounds on correlation decay rates that fail to be sharp , compared to the direct computation via young towers detailed in @xcite .
the range of parameters in our family leading to a clt is similarly underestimated by the hyperbolic times analysis .
remark [ rem : compare ] at the end of section [ sec.1 ] provides a comparative analysis of these two approaches .
the class @xmath15 is presented in section [ sec.1 ] , hyperbolic times are reviewed and lower bounds are derived in section [ sec.2 ] and the upper bounds are established ( via a large deviations principle of melbourne and nicol @xcite on a suitable young tower @xcite ) in section [ sec.3 ] . in section [ sec.concl ] we discuss our results in the context of the existing literature on hyperbolic times .
some technical estimates are placed in an appendix ( section [ sec.4 ] ) .
_ notation : _ we write @xmath29 to mean there is a constant @xmath30 such that @xmath31 and @xmath32 to mean @xmath29 and @xmath33 . |
understanding the mechanisms that drive the transition to the deconfined regime is one of the great open problems of qcd . in particular with the expected new results from the experiments at the rhic , lhc and gsi facilities
also the theoretical side is challenged to contribute to our understanding of confinement and the transition to the deconfined phase . since
phase transitions are non - perturbative phenomena , the applied methods must be non - perturbative approaches .
a particularly powerful technique is the lattice formulation of qcd , where numerical simulations have become reliable quantitative tools of analysis . an interesting idea , which is partly rooted in the lattice formulation , is the svetitsky - jaffe conjecture @xcite which links the deconfinement transition of an su(@xmath0 ) gauge theory in @xmath1 dimensions to the magnetic transition of a @xmath2-dimensional spin system which is invariant under the center group @xmath3 .
the spins of the system are related @xcite to the local polyakov loops , which are static quark sources in the underlying gauge theory .
having identified an effective spin system which describes su(@xmath0 ) gauge theory at the deconfinement transition , it is natural to ask whether one can turn the argument around and identify characteristic features of a spin system directly in the corresponding gauge theory .
furthermore one may analyze whether the gauge - spin relation holds only at the critical temperature @xmath4 or also in a finite range of temperatures around @xmath4 .
a particular property of many discrete spin systems is the percolation of suitably defined clusters of spins at the magnetic transition .
since the spin systems relevant for gauge theories have the discrete @xmath3 invariance , one may expect to find some kind of percolation phenomenon for center degrees of freedom at the deconfinement transition of the gauge theories .
indeed , for the case of su(2 ) lattice gauge theory studies of percolation properties can be found in the literature @xcite , and more recently first results for su(3 ) @xcite as well as full qcd @xcite were presented .
establishing finite clusters below @xmath4 and percolating infinite clusters above @xmath4 gives rise to a tempting interpretation of the deconfinement transition : the size of finite clusters in the confining phase might be related to the maximal distance one can place a quark and an anti - quark source such that they still have a non - vanishing vacuum expectation value . for larger distances the two sources always end up in different clusters and average to zero independently . above @xmath4
there exists an infinite cluster and with a finite probability the two sources are correlated also at arbitrary distances such that they can move freely
. however , the above sketched picture hinges crucially on the scaling properties of the center clusters a question that so far has not been addressed in the literature .
a spin system has an intrinsic scale : the lattice constant of the underlying grid . in lattice gauge theory
the situation is different : there one is interested in studying the system for finer and finer lattices in order to learn about the continuum limit . for the percolation picture
this implies that when measured in lattice units , the clusters have to be larger for finer lattices
. only then the size of the cluster in physical units , e.g. , the diameter of the cluster multiplied with the lattice constant in fm can approach a finite value and can be assigned a physical meaning .
if no such scaling behavior can be established the clusters are merely lattice artifacts . in this article
we compare for su(3 ) and su(2 ) lattice gauge theory the evidence for the existence of center clusters and their percolation at @xmath4 .
particular focus is put on the analysis of the scaling behavior of the clusters .
we study the flow of the cluster parameters as a function of the lattice spacing and demonstrate that a continuum limit for the cluster picture is plausible .
in our analysis we explore pure su(3 ) and su(2 ) lattice gauge theory at temperatures below and above the deconfinement transition .
the basic observable we analyze is the local polyakov loop @xmath5 defined as @xmath6 @xmath5 is the ordered product of the su(3 ) or su(2 ) valued temporal gauge variables @xmath7 at a fixed spatial position @xmath8 , where @xmath9 is the number of lattice points in time direction and @xmath10 denotes the trace over color indices .
the loop @xmath5 thus is a gauge transporter that closes around compactified time .
often also the spatially averaged loop @xmath11 is considered , which we define as @xmath12 where @xmath13 is the spatial volume . due to translational invariance @xmath11 and
@xmath5 have the same vacuum expectation value .
the polyakov loop corresponds to a static quark source and its vacuum expectation value is ( after a suitable renormalization ) related to the free energy @xmath14 of a single quark , @xmath15 , where @xmath16 is the temperature ( the boltzmann constant is set to 1 in our units ) . below the critical temperature @xmath4 quarks are confined and @xmath14 is infinite , implying @xmath17 . the transition from the confined to the deconfined phase is of first order for the case of su(3 ) , while it is second order for su(2 ) gauge theory .
the deconfinement transition of pure yang - mills theory may also be interpreted as the spontaneous breaking of center symmetry . for su(3 ) the elements @xmath18 of the center group @xmath19 are a set of three phases , @xmath20 , while for su(2 ) we have the center group @xmath21 with @xmath22 . in a center transformation all temporal links in a fixed time slice are multiplied with an element @xmath18 of the center group . while the action and the path integral measure are invariant under a center transformation , the local and averaged polyakov loops transform non - trivially as @xmath23 the non - vanishing expectation value @xmath24 , which we find above @xmath4 , thus signals the spontaneous breaking of the center symmetry . in our study
we analyze the behavior of the local polyakov loops @xmath5 using quenched su(3 ) and su(2 ) configurations at finite temperature . for both gauge groups we use the lscher - weisz gauge action @xcite on lattices with different sizes ranging from @xmath25 to @xmath26 . for su(3 )
the lattice constant @xmath27 was determined in @xcite using the sommer parameter , while for su(2 ) we express dimensionful quantities using suitable powers of the string tension @xmath28 at zero temperature . for su(3 )
the critical temperature determined in @xcite is used , for su(2 ) it was determined using the polyakov loop susceptibility @xcite .
all errors we show are statistical errors determined with single elimination jackknife .
we begin our analysis with studying the distribution properties of the local polyakov loops @xmath5 defined in ( [ ploopdef ] ) . while @xmath5 is a real number for the gauge group su(2 ) it is complex for su(3 ) . for the latter case
we decompose the local polyakov loops into modulus and phase , @xmath29 in @xcite it was demonstrated that the distribution of the modulus @xmath30 is a rather unspectacular quantity .
it is very well described by the haar measure distribution @xmath31 .
in particular the distribution of the modulus is almost entirely insensitive to temperature , lattice volume and resolution scale @xcite .
thus we conclude that the first order transition of su(3 ) lattice gauge theory into the deconfined phase is not driven by a change in the distribution of the modulus .
su(3 ) ensembles and we compare two temperatures , @xmath32 ( lhs . plot ) and @xmath33 ( rhs . ) .
the curve superimposed on the low temperature histograms is the haar measure distribution @xmath34 .
the high temperature results are for the sector of gauge configurations characterized by the su(3 ) center element @xmath35 .
[ ploophistosu3],width=472 ] the situation is different for the distribution of the phase @xmath36 , where indeed we observe a strong change in the distribution as one crosses into the deconfined phase .
this is illustrated in fig .
[ ploophistosu3 ] , where we show histograms @xmath37 $ ] for the distribution of the phases @xmath36 comparing the two temperatures @xmath38 and @xmath39 . below @xmath4 ( lhs .
plot ) , the distribution shows three pronounced peaks located at the center phases @xmath40 and @xmath41 .
the whole @xmath38 histogram perfectly matches the haar measure distribution @xmath42 which we superimpose as a full curve on the histograms in the lhs .
plot of fig .
[ ploophistosu3 ] . above @xmath4 ( rhs .
plot ) the distribution changes : one of the peaks has grown , while the other two peaks have shrunk .
we stress at this point , that the selection which peak becomes enhanced is a matter of spontaneous symmetry breaking . in the rhs .
plot of fig .
[ ploophistosu3 ] the system has chosen the sector which is characterized by a phase @xmath43 of the spatially averaged polyakov loop @xmath11 ( compare eq .
( [ averloop ] ) ) . in case one of the other two center sectors
is selected , the whole distribution in the rhs .
plot is shifted periodically by @xmath44 .
it is interesting to note , that also above @xmath4 we still see the subdominant peaks which correspond to the center sectors that were not selected in the act of spontaneous symmetry breaking .
this already hints at the possibility that spatial bubbles with phases @xmath36 near the subdominant center phases might exist also above @xmath4 .
su(2 ) ensembles and we compare two temperatures , @xmath32 ( lhs . plot ) and @xmath45 ( rhs . ) .
the curve superimposed on the low temperature histograms is the haar measure distribution @xmath46 .
the high temperature results are for the sector of gauge configurations characterized by the su(2 ) center element @xmath47 .
[ ploophistosu2],width=472 ] let us now come to the case of su(2 ) .
there the local polyakov loops @xmath5 are real and we can directly look at their distribution . in fig .
[ ploophistosu2 ] we show histograms @xmath48 $ ] for the distribution of @xmath5 , again comparing two temperatures , @xmath49 ( lhs . plot ) and @xmath50 ( rhs . ) .
similar to the su(3 ) case we find that below @xmath4 the distribution very closely follows the haar measure distribution @xmath46 , which we superimpose as a full curve in the corresponding plot .
above @xmath4 we observe a deformation of the distribution favoring positive values .
however , again we stress that this is a manifestation of spontaneous symmetry breaking , since here we use high temperature configurations which are characterized by a positive value of the spatially averaged polyakov loop @xmath11 . in case the system spontaneously selects the sector of the other center element @xmath51 , which is characterized by negative values of @xmath11 above @xmath4 , the distribution in the rhs .
plot of fig .
[ ploophistosu2 ] would display its peak at negative values of @xmath5 . the spontaneous symmetry breaking , which leads to a non - vanishing expectation value of the polyakov loop above @xmath4 ,
can now be directly related to the distributions of the local polyakov loops @xmath5 .
in particular the relative distributions in the three ( two ) center sectors for gauge group su(3 ) ( gauge group su(2 ) ) drive the change of the expectation value of the polyakov loop .
the three sectors of su(3 ) are here defined by the three intervals @xmath52 $ ] , @xmath53 $ ] and @xmath54 $ ] , the phases @xmath36 can fall into .
the two sectors for the case of su(2 ) are distinguished by the sign of @xmath5 . of sites in each sector normalized by the spatial volume @xmath13 as a function of temperature .
the results are for our @xmath55 lattices and we compare the case of su(3 ) ( lhs .
plot ) with su(2 ) gauge theory ( rhs . ) .
[ abundance],title="fig:",height=245 ] of sites in each sector normalized by the spatial volume @xmath13 as a function of temperature .
the results are for our @xmath55 lattices and we compare the case of su(3 ) ( lhs .
plot ) with su(2 ) gauge theory ( rhs . ) .
[ abundance],title="fig:",height=245 ] below @xmath4 all three ( two ) sectors are populated with an equal number of sites @xmath8 .
when all center sectors are equally populated , the polyakov loop expectation value vanishes as the center elements add up to zero , i.e. , @xmath56 for su(3 ) , and @xmath57 for su(2 ) .
above @xmath4 one of the center sectors is more populated and thus the contributions do no longer add up to zero .
in fig . [ abundance ]
we show the abundance @xmath58 of sites in each sector normalized by the spatial volume @xmath13 as a function of temperature and compare su(3 ) ( lhs .
plot ) and su(2 ) ( rhs . ) .
for low temperature each of the sectors is for the case of su(3 ) populated with roughly a third of the lattice sites ( half of the lattice sites for su(2 ) ) .
near @xmath4 one of the sectors starts to increase the abundance of sites , while the other sectors become depleted .
it is interesting to note that for the case of su(3 ) , where the transition is first order , the curves show a rather sudden change near @xmath4 , while for su(2 ) , where the transition is of second order , the behavior is smoother ( as expected ) .
in the previous section we have studied the distribution of the local polyakov loops @xmath5 . when analyzing the abundance of sites assigned to the different center sectors we found that near @xmath4 one of the sectors starts to become more populated , thus giving rise to the non - vanishing expectation value of the polyakov loop .
in this section we now analyze the connectedness properties of sites @xmath59 where the values of the local loops @xmath60 fall in the same center sector . for that purpose
we assign sector numbers @xmath61 to the sites @xmath8 . for the case of su(3 ) the sector numbers can have three values @xmath62 assigned according to @xmath63 \ ; , \\ 0 & \ ; \mbox{for } \;\;\;\ ; \varphi(\vec{x } ) \ , \in \ , [ \,-\pi/3 + \delta \ , , \ , \pi/3 - \delta \ , ] \ ; , \\ + 1 & \ ; \mbox{for } \;\;\;\ ; \varphi(\vec{x } ) \ , \in \ , [ \,\pi/3 + \delta \ , , \ , \pi - \delta \ , ] \ ; , \end{array } \right .
\label{sectornumbers}\ ] ] while for su(2 ) we have two possibilities , @xmath64 , with @xmath65 if the phase @xmath36 ( the local loop @xmath5 for su(2 ) ) is not contained in one of the intervals , no sector number is assigned to the site @xmath8 , which then is no longer taken into account in the subsequent analysis .
we can now define center clusters by assigning two neighboring sites @xmath8 and @xmath66 to the same cluster if @xmath67 .
let us comment on the role which the parameter @xmath68 plays in our construction of the center clusters .
the parameter @xmath68 allows one to cut those lattice sites where the corresponding local polyakov loop does not clearly lean towards one of the center elements . for the case of su(3 ) sites @xmath8 with phases near the minima of the distributions shown in fig .
[ ploophistosu3 ] are cut , while for su(2 ) the cut leads to a removal of sites where @xmath5 is close to zero . in order to have a more accessible definition of how much we cut , instead of quoting a value of the parameter @xmath68 , from now on we express the cut in percent of sites that are not assigned a sector number @xmath61 , i.e. , the percentage of sites that are removed from the analysis . to allow for a comparison of the values @xmath68 and the cut in % , we list the corresponding numbers for @xmath69 at three temperatures in table 1 .
the percentage of points that are cut for a given @xmath68 shows a mild temperature dependence , and when refering to these numbers in the subsequent text we quote an average value .
the volume dependence is negligible .
.comparison of the cut parameter values @xmath68 and the fraction of cut sites in @xmath70 .
we show the su(3 ) results for our @xmath71 lattices and compare different temperatures . [ cols="^,^,^,^",options="header " , ] introducing such a cut for the analysis of cluster properties of course immediately raises the question whether such a cut does not destroy the physics one wants to study . in order to address this question we show in the lhs .
plot of fig .
[ ploopcut ] the results for the polyakov loop as a function of temperature for different amounts of lattice sites removed when turning on @xmath68 ( we show results for su(3 ) on @xmath55 ) .
it is obvious that even a cut of almost 40 % of lattice points leads to only a small reduction of the expectation value of the polyakov loop , indicating that the cut indeed removes only the rather unimportant fluctuations between the center values .
one can even go one step further and replace the local loop @xmath5 by the nearest center element .
the result is shown in the rhs .
plot of fig .
[ ploopcut ] , again comparing different values for the cut .
it is obvious that the center elements alone are sufficient to reproduce most of the polyakov loop expectation value .
repeating the same analysis for su(2 ) leads to equivalent results @xcite . ) .
rhs . : expectation value of the polyakov loop when using the nearest center elements instead of the local loop @xmath5 .
both figures are for su(3 ) on @xmath55 lattices and we compare different values of the cut . [ ploopcut],title="fig:",height=222 ] ) .
rhs . : expectation value of the polyakov loop when using the nearest center elements instead of the local loop @xmath5 .
both figures are for su(3 ) on @xmath55 lattices and we compare different values of the cut .
[ ploopcut],title="fig:",height=222 ] having convinced ourselves that the cut does not destroy the physics we want to analyze , let us add a few more comments on the role of the cluster parameter @xmath68 .
it is obvious that a small value of @xmath68 will produce denser clusters as more points are available for forming the clusters .
as one increases @xmath68 the clusters become thinner and smaller .
one of the motivations of this analysis is to study a possible characterization of the deconfinement transition as a percolation phenomenon .
this is an idea that has been widely explored in the context of spin systems , where for many models the magnetic transition may be characterized by the percolation of suitably defined clusters . for these systems
it is well known that a naive cluster definition , where neighboring sites with equal spin values are assigned to the same cluster , gives rise to clusters that are too dense , such that the percolation- and the magnetic transitions do not coincide .
only if the clusters are `` thinned out '' the two critical temperatures will agree .
in particular one may use the fortuin - kasteleyn cluster construction @xcite , which is now understood to give the correct cluster description of the magnetic transition in potts models @xcite .
the parameter @xmath68 which we introduce in our cluster construction plays exactly the same role as the more educated cluster definitions in spin systems , such as the fortuin - kasteleyn construction . as a matter of fact
the introduction of a free parameter for controlling the cluster density , similar to our prescription , has been discussed in the literature @xcite .
there is , however , an important difference between the analysis of percolation in spin systems and in lattice gauge theory .
while in the former case there is a fixed scale , the lattice spacing of the ferromagnet one studies , in the analysis of lattice gauge theories one is interested in performing the continuum limit , i.e. , the limit of vanishing lattice constant @xmath72 .
the cluster picture only has a chance for a reasonable continuum limit , if the cluster diameter in lattice units diverges as one approaches @xmath73 . only in that case
the cluster diameter in physical units , which is obtained as the product of the diameter in lattice units with the lattice constant @xmath27 , can have a finite limit . as we will demonstrate in the next section ,
our cluster definition with the parameter @xmath68 is suitable for such an analysis of a possible continuum limit . of the largest cluster normalized by the volume as a function of temperature .
we compare the case of su(3 ) ( lhs .
plot ) with su(2 ) gauge theory ( rhs . ) and display data for three different volumes .
the cuts we use are 39 % of sites for su(3 ) and 48 % for su(2 ) .
[ weight_maxcluster],title="fig:",height=226 ] of the largest cluster normalized by the volume as a function of temperature .
we compare the case of su(3 ) ( lhs .
plot ) with su(2 ) gauge theory ( rhs . ) and display data for three different volumes .
the cuts we use are 39 % of sites for su(3 ) and 48 % for su(2 ) .
[ weight_maxcluster],title="fig:",height=226 ] having discussed our cluster construction and the role of the parameter @xmath68 , we now have a first look at a possible percolation behavior near @xmath4 . for that purpose
we analyze the weight @xmath74 of the largest cluster , i.e. , the number of sites in the largest cluster , as a function of temperature . in fig .
[ weight_maxcluster ] we display the expectation value of @xmath74 normalized by the spatial volume @xmath13 as a function of @xmath16 .
we show the results for su(3 ) in the lhs .
plot ( for a cut of 39 % ) and the results for su(2 ) on the rhs .
( for a cut of 48 % ) . in both cases
we compare three different volumes .
below @xmath4 we find that @xmath75 depends on the spatial volume @xmath13 .
the observation that for a fixed lattice constant @xmath75 decreases with increasing volume @xmath13 suggests that below @xmath4 the clusters have a finite maximal size @xmath76 . above @xmath4 the volume dependence is gone and the largest cluster fills a certain fraction of the total volume with a finite density that keeps growing as one further increases the temperature .
this absence of a volume dependence indicates that the clusters are percolating above @xmath4 . for a direct analysis of the percolation probability as function of temperature
see @xcite .
in the previous section we have demonstrated that for suitably defined center clusters percolation sets in at @xmath4 .
so far the free parameter @xmath68 which we introduced in our cluster construction was chosen arbitrarily .
now we change our approach and use a physical scale to set the cluster parameter @xmath68 .
the scale we use is the diameter of the clusters in physical units .
with such a prescription we can compare properties of center clusters on lattices with different resolution @xmath27 . in order to define the diameter of the clusters we consider the correlation functions of points within the same cluster , belonging to the center sector characterized by the sector number @xmath77 , @xmath78
the sector number @xmath77 can have the three values @xmath79 for su(3 ) , while for su(2 ) we have @xmath80 . by @xmath81
we denote the total number of sites @xmath8 with section number @xmath82 .
the first sum runs over all lattice sites , the second one over all three spatial directions @xmath83 and by @xmath84 we denote the corresponding unit vector .
the parameter @xmath85 assumes values @xmath86 , i.e. , in ( [ correlatordef ] ) we consider the correlation along the coordinate axes .
the function @xmath87 is defined through @xmath88 as a function of @xmath85 ( symbols ) , and the corresponding fit with the functional form of ( [ fitfunct ] ) ( full curve ) .
the data are for su(3 ) , @xmath26 at @xmath89 with a cut of 18 % .
[ correlator],height=215 ] up to some correction at short distances , these correlation functions decay exponentially in @xmath85 .
similar to euclidean correlators in lattice spectroscopy , for sufficiently large @xmath85 we fit them with the function @xmath90 where @xmath58 and @xmath91 are two real fit parameters and @xmath92 is the number of lattice points in the spatial direction . as always
, for correlation functions in a finite volume the hyperbolic cosine appears due to the periodic spatial boundary conditions which we use .
an example of the correlation function and the corresponding fit is shown in fig .
[ correlator ] for the case of su(3 ) .
the value of @xmath93 is now used as the diameter in lattice units .
it is converted to the diameter in physical units @xmath94 by multiplication with the lattice constant @xmath27 , such that we obtain the diameter @xmath95 in fm ( or in units of the string tension for su(2 ) ) .
we now use the physical scale @xmath94 to set the value for the cluster parameter @xmath68 .
for example we may decide to analyze clusters that have a fixed diameter of @xmath96 fm defined at some fixed temperature , e.g. , @xmath38 .
then , working on the @xmath97 ensembles , we adjust the parameter @xmath68 such , that our diameter @xmath94 has the desired value of @xmath96 fm .
this procedure can now be repeated on lattices with different lattice constant @xmath27 and we thus can study scaling effects and the continuum limit .
of the clusters in physical units as a function of temperature for the case of su(3 ) ( lhs .
plot ) and su(2 ) ( rhs . ) .
the inserts are a zoom into the low temperature data . in both cases
we compare the results for two different values of the lattice constant @xmath27 .
[ physical_size],title="fig:",height=246 ] of the clusters in physical units as a function of temperature for the case of su(3 ) ( lhs .
plot ) and su(2 ) ( rhs . ) .
the inserts are a zoom into the low temperature data . in both cases
we compare the results for two different values of the lattice constant @xmath27 .
[ physical_size],title="fig:",height=246 ] based on that physical definition of the center clusters we now address the question how the average diameter of the clusters in physical units behaves as a function of temperature . for that purpose we proceed as described in the last paragraph and for a given lattice constant @xmath27 we adjust the cluster parameter @xmath68 such that the clusters have an average diameter of @xmath96 fm ( which is a reasonable mesonic scale for very heavy quarks ) .
we then keep this value of the cluster parameter @xmath68 fixed and change the temperature , to study how the physical cluster diameter changes with @xmath16 . keeping @xmath68
fixed simply means that we work with the same cluster construction for all temperatures , i.e. , the percentage of points we cut is almost the same for all @xmath16 ( compare table 1 ) .
the result of this analysis is presented in fig .
[ physical_size ] where we show the cluster diameter @xmath94 in fm for the case of su(3 ) ( lhs .
plot ) , and for su(2 ) in units of the string tension ( rhs . ) , as a function of temperature .
for both cases we find that below @xmath4 the cluster diameter in physical units remains essentially constant at the value of @xmath94 that we dialed in at @xmath38 . at the critical temperature
the cluster diameter starts to rise quickly indicating that the clusters start to percolate .
we stress at this point that on an infinite lattice the curve for the diameter would jump to infinite slope at @xmath4 . on a finite lattice with @xmath92 lattice points in the spatial directions , the cluster diameter we can determine using the correlator @xmath98
is always bounded by the finite spatial box size , giving rise to a finite slope in fig .
[ physical_size ] above @xmath4 . comparing different values of the box size @xmath99 in physical units , we found that the results for @xmath94 at a fixed value of @xmath100 increase with @xmath92 .
we stress at this point that while it is obvious that at fixed @xmath16 the values for @xmath94 increase with @xmath92 , it is not a priori clear that for fixed @xmath92 the value of @xmath94 is an increasing function of @xmath16 , because we drive the temperature by changing the lattice constant @xmath27 , such that also the physical volume shrinks .
the fact that above @xmath4 @xmath94 keeps growing with @xmath16 is due to an increase of the density of the percolationg cluster which can , e.g. , be seen from fig . 5 , which shows that the number of sites in the largest cluster keeps growing above @xmath4 .
the crucial check for our analysis of the cluster diameter in physical units is the comparison of the results obtained on ensembles with different lattice spacing @xmath27 . for that purpose in fig .
[ physical_size ] we compare the curves for the cluster diameters on @xmath69 and @xmath101 lattices . in both cases
we followed the above described procedure , adjusted the diameter to a fixed physical value @xmath94 at @xmath38 and kept the corresponding cluster parameter @xmath68 for all temperatures . if the cluster picture and the percolation properties we found have a meaning in physical units , the two curves should fall on top of each other .
the plots show that this is the case for both su(3 ) and su(2 ) .
having analyzed the percolation picture in physical units and finding universal behavior when comparing different lattice spacings , let us now turn to another conundrum related to the continuum limit of the center cluster picture .
[ physical_size ] suggests that the center clusters have a fixed physical diameter below @xmath4 ( at least in the range of temperatures we studied ) , and that this behavior can be seen on lattices with different resolution @xmath27 .
this poses the questions how below @xmath4 the physical diameter @xmath102 can remain finite in the limit @xmath103 .
equation ( [ dphys ] ) implies that for the construction of the continuum limit the cluster diameter in lattice units @xmath104 has to diverge , in other words the clusters must become infinite in that limit .
we address this question by comparing clusters of the same physical size on lattices with different resolution @xmath27 .
this comparison is done at the fixed temperature @xmath38 . for each value of @xmath27 we set the cluster parameter @xmath68 such that the physical cluster diameter @xmath94 has the desired size , e.g. , @xmath96 fm . for different values of the lattice spacing @xmath27 different values of @xmath68
are necessary to obtain the desired @xmath94 : for finer lattices we need larger clusters and thus can cut only fewer points than for coarse lattices . by @xmath105
we denote the number of lattice sites that are removed when constructing the clusters according to eqs .
( [ sectornumbers ] ) and ( [ sectornumberssu2 ] ) .
we measure the influence of the parameter @xmath68 by defining the following fraction @xmath106 this fraction measures how many sites are available (= the numerator @xmath107 ) , per volume ( factor @xmath108 ) , per center sector ( factor @xmath109 , where @xmath110 is the number of center elements ) . in other words
@xmath111 measures the fraction of sites available for clusters in a given center sector . in fig .
[ fractionavailable ] we show the flow of @xmath111 with @xmath112 .
the continuum limit is reached for @xmath113 .
we compare the results for different values of @xmath94 , which for the case of su(3 ) we measure in fm ( lhs .
plot ) , while for su(2 ) ( rhs . )
we express @xmath94 in units of the string tension .
of points available for clusters per center sector for the case of su(3 ) ( lhs .
plot ) and su(2 ) ( rhs . ) .
we compare different physical cluster diameters , which for the case of su(3 ) we give in fm , while for su(2 ) the cluster diameter is given in units of the string tension .
[ fractionavailable],title="fig:",height=251 ] of points available for clusters per center sector for the case of su(3 ) ( lhs . plot ) and su(2 ) ( rhs . ) .
we compare different physical cluster diameters , which for the case of su(3 ) we give in fm , while for su(2 ) the cluster diameter is given in units of the string tension .
[ fractionavailable],title="fig:",height=251 ] it is obvious from the plots that for a given @xmath94 the values for the available fraction @xmath111 at different @xmath112 fall on almost perfect straight lines .
moreover , all the lines extrapolate to roughly the same value in the continuum limit , namely @xmath114 .
we use shaded bands to mark the straight lines and their extrapolation to the continuum limit in order to indicate our estimate in the uncertainty of the analysis .
we stress at this point , that the number @xmath114 has nothing to do with the number of colors in su(3 ) the same value is obtained also for the case of su(2 ) .
let us now try to understand the significance of the value @xmath114 . for site percolation in three dimensions
the critical value for the occupation probability is @xmath115 .
the fraction @xmath111 of points available for clusters in each of the sectors extrapolates in the continuum limit to a value of @xmath114 which is just above the percolation threshold @xmath115 . tries to extrapolate to exactly the percolation threshold 0.316 , which would be an extremely beautiful result .
however , with the resources currently available to us we can not perform a calculation which is accurate enough for a determination of @xmath116 with sufficient precision to test this hypothesis . ]
this suggests that the clusters indeed grow to infinity in lattice units as we approach @xmath73 , and a continuum limit of the center clusters may be possible .
we stress again that our calculation should be viewed only as a first indication that a continuum limit might be possible and certainly needs to be substantiated with a ( unfortunately very expensive ) high precision analysis .
, height=168 ]
in this paper we analyze distribution properties of the local polyakov loop in su(3 ) and su(2 ) lattice gauge theory at finite temperature . at each spatial lattice site
we identify the nearest center element and study cluster properties of neighboring sites in the same center sector .
while all center sectors are populated equally below @xmath4 , above @xmath4 one of the sectors is selected spontaneously and becomes more populated at the expense of the other , subdominant sectors . after introducing the cluster parameter @xmath68 one can construct suitable clusters which start percolating at the deconfinement transition .
we address the question whether a continuum limit for the clusters is possible by using the cluster diameter in physical units as a physical scale . working on lattices with different lattice constant @xmath27
we adjust the cluster parameter @xmath68 such that the cluster diameter has a fixed value in physical units . when approaching the continuum limit we find that the fraction of sites which are available for clusters extrapolates to a value just above the percolation threshold .
this implies that in lattice units the clusters become infinite in the continuum limit , while in physical units they remain constant ( below @xmath4 ) .
this finding indicates that the cluster picture could indeed have a well defined continuum limit .
when analyzing the cluster diameter in physical units as a function of temperature , we find it is essentially constant below @xmath4 and becomes infinite above . the center clusters and their percolation at the deconfinement temperature give rise to a simple picture for confinement and the deconfinement transition , which we illustrate in fig .
[ mechanism ] .
below @xmath4 ( lhs . and center panels in fig . [ mechanism ] )
the center clusters have a characteristic finite size . if one places two static sources @xmath5 and @xmath117 at a distance @xmath118 which is small enough to fit into one of the clusters ( lhs .
panel ) , the center phase information is the same at both positions @xmath8 and @xmath66 and cancels in the correlator @xmath119 which thus can have a non - vanishing expectation value . if the distance @xmath118 is too large to fit into a single cluster ( center panel ) , then the two sources will always end up in different clusters .
consequently @xmath5 and @xmath117 will be subject to independent fluctuations of the center phase such that @xmath119 averages to zero .
this averaging does not imply that the correlator vanishes abruptly above a fixed value of @xmath118 , because the sizes of the clusters fluctuate giving rise to the well known exponential decay of @xmath119 . above @xmath4 ( rhs.panel in fig .
[ mechanism ] ) , the center clusters percolate and thus provide a coherent center information for arbitrary large distances . as a consequence , above @xmath4
there are non - vanishing contributions to @xmath119 at arbitrary large distances @xmath118 , and the sources are deconfined . for a further analysis of the center cluster picture
several directions need to be explored .
as already discussed , the numerical evidence presented here can be improved considerably by a large scale precision study of the behavior of the center clusters using ensembles in a wide range of the lattice constant @xmath27 .
in particular the behavior of the fraction @xmath111 of available sites for clusters per center sector should be explored closer to the continuum limit in order to precisely determine the value it extrapolates to .
a second highly important question is how the center clusters change when full qcd with dynamical fermions is considered . in this case
the deconfinement transition turns into a crossover and this different behavior should be reflected in the cluster properties .
preliminary results for that case can be found in @xcite and a more detailed account on the dynamical case is in preparation .
* acknowledgments : * the authors thank szabolcs borsanyi , julia danzer , george flemming , zoltan fodor , michael ilgenfritz , christian lang and axel maas for valuable comments .
the numerical calculations were done at the zid clusters of the university graz .
yaffe , b. svetitsky , phys .
d * 26 * ( 1982 ) 963 ; nucl .
b * 210 * ( 1982 ) 423 .
j. polonyi , k. szlachanyi , phys .
b * 110 * ( 1982 ) 395 ; m. ogilvie , phys .
lett . * 52 * ( 1984 ) 1369 ; f. green , f. karsch , nucl .
b * 238 * ( 1984 ) 297 ; c. wozar , t. kaestner , a. wipf , t. heinzl , b. pozsgay , phys . rev .
d * 74 * ( 2006 ) 114501 ; a. wipf , t. kaestner , c. wozar , t. heinzl , sigma * 3 * ( 2007 ) 006 ; c. wozar , t. kaestner , a. wipf , t. heinzl , phys .
d * 76 * ( 2007 ) 085004 . s. fortunato , h. satz , phys . lett .
b * 475 * ( 2000 ) 311 ; s. fortunato and h. satz , nucl .
a * 681 * ( 2001 ) 466 ; s. fortunato , f. karsch , p. petreczky and h. satz , phys .
b * 502 * ( 2001 ) 321 .
s. fortunato , j. phys .
a * 36 * ( 2003 ) 4269 .
a. schmidt , master thesis , karl - franzens university graz 2010 , http://physik.uni-graz.at/itp/files/schmidt/masterthesis.pdf c. gattringer , phys .
b * 690 * ( 2010 ) 179 .
s. borsanyi , j. danzer , z. fodor , c. gattringer and a. schmidt , proceedings of confinement 2010 , aip ( 2011 ) , [ arxiv:1007.5403 [ hep - lat ] ] ; j. danzer , c. gattringer , s. borsanyi and z. fodor , pos ( lattice 2010 ) 176 ( 2010 ) , [ arxiv:1010.5073 [ hep - lat ] ] . m. lscher , p. weisz , commun .
* 97 * ( 1985 ) 59 [ err . :
* 98 * ( 1985 ) 433 ] ; g. curci , p. menotti , g. paffuti , phys .
b * 130 * ( 1983 ) 205 [ err . :
b * 135 * ( 1984 ) 516 ] . c. gattringer , r. hoffmann , s. schaefer , phys .
d * 65 * ( 2002 ) 094503 . c. gattringer , p.e.l .
rakow , a. schfer , w. sldner , phys .
d * 66 * ( 2002 ) 054502 .
fortuin and p. w. kasteleyn , physica * 57 * ( 1972 ) 536 .
a. coniglio and w. klein , j. phys .
a * 13 * ( 1980 ) 2775 . | properties of local polyakov loops for su(2 ) and su(3 ) lattice gauge theory at finite temperature are analyzed .
we show that spatial clusters can be identified where the local polyakov loops have values close to the same center element . for a suitable definition of these clusters
the deconfinement transition can be characterized by the onset of percolation in one of the center sectors .
the analysis is repeated for different resolution scales of the lattice and we argue that the center clusters have a continuum limit . center clusters in the yang - mills vacuum christof gattringer and alexander schmidt institut fr physik , universitt graz , + universittsplatz 5 , 8010 graz , austria |
the theory of second order phase transitions is based on the assumption that at temperatures close to the critical temperature @xmath6 there is only one dominating length scale associated with the critical behavior of the system , the correlation length . since the correlation length diverges as the critical temperature is approached
the microscopic details of the system are irrelevant for its critical behavior .
this intuitive picture has its foundation in the renormalization group treatment of second order phase transitions . within the renormalization group treatment
it becomes evident that the critical behavior can be divided into different universality classes .
each universality class is characterized by a set of critical exponents which describe the singular behavior of physical quantities in terms of the reduced temperature @xmath11 , e.g. for a three - dimensional bulk system the correlation length @xmath12 diverges close to @xmath6 as @xmath13 .
if the system is confined in a finite geometry ( e.g. a cubic or film geometry ) the singularities in the physical quantities are smoothed out or a crossover to lower - dimensional critical behavior takes place .
finite - size scaling theory @xcite is thought to describe well the behavior of the system at temperatures close to @xmath6 .
the intuitive idea behind the finite - size scaling theory is that finite - size effects can be observed when the bulk correlation length becomes of the order of the system size ( for a film geometry this is the film thickness @xmath3 ) . for a physical quantity @xmath14
this statement can be expressed as follows @xcite : @xmath15 @xmath16 is a universal function depending only on the geometry and the boundary conditions applied . though earlier experiments on superfluid helium films of finite thickness @xcite seemed to confirm the validity of the approach outlined above , in a recent experiment rhee , gasparini , and bishop @xcite showed that their data for the superfluid density of thick helium films do not satisfy eq .
( [ opfss ] ) when the expected value @xmath17 is used .
( for a comprehensive review of experiments on @xmath18he to test the finite - size scaling theory cf .
ref.@xcite . ) as an attempt to understand these discrepancies between theory and experiment renormalization group calculations for the standard landau - ginzburg free energy functional in different geometries with dirichlet boundary conditions ( vanishing order parameter at the boundary ) have been carried out @xcite .
new specific heat measurements @xcite and also a reanalysis @xcite of the old specific heat data @xcite show good agreement between the renormalization group calculations reported in @xcite and those data .
these calculations demonstrated the important role played by the boundary conditions .
in particular , periodic boundary conditions were shown to be inadequate compared to dirichlet boundary conditions to describe the experimental specific heat data .
the renormalization group calculations have determined the specific heat for that range of the scaling variable where the surface contribution to the specific heat is dominant @xcite ( c.f . also @xcite ) .
such field theoretical calculations are not available for the case of the superfluid density and the lack of scaling in the case of the superfluid density of helium films is not understood .
furthermore , new experiments on liquid @xmath19 under microgravity conditions are planned @xcite to examine the finite - size scaling properties of the specific heat . in order to test the renormalization group calculations and
because of the reasons above , numerical investigations of the finite - size scaling properties of the superfluid density @xcite and the specific heat @xcite of thin helium films have been carried out . in refs .
@xcite , we used the @xmath0 model with _ periodic boundary _ conditions in the direction of the film thickness @xmath3 to compute the superfluid density and the specific heat of thin helium films .
we demonstrated _ scaling _ with respect to the film thickness using the expected values for the critical exponents of the superfluid density and the specific heat , thus confirming the validity of the finite - size scaling theory .
however , the obtained universal function for the specific heat does not match the experimentally determined universal function of ref .
@xcite , indicating that periodic boundary conditions are only a poor approximation of the correct physical boundary conditions as was already demonstrated in ref.@xcite .
later we employed staggered - spin boundary conditions in the top and bottom layers of the film which improves the agreement between the numerically computed scaling function and the experimentally determined scaling function of the specific heat @xcite .
another example where the boundary conditions play a role in the scaling behavior comes from ref .
@xcite where the villain model , which also belongs to the @xmath0 universality class , was studied in a film geometry with open boundary conditions in the direction of the film thickness .
the authors of ref .
@xcite extracted the thickness dependent critical temperature from the temperature dependence of the correlation length in the disordered phase and found for the critical exponent @xmath20 the value @xmath21 which is different from its value of @xmath22 known from experiments on liquid helium@xcite . in this paper
we intend to study the effect of staggered - spin boundary conditions ( dirichlet like boundary conditions , i.e. vanishing order parameter on the film boundaries ) on the finite - size scaling behavior of the superfluid density and the specific heat of @xmath19 in a film geometry in detail .
dirichlet like boundary conditions are believed to approximate the physical boundary conditions more closely @xcite . throughout our numerical calculations we are going to describe superfluid @xmath19 near the @xmath23-critical point by another form of the standard landau - ginzburg free energy functional : the @xmath0 model ( cf .
@xcite ) . in the pseudospin notation
the @xmath0 model takes the following form : @xmath24 where the summation is over all nearest neighbors , the two - component vector @xmath25 , and @xmath26 sets the energy scale .
the angle @xmath27 corresponds to the phase of the expectation value of the helium atom creation operator which is defined in a volume whose linear extensions are much larger than the interparticle spacing and much smaller than the correlation length . in this article
we study the superfluid density which corresponds to the helicity modulus in the pseudospin notation and the specific heat of the @xmath0 model in a film geometry , i.e. on @xmath28 lattices with @xmath2 .
the top and bottom layers are coupled to a static staggered spin configuration , playing the role of the `` substrate '' layers , so that the magnetization in these layers is exactly zero .
the crucial difference between these boundary conditions and periodic boundary conditions is that the superfluid density develops a profile in the @xmath3-direction , whereas it is completely homogeneous for periodic boundary conditions ( cf . also the magnetic profile for the ising model in a film geometry with different boundary conditions in ref .
we applied periodic boundary conditions in the @xmath4-directions because we intend to take the limit @xmath29 .
in the temperature range where the model behaves effectively two - dimensionally we used the kosterlitz - thouless - nelson renormalization group equations to compute the values for the helicity modulus in the @xmath30 limit .
this way we eliminated the @xmath4-dependence of our data for the helicity modulus and were able to extract the kosterlitz - thouless transition temperature @xmath31 for different films .
we investigated the validity of finite - size scaling for the superfluid density and the specific heat of such films of infinite planar dimension and finite @xmath3 .
we shall also discuss scaling of the experimental results for the specific heat and superfluid density with respect to the film thickness and compare the universal scaling functions to those obtained from our theoretical investigation .
the rest of this paper is organized as follows . in the next section
we introduce the physical observables defined for the @xmath0 model and the numerical method we applied to carry out the calculations . in section [ sec3 ] , we discuss the finite - size scaling theory and boundary effects . in section [ sec4 ]
we discuss our results and the last section briefly summarizes the work described in this paper .
for the @xmath0 model on a cubic lattice the helicity modulus is defined as follows @xcite : @xmath32 where @xmath33 is the volume of the lattice , @xmath34 , @xmath35 is the unit vector in the corresponding bond direction , and @xmath36 is the vector connecting the lattice sites @xmath37 and @xmath38 . in the following
we will omit the vector index since we will always refer to the @xmath39-component of the helicity modulus .
note that , because of isotropy , we have @xmath40 .
the connection between the helicity modulus and the superfluid density @xmath41 is established by the relation @xcite @xmath42 where @xmath43 denotes the mass of the helium atom .
the specific heat @xmath44 is obtained from the energy @xmath45 through @xmath46 where the energy is defined as : @xmath47 and @xmath48 is the number of spins contributing to the specific heat .
the thermal averages denoted by the angular brackets are computed according to @xmath49 \exp ( - \beta e ) .
\label{ev}\end{aligned}\ ] ] @xmath50 $ ] denotes the dependence of the physical observable @xmath14 on the configuration @xmath51 , the partition function @xmath52 is given by @xmath53 the multi - dimensional integrals in the expressions ( [ ev ] ) and ( [ z ] ) are computed by means of the monte carlo method using wolffs 1-cluster algorithm @xcite .
we computed the helicity modulus and the specific heat on @xmath28 lattices , where @xmath54 and @xmath55 .
we applied periodic boundary conditions in the @xmath4-directions , whereas the first and the @xmath3-th layer are coupled to a boundary layer which consists of a staggered spin configuration , i.e. @xmath56 where @xmath57 $ ] and label the integer coordinates of the lattice sites in a plane perpendicular to the @xmath3-direction .
thus , we have @xmath58 with @xmath59 denoting the lattice spacing and @xmath60 .
we carried out of the order of @xmath61 thermalization steps and of the order of @xmath62 measurements .
the calculations were performed on a heterogeneous environment of computers including sun , ibm rs/6000 and dec alpha axp workstations and a cray - ymp .
here we shall discuss the finite - size scaling theory of the superfluid density . in ref .
@xcite , we studied the helicity modulus @xmath63 for the @xmath0 model in a film geometry with _ periodic _ boundary conditions in the @xmath3-direction and we have shown the following steps . in a certain temperature range around the bulk critical temperature @xmath6 where the bulk correlation length @xmath64 becomes of the order of the film thickness @xmath3 the quantity @xmath65 exhibits effectively two - dimensional behavior and a kosterlitz - thouless phase transition takes place at a temperature @xmath66 .
the critical temperature @xmath31 approaches @xmath6 in the limit @xmath67 as @xmath68 where for periodic boundary conditions we found that @xmath69 using for the critical exponent @xmath20 the experimental value @xmath70 @xcite and for @xmath6 the value @xmath71 @xcite .
the quantity @xmath65 is a function of the ratio @xmath72 , i.e. @xmath73 the universal function @xmath74 has the properties @xcite @xmath75 in the limit @xmath76 the function @xmath74 can be written as : @xmath77 where for periodic boundary condition we found that @xmath78 .
this form of the universal function reconciles the scaling expression ( [ tyh ] ) with the two - dimensional behavior @xcite , i.e. as @xmath79 @xmath80 where @xcite @xmath81 the results stated above confirm the theoretical expectations about scaling given by ambegaokar et al . in ref .
@xcite and agree with the experimental findings of bishop and reppi @xcite and rudnick @xcite . in the case of periodic boundary conditions in the @xmath3-direction
the validity of the finite - size scaling form ( [ tyh ] ) can already be observed for films of thicknesses @xmath82 @xcite .
if nonperiodic boundary conditions are introduced privman argued that the general scaling form ( [ tyh ] ) has to be altered into @xcite @xmath83 where @xmath84 and @xmath85 are constants depending on the boundary conditions .
the monte carlo data for the helicity modulus obtained from a monte carlo simulation of the @xmath0 model on cubes @xmath86 where the spins in the boundary layers were all parallel ( pinned - surface - spin boundary conditions ) , were found to be consistent with the presence of the logarithmic term in the scaling form ( [ yhtdbc ] ) @xcite .
let us now investigate the consequences of the logarithmic term in ( [ yhtdbc ] ) .
again we have to reconcile the two - dimensional behavior ( [ kinf ] ) and the general scaling form ( [ yhtdbc ] ) for temperatures close to the critical temperature @xmath31 for thick enough films . introducing the expression ( [ tch ] ) for the @xmath3-dependent critical temperature into eq .
( [ kinf ] ) and keeping only terms up to @xmath87 under the square root leads to @xmath88 where we have absorbed the factor of @xmath89 in the definition of @xmath90 . with the assumption ( [ bh ] )
we may make the identification @xmath91 in order to account for the logarithmic term in ( [ yhtdbc ] ) we have to abandon the universal jump at @xmath31 @xmath92 instead we have to assume @xmath93 i.e. the jump becomes @xmath3-dependent .
( the numerical values for @xmath94 and @xmath95 will be different from the values given above as they depend on the boundary conditions . ) eq .
( [ hjum ] ) means that for nonperiodic boundary conditions in the @xmath3-direction expression ( [ kinf ] ) has to be generalized to @xmath96 where @xmath97 for the finite - size scaling of the specific heat one can use similar scaling expressions to eq .
( [ opfss ] ) which were examined in detail in ref . @xcite .
the finite - size scaling expression for the specific heat @xmath44 can also be written in an equivalent way as @xcite : @xmath98 the function @xmath99 is universal and @xmath70 . at the reduced temperature @xmath100 the correlation length is equal to the film thickness @xmath3 , i.e. @xmath101 with @xmath102 @xcite
this scaling form has been used to analyze the experimental data and , thus , we shall also discuss the scaling of the specific heat using this form in order to compare to published experimental results for the universal function @xmath10 .
we have @xmath103 where we have found that @xmath104 , @xmath105 @xcite and @xmath106 because of the hyperscaling relation @xmath107 . in ref .
@xcite we demonstrated that our numerical results for the specific heat of the @xmath0 model on a film geometry with periodic boundary conditions follow the finite - size scaling form ( [ cimph ] ) for the thicknesses as small as @xmath82 .
here we shall present our results for the superfluid density and the specific heat and our analysis for the case of _ staggered - spin _ boundary conditions as defined in [ sec2 ] . in this section
we would like to determine the values of the ratio @xmath108 in the limit @xmath109 and find estimates for the critical temperatures @xmath31 and the parameters @xmath110 and @xmath90 ( cf .
( [ yht2dg ] ) ) . in order to do this
we follow closely the procedure described in refs .
@xcite .
[ fig0 ] displays the monte carlo data for the helicity modulus in units of the lattice spacing @xmath59 and the energy scale @xmath111 for the film of fixed thickness @xmath112 .
this figure demonstrates that staggered boundary conditions for the top and bottom layers of the film strongly suppress the values of the helicity modulus with respect to the case of periodic boundary conditions . as a consequence films with staggered boundary conditions have a smaller critical temperature than films with periodic boundary conditions . at the temperature @xmath113 we computed the helicity modulus on a @xmath114 lattice for each layer separately and plotted the result @xmath115 in fig .
[ fig0a ] , where @xmath116 enumerates the layers .
the layered helicity modulus is symmetric with respect to the middle layer where it reaches its maximum and decreases when the boundaries are approached .
although the helicity modulus @xmath117 is not the average of the quantity @xmath115 over all layers ( this is due to the second nonlinear term in expression ( [ lathel ] ) ) , the curve in fig .
[ fig0a ] is an approximation to the profile the superfluid density develops in thin films .
let us turn now to the computation of the values for the ratio @xmath118 in the @xmath109 limit .
for a fixed thickness @xmath3 and at temperatures @xmath119 below but sufficiently close to the critical temperature @xmath31 the system behaves like a two - dimensional system@xcite .
in this regime we demonstrated@xcite that the dimensionless ratio @xmath120 obeys the kosterlitz - thouless - nelson renormalization group equations @xcite : @xmath121 @xmath122 is the chemical potential to create a single vortex , @xmath123 denotes the size of the core radius of a vortex .
these equations contain the universal jump @xmath124 . in order to adjust the above equations to the possibility of an @xmath3-dependent jump of the ratio @xmath120 at @xmath31 we generalize equations ( [ rg1 ] ) and ( [ rg2 ] ) to : @xmath125 where @xmath126 and @xmath127 are @xmath3-dependent constants .
after eliminating the variable @xmath128 from the coupled system of differential equations we obtain : @xmath129 where @xmath130 is a constant which satisfies the condition @xmath131 .
this condition allows for the existence of roots of the right hand side of eq .
( [ dglk ] ) .
if we identify the scale @xmath132 with @xmath133 up to a constant we can use eq .
( [ dglk ] ) to extrapolate the computed values @xmath134 obtained on lattices of finite planar dimension @xmath4 to the @xmath135 limit .
namely , at @xmath135 the left hand side of eq .
( [ dglk ] ) vanishes @xcite and @xmath136 is given by the root of the right hand side of eq .
( [ dglk ] ) .
the parameters @xmath127 and @xmath130 are found by fitting the monte carlo data for @xmath134 at a fixed @xmath3 to the numerical solution to eq .
( [ dglk ] ) .
table [ ta1 ] contains our fitting results for the fitting parameters @xmath127 and @xmath130 and the values for @xmath137 for the thicknesses @xmath138 $ ] and fig .
[ fig1 ] shows a typical fit .
we were not able to explore the two - dimensional region for the film with @xmath139 because the temperature range where the film behaves two - dimensional becomes very narrow and our computer resources did not allow an accumulation of data accurate enough to resolve this region . in fig .
[ fig1a ] we plot @xmath108 versus @xmath140 for the thicknesses @xmath141 to check the validity of the scaling form ( [ tyh ] ) where @xmath70 . the data for the helicity modulus used in fig .
[ fig1a ] have completely lost their @xmath4-dependence .
we do not obtain a universal scaling curve , thus scaling according to the expression ( [ tyh ] ) is not valid for the films with thicknesses up to @xmath139 .
therefore we will try to employ the scaling form ( [ yhtdbc ] ) which requires the knowledge of @xmath110 .
since @xmath142 satisfies eq .
( [ yht2dg ] ) for a fixed @xmath3 and temperatures close enough to the critical temperature @xmath31 we can fit the obtained results for @xmath142 to eq .
( [ yht2dg ] ) and find an estimate for @xmath31 and the parameters @xmath90 and @xmath110 . in table [ ta2 ] we present our fitting results and fig .
[ fig2 ] shows the fit to the data for @xmath108 at @xmath112 .
it is interesting to note that the @xmath3-dependence of the parameter @xmath143 can be described by the formula @xmath144 for @xmath138 $ ] .
this is consistent with privman s prediction @xcite . in fig .
[ fig3 ] we plot @xmath145 versus @xmath140 for @xmath146 where @xmath110 is given in table [ ta2 ] , the bulk critical temperature @xmath71 . also the scaling form ( [ yhtdbc ] ) does not collapse our data points onto one universal curve .
this situation is the same as the one rhee , gasparini and bishop encountered when they tried to verify finite - size scaling for their data of the superfluid density @xcite .
their data of the superfluid density did not fall onto one universal curve when the scaling form ( [ tyh ] ) was employed , neither did the inclusion of a logarithmic term as in eq .
( [ yhtdbc ] ) help to achieve data collapse @xcite . of course , one reason for the failure of scaling of our data of the helicity modulus according to the expressions ( [ tyh ] ) or ( [ yhtdbc ] ) could be that our thicknesses are still too small . on
the other hand rhee et al .
use films of macroscopic sizes and do not confirm scaling .
furthermore , for films with periodic boundary conditions scaling of the helicity modulus occurs already for thicknesses as small as @xmath147 @xcite .
let us therefore pursue another line of thought @xcite which we borrow from the mean field treatment of thin ferromagnetic films @xcite .
the reduced critical temperature of a ferromagnetic film @xmath148 ( @xmath149 where @xmath150 is the 3d bulk critical temperature of the ferromagnet ) can be obtained from the following set of equations @xcite : @xmath151 the lattice spacing is denoted by @xmath59 and @xmath23 is the extrapolation length @xmath23 ( cf . also ref .
@xcite ) . let us compute @xmath148 in the limit @xmath152 . from fig.2 of ref.@xcite
it is clear that @xmath153 in the limit @xmath154 .
thus , writing @xmath155 with @xmath156 eq.([e1 ] ) turns into @xmath157 solving for @xmath158 yields @xmath159 inserting this into eq.([e2 ] ) we obtain finally @xmath160 thus , within the mean field treatment the reduced critical temperature scales with the correct critical exponent @xmath161 but with an effective thickness @xmath162 .
it is interesting that @xmath163 appears in eq.([e5 ] ) .
this means that we have to add twice the extrapolation length ( for each side of the film ) to the actual thickness .
furthermore , if the magnetization is suppressed close to the boundaries @xmath164 the critical temperature @xmath165 is smaller than the 3d bulk critical temperature and we have to add @xmath163 to the actual thickness .
the lack of scaling of our data of the helicity modulus with the expected critical exponent @xmath70 indicates that the critical temperatures @xmath31 do not satisfy eq .
( [ tch ] ) . instead , due to the profile of the superfluid density we may expect an effective film thickness @xmath166 which enters the scaling expressions ( [ tch ] ) and ( [ tyh ] ) . in close analogy to the mean field treatment of ferromagnetic films discussed in the paragraph above
, we assume that @xmath167 where @xmath8 is a constant . indeed ,
for the film thicknesses @xmath168 we obtain @xmath169 and @xmath170 with @xmath70 . in fig .
[ fig4 ] we plot @xmath171 as a function of @xmath172 for films with @xmath141 where @xmath70 . the data for the helicity modulus collapse reasonably well onto a single curve
. we can understand the increment @xmath8 as a scaling correction which renders the scaling relations ( [ tch ] ) and ( [ tyh ] ) valid even for very thin films . for large thicknesses @xmath3
the increment @xmath8 can be neglected and we recover the conventional scaling forms .
of course , it is possible to invent scaling forms different from the structure ( [ e5 ] ) which yield the conventional scaling expressions in the limit @xmath67 .
for example , we have obtained similarly good fitting results using the expression @xmath173 which is also motivated by mean field theory ( cf .
ref.@xcite ) .
however , since we find eq.([e5 ] ) physically appealing we continue to describe the effects of boundary conditions by an effective thickness as we have done in this paragraph . in order to test further the assumption that the boundaries introduce an effective thickness into the scaling expression ( [ tyh ] ) we try to describe the thickness dependence of the kosterlitz - thouless transition temperature of the villain model with open boundary conditions ( interactions of the top and bottom layer only with the interior film layers ) @xcite by eq .
( [ tch ] ) , where @xmath3 is replaced by the effective thickness @xmath174 .
indeed , taking @xmath70 we find @xmath175 thus @xmath176 and @xmath177 .
the function ( [ tchjanke ] ) is the solid line in fig .
again for this case , the increment @xmath8 is a correction which makes the scaling relations ( [ tch ] ) and ( [ tyh ] ) valid even for very thin films .
the result ( [ tchjanke ] ) means that the film thicknesses considered in ref .
@xcite were still too small to extract the expected value of the critical exponent @xmath20 from the @xmath3-dependence of the critical temperature ( [ tch ] ) without the help of an effective thickness @xmath178 . in fig .
[ fig6 ] we achieve approximate collapse of the experimental data of rhee et al.@xcite for the superfluid density @xmath41 for films of various thickness @xmath85 ( @xmath85 is in @xmath179 ) by plotting @xmath180 versus @xmath181 with @xmath70 and @xmath182 .
this value of the effective thickness was found by examining the reduced temperatures @xmath183 where finite - size effects set in . according to finite - size scaling theory
@xmath184 has to fulfill the relation @xmath185 , thus in our case @xmath186 .
the data corresponding to the film with @xmath187 deviate from the universal curve and we attribute this to the anomalous behavior of these data .
namely , in general @xmath188 if @xmath189 , but this is not the case for @xmath190 and @xmath191 ( cf .
@xcite ) . let us compare the increments over the film thickness for the three cases of film geometry considered above .
for the villain model with open boundary conditions we obtained @xmath176 while for the @xmath0 model with staggered boundary conditions we obtained @xmath192 .
all increments are expressed in lattice spacing units .
the difference in these values of the increments reflect how severe the effect of the boundary conditions is .
open boundary conditions are less demanding on the order parameter at the boundary compared to staggered boundary conditions used in our simulations .
the value of the increment in the case of @xmath18he on silicon is large , @xmath193 in lattice spacing units @xmath59 ( @xmath194 @xcite ) .
this might cast some doubts on our proposed scaling form for the superfluid data of rhee et al .
it is possible , however , to imagine that on the surface of these films vortices are pinned by impurities or other forms of disorder , which make the order parameter vanish at the boundary and which introduce an effective length scale of such a magnitude .
so far we have seen that different boundary conditions create different effective thicknesses .
the influence of the boundaries vanishes for thick enough films and only in a certain small range of film thickness the influence of the boundary conditions has to be taken into account .
according to our findings the scaling form ( [ unif ] ) for the helicity modulus in the limit @xmath79 takes the following form now : @xmath195 where @xmath95 , @xmath94 and @xmath196 are constants depending on the boundary conditions . especially for @xmath197
we should have @xmath198 and this value of @xmath199 is not necessarily equal to @xmath89 . in table
[ ta3 ] we give the values for @xmath196 found by our fitting procedure .
we still have a slight @xmath3-dependence in @xmath196 but for the thicknesses @xmath200 the value for @xmath196 seems to saturate at @xmath201 .
since we can neglect the effective thickness for very large film thicknesses , this means that films with staggered boundary conditions in the @xmath3-direction of the film exhibit a jump in the quantity @xmath202 that is different from @xmath89 which was found for films with periodic boundary conditions @xcite .
therefore , assuming that our extrapolation to large film thicknesses from small size films using the idea of the effective thickness is valid , we have to conclude that the jump @xmath202 depends on the boundary conditions . in principle
there is nothing wrong with this conclusion because the universal functions ( and the jump is a particular feature of a particular universal function ) depend very importantly ( especially near the critical temperature ) on the boundary conditions .
the scaling function should not be confused with the critical exponents which are independent of the geometry and boundary conditions .
the scaling functions for given universality , given geometry and boundary conditions are universal .
this leads us to the conclusion that this jump in the experimental findings should depend on the substrates ( cf .
also ref.@xcite ) .
this influence of the substrate must be mediated by the vortices whose generation is enhanced close to the boundaries due to the effect of the boundary ( for a more detailed discussion cf .
section [ disc ] ) . in the experiments
the value of @xmath89 was found @xcite , however , the thicknesses of these films are much smaller than the above length scale @xmath8 found to fit the data of rhee et al .
we believe that the vortex density was almost homogeneous throughout the films used for these measurements .
this situation corresponds to the @xmath0 model in a film geometry with periodic boundary conditions along the film direction where the vortex density is the same everywhere . in this section
we would like to investigate the finite - size behavior of the specific heat @xmath203 . since we do not possess an easily handable procedure to take the @xmath109 limit for the values of the specific heat @xmath204 computed on finite lattices @xmath205 we approximate films with infinite planar dimension by @xmath206 lattices .
this seems justified because the specific heat appears independent of @xmath4 for @xmath207 ( cf .
[ fig8 ] ) .
furthermore , we do not expect the maximum of the specific heat to grow dramatically with increasing values of @xmath4 because for temperatures in the range @xmath208 the behavior of the specific heat can be described by the kosterlitz - thouless theory which leads to a finite value of this maximum . in order to illustrate this argument we show in fig .
[ fig11 ] the size dependence of the specific heat @xmath209 computed on pure two - dimensional lattices @xmath210 with periodic boundary conditions .
the @xmath4-dependence of the specific heat can be neglected for values of @xmath211 . in fig .
[ fig13a ] we compare the specific heat of films with @xmath212 to the bulk specific heat @xmath213 taken from ref .
the specific heat values for films of thickness @xmath168 lie above the bulk curve for @xmath214 .
such a behavior is also found in experiments on helium films about 30 thick @xcite .
this crossing effect is due to the large shift of the temperature @xmath215 where the specific heat for a certain film thickness takes its maximum down to temperatures below the bulk critical temperature @xmath6 . for
thicker and thicker films the maximum temperature @xmath215 approaches @xmath6 , thus the confined specific heat data will fall below the bulk curve which is expected from the field theoretical calculations@xcite .
the film thicknesses used in our monte carlo calculation range from @xmath216 to @xmath217 ( the lattice spacing @xmath194 @xcite ) which is comparable to the experimental film thicknesses ( 30 ) where this crossing effect can be observed . fig .
[ fig13b ] shows the specific heat of the film with @xmath139 alone , indicating that the effect of crossing the bulk curve indeed vanishes for thicker and thicker films .
unfortunately it is beyond our means to carry out the necessary analysis for thicker films than were treated here .
it is interesting to note that we find the same qualitative behavior of the specific heat in the case of the @xmath0 model in a cylindrical geometry @xcite . in fig .
[ fig8b ] we plot the scaling function @xmath99 given by expression ( [ cimph ] ) .
according to the previous discussion this scaling function can only be an approximation to the correct one which one would need to compute from films with @xmath218 and @xmath219 .
we find that our data for the specific heat for films of various thicknesses collapse approximately onto a single curve .
it seems that the specific heat is rather insensitive to the boundary effect of introducing an effective thickness and a very high accuracy in the computation of the specific heat is needed to detect the presence of the effective thickness in the scaling function @xmath99 .
for example , one could determine the temperatures @xmath220 where the specific heat reaches its maximum and examine the validity of eq .
( [ tch ] ) for @xmath220 , because the @xmath3-dependence of @xmath220 is also given by expression ( [ tch ] ) ( with a different value for @xmath94 than for the critical temperatures @xmath31 ) .
this can be done more easily in experiments because in monte carlo calculations an extrapolation procedure for the values of the specific heat at finite planar dimensions to the values at infinite planar dimensions is needed ( and which is not available at present ) whereas the films used in experiments represent films with infinite planar dimensions .
we can directly compare our function @xmath221 to the experimentally determined scaling function @xmath99 given in refs .
@xcite by expressing all lattice units in physical units using the conversion formula : @xmath222 where @xmath223 is the molar volume of @xmath19 at saturated vapor pressure at @xmath6 , the lattice spacing @xmath194 @xcite and the film thickness is measured in . in fig .
[ fig12 ] we compare the functions @xmath99 obtained from monte carlo calculations of the specific heat of films with periodic boundary conditions and staggered boundary conditions in the direction of the film thickness to the experimentally determined function @xmath99 and to the function @xmath99 obtained from field theoretical calculations @xcite .
this figure clearly shows the influence of the boundary conditions on the shape of the universal function as was already demonstrated by the field theoretical calculations reported in ref .
@xcite . in fig .
[ fig12 ] we see that the scaling function @xmath99 for films with staggered boundary conditions crosses the scaling function @xmath99 for films with periodic boundary conditions ( cf .
also @xcite ) in the range @xmath224 ( cf .
[ fig12 ] ) with @xmath3 measured in .
this crossing is due to the relative smallness of our film thicknesses ( see the discussion above ) and does not occur if we had used much thicker films in our monte carlo calculations to deduce the scaling function @xmath225 @xcite .
we expect our function @xmath225 to be slightly modified in the range @xmath224 ( cf .
fig.[fig12 ] ) when it is computed from much thicker films .
we believe that the wings of our curve , however , will remain unchanged .
unfortunately , it is beyond our computational means to repeat the calculations for thicknesses larger than 24 .
our findings suggest that it is possible to introduce an effective thickness @xmath167 into the scaling expressions for the superfluid density ( [ tyh ] ) and achieve scaling ( i.e. data collapse ) even for rather thin films .
the increment @xmath8 over the film thickness @xmath3 can be understood as an effective correction to scaling .
the appearance of the effective thickness in our scaling - function can be understood as follows .
the superfluid density has a finite value in the first layers next to the boundary layers of the film , and its rise from this finite value to its bulk value @xmath226 ( for @xmath227 ) inside the film can be divided into two regions .
let us assume that @xmath119 is very close to @xmath165 where the correlation length is very large .
there is a rather narrow region of thickness @xmath228 of the film which is in contact with the boundary wall where the superfluid density rises very fast to attain some value @xmath229 .
then it rises with a much slower rate over a length scale of the order of the correlation length to reach its value of @xmath230 .
the reason for the initial fast rise are the correlations over length scales much smaller than the correlation length .
one might think that one then has to exclude the region of the film where the superfluid density rises very sharply and this leads to a negative value for @xmath8 .
however , this initial rise of @xmath231 occurs very fast over a length scale @xmath228 which is smaller than a length @xmath232 which would have been required in order for the superfluid density to reach the same value if this rise would have occurred over larger distances over which the spin - spin correlations are governed by the correlation length which controls the long distance behavior of the correlation function .
this implies that the required increment to the thickness is @xmath233 , which is positive . only for films with thicknesses
@xmath3 which fulfill @xmath234 scaling with @xmath3 can be observed . for periodic boundary conditions we have @xmath235 @xcite
, open boundary conditions seem to yield @xmath236 and for staggered boundary conditions we obtain @xmath237 . due to their structure staggered boundary conditions
support vortex formation close to the boundaries , thus the superfluid density decreases from its maximum in the middle of the film towards the boundaries .
this effect is less pronounced for open and absent for periodic boundary conditions .
our results imply that the more the superfluid density is suppressed near the boundaries the larger is the value of @xmath8 . the data for the superfluid density of rhee et al .
@xcite which correspond to @xmath19 on si require a large value of @xmath238 .
thus , si should suppress the superfluid density dramatically close to the boundary .
since @xmath8 is so large , only films with @xmath239 should allow scaling with @xmath3
. it would be interesting to investigate the scaling behavior of the superfluid density of @xmath19 on different substrates ( which represent different types of boundary conditions ) in a wide range of film thicknesses to check our hypothesis .
a consequence of scaling our data of the helicity modulus ( or superfluid density ) using an effective thickness is that the jump in the quantity @xmath202 depends on the boundary conditions in the top and bottom layers of the film , i.e. on the substrates in real helium experiments .
the value of the jump is only @xmath89 ( in lattice units ) for film thicknesses small compared to @xmath8 as is the case in the experiments reported in refs .
the same value of the jump was found is the case for @xmath0 films with periodic boundary conditions @xcite .
thus , experiments could be also used to determine the jump in the areal superfluid density at the kosterlitz - thouless temperature and determine its substrate dependence . in this work
we represented the substrate by a staggered spin configuration coupled to the top and bottom layers of the film to simulate dirichlet - like boundary conditions ( zero order parameter in the boundary ) in the substrate .
for the staggered spin configuration the local magnetization is exactly zero on a plaquette , i.e. on a domain of the size @xmath240 where @xmath59 denotes the lattice spacing .
dirichlet - like boundary conditions in the substrate are , however , also realized by the following spin configuration : the spins are parallel in a square domain of linear dimension @xmath241 , but any two spins representing two adjacent domains are antiparallel to each other .
thus , an additional length scale associated with a finite value of the local magnetization over the length scale @xmath241 is introduced and influences the scaling behavior of the helicity modulus and the specific heat .
since disorder in the boundaries supports vortex formation close to the boundaries , vortices should play an active role in creating the boundary effect described above .
we have investigated the finite - size scaling properties of the specific heat @xmath44 and the helicity modulus @xmath63 of the @xmath0 model in a film geometry , i.e. on @xmath1 lattices with @xmath2 where staggered and periodic boundary conditions where applied in the @xmath3-direction and the @xmath4-directions of the film , respectively .
we found that a kosterlitz - thouless phase transition takes place at the @xmath3-dependent critical temperatures @xmath31 , however , for the films used in our calculations the jump @xmath242 appears to be @xmath3-dependent .
furthermore , scaling of the helicity modulus according to eq .
( [ tyh ] ) is not valid for our film thicknesses , neither is scaling according to eq .
( [ yhtdbc ] ) which was derived following a suggestion of privman @xcite . introducing an effective thickness @xmath243 into the scaling expression ( [ tyh ] )
we are able to collapse our data as well as the data of rhee et al .
@xcite reasonably well onto one universal curve .
our results suggest that the boundary effect of creating an effective thickness @xmath166 depends on the boundary conditions which can be realized in experiments by different substrates and is negligible for thicknesses @xmath3 which fulfill @xmath244 .
we argue that the jump in the quantity @xmath202 depends on the boundary conditions and is @xmath89 only for certain ideal boundary conditions such as the periodic boundary conditions . within error
bars scaling of the specific heat does not require an effective thickness and the scaling function @xmath99 for the specific heat agrees rather well with the experimentally determined scaling function @xmath99 and with the result of the renormalization group calculations reported in @xcite .
however , we found that monte carlo simulations of much thicker films than we have used have to be performed to determine the position of the maximum of the scaling function accurately . at present
this is unfortunately beyond our computer resources .
we would like to thank prof .
dohm for interesting and useful discussions .
n. s. would like to thank the hchsleistungsrechenzentrum jlich for the opportunity of using their computing facilities .
this work was supported by the national aeronautics and space administration under grant no .
nag3 - 1841 and by sonderforschungsbereich 341 der deutschen forschungsgemeinschaft .
m. e. fisher and m. n. barber , phys .
* 28 * , 1516 ( 1972 ) ; m. e. fisher , rev .
phys . * 46 * , 597 ( 1974 ) ; v. privman , finite size scaling and numerical simulation of statistical systems , singapore : world scientific 1990 .
e. brezin , j. physique * 43 * , 15 ( 1982 ) .
j. maps and r. b. hallock , phys .
lett * 47 * , 1533 ( 1981 ) .
i. rhee , f. m. gasparini , and d. j. bishop , phys .
lett . * 63 * , 410 ( 1989 ) .
i. rhee , d. j. bishop , and f. m. gasparini , physica * b165&166 * , 535 ( 1990 ) .
f. m. gasparini and i. rhee , prog . low temp .
xiii * , 1 ( 1992 ) .
r. schmolke , a. wacker , v. dohm , and d. frank , physica * b165 & 166 * , 575 ( 1990 ) . v. dohm , physica scripta * t49 * 46 ( 1993 ) .
p. sutter and v. dohm , physica * b194 - 196 * , 613 ( 1994 ) . w. huhn and v. dohm , phys .
. lett . * 61 * , 1368 ( 1988 ) .
m. krech and s. dietrich , phys . rev . * a46 * 1886 ( 1992 ) , 1922 ( 1992 ) . j. a. nissen , t. c. p. chui , and j. a. lipa , j. low temp . phys . *
92 * , 353 ( 1993 ) , physica * b194 - 196 * , 615 ( 1994 ) .
a. wacker and v. dohm , physica * b194 - 196 * 611 ( 1994 ) . t. chen and f. m. gasparini , phys .
lett . * 40 * , 331 ( 1978 ) ; f. m. gasparini , t. chen , and b. bhattacharyya , phys . rev . * 23 * , 5797 ( 1981 ) .
m. pitz , diploma thesis , rwth aachen , germany ( 1996 ) .
m. coleman and j. a. lipa , czech .
* 46 * ( suppl .
s1 ) , 183 ( 1996 ) .
j. a. lipa , private communications .
n. schultka and e. manousakis , phys . rev . *
b51 * , 11712 ( 1995 ) .
n. schultka and e. manousakis , phys . rev . *
b52 * , 7528 ( 1995 ) .
n. schultka and e. manousakis , phys .
* 75 * , 2710 ( 1995 ) .
w. janke and k. nather , phys .
* b48 * 15807 ( 1993 ) . l. s. goldner and g. ahlers , phys . rev . *
b45 * , 13129 ( 1992 ) . v. l. ginzburg and l. p. pitaevskii , soviet physics , j. e. t. p. , * 34 * , 858 ( 1958 ) .
v. l. ginzburg , soviet physics , j. e. t. p. * 2 * , 170 ( 1956 ) .
h. kleinert , gauge fields in condensed matter , singapore : world scientific 1989 .
s. dasgupta , d. stauffer , and v. dohm , physica * a213 * , 368 ( 1995 ) .
s. teitel and c. jayaprakash , phys . rev . *
b27 * , 598 ( 1983 ) . y .-
h . li and s. teitel , phys .
* b40 * , 9122 ( 1989 ) .
m. e. fisher , m. n. barber and d. jasnov , phys . rev .
* b16 * , 2032 ( 1977 ) .
u. wolff , phys .
lett . * 62 * , 361 ( 1989 ) .
w. janke , phys .
a148 * , 306 ( 1992 ) .
v. ambegaokar , b. i. halperin , d. r. nelson and e. d. siggia , phys . rev . *
b21 * 1806 ( 1980 ) .
d. r. nelson and j. m. kosterlitz , phys .
lett . * 39 * , 1201 ( 1977 ) .
r. g. petschek , phys .
* 57 * , 501 ( 1986 ) .
d. j. bishop and j. d. reppy , phys .
40 * , 1727 ( 1978 ) .
i. rudnick , phys .
lett . * 40 * , 1454 ( 1978 ) .
v. privman , j. phys .
* a23 * , l711 ( 1990 ) .
k. k. mon , phys . rev . * b44 * , 2643 ( 1991 ) .
a. p. gottlob and m. hasenbusch , physica * a201 * , 593 ( 1993 ) .
n. schultka and e. manousakis , phys . rev . *
b49 * , 12071 ( 1994 ) .
j. v. jose , l. p. kadanoff , s. kirkpatrick , and d. r. nelson , phys . rev .
* b16 * , 1217 ( 1977 ) j. m. kosterlitz , j. phys . *
c7 * , 1046 ( 1974 ) ; j. m. kosterlitz and d. j. thouless , j. phys . * c6 * , 1181 ( 1973 ) . n. schultka and e. manousakis , j. low temp . phys .
* 105 * , 3 ( 1996 ) .
m. i. kaganov and a. n. omelyanchuk , sov .
jetp * 34 * , 895 ( 1972 ) . k. binder and p. c. hohenberg , phys . rev . *
b 6 * , 3461 ( 1972 ) .
n. schultka and e. manousakis , czech .
j. phys . * 46 ( suppl .
s1 ) * , 451 ( 1996 ) .
v. dohm , private communication .
n. schultka and e. manousakis , in preparation .
j. rudnick and g. gaspari , phys .
b*32 * , 7594 ( 1985 ) .
.[ta1 ] fitted values of the parameters @xmath127 and @xmath130 of the expression ( [ dglk ] ) and the extrapolated values @xmath137 at various temperatures @xmath119 and different thicknesses @xmath3 . @xmath245 and the goodness of the fit @xmath246 are also given .
[ cols= " < , < , < , < , < , < , < " , ] | we have studied the superfluid density and the specific heat of the @xmath0 model on lattices @xmath1 with @xmath2 ( i.e. on lattices representing a film geometry ) using the cluster monte carlo method . in the @xmath3-direction we applied staggered boundary conditions so that the order parameter on the top and bottom layers is zero , whereas periodic boundary conditions were applied in the @xmath4-directions .
we find that the system exhibits a kosterlitz - thouless phase transition at the @xmath3-dependent temperature @xmath5 below the critical temperature @xmath6 of the bulk system .
however ,
right at the critical temperature the ratio of the areal superfluid density to the critical temperature is @xmath3-dependent in the range of film thicknesses considered here
. we do not find satisfactory finite - size scaling of the superfluid density with respect to @xmath3 for the sizes of @xmath3 studied .
however , our numerical results can be collapsed onto a single curve by introducing an effective thickness @xmath7 ( where @xmath8 is a constant ) into the corresponding scaling relations .
we argue that the effective thickness depends on the type of boundary conditions .
scaling of the specific heat does not require an effective thickness ( within error bars ) and we find good agreement between the scaling function @xmath9 calculated from our monte carlo results , @xmath9 calculated by renormalization group methods , and the experimentally determined function @xmath10 . |
a significant number of quantum computation and quantum information protocols use the quadrature amplitudes of the quantized electromagnetic field @xcite , @xcite . the essence of all these protocols rely in the possibility of producing entangled photons in a nonlinear medium @xcite .
the parametric down conversion ( pdc ) @xcite is one of the most often used processes for production of entangled pairs of photons .
however , as for many nonlinear optical processes , pdc is ruled by phase matching whose importance was early recognized @xcite , theoretical treatment of a phase mismatch that occurs during the propagation in the nonlinear medium is complicated and several approximations have been made @xcite , in order to relate the phase mismatch , with the amount of quantum correlations in the spontaneous emission of photons . at room temperatures
the average number of photons at optical frequencies is very small ( on the order of @xmath0 ) .
however , in the microwave part of the spectrum the number of thermal photons initially present in the nonlinear medium can not be ignored and we expect that they will play a determinant role in the amount of entanglement that can be extracted from the interaction . to our knowledge a comprehensive study of pdc processes with a phase mismatch and at finite temperature of the nonlinear medium has not been addressed . in this paper
we describe the pdc process with a phase mismatch with a simple model that is the quantum transcription of the classical pdc approach derived by bloembergen @xcite .
this model is valid when the interaction is weak , and when the complex amplitudes of the interacting waves vary slowly with the propagating distance in the nonlinear medium .
our purpose is to relate the joint effect of the temperature of the nonlinear medium and of the phase mismatch , with the degree of entanglement of the down - converted photon states . to quantify the degree of entanglement in terms of the phase mismatch and of the temperature we use the logarithmic negativity @xcite .
we show that for finite temperatures there is a minimum time of interaction needed , the _ birth time of entanglement _ ( bte ) , before the system starts to be entangled and that it varies with the temperature and with the phase mismatch . for a given phase mismatch
, we compute the degree of entanglement with the temperature in order to obtain the critical temperature , above which the quantum state becomes separable .
this study allows us to identify what are the regions of joint values of temperature and phase mismatch , where the states of the system are entangled and where they are separable .
finally , we propose a feasible entanglement witness that is directly connected with a dynamical invariant of the system .
the plan of the paper is the following : in sec.ii we introduce the hamiltonean of the system with non zero phase matching , and obtain the time dependent bosonic operators . in sec .
iii we compute the _ birth time of entanglement _ and the degree of entanglement in terms of the phase mismatch and of the temperature and , an entanglement witness is proposed . finally in sec.iv
we present the conclusions .
the parametric down conversion is a process where an intense pump wave of frequency @xmath1 , propagates in a nonlinear medium and gives rise to two other modes of smaller frequencies @xmath2 ( the signal ) and @xmath3 ( the idler ) obeying the energy conservation rule @xmath4 . when the interaction is weak , and when the complex amplitudes of the interacting waves vary slowly with the propagating distance so that they can be assumed approximately constant within a distance of a wavelength , it is possible to use the slowly varying envelope approximation derived by bloembergen @xcite in a classical context , to describe the nonlinear interaction with a phase mismatch @xmath5 .
the quantum version associated to the classical hamiltonean describing this process , is @xmath6 } \hat{a}_1^\dag \hat{a}_2^\dag+h.c.)\ ] ] where the intense pump mode is treated classically as a coherent undepleted field of complex amplitude @xmath7 } $ ] and an arbitrary phase @xmath8 .
modes @xmath9 and @xmath10 are described by the bosonic operators @xmath11 and @xmath12 .
the interaction time @xmath13 is the propagation time in the nonlinear medium and the coupling constant @xmath14 is proportional to the second order susceptibility of the medium and to the modulus @xmath15 of the pump wave . solving the heisenberg equations of motion of the system (
see the appendix ) , we obtain the time dependent bosonic operators @xmath16\ ] ] @xmath17\ ] ] where @xmath18 is an effective dimensionless interaction time , @xmath19 is the dimensionless mismatch parameter with values in the interval @xmath20 , @xmath21 is the dimensionless effective frequency of oscillation of mode @xmath22 and @xmath23 .
the subscript zero refers to initial time .
the coefficients @xmath24 and @xmath25 are defined by @xmath26 @xmath27 equations ( [ time1 ] ) and ( [ time2 ] ) extend the well known result derived in @xcite to take into account the influence of the phase mismatch , and reduces to it when @xmath28 .
because the pump amplitude @xmath29 , is treated as constant , the solution to the parametric interaction ceases to be valid when the number of generated pair of photons is such that determines an appreciable depletion of the pump mode .
the argument of the hyperbolic functions is the squeezing parameter , @xmath30 it depends not only on the dimensionless interaction time @xmath31 , but also on the mismatch .
making @xmath32 in the last equation we obtain the usual squeezing parameter @xmath33 @xcite . for a given interaction time the squeezing parameter decreases with increasing mismatch and we expect that the efficiency of the correlated pair production also decreases with @xmath34 .
in this work , we are interested in the amount of entanglement we can extract from the correlated pair production in terms of the phase mismatch and at finite temperature @xmath35 . in the spontaneous pdc emission the signal * 1 * and the idler * 2 * modes are initially in the vacuum state @xmath36 ( @xmath37 ) .
the initial state of the composite system is the pure and separable state given by the tensor product @xmath38 .
when the nonlinear medium is in thermal equilibrium at temperature t , the initial state of the two modes is a separable mixed state given by the density operator @xmath39 , where @xmath40 is the thermal field density operator of mode @xmath22 , with @xmath41 and @xmath42 is the boltzmann constant .
the initial mean number of photons in mode @xmath22 is @xmath43 . at room temperature and
in the optical part of the electromagnetic spectrum @xmath44 , however , in the microwave part of the spectrum @xmath45 and we can not ignore the presence of thermal photons in the nonlinear medium . the initial average number of photons in the limit @xmath46 is @xmath47 , and we recover the vacuum state .
both vacuum and thermal states are gaussian states and since the hamiltonian @xmath48 is bilinear in the field modes , the overall output state is also gaussian and the separability criterion @xcite is completely characterized by the first and the second statistical moments of the quadrature phase operators @xmath49 and @xmath50 , @xmath51 .
these operators can be grouped together in a vector of operators @xmath52 .
the first and the second statistical moments of the quadrature phase operators , will be denoted , respectively , by the vector @xmath53 and by the covariance matrix ( cm ) @xmath54 , of elements @xmath55 .
the first moments can be adjusted to zero by local unitary operations which leave invariant entropy and entanglement .
the time dependent covariance matrix @xmath56 can be expressed in terms of the three @xmath57 block matrices @xmath58 , @xmath59 and @xmath60 @xmath61 the diagonal blocks @xmath58 , @xmath62 are the local cm of modes * 1 * and * 2 * , respectively .
the off - diagonal block , @xmath63 , encode the intermodal correlations ( quantum and classical ) between the two modes .
hamiltonians that are quadratic in the bosonic operators , possess universal quantum invariants , i.e. , certain combinations of variances which are conserved in time independently of the concrete form of coefficients of the hamiltonian @xcite .
these invariants exist due to the symplectic structure of the transformation relating initial and time - dependent values of the quadrature components operators .
concerning the statistical moments , these universal invariants are @xcite @xmath64 defining the quantities @xmath65 the entanglement criterion can be written in a simple form @xmath66 for both vacuum and thermal initial states , we prove in the appendix , that @xmath67 , with @xmath68 . for the initial vacuum state , @xmath69 and @xmath70 . using the entanglement criterion ( [ criterion ] ) ,
we conclude that the two modes are entangled for interaction times @xmath71 obeying the inequality @xmath72 is an increasing function of @xmath31 .
the instant @xmath73 , such that @xmath74 is the minimum interaction time needed before entanglement appears , for obvious reasons it will be named the _ birth time of entanglement _ ( bte ) . for the vacuum initial state , @xmath75 , therefore the two modes are entangled since the very beginning of the interaction , this is , @xmath76 for any value of phase mismatch @xmath34 . on the contrary , for an initial thermal state
, the condition ( [ ent ] ) is attained only after a finite time of interaction , @xmath73 , has passed .
the _ birth time of entanglement _ depends on the temperature @xmath35 , through the initial average number of photons @xmath77 and @xmath78 . to compute the initial average number of photons we assume two typical frequencies in the microwave region of the eletromagnetic spectrum : @xmath79hz and @xmath80 and for the coupling constant we assume @xmath81 .
the dependence of the dimensionless bte , @xmath73 , over the temperature is displayed in fig.1 , for the dimensionless frequencies @xmath82 and for different values of the dimensionless mismatch parameter @xmath34 : @xmath83 ( triangles ) , @xmath84 ( circles ) and @xmath85 ( stars ) .
we observe that , for a given mismatch parameter @xmath34 , the bte increases with the temperature although this growth slows down at higher temperatures . as we will show , only when the average number of correlated photon pairs is higher than a threshold defined by the uncorrelated initial thermal photons is the _ break - even _ for entanglement attained .
, over the temperature , for the dimensionless frequencies @xmath86 and for different values of the dimensionless mismatch parameter @xmath34 : @xmath83 ( triangles ) , @xmath84 ( circles ) and @xmath85 ( stars ) .
_ _ ] in fig.2 , the bte is displayed as a function of the dimensionless mismatch parameter @xmath34 , for different temperatures : @xmath87 ( triangles ) , @xmath88 ( crosses ) and @xmath89 ( circles ) .
we observe that , for @xmath90 , the bte increases with @xmath34 , and the growth rate is higher at higher temperatures .
this is not surprising since the squeezing parameter , responsible for the efficiency of the photon pair production , decreases with increasing mismatch introducing a randomness in the interaction . , for different temperatures : @xmath87 ( triangles ) , @xmath88 ( crosses ) and @xmath89 ( circles ) . _ ] the logarithmic negativity @xmath91 @xcite provides a proper quantification of entanglement in two - mode gaussian states and is a computable measure of entanglement . in fig.3 the degree of entanglement @xmath91
is displayed in terms of @xmath34 for @xmath92 and for @xmath93 ( crosses ) , @xmath87 ( triangles ) and @xmath89 ( circles ) . the same dimensionless frequencies as in fig.2 are assumed .
we observe that , for a given temperature , the degree of entanglement decreases with increasing @xmath34 .
moreover , for @xmath89 and for @xmath94 , @xmath95 , this is , the two modes are in a separable state though for @xmath87 the system is still entangled for @xmath96 . , in terms of @xmath34 for @xmath92 and for @xmath93 ( crosses ) ,
@xmath87 ( triangles ) and @xmath89 ( circles ) . the same dimensionless frequencies as in fig.2 are assumed . _ _ ] as a matter of fact , when the initial temperature of the nonlinear medium rises , there are initially more uncorrelated thermal photons in the modes .
we expect intuitively that , for a given interaction time , any entanglement should vanish at some critical temperature @xmath97 .
we have computed @xmath91 , for @xmath98 , in terms of the temperature and for different values of the mismatch parameter @xmath34 , in order to find the critical temperature for each @xmath34 .
the results are displayed in fig.4 where the points @xmath99 are marked with circles .
the solid curve was obtained by interpolation .
the quantum states of the two modes characterized by the parameters @xmath34 and @xmath35 in the shaded region under the curve are entangled states . on the contrary , above the curve these parameters define separable states . we observe that there is a kind of trade - of between the phase mismatch and the critical temperature : greater is the phase mismatch smaller is the critical temperature and vice versa . for different values of @xmath34 ( circles ) .
the solid curve was obtained by interpolation .
the shaded region under the curve correspond to values of parameters @xmath100 where the states are entangled . above the curve
these parameters define separable states . _
_ ] produced in the pcd process after a time of interaction @xmath98 , in terms of the temperature @xmath35 for different values of the mismatch parameter : @xmath101 ( dashed - dot line ) , @xmath102 ( dashed line ) and @xmath103 ( dotted line ) . the invariant @xmath104 ( solid line ) is displayed in terms of the temperature @xmath35 .
another interesting conclusion is that it is possible to have entangled states at room temperature for small enough phase mismatch .
for instance , in the case considered in the fig.4 , the modes are entangled at @xmath89 for @xmath105 .
in fig.5 is displayed the average number of photons @xmath106 produced in the pcd process after a time of interaction @xmath98 , in terms of the temperature @xmath35 and for different values of the mismatch parameter @xmath101 ( dashed - dot line ) , @xmath102 ( dashed line ) and @xmath103 ( dotted line ) .
it is also displayed the invariant @xmath104 ( solid line ) in terms of the temperature .
the first three lines intersect @xmath104 at different values of the temperature @xmath35 .
computing the temperature at the intersection points we find out that they coincide with the corresponding critical temperature @xmath97 for each @xmath34 . for temperatures such that the average number of photons is in the shaded region of the figure , the system is entangled . under the solid curve the average number of correlated photons produced in the interaction are not enough to attain the _ break - even _ , given by the square root of the invariant @xmath107 , this is , they are not enough to exceed the uncorrelated photons present in the initial thermal state , and the system is separable
. we can then define a simple entanglement witness @xmath108 as usual , the negative value of @xmath109 indicates entanglement . to check wether
the two modes are entangled , it will be enough to measure the average number @xmath110 of correlated photons generated in the two modes and the equilibrium temperature @xmath35 of the nonlinear medium , since this define the initial average number of photons @xmath111 in each mode .
we have computed the joint effect of a phase mismatch and of the temperature in the degree of entanglement in pdc processes described by hamiltonean ( [ hamiltonean ] ) . using the time invariants associated to that hamiltonean we computed the minimum interaction time needed before entanglement appears and have shown how it behaves with the phase mismatch and temperature .
we have also shown that for each phase mismatch there is a critical temperature , above which the entanglement disapears .
this study enables us to draw the border line between separable and entangled states in the plane of the parameters , phase mismatch and temperature .
depending on the interaction time , and on the phase mismatch it is possible to obtain entanglement at room temperature .
finally we propose a feasible entanglement witness .
the heisenberg equations of motion for modes @xmath9 and @xmath10 are @xmath112}\\ -ig e^ { i [ 2 ( \omega + \delta ) t + \varphi ] } & i \omega_2 \\ \end{array } \right ) \left ( \begin{array}{c } \hat{a}_1\\ \hat{a}_2^{\dag}\\ \end{array } \right)\ ] ] jointly with their hermitian conjugates , which are decoupled from these two . in order to obtain an autonomous system of equations
we define new annhilation operators @xmath113 and @xmath114 , by the canonical transformation @xmath115 physically this corresponds to move to a reference frame that rotates with the frequency @xmath116 .
the heisenberg equations for the new operators @xmath113 and @xmath117 are then given by @xmath118 they are linear time independent differential equations and can be readily integrated .
the only solution of eq.([heis1 ] ) with physical interest is given by eqs.([time1 ] , [ time2 ] ) of section 2 , and corresponds to the condition @xmath119 where the bandwidth is small compared with the coupling parameter . using the time dependent bosonic operators given in eqs.([time1 ] ) and ( [ time2 ] ) we compute the entries of the covariance matrix @xmath121 for the thermal initial state @xmath122 @xmath123 @xmath124 @xmath125\ ] ] @xmath126\ ] ] where @xmath127 and @xmath128 , @xmath37 . | in this work the degree of entanglement of two - photon states generated by parametric - down conversion with a phase mismatch is computed in terms of the temperature of the nonlinear medium where the interaction takes place .
the minimum interaction time , named _ birth time of entanglement _ , needed before entanglement is attained , is computed in terms of the temperature and of the phase mismatch .
our results show that , for each value of the phase mismatch , there is a critical temperature above which entanglement disappears .
we are able to identify the regions of parameters , temperature and mismatch , where the states are entangled and where they are separable .
we also propose a entanglement witness based in a dynamical invariant of the nonlinear interaction .
pacs number(s ) 03.67.a , 42.50-p . , 42.65.lm |
hydrogen is an ubiquitous element , present even in ultra - high - vacuum chambers , although very difficult to detect .
it shows a large sticking probability in most transition metal surfaces @xcite , among them co. adsorption of h on transition metals was intensively studied in the past decades , mainly motivated by the understanding of the action of residual gases in catalytic and embrittlement processes .
much less attention has been paid to the influence of h on the magnetic properties , though it is known to have a remarkable effect .
for example , the adsorption of h@xmath0 molecules on the surface of magnetic transition metals produces a reduction of surface magnetism @xcite .
in addition , h coverage shifts the critical thickness for the spin reorientation transition ( srt ) of ni films @xcite , and a reversible manipulation of the magnetic exchange coupling of fe / v superlattices can be achieved upon loading the v layers with h @xcite . previous experimental results characterizing the structure and energetics of h adsorption on different co surfaces and nanostructures with two - dimensional ( 2d ) 3-fold symmetry , as co(0001 ) @xcite or co nanoislands on cu(111 ) @xcite , indicate that h prefers adsorption sites preserving the symmetry of the 2d lattice .
this is in agreement with _
ab initio _ calculations of the hcp co(0001 ) termination @xcite . even for co(@xmath1 )
the formation of superstructures leading to adsorption sites of 3-fold symmetry has been reported @xcite .
the evaporation of molecular h@xmath0 on different co surfaces results in the dissociative chemisorption of h atoms up to a monolayer ( ml ) completion @xcite . for higher doses , molecular h@xmath0 physisorption occurs .
the desorption of h from co nanoislands on cu(111 ) takes place locally and leaves the co surface as it was prior the h evaporation , in contrast to the substrate exposed areas @xcite .
this surface character of the h - co interaction is also supported by total energy calculations , which show a negligible weight of the h induced states even in subsurface co atoms @xcite . in addition , no dependence of the binding energy on the h coverage is obtained @xcite . moreover , h subsurface sites are highly unstable , indicating a low probability for h diffusion to the bulk . as compared to other transition metal surfaces , there are only a few studies of the influence of h on the electronic and magnetic properties of co surfaces and thin films .
spectroscopic measurements indicate a quenching of the @xmath2-like surface states and a shift of the @xmath3 surface states downward in energy upon h adsorption @xcite . moreover , the co electronic properties can be tuned individually controlling the amount of adsorbed h @xcite .
the calculated density of states ( dos ) also indicates that h states appear below the @xmath3 valence band ( vb ) @xcite .
when spin - polarization is included , it can be shown that this results in the quenching of the surface - induced enhancement of the co spin moment , at least for 1 ml of h coverage @xcite .
recently , we studied the magnetic properties of ultrathin co films grown on ru(0001 ) @xcite .
this is a unique system showing a double srt linked to the completion of co atomic layers . the srt is intimately related to the structural distortion of the ultrathin co film .
both co and ru are hcp metals , but with very different lattice parameters : 2.51 and 2.71 , respectively .
while 1 ml of co grows epitaxially , the second layer starts compressing towards the co lattice , leading to an intermediate 2d lattice parameter of 2.6 . in this work
we study the influence of h adsorption on the structural , electronic and magnetic properties of these ultrathin co / ru(0001 ) films , with special emphasis on the 2 ml thick ones which show perpendicular magnetic anisotropy
. the study of the magnetic anisotropy will be the subject of a forthcoming paper .
our results indicate that h bonds strongly to the surface and tends to form a complete overlayer leading to a very stable surface system .
partial h coverage , although thermodinamically stable , presents slightly smaller energies of adsorption . as we will show , while 1 ml of h quenches the surface magnetism , lower h coverage may enhance it .
this can be a crucial factor regarding the role of residual h in the measured magnetic properties of co surfaces .
we have performed _ ab initio _ calculations within the local density approximation ( lda ) , combining two different approaches .
first , we have performed an exhaustive search of the equilibrium adsorption positions of h using slab models , a plane - wave basis set and the projector - augmented wave ( paw ) method @xcite as implemented in the vienna ab - initio simulation package ( vasp ) @xcite . for the most stable configurations
, we have then determined the electronic and magnetic properties within the fully - relativistic framework of the screened korringa - kohn - rostoker ( skkr ) method @xcite , which allows for a smooth matching of the surface region to the semiinfinite bulk and lacks of the quantum size effects inherent to slab models . for the calculations with vasp , an energy cutoff of 350 ev has been used .
the structures are modeled by periodically repeated slabs of 6 to 10 ru layers covered by 2 co mls on one side , and adding 1/4 ml or 1 ml of h on top of or underneath the outermost co layer .
a vacuum region of 12 is left between both slab surfaces which prevents any spurious interaction between them .
this model also guarantees the recovery of the bulk properties at the inner ru layers of the slab .
the search for the most stable h adsorption sites has been performed starting from different h geometries and allowing full relaxation of the atomic positions of h , co and the outermost ru layer .
a uniform sampling of the k - mesh centered in @xmath4 has been used , and convergence for the most stable structures has been obtained using k - samplings of up to ( @xmath5 ) for @xmath6 geometries and ( @xmath7 ) for the @xmath8 unit cells corresponding to adsorption of 0.25 ml of h. under these conditions , convergence in total energy differences below 1 mev is achieved . by using the optimized geometry with the most stable h adsorption sites , the detailed electronic structure and the magnetic moments were calculated in terms of the fully relativistic skkr method.@xcite the skkr formalism provides , in a natural way , layer - resolved physical quantities such as charges and magnetic moments .
it , thus , manifests a perfect tool for the analysis of the h - co interactions , and of the influence of the co film thickness in the h induced properties . for this purpose
, we have extended our calculations to films of 3 to 6 co mls .
details of the skkr calculations for the co / ru system can be found elsewhere @xcite .
the h atom is modeled by using an atomic sphere of radius 0.59 , placed at the equilibrium distance to co determined by vasp .
the stable adsorption site of h on 2co / ru(0001 ) has been explored starting from different geometries , depicted in figure [ hposit ] : top , bridge , hollow ( either fcc or hcp ) and off - symmetry positions , both at surface and subsurface sites .
in addition , for 1 ml coverage , the influence of structural modifications present in the co film has been considered , such as the variation of the 2d lattice parameter , or the existence of stacking faults which alter the co / ru stacking sequence from the hcp ab / ab to bc / ab @xcite . a summary of the results is compiled in table [ tstruc ] . in all cases , the most stable adsorption site corresponds to h at fcc hollow surface positions .
the table shows the energy difference with respect to the equilibrium configuration only for the most stable structures .
the rest of geometries considered , including on - top positions , provide total energies larger by more than 1 ev . our results are in good agreement with previous ones relative to different co surfaces and thin films @xcite : h tends to stick at the surface trying to preserve the 3-fold symmetry of the co lattice , independent of the detailed co film structure or the h coverage .
the energy difference between adsorption sites provides an estimate of the diffusion barriers for h in the surface .
although the trends are similar for all surfaces studied , partial h coverage or alteration of the stacking sequence at the co film may modify the diffusion barriers significantly .
regarding subsurface sites , the barrier for the diffusion of h into the co film decreases as the h coverage increases , but it is always large enough to support the low probability of the penetration of h into the film , in good agreement with the experiments of thermal desorption through formation of h@xmath0 @xcite .
the results indicate that the adsorption of h is governed by the local surface formation of strong h - co bonds .
supporting this , for all structures considered the h - co distance remains around 1.7 , independent of the h coverage and the co film structure .
the largest variations of the h - co distance are found to depend on the h adsorption site , where h is coordinated to a different number of co atoms : from 1.5 for the on - top surface adsorption ( coordination 1 ) , to 1.9 for hollow subsurface sites ( coordination 6 ) .
.energy ( in mev ) per h atom with respect to the most stable fcc hollow site for different h adsorption sites on 2 ml co on ru(0001 ) , both for 1 ml and 0.25 ml h coverage .
the dependence on the in - plane lattice parameter a@xmath9 and the stacking sequence of the co film are shown for the high h coverage . [ cols="^,<,^,^,<,^,^,^ " , ] [ mom ] finally , we discuss the h induced effects for thicker co films . to this end , we performed skkr calculations for co films of up to 6 mls covered by 1 ml of h. figures [ fmoms ] and [ fmoml ] show respectively the layer resolved values of the spin ( m@xmath10 ) and orbital ( m@xmath11 ) contributions to the magnetic moment for the different co thicknesses , together with those at the corresponding bare surfaces . upon h adsorption
there is always a strong reduction of both m@xmath10 and m@xmath11 at the surface co atoms . on average ,
the reduction is around 0.25 @xmath12 for m@xmath10 and 0.04 @xmath12 for m@xmath11 .
by contrast , at the subsurface co layer the reduction of the magnetic moment is low , becoming negligible for the layers underneath .
although the specific values of m@xmath10 and m@xmath11 at each layer involve in a complex way the oscillatory layerwise variation of the electronic properties , the effect of h does not depend on the co thickness .
furthermore , it is very similar for both components of the magnetic moment . not shown in the figure , in all cases
the h atom acquires a m@xmath10 of 0.01 @xmath12 , and no orbital polarization .
similarly , the components of the moment induced at the interface ru atom are around -0.02 @xmath12 ( m@xmath10 ) and -0.001 @xmath12 ( m@xmath11 ) for all co thicknesses , without any significant variation as compared to the bare film .
the layer resolved dos ( not shown here ) are consistent with this conclusion , with h induced features almost restricted to the surface co layer in each case .
thus we can conclude that the h induced effects are essentially restricted to the surface and independent of the co thickness .
( color online ) layer - resolved co spin magnetic moment for co films of 2 to 6 mls ( from top to bottom ) , either bare ( empty circles ) or covered by h ( solid circles ) .
layers are numbered from the ru interface . ]
( color online ) same as figure [ fmoms ] for the orbital contributions to the magnetic moment . ]
in terms of _ ab initio _ calculations , we performed a detailed theoretical analysis of the stability , electronic structure and magnetism of co / ru(0001 ) films upon h adsorption .
our main observations are as follows : h adsorbs on co at the hollow sites by preserving the threefold symmetry of the system .
the adsorption is endothermic for all h coverages , with the largest binding energy corresponding to the complete h overlayer .
furthermore , the probability of h desorption or segregation is low .
concerning the structural , electronic and magnetic properties , the effect of h has a local surface character and is almost independent of the detailed co thickness and structure .
h always attenuates surface effects .
however , for partial h coverages , this leads to the enhancement of the surface induced features for co atoms not bonded to h , and in particular to an enhancement of the net surface magnetic moment .
our results point to the importance of residual h in magnetic measurements , and to the possibility for the reversible manipulation of the properties of ultrathin co films and surfaces by tuning h adsorption .
this work was supported by the spanish ministry of science and technology under contracts mat2006 - 05122 , hh2006 - 0027 and mat2009 - 14578-c03 - 03 , and by the hungarian research foundation under contracts otka k68312 and k77771 . | by means of _ ab initio _ calculations , we have investigated the effect of h adsorption on the structural , electronic and magnetic properties of ultrathin co films on ru(0001 ) .
our calculations predict that h occupies hollow sites preserving the two - dimensional 3-fold symmetry .
the formation of a complete h overlayer leads to a very stable surface with strong h - co bonds .
h tends to suppress surface features , in particular , the enhancement of the magnetic moments of the bare film .
the h - induced effects are mostly confined to the co atoms bonded to h , independent of the h coverage or of the thickness and the structure of the co film . however , for partial h coverages a significant increase occurs in the magnetic moment for the surface co atoms not bonded to h , leading to a net enhancement of surface magnetism . |
early studied in the 1930 s @xcite , the counter - rotating machines arouse a greater interest in the turbomachinery field , particularly for their potential improvement of the efficiency with respect to conventional machines by recovering kinetic energy from the front rotor exit - flow and by adding energy to the flow .
the first counter - rotating machines have appeared in aeronautic @xcite and marine applications @xcite in open configuration .
conventional designs of high speed counter - rotating fans are based on quite expensive methods and require a systematic coming and going between theoretical methods such as the lifting line theory or the strip - analysis approach @xcite and cfd analysis @xcite .
moreover , the axial spacing , which has a major role on the rotors interaction and consequently on the noise @xcite , is a key parameter to find a compromise between high aerodynamic and good acoustic performance for high speed fans @xcite . in order to reduce this interaction ,
the axial spacing of high speed fans has to be relatively large , resulting in a decrease in the aerodynamic performance @xcite .
for the same reason , the rear rotor ( rr ) diameter has to be smaller ( about 10@xmath2 according to @xcite ) than the front rotor ( fr ) diameter to reduce interaction between the fr tip vortex and the rr blade tip .
contrary to that , in the case of low speed fans axial spacing could be shortened using the benefit of a relatively low rotor interaction .
therefore these machines see a revival of interest in several distinct configurations open and ducted flows , shrouded or not shrouded rotors in various subsonic regime applications @xcite .
recent research work dealt with the effects of global parameters like rotation speed ratio @xcite , local phenomena such as tip vortex flows @xcite and improvement of cavitation performance for pumps @xcite .
all previous studies have shown the benefit of rr in improving the global efficiency and in increasing the operating flow - rate range while maintaining high efficiency .
the counter - rotating systems ( crs ) moreover allow to reduce the fans diameter and/or to reduce the rotation rate .
more axial spacing is needed compared to one simple fan , but not much more than a rotor - stator stage .
however , it requires a more complex shaft system .
another interesting feature of crs is that it makes it possible to design axial - flow fans with very low angular specific speed @xmath3 with @xmath4 the mean angular velocity , @xmath5 the flow rate , @xmath6 the total pressure rise , and @xmath7 the fluid density . with such advantages ,
the crs becomes a very interesting solution and the interaction between the rotors needs to be better understood in order to design highly efficient crs .
however , only a few studies have been concerned with , on the one hand , the effect of the axial spacing , and , on the other hand , the design method @xcite , particularly with rotors load distribution for a specified design point .
this paper focuses on two major parameters of ducted counter - rotating axial - flow fans in subsonic regime : the rotation rate ratio , @xmath8 and the relative axial spacing , @xmath9 . in some cases ,
these systems are studied by using two identical rotors or the rr is not specifically designed to operate with the fr . in this study , the fr is designed as conventional rotor and the rr is designed on purpose to work with the fr at very small axial spacing . in this first design , the total work to perform by the crs was arbitrarily set up approximately to two halves one half respectively for the fr and rr . in [ sec : design ] the method that has been used to design the front and the rear rotors is firstly described .
the experimental set - up is presented in [ sec : setup ] .
then the overall performances of the system in its default configuration and the effects of varying the rotation ratio and the relative axial spacing between the rotors are discussed in [ sec : results ] .
the design of the rotors is based on the use of the software mft ( mixed flow turbomachinery ) , a 1d code developed by the dynfluid laboratory @xcite based on the inverse method with simplified radial equilibrium to which an original method has been added specifically for the design of the rr of the counter - rotating system . from the specified total pressure rise , volume flow - rate and rotating speed , optimal values of the radii @xmath10 and @xmath11
are first proposed . in a second step
, the tip and the hub radii as well as the radial distribution of the circumferential component of the velocity at the rotor outlet , @xmath12 , could be changed by the user .
the available vortex models are the free vortex ( @xmath13 ) , the constant vortex ( @xmath14 ) and the forced vortex ( @xmath15 ) .
the velocity triangles are then computed for @xmath16 radial sections , based on the euler equation for perfect fluid with a rough estimate of the efficiency of @xmath17 and on the equation of simplified radial equilibrium ( radial momentum conservation ) .
the blades can then be defined by the local resolution of an inverse problem considering a 2d flow and searching for the best suited cascade to the proposed velocity triangles by the following parameters : @xmath18 the stagger angle , computed from the incidence angle , @xmath19 giving the lower pressure variation on the suction surface of the blade using equations [ eq : gamma ] and [ eq : a ] .
the solidity , @xmath20 and the chord length , @xmath21 are thus computed at the hub and at the tip using equations [ eq : sigma ] and [ eq : c ] where @xmath22 denotes the lieblein s diffusion factor@xcite .
the intermediate chords are obtained by linearisation . finally , the camber coefficients @xmath23 are computed using equation [ eq : coef_port ] . @xmath24
these empirical equations have been validated for naca-65 cascades @xcite , for @xmath25 and @xmath26 .
velocity triangles for the crs .
the fluid is flowing from left to right .
, width=294 ] the behaviour of the designed machine resulting from the above method can then be analysed using a direct method in order to determine whether the design point is achieved and what are the characteristics of the machine at the neighbourhood of the design point .
the effects due to real fluid are taken partially into account with in - house loss models and the introduction of an axial - velocity distribution which considers the boundary layers at the hub and casing .
thus , the characteristics of the machine can be obtained in the vicinity of the design - point discharge . regarding the crs , the geometrical dimensions , the number of blades of fr and of rr and their rotation rates are imposed . in particular , the number of blades of each rotor was chosen in order to prevent to have the same blade passing frequency or harmonics for both rotors in the lower frequencies range .
the system that is presented here has moreover been designed to have a pure axial exit - flow .
an iterative procedure is then performed .
the pressure rise of the fr is initially chosen and then designed and quickly analysed as explained .
an estimate of the pressure rise that rr would made is then performed , based on this analysis .
if the total pressure rise of the crs is not met , the design pressure rise of fr is varied and the calculus are made again . in this method ,
losses and interactions in - between the two rotors are not taken into account .
any recirculation happening near the blade passage or near the blade hub or tip is not predicted by mft as it is based on simplified radial equilibrium . .
.[tab : speci]design point of the counter - rotating system for air at @xmath27 [ cols= " < , < , < , < " , ] fans characteristics : ( a ) static pressure rise @xmath28 _ vs _ flow rate @xmath5 ; ( b ) static efficiency @xmath29 _ vs _ flow rate @xmath5 .
the axial spacing is @xmath30 .
@xmath31 : fr rotating alone at @xmath32 rpm ( rr has been removed ) , @xmath33 : rr rotating alone at @xmath34 rpm ( fr has been removed ) and @xmath35 : crs at @xmath32 rpm and @xmath36 .
the @xmath37 and the dashed lines stand for the design point of the crs ] the fr rotating alone has a very flat curve ( @xmath31 in fig .
[ fig : caract_r1_r2_r12_nominal ] ) .
the nominal flow - rate of fr is slightly greater than the design point it is @xmath38 greater .
the measured static pressure rise at the design point is @xmath39 pa , with a relatively low static efficiency of @xmath40 .
this is not surprising with no shroud and a large radial gap .
moreover , this is consistent with the estimated static pressure rise by mft , which is around @xmath41 pa .
the rr rotating alone has a steeper curve ( @xmath33 in fig .
[ fig : caract_r1_r2_r12_nominal ] ) and its nominal flow - rate @xmath42 [email protected]@xmath44 is lower than the design flow - rate of fr and crs .
this is consistent with the bigger stagger angle of the blades ( see tab . [ tab : geospeci ] ) and can be explained by examining the velocity triangles in fig .
[ fig : trianglevit ] and considering the case with the fr coupled to the rr : the incoming velocity @xmath45=@xmath46 has an axial component as well as a tangential component .
hence , the flow angle in the relative reference frame reads : @xmath47 now the case without the fr is considered and it is assumed that the flow through the honeycomb is axial . since the tangential component does not exist any more , @xmath48 .
mft estimates @xmath49 m.s@xmath44 , @xmath50 m.s@xmath44 and @xmath51 m.s@xmath44 , which leads to @xmath52 at the blade mid - span .
supposing that rr rotating alone reaches its maximum efficiency for @xmath52 , equation [ eq : tanbeta ] implies that @xmath53 m.s@xmath44 , _ i.e. _ @xmath54 [email protected]@xmath44 .
this is exactly the nominal flow - rate of rr rotating alone ( see fig .
[ fig : caract_r1_r2_r12_nominal ] and tab . [ tab : rendmax ] ) .
it is clear from the above analysis why the nominal flow - rate of rr is lower than the design flow - rate . the characteristic curve of the crs ( @xmath35 in fig .
[ fig : caract_r1_r2_r12_nominal ] ) is steeper than the characteristic curve of fr .
it is roughly parallel to the rr curve .
the nominal flow - rate of the crs matches well with the design flow - rate , _
@xmath55 [email protected]@xmath44 .
the static pressure rise at the nominal discharge ( @xmath56 pa ) is @xmath57 lower than the design point ( @xmath58 pa ) , which is not so bad in view of the rough approximations used to design the system .
please notice that the static pressure rise of the crs is not equal to the addition of the static pressure rise of the fr with the pressure static rise of the rr , taken separately .
the crs has a high static efficiency ( @xmath59 ) compared to a conventional axial - flow fan or to a rotor - stator stage with similar dimensions , working at such reynolds numbers @xcite .
the gain in efficiency with respect to the fr is @xmath60 points , whilst an order of magnitude of the maximum gain using a stator is typically @xmath61 points@xcite .
awaiting for more accurate local measurements of the flow angle at the exit of the crs , a simple test of flow visualization with threads affixed downstream of the crs was performed .
it has been observed that without the rr the flow is very disorganized .
when the rr is operating , at the design configuration ( @xmath36 and @xmath62 ) , the flow is less turbulent , the threads are oriented with a small angle at the exit .
this small angle seems , however to decrease when @xmath8 is increased between @xmath55 and @xmath63 .
this is consistent with the results in section [ subsec : influence ] where it is found that the nominal operating point is observed for a value of @xmath8 higher than the design value .
the flow - rate range for which the static efficiency lays in the range @xmath64 is : @xmath65 [email protected]@xmath44 , that is from @xmath66 of the nominal flow - rate up to @xmath67 of the nominal flow - rate .
one open question is to what extent the global performances of the crs are affected by the axial spacing and the speed ratio , and whether the efficient range could be extended by varying the speed ratio .
crs characteristics at @xmath32 rpm , @xmath30 and @xmath68 $ ] : ( a ) static pressure rise @xmath28 _ vs _ flow rate @xmath5 ; ( b ) static efficiency @xmath29 _ vs _ flow rate @xmath5 .
@xmath69 : @xmath70 , @xmath71 : @xmath72 , @xmath73 : @xmath74 , @xmath75 : @xmath76 , @xmath77 : @xmath36 , @xmath78 : @xmath79 , @xmath80 : @xmath81 , @xmath31 : @xmath82 , @xmath83 : @xmath84 , @xmath35 : @xmath85 and @xmath86 : @xmath87 .
the blue @xmath37 and the dashed lines stand for the design point of the crs ] maximal static efficiency @xmath29 _ vs _
@xmath8 for the crs with @xmath32 rpm and @xmath30 . ] in this paragraph , the rotation rate of fr is kept constant at @xmath32 rpm , and the rotation rate of rr is varied from @xmath88 to @xmath89 rpm .
the corresponding @xmath8 are @xmath90 .
the axial spacing is @xmath30 .
the overall performances of the crs in these conditions are plotted in fig .
[ fig : caractrend_n1_2000 ] . as expected , the more the rotation rate of rr increases , the more the static pressure rise of the crs increases and the nominal flow - rate of the crs increases . the maximal efficiency as a function of @xmath8
is plotted in fig .
[ fig : etadetheta ] . for very low rotation rates of rr , _
i.e. _ for @xmath70 ( @xmath69 in fig .
[ fig : caractrend_n1_2000 ] ) and @xmath72 ( @xmath71 in fig .
[ fig : caractrend_n1_2000 ] ) , the system is very inefficient : in the first case when the rr is at rest the maximum efficiency hardly reaches @xmath91 which is below the maximal efficiencies of both fr and rr alone .
the maximum flow - rate that can be reached is moreover very low in both cases compared to the discharge goal of @xmath92 [email protected]@xmath44 . in the range @xmath93 $ ] , _ i.e. _ @xmath94 $ ] rpm , the system is highly efficient . the maximum efficiency increases with @xmath8 to reach a maximum value of @xmath95 for @xmath82 and is then quasi - constant ( @xmath96 for @xmath97 ) .
this is a very interesting feature of the counter - rotating system .
one could imagine , simply by varying the rr rotation rate , to work at a constant pressure rise with an efficiency greater than @xmath98 for a large flow - rate range .
for instance in the present case , the system could give a constant static pressure rise of @xmath99 pa with @xmath100 for @xmath101 [email protected]@xmath44 with @xmath32 rpm , @xmath30 and @xmath102 $ ] .
one could also imagine to work at a constant flow - rate with high static efficiency .
for instance in the present case , the system could give a constant flow - rate of @xmath92 [email protected]@xmath44 with @xmath100 for @xmath103 pa with @xmath32 rpm , @xmath30 and @xmath93 $ ] .
crs characteristics at various axial spacing : ( a ) static pressure rise @xmath28 _ vs _ flow rate @xmath5 ; ( b ) static efficiency @xmath29 _ vs _ flow rate @xmath5 .
the rotation ratio of fr is @xmath104 and @xmath36 .
@xmath35 : @xmath30 , @xmath77 : @xmath105 , @xmath106 : @xmath107 , @xmath69 : @xmath108 , @xmath109 : @xmath110 and @xmath83 : @xmath111 .
the blue @xmath37 and the dashed lines stand for the design point of the crs ] figure [ fig : caract_distance ] shows the characteristics curves at the design rotation rates , i.e. , @xmath32 rpm and @xmath36 .
regarding @xmath112 $ ] , the overall performances do not change significantly and the variation is in the uncertainty range .
the efficiency does not vary significantly either .
in other studies@xcite it was reported that the axial spacing had a more significant influence on the overall performances .
this was noticed as well in this study . for a=@xmath113 and a=@xmath114 ,
the global performances are decreased by @xmath115 pa ( @xmath116 ) comparing to the other spacings . however , even for a=@xmath114 , the crs still shows good performances with high efficiency compared to the conventional fan systems .
a counter - rotating axial - flow fan has been designed according to an iterative method that is relatively fast .
it is based on semi - empirical modelization that partly takes into account the losses , boundary layers at hub and casing , and the effects of low reynolds numbers ( below @xmath117 ) .
the overall performances at the nominal design point are slightly lower than predicted , with a static pressure rise @xmath57 lower .
the static efficiency is however remarkably high ( @xmath118 ) and corresponds to a @xmath60 points gain in efficiency with respect to the fr maximal efficiency and to a @xmath61 points gain with respect to the rr .
the overall measurements give first clues that allow to validate the design method .
the counter - rotating system has a very flexible use that allows to work at constant flow - rate on a wide range of static pressure rises or to work at constant pressure rise on a wide range of flow - rates , with static efficiency bigger than @xmath98 , simply by varying the rr rotation rate .
one could thus imagine an efficient closed - loop - controlled axial - flow fan .
the overall performances moreover do not significantly vary with the axial spacing in the range @xmath119 $ ] . however , for @xmath110 and @xmath120 the overall performances slightly decrease .
10 bechet , s. , negulescu , c. , chapin , v. , and simon , f. , 2011 .
`` integration of cfd tools in aerodynamic design of contra - rotating propellers blades '' . in 3rd
ceas conference ( council of european aerospace societies ) .
blandeau , v. p. , joseph , p. f. , and tester , b. j. , 2009 .
`` broadband noise prediction from rotor - wake interaction in contra - rotating propfans '' . in 15th aiaa / ceas aeroacoustics conference -30th aiaa aeroacoustics conference ,
aiaa 20093137 .
shigemitsu , t. , furukawa , a. , watanabe , s. , and okuma , k. , 2005 . `` air / water two - phase flow performance of contra - rotating axial flow pump and rotational speed control of rear rotor '' . in asme 2005 fluids engineering division summer meeting
june 19 - 23 , 2005 , houston , texas , usa , pp .
10691074 .
cho , l. , choi , h. , lee , s. , and cho , j. , 2009 .
`` numerical and experimental analyses for the aerodynamic design of high performance counter - rotating axial flow fans '' . in proceedings of the asme 2009 fluids engineering division summer meeting - fedsm2009 , colorado usa , pp .
fedsm200978507 .
noguera , r. , rey , r. , massouh , f. , bakir , f. , and kouidri , s. , 1993 . `` design and analysis of axial pumps '' . in asme fluids engineering , second pumping machinery symposium ,
washington , usa . , pp .
95111 .
lieblein , s. , schwenk , f. c. , and broderick , r. l. , 1953 .
diffusion factor for estimating losses and limiting blade loading in axial - flow - compressor blade elements .
tech . rep .
tm e53d01 , national advisory committee for aeronautics .
sarraf , c. , nouri , h. , ravelet , f. , and bakir , f. , 2011 .
`` experimental study of blade thickness effects on the global and local performances of a controlled vortex designed axial - flow fan '' . , p. 684 | _ an experimental study on the design of counter - rotating axial - flow fans was carried out .
the fans were designed using an inverse method . in particular ,
the system is designed to have a pure axial discharge flow .
the counter - rotating fans operate in a ducted - flow configuration and the overall performances are measured in a normalized test bench .
the rotation rate of each fan is independently controlled .
the relative axial spacing between fans can vary from @xmath0 to @xmath1 .
the results show that the efficiency is strongly increased compared to a conventional rotor or to a rotor - stator stage .
the effects of varying the rotation rates ratio on the overall performances are studied and show that the system has a very flexible use , with a large patch of high efficient operating points in the parameter space .
the increase of axial spacing causes only a small decrease of the efficiency _ |
in the study of interconnection and communication networks , the theory of digraphs plays a key role because , in many cases , the links between nodes are unidirectional . in this theory , there are three concepts that have shown to be very fruitful to construct good and efficient networks .
namely , those of quotient digraphs , voltage digraphs and ( partial ) line digraphs . roughly speaking ,
quotient digraphs allow us to obtain a simplified or ` condensed ' version of a bigger digraph , while the voltage and line digraph techniques do the converse by ` expanding ' a smaller digraph . from this point of view , it is natural that the three techniques have close relationships . in this paper
we explore some of such interrelations by introducing a general construction that we call expanded digraphs .
these digraphs are obtained from a base graph whose vertices become vertex sets in the new graph , and the adjacencies are defined from a set of mappings .
a special case is obtained when such mappings are defined within a group , so obtaining the lifted graphs of base graphs with assigned voltages ( elements of the group ) on its arcs . in this context , we show that de bruijn and kautz digraphs can be defined as lifted digraphs of smaller de bruijn digraphs .
moreover , it is proved that , under some sufficient conditions , the lift of a base graph that is a line digraph is again a line digraph . in the more general case of nonrestricted maps
, we consider the quotient digraphs , and the equivalent constructions of vertex - split digraphs and partial line digraphs . here
, it turns out that the line digraph and quotient operations commute .
finally , it is proved that the techniques of vertex - splitting and partial line digraph are equivalent , and some consequences are derived .
let us first recall some basic terminology and notation concerning digraphs
. for the concepts and/or results not presented here , we refer the reader to some of the basic textbooks on the subject ; see , for instance , chartrand and lesniak @xcite or diestel @xcite . through this paper , @xmath0 denotes a digraph , with vertex set @xmath1 and arc set @xmath2 . an arc from vertex @xmath3 to vertex @xmath4
is denoted by either @xmath5 , @xmath6 , or @xmath7 .
we allow _ loops _ ( that is , arcs from a vertex to itself ) , and _
multiple arcs_. the set of vertices adjacent to and from @xmath8 is denoted by @xmath9 and @xmath10 , respectively .
such vertices are referred to as _ in - neighbors _ and _ out - neighbors _ of @xmath4 , respectively .
moreover , @xmath11 and @xmath12 are the _ in - degree _ and _ out - degree _ of vertex @xmath4 , and @xmath13 is _ @xmath14-regular _ when @xmath15 for any @xmath8 .
expanded digraphs are , in fact , a type of compounding that consists of connecting together several copies of a ( di)graph by setting some ( directed ) edges between any two copies .
let @xmath0 be a ( base ) digraph on @xmath16 vertices .
as said before , we allow loops and multiple arcs .
assume that each vertex @xmath8 has assigned a vertex set @xmath17 , and each arc @xmath18 has assigned a mapping @xmath19 .
let @xmath20 .
the _ expanded digraph _
@xmath21 of @xmath13 with respect to @xmath22 has vertex set @xmath23 , and there is an arc from @xmath24 to @xmath25 whenever @xmath26 and @xmath27 .
some important particular cases of this construction are obtained when the mappings in @xmath22 are defined from a group : * * cayley digraphs*. let @xmath28 be a group with generating set @xmath29 having @xmath30 elements . if @xmath13 is a singleton with assigned vertex set @xmath28 , @xmath30 loops , and each loop @xmath31 has the mapping @xmath32 , with @xmath33 , then the expanded digraph @xmath21 is the _ cayley digraph _ @xmath34 . *
* coset digraphs*. let @xmath28 be a group with generating set @xmath29 having @xmath30 elements and with subgroup @xmath35 .
if @xmath13 is a singleton with assigned vertex set @xmath36 , @xmath30 loops , and each loop @xmath31 has the mapping @xmath37 , with @xmath33 , then the expanded digraph @xmath21 corresponds to the _ coset digraph _ @xmath38 .
two natural generalizations of these concepts are the following ( as far as we know , the second one is a new proposal ) : * * lifted ( of voltage ) digraphs * or * expanded cayley digraphs*. let @xmath28 be a group with generating set @xmath29 having @xmath30 elements .
if each vertex of @xmath13 is assigned to the vertex set @xmath28 , and each arc @xmath31 has the mapping @xmath39 , with @xmath33 , then the expanded digraph @xmath21 is the so - called _ lifted digraph _
( or simply _ lift _ ) @xmath40 ( see section [ sec : voltage ] ) . * * expanded coset digraphs*. let @xmath28 be a group with generating set @xmath29 having @xmath30 elements and with subgroup @xmath35 .
if each vertex of @xmath13 is assigned to the vertex set @xmath36 , and each arc @xmath31 has the mapping @xmath37 , with @xmath33 , then we refer to the corresponding expanded digraph @xmath21 as the _ expanded coset digraph _ @xmath40 .
when a group is involved in the setting of the mappings , the symmetry of the obtained constructions yield digraphs with large automorphism groups . to our knowledge ,
one of the first papers where voltage ( undirected ) graphs were used for construction of dense graphs was that of alegre , fiol and yebra @xcite , but without using the name of ` voltage graphs ' .
this name was coined previously by gross @xcite . for more information ,
see gross and tucker @xcite , baskoro , brankovi , miller , plesnk , ryan and ir @xcite , and miller and ir @xcite .
let @xmath13 be a digraph with vertex set @xmath41 and arc set @xmath42 .
then , given a group @xmath28 with generating set @xmath29 , a voltage assignment of @xmath13 is a mapping @xmath43 .
the lift @xmath44 is the digraph with vertex set @xmath45 and arc set @xmath46 , where there is an arc from vertex @xmath47 to vertex @xmath48 if and only if @xmath49 and @xmath50 . such an arc is denoted by @xmath51 .
recall that the _ de bruijn digraph _ @xmath52 has vertices @xmath53 , where @xmath54 for @xmath55 , and adjacencies @xmath56 in contrast with that , the _ kautz digraph _ @xmath57 has vertices @xmath53 , where @xmath58 , @xmath59 for @xmath60 , and adjacencies : @xmath61 now , if we consider the mapping @xmath62 two alternative definitions are the following : the de bruijn digraph @xmath52 has vertices @xmath63 , where @xmath64 , and adjacencies : @xmath65 similarly , the kautz digraph @xmath57 has vertices @xmath66 , where @xmath67 , @xmath68 , and adjacencies : @xmath69 notice that the mapping @xmath70 where @xmath71 , is a homomorphism from the kautz digraph @xmath57 to the de bruijn digraph @xmath72 .
also , note the inverse mapping @xmath73 now we show that both de bruijn and kautz digraphs can be seen as lifts of smaller de bruijn digraphs with appropriate voltage assignments .
[ lemmadb ] the equality @xmath74 holds with @xmath75 the vertices of the arc @xmath76 in @xmath52 , with voltage @xmath77 , give rise to the vertex subsets @xmath78 and @xmath79 in @xmath80 .
moreover , the elements of @xmath81 can be written as @xmath82 and @xmath83 thus , if we define the mapping @xmath84 in such a way that @xmath85 , we get the adjacencies @xmath86 with @xmath87 , which correspond to those of the de bruijn digraph @xmath88 , as claimed . as an example , figure [ voltage - debruijn - dibuix ] shows how to obtain @xmath89 as a lifted digraph of @xmath90 . as a lifted digraph of the base digraph @xmath90.,width=529 ]
-14 cm similarly , the following result shows that kautz digraphs can be seen as lifted digraphs of de bruijn digraphs .
the equality @xmath91 holds with @xmath92 where @xmath93 .
the proof is similar to that of proposition [ lemmadb ] .
indeed , the vertices of the arc @xmath76 in @xmath52 , with @xmath93 and voltage @xmath77 , give rise to the vertex subsets @xmath94 and @xmath95 in @xmath96 .
moreover , the elements of @xmath97 can be written as @xmath98 and @xmath99 thus , if we define the mapping @xmath100 in such a way that @xmath85 , we get the adjacencies @xmath86 with @xmath101 , which correspond to those of the kautz digraph @xmath102 .
this completes the proof . by way of example , in figure [ k(2,3)voltage - dibuix ] we show how @xmath103 can be seen as a lifted digraph of @xmath90 . as a lifted digraph of the base digraph @xmath90.,width=529 ] -13.5 cm
in the more general context of nonrestricted maps , we now consider the quotient digraphs , and the equivalent constructions of vertex - split digraphs and partial line digraphs .
let @xmath0 be a digraph with @xmath16 vertices .
a partition @xmath104 of its vertex set @xmath105 , for @xmath106 , is called _ regular _ if the number @xmath107 of arcs from a vertex @xmath108 to vertices in @xmath109 only depends on @xmath110 and @xmath111 .
the numbers @xmath107 are usually called _ intersection parameters _ of the partition .
the _ quotient digraph _ of @xmath13 with respect to @xmath104 , denoted by @xmath112 , has vertices the subsets @xmath113 , for @xmath114 , and @xmath107 parallel arcs from vertex @xmath113 to vertex @xmath109 . in the _ line digraph _ @xmath115 of a digraph @xmath13
, each vertex represents an arc of @xmath13 , that is , @xmath116 , and a vertex @xmath6 is adjacent to a vertex @xmath117 when the arc @xmath5 is adjacent to the arc @xmath118 : @xmath119 .
line digraphs have shown to be very interesting structures in the study of dense digraphs ( that is , digraphs with a large number of vertices for given degree and diameter ) . moreover ,
it is know that the iteration of the line digraph technique yields digraphs with maximum connectivity . for more details ,
see , for instance the papers by fiol , yebra , and alegre @xcite , and fbrega and fiol @xcite .
furthermore , by the heuchenne s condition @xcite , a digraph @xmath13 is a line digraph if and only if , for every pair of vertices @xmath3 and @xmath4 , either @xmath120 or @xmath121
. let @xmath124 be a regular partition of @xmath13 with intersection parameters @xmath107 .
then , consider the induced partition of its arcs ( or vertices of @xmath115 ) @xmath125 with sets @xmath126 . this partition is also regular since , from the definition of line digraph , the number of arcs from a vertex @xmath127 to any vertex @xmath128 is @xmath129 . to prove the digraph isomorphism ,
assume that @xmath130 has the arc @xmath131 .
this means that in @xmath13 there is the path @xmath132 , where @xmath133 , @xmath134 and @xmath135 . but
@xmath127 and @xmath136 so that , in @xmath137 , there is the arc @xmath138 and this concludes the proof .
[ prop : lift - is - a - ld ] let @xmath139 be a digraph endowed with a voltage assignment @xmath140 . if @xmath13 is a line digraph , and for every pair of vertices @xmath3 and @xmath4 with common out - neighbor sets @xmath141 , we have @xmath142 then , the lifted digraph @xmath40 is again a line digraph .
it suffices to prove that @xmath40 satisfies heuchenne s condition . with this aim ,
let @xmath143 , so that in @xmath40 the vertex @xmath47 is adjacent to the vertices @xmath144 and @xmath145 .
now , if there is a vertex @xmath8 such that @xmath146 , we also have @xmath147 ( because @xmath13 is a line digraph ) .
but the former implies that vertex @xmath48 , where @xmath148 is adjacent to vertex @xmath149 moreover , from ( [ line - condition ] ) interchanging @xmath110 and @xmath111 , we get that @xmath150 .
let us now consider what we call _ the vertex - splitting method _ to `` blow up '' a digraph . given a digraph @xmath139 on @xmath16 vertices and @xmath152 arcs , the vertex - split digraph @xmath153 , where @xmath154 is constructed as follows .
every vertex @xmath8 is split into @xmath155 vertices @xmath156 , where @xmath157 .
thus the order of @xmath153 is @xmath158 satisfying @xmath154 .
moreover , for each arc @xmath159 , we choose any vertex , say @xmath160 , with @xmath161 , and set the arcs @xmath162 for every @xmath163 .
we proceed in this way with all the arcs of @xmath13 , with the condition that , in the end , all vertices of @xmath153 must have nonzero indegree .
( notice that this is always possible , as @xmath164 for every @xmath165 . )
the above method is shown to be equivalent to the partial line digraph technique , that was proposed by fiol and llad in @xcite , which is as follows . given the digraph @xmath166 as above , let @xmath167 a subset of @xmath168 arcs satisfying @xmath169 , so that @xmath170 . then in the _ partial line digraph _ of @xmath13 , denoted by @xmath171 , each vertex represents an arc of @xmath172 , and a vertex @xmath6 is adjacent to the vertices @xmath173 for each @xmath174 , where let the partial line digraph @xmath180 be constructed from the arc set @xmath172 .
then , in constructing @xmath153 , every vertex @xmath4 of @xmath13 is split into the vertices @xmath181 if and only if @xmath182 for every @xmath183 . now , for every @xmath8 , assume that @xmath184 . then ,
in @xmath153 we have @xmath185 for some @xmath186 . assuming that @xmath159
, we have to consider two cases : * if @xmath176 , then in @xmath153 we choose every vertex @xmath187 , for @xmath188 , to be adjacent to the vertex @xmath189 . * otherwise , in @xmath180 we choose the vertex @xmath6 to be adjacent to the vertex @xmath173 , where @xmath190 ( for the chosen vertex @xmath191 in @xmath153 ) .
let @xmath13 be a digraph different from a cycle , with diameter @xmath194 and mean distance @xmath195 .
then , the diameter @xmath196 and mean distance @xmath197 of the vertex - split digraph @xmath153 , with @xmath198 , satisfy : @xmath199 * acknowledgments .
* this research is supported by the _ ministerio de ciencia e innovacin _ , spain , and the _
european regional development fund _ under project
mtm2014 - 60127-p , and the _ catalan research council _ under project 2014sgr1147 ( c. d. and m. a. f. ) . | in this note we present a general approach to construct large digraphs from small ones .
these are called expanded digraphs , and , as particular cases , we show the close relationship between lifted digraphs of voltage digraphs and line digraphs , which are two known ways to obtain dense digraphs . in the same context , we show the equivalence between the vertex - splitting and partial line digraph techniques . then , we give a sufficient condition for a lifted digraph of a base line digraph to be again a line digraph .
some of the results are illustrated with two well - known families of digraphs .
namely , the de bruijn and kautz digraphs , where it is shown that both families can be seen as lifts of smaller de bruijn digraphs with appropriate voltage assignments .
_ keyword : _ digraph , adjacency matrix , regular partition , quotient digraph , voltage digraphs , lifted digraph , partial line digraphs , vertex - split digraphs .
_ mathematics subject classifications : _ 05c20 , 05c50 , 15a18 . |
the dynamics of confined cavity fields interacting with the cavity wall is of great interest for the understanding of a variety of problems such as hadron bag models @xcite , sonoluminescence @xcite , cavity qed @xcite and black hole radiations @xcite .
most previous studies on dynamical cavity concentrated on scalar or photon fields @xcite , which despite the apparent simplicity , exhibit rich and complex dynamics .
in contrast , the system of a schrdinger particle in an oscillating spherical cavity has not gained as much attention . in this article
, we study the generalized quantum geometric phase of a particle in resonance with the vibrating cavity .
we will show that the geometric phase acquires sudden @xmath0-jumps when the particle makes transitions .
the geometric phase of a quantum system has drawn much attention since it was first proposed by berry in 1984 @xcite .
it reveals the geometric structure of the hilbert space of the system @xcite , and its existence in physical systems has been verified in many experiments @xcite , including electron diffraction from a screw dislocation @xcite and neutron interferometry @xcite .
the phase effects in molecular physics can lead to energy splittings and shift quantum numbers .
the geometric phase has also been shown to be intimately connected to the physics of fractional statistics , the quantized hall effect , and anomalies in gauge theory @xcite . as far as we know , our study represents the first calculation of the geometric phase of a resonating system , which evolves non - adiabatically and non - cyclically .
we consider an infinite cylindrical or spherical potential well with oscillating boundary @xcite : @xmath1 where @xmath2 , with @xmath3^{-1}$ ] .
the coordinates can be transformed to a fixed domain via @xmath4 , and , to preserve unitarity , the wavefunction is renormalized through @xmath5 , where @xmath6 ( 3/2 ) , for a cylindrical ( spherical ) well .
since the full hamiltonian @xmath7 commutes with @xmath8 and @xmath9 , the wavefunction can be factorized : @xmath10 , where @xmath11 depends only on the angular variables . inside the cavity ,
the radial wavefunction satisfies @xmath12y,\ ] ] where @xmath13 is the particle mass , and @xmath14 , 2 and @xmath15 for cylindrical and spherical wells respectively . in this paper
, we only consider the @xmath16 sector .
the wavefunction described by eq .
[ d.r.e ] evolves in time and acquires a time - dependent phase , which in general consists of a dynamical phase and a geometric one @xcite .
when the dynamics is adiabatic and/or the evolution is cyclic , the geometric phase , or berry s phase , has been studied for many systems . since we are interested in the geometric phase of a non - adiabatic , non - cyclic system , we have to resort to a generalized method .
following ref .
@xcite , we first remove the dynamical phase from the wavefunction of the system : @xmath17 , the dynamical phase is @xmath18 where @xmath19 .
the general setting geometric phase , or the pancharatnam phase @xmath20 is defined as the relative phase between the state at time @xmath21 and that at time @xmath22 , both with the dynamical phase removed .
it can be obtained from the inner product of these two states @xmath23 our main goal in this paper is to study this geometric phase for a simple but nontrivial dynamical system .
we have solved eq . [ d.r.e ] numerically and checked that the solution is stable and converges very well . in fig . 1 , we show the maximum energy of the particle as a function of the driving frequency @xmath24 , both having been scaled to be dimensionless : @xmath25 , and @xmath26 .
the particle is initially in the ground state , and it is in resonance at specific values of @xmath27 . in each period of cavity vibration @xmath28
, the geometric phase acquires some changes , @xmath29 , and it exhibits oscillations at these resonances .
we show three examples in fig . 2 for @xmath30 and 22.21227 , where the geometric phase acquires periodic changes of @xmath31 respectively .
all the resonances shown in fig .
1 are associated with oscillations of @xmath32 with amplitudes of @xmath33 , @xmath34 . to understand the resonances and the associated geometric phases , we limit ourselves to the parameter regime where the hamiltonian can be truncated to a two - level system . specifically ,
if @xmath35 , and the initial state is the @xmath36 unperturbed eigenstate @xmath37 , with eigenenergy @xmath38 , then when the driving frequency corresponds to the energy difference between the initial state and another unperturbed eigenstate @xmath39 , with eigenenergy @xmath40 , _ i.e. _ , @xmath41 , the particle is expected to behave as in a two - level system .
then the problem simplifies considerably , and we have obtained its solution with two analytic approaches : the su(2 ) method and the rotating - wave approximation ( rwa ) . following cheng _
et al._@xcite we first expand the time - dependent hamiltonian @xmath7 of a two - level system in the identity operator @xmath42 , the raising and lowering operators @xmath43 , and pauli spin matrix @xmath44 @xmath45 , \label{eq.h(t)}\ ] ] where @xmath46 , are in general complex functions of time . the evolution operator can then be written as @xmath47 where @xmath48 , and @xmath49 satisfies @xmath50 with the initial conditions @xmath51 for @xmath521 , 2 , or 3 .
suppose that the two levels in which the system oscillates are @xmath37 and @xmath39 , so that the wavefunction is @xmath53 .
furthermore , if the initial conditions are @xmath54 and @xmath55 , then @xmath56 the su(2 ) method is exact and reduces the problem to solving the ode s for @xmath57 .
the spirit of the rwa is to retain only those terms in the hamiltonian that correspond to the resonance frequency .
we first separate out the fast phase factors from the wavefunction : @xmath58 with @xmath59 . substituting eq .
[ 2.level.phi ] in the schrdinger equation , we have @xmath60 where @xmath61,\ ] ] and @xmath62 is a constant that depends only on the states involved . as @xmath63 is small , we can expand @xmath64 as a series of @xmath63 . retaining all terms up to third - order
, we have : @xmath65 \right . \nonumber \\ & + & { \epsilon^{2 } \over 4\omega } \left [ \left(\omega ^2 - 6\omega_{nk}^2 \right ) \cos \omega t - \left(\omega^2 + 2\omega_{nk}^2\right ) \right .
\cos 3\omega t \nonumber \\ & - & \left .
i\frac{7}{2 } \omega_{nk}\omega(\sin \omega t + \sin 3\omega t ) \right ] \right\ } .
\label{w}\end{aligned}\ ] ] note that @xmath64 consists of oscillatory terms with various frequencies depending on @xmath24 . in the spirit of rwa
, we keep only terms with the _ lowest _ frequency for each @xmath24 , the rationale being that @xmath66 and @xmath67 vary slowly in time and the contributions to @xmath68 from high frequency terms cancel on average over such a long time scale .
it is clear then from eq .
[ w ] that @xmath68 is largest if @xmath69 , with @xmath70 an integer , because of the emergence of zero frequency terms . for these driving frequencies
, @xmath68 is large and effective in inducing transitions , and we have resonances . at or close to a resonance , eq .
[ w ] simplifies tremendously , and closed - form solutions for eq .
[ matrix.eq.b ] can be obtained easily .
for example , when @xmath71 , @xmath72 , and we have rabi oscillations @xcite : @xmath73 , \\ \label{c_n } c_{n}(t ) & = & \gamma e^{-i\delta \omega t/2}\sin \chi t /\chi , \end{aligned}\ ] ] where we took @xmath74 , @xmath75 , and @xmath76 .
the maximum value of @xmath77 , @xmath78 $ ] , is a lorentzian with a fwhm of @xmath79 .
the solutions for higher @xmath70 resonances differ only in the widths @xmath80 , where @xmath81 with @xmath82 .
the fwhm of the resonances @xmath83 are narrower for larger @xmath70 , and the oscillation periods @xmath84 are longer .
the energy of the system is given in rwa by @xmath85 , \end{aligned}\ ] ] where @xmath86/4\chi^2 $ ] .
the energy of the particle therefore exhibits fast but small oscillations due to the @xmath87 factor , modulated by a large but slow oscillation of period @xmath88 .
it follows that @xmath89 which is again a universal lorentzian for all resonances .
the rwa results are virtually identical with those of the su(2 ) method . fig .
1 can be understood completely in terms of overlapping series of the @xmath90 resonances , and the positions of the peaks are in almost exact agreement with the rwa or su(2 ) predictions .
the resonance line shapes are well fitted by the lorentzian eq .
[ transform.emax ] , though the widths are underpredicted for the @xmath91 resonances , as shown in table 1 .
for example , the @xmath92 resonance has a much larger width than predicted by rwa and su(2 ) , which is mainly due to the involvement of other states during transition , such as a @xmath93 process , which is second order in perturbation theory and so affects the @xmath94 transitions more severely .
occasionally , resonances at similar frequencies may overlap and lead to broadened widths . at a resonance , the rabi oscillations of the particle as predicted in rwa ( eq . [ approx.et.1 ] )
can be seen explicitly in fig . 2 , where the energy of the particle vs. time is shown for @xmath95 and @xmath96 .
the rwa also predicts that at exact resonances , @xmath97 , and therefore the oscillation periods should be inversely proportional to the widths of the resonances , @xmath98 .
as shown in the last column of table 1 , this is clearly borne out in the numerical data , where @xmath99 is listed and is found to be closed to 1 for all resonances , deviating by less than 10% for even those with strong mixing . using the rwa and two - level approximation
, we can calculate the dynamical phase easily . removing the dynamical phase with the help of eq .
[ approx.et.1 ] , we have @xmath100 where @xmath101 . for @xmath69 resonances , if we choose @xmath102 to be an integral multiple , @xmath103 , of the cavity oscillation period @xmath104 , we get @xmath105 , \nonumber\ ] ] where @xmath106 . in particular ,
if @xmath107 , @xmath108 where @xmath109 , with @xmath110 .
there are sudden approximate @xmath0 jumps in @xmath111 at @xmath112 , as shown for example in fig . 3 for @xmath113 . since @xmath114 ,
the phase change in each cycle ( @xmath115 ) is @xmath116 at an exact @xmath117 resonance , @xmath118 and @xmath119 , and so @xmath120.\ ] ] therefore , the geometric phase oscillates with an amplitude of @xmath33 and period of @xmath121 .
both of these rwa predictions are in excellent agreement with the numerical data , as shown in fig . 2 .
note that the @xmath0-jumps and the functional forms of the geometric phases are independent of @xmath63 and @xmath122 , as long as they are nonzero .
to gain more insight into the geometric phase for a two - level system , we have studied a simple model of a magnetic field rotating around a spin-1/2 particle .
suppose an electron of charge @xmath123 and mass @xmath124 is placed at the origin , in the presence of a magnetic field @xmath125\;,\ ] ] which has a constant magnitude @xmath126 but its direction sweeps out a cone with an opening angle @xmath127 , @xmath128 , at a constant angular speed @xmath24 .
the hamiltonian of the system is given by @xmath129 where @xmath130 and * s * is the spin matrix .
the system can be solved analytically @xcite .
the instantaneous eigenspinors of @xmath7 with eigenenergies @xmath131 respectively are @xmath132 and @xmath133 suppose the electron spin is initially parallel to @xmath134 , and we consider the case when the particle makes a transition to spin down along the instantaneous direction of @xmath135 .
this happens with unit probability if @xmath136 , provided that @xmath137 .
the state vector at any time is then @xcite @xmath138 where @xmath139 the pancharatnam phases comparing the initial state with the state at time @xmath22 for different values of @xmath127 are shown in fig .
( [ ebb0k100w60]-[ebb0k100.2w90 ] ) .
the sudden @xmath0-jump and a two - period oscillation can be seen in the case when @xmath140 , @xmath141 being an even integer , such as the case in fig .
( [ ebb0k100w60 ] ) .
when @xmath141 is an odd integer , the pancharatnam phase performs a single - period oscillation with no @xmath0-jump , as shown in fig .
( [ ebb0k101w61 ] ) .
it has neither single - period nor two - period oscillation when other values of @xmath127 are used , fig .
( [ ebb0k100.2w90 ] ) . if @xmath142 , with @xmath103 an integer and @xmath143 , @xmath144 which has the same form as eq .
[ beta0.vib ] , where @xmath124 is an integer , @xmath145 , and @xmath146 for the case @xmath147 and @xmath148 , @xmath149 \ , & { \rm for}\ ; 0<\sin\alpha<1/2,\\ -{1\over 2 } \left [ \omega ( t_1+\tau ) - \omega ( t_1 ) \right ] \pm \pi\ , & { \rm for}\ ; 1/2<\sin\alpha<1\;. \label{beta1.k1 } \end{array } \right.\ ] ] in the limit @xmath150 , @xmath151 which has the same form as eq .
[ beta1 ] . in the cyclic limit , _
i.e. _ , @xmath152 , @xmath153 and thus the geometric phases are @xmath154 and @xmath155 .
since the geometric phases in the two models - an electron in a rotating magnetic field and a particle in a vibrating cavity - are remarkably similar , we conjecture that the main features of the generalized geometric phase we calculated , especially the @xmath0-jumps , are universal for a particle in transition from one state to another .
similar features about the geometric phase have also been obtained in @xcite .
here we present a geometrical model that helps to visualize the evolution of the geometric phase of a two - level system .
it is clear from eq .
[ simp.psi ] that the state vector traces out a path on the unit sphere defined by the angular coordinates @xmath156 and @xmath157 .
note that we have used a time - dependent basis @xmath158 , @xmath159 , which mark the north and south poles respectively on the unit sphere .
the excitation condition of our system therefore gives us the trajectory of the state on this unit sphere , @xmath160 which is a spiral curve .
the solid angle subtended at the origin by the spiral curve and the geodesic , @xmath161 , up to any @xmath162 is @xmath163 which coincides with eq .
[ omega.spin ] , in the limit @xmath164 .
therefore , we see from eq .
[ beta_0.spin ] and eq .
[ beta1.k1 ] that the geometric phases @xmath111 and @xmath165 are simply the solid angles subtended by the spirial curve , @xmath166 , from @xmath167 and @xmath168 up to @xmath169 respectively , divided by @xmath170 .
this picture is a generalization of berry s @xcite for adiabatic and cyclic evolution corresponding to trajectories with constant @xmath171 , and the solid angle is simply @xmath172 .
even the @xmath0-jump can be represented in this model by a jump of the particle from one sphere to another if its trajectory happens to reach the south pole , which depends on the value of the angle @xmath127 .
the double - sphere picture reminds one of the underlying su(2 ) structure .
in summary , we have reported the first calculation of the quantum geometric phase of a physical system in resonance - that of a particle in a vibrating cylindrical or spherical cavity , and we have shown that it acquires sudden @xmath0-jumps when the particle makes transitions from one state to another .
we have derived analytic expressions in the rwa and su(2 ) methods , which give excellent description of the energy , wavefunctions , and the geometric phases at these resonances .
we found remarkably similar properties of the geometric phases for the simple system of an electron in a rotating magnetic field , which led us to conjecture that the main features of the generalized geometric phases and especially the @xmath0-jump we found are universal .
we have also developed a geometrical model to help visualize these phases .
moore , j. math .
phys . * 11 * , 2679 ( 1970 ) ; p.w .
milonni , _ the quantum vacuum _ ( academic press , new york , 1993 ) ; n.d .
birrell and p.c.w .
davies , _ quantum fields in curved space _
( cambridge university press , cambridge , 1982 ) .
for example , c.k .
law , phys .
lett . * 73 * , 1931 ( 1994 ) ; v.v .
dodonov and a.b .
klimov , phys .
a * 53 * , 2664 ( 1996 ) , and references therein ; a. lambrecht , m.t .
jaekel and s. reynaud , phys . rev .
lett . * 77 * , 615 ( 1996 ) ; c.k . cole and w.c .
schieve , phys .
a * 52 * , 4405 ( 1995 ) ; p. meystre _ et al .
_ , j. opt .
b * 2 * , 1830 ( 1985 ) ; ying wu _ et al .
_ , phys .
a * 59 * , 3032 ( 1998 ) ; ying wu _ et al .
_ , phys .
a * 59 * , 1662 ( 1998 ) ; k.w .
chan , master thesis , the chinese university of hong kong ( unpublished ) , 1999 ; k. colanero and m .- c .
chu , phys . rev .
e * 62 * , 8663 ( 2000 ) .
.scaled widths ( @xmath173 ) of the lowest @xmath174 resonances for cylindrical cavity with @xmath175 .
full numerical results are compared to the rwa values .
the widths multiplied by the rabi oscillation period , both extracted from the numerical data , are also shown in the last column .
corresponding numbers for spherical cavity with @xmath176 are shown in parentheses . [ cols="^,^,^,^,^",options="header " , ] | we study the general - setting quantum geometric phase acquired by a particle in a vibrating cavity . solving the two - level theory with the rotating - wave approximation and the su(2 ) method
, we obtain analytic formulae that give excellent descriptions of the geometric phase , energy , and wavefunction of the resonating system .
in particular , we observe a sudden @xmath0-jump in the geometric phase when the system is in resonance .
we found similar behaviors in the geometric phase of a spin-1/2 particle in a rotating magnetic field , for which we developed a geometrical model to help visualize its evolution . |
study of atomic and molecular polarizabilities remains an important task in atomic physics . the atomic polarizability @xmath9 depends on electric dipole matrix elements @xmath10 @xcite which also describe transition strengths , state lifetimes , van der waals interactions , and scattering cross - sections .
here @xmath11 denotes an electric dipole operator , @xmath12 the level energy with quantum number @xmath13 , and @xmath14 its wave functions .
accurate measurements of polarizability facilitate progress in sophisticated atomic structure calculations and the theory of heavy atoms which results in more precise predictions for other important atomic parameters ( see e.g. @xcite ) .
measurements of polarizabilities become even more crucial in applications for modern optical atomic clocks .
predictions of the `` magic '' wavelength in optical lattice clocks @xcite and accurate estimation of the blackbody radiation shift require precise knowledge of static and dynamic polarizabilities @xcite .
measurement of static polarizabilities provides an important benchmark for calculations resulting in significant improvement of optical clock performance @xcite .
no less important are polarizability measurements for the ground state hyperfine components of the alkali atoms used in microwave atomic clocks ( see , e.g. , @xcite ) . for alkalis in the ground state
the uncertainty in the theoretical prediction for the polarizability is about 0.1% @xcite while the measurement uncertainty is typically 0.5 - 1.0% ( see @xcite ) .
the lowest uncertainty is demonstrated by using laser cooled atoms and atomic interferometers providing high sensitivity to electric fields @xcite .
ground state atoms are relatively easy to prepare in a particular hyperfine and magnetic quantum state while the natural decay does not pose any limitation for the experiment .
on the other hand , relatively long - lived rydberg atoms are highly sensitive to electric fields @xcite which simplifies interpretation of the experimental results .
polarizability measurements were performed in atomic vapor cells @xcite and on laser cooled atoms @xcite with relative uncertainties of 0.1 - 3% depending on the state .
asymptotic theory of rydberg atoms is well understood and shows good agreement with experimental observations .
however , atoms in intermediately excited states pose a challenge both for experiment and theory . they are typically short - lived and difficult to address , while the response to an electric field is small compared to the rydberg states . for example , the intermediate states in rb and cs ( @xmath15 ) were studied previously using atomic beams ( see , e.g. , @xcite ) . in the cited reference
a scalar polarizability was measured with a relative uncertainty of about 5% .
calculations of these states are also less accurate since the sum ( [ eq0 ] ) contains terms of alternating signs cancelling each other while a numerical error accumulates .
in this paper we report an accurate measurement of the static scalar and tensor polarizabilities of the @xmath16 and @xmath17 levels in rb-87 using spectroscopy of laser cooled atoms in a dc electric field . to our knowledge , the polarizability of the 5d level in rb has not been measured to date .
the 5d level in rb is used in metrology @xcite because the frequency of the 5s-5d transition is recommended by the international committee for weights and measures ( cipm ) for the practical realization of the definition of the meter @xcite .
knowledge of the 5d level polarizability is essential for an accurate evaluation of systematic shifts .
however , published calculations show considerable discrepancy .
two approaches were implemented to calculate the polarizabilities of the 5d level in rb : the method of model potential @xcite and the regular second order perturbation theory with direct summation of matrix elements and integration over the continuous spectrum @xcite . in the latter case
the transition probabilities were calculated by the program atom @xcite partly relying on an accurate experimental input .
the calculated results @xcite and @xcite differ 30% in the scalar polarizability and more than 100% in its tensor component as shown in table [ table1 ] .
although this discrepancy can be readily explained by the intrinsic uncertainty of the theoretical approach @xcite , an accurate experimental measurement of the polarizability components is highly desirable .
[ t ! ] c|c|c|c|c ref . &
@xmath18 & @xmath19 & @xmath20&@xmath21 + @xcite&21 110&-2871&20 670 & -3387 + @xcite&16 600&-1060&16 200 & -909 + using laser cooled rb atoms placed in the center of a plane capacitor we managed to reach a relative uncertainty for the scalar polarizability of 0.4% which is comparable to measurements in the ground state .
optical pumping of atoms to a certain magnetic sublevel allowed us to measure the tensor polarizability component with an uncertainty of 4% .
the measured values allow for distinction between the results of calculations and may facilitate further theoretical progress .
if an atom is placed in an external electric field , it becomes polarized and its energy levels are shifted according to @xcite : @xmath23 here @xmath0 and @xmath1 are the scalar and tensor polarizabilities , respectively , while for alkali atoms the parameter @xmath24 can be written as : @xmath25[3q(q-1)-4f(f+1)j(j+1)]}{(2f+3)(2f+2)f(2f-1)j(2j-1)}\,\ ] ] with @xmath26 . here
@xmath27 is the magnetic quantum number , and @xmath28 , @xmath29 , @xmath30 are the total magnetic moment , the electron magnetic moment and the nuclear spin quantum numbers , respectively .
the tensor component describes the relative splitting of magnetic sublevels in the multiplet and equals 0 for states with @xmath31 and @xmath32 . to measure both scalar and tensor polarizabilities
one should control the atomic state and address different magnetic and hyperfine sublevels .
if laser spectroscopy is used to probe the stark effect , both ground and excited levels are shifted in the external electric field . in that case the resonance frequency will be shifted according to @xmath33 where @xmath34 and @xmath35 stand for the ground and excited states , respectively , and @xmath36 is the shift of the resonance frequency . for non - degenerate states ,
the contributions of the individual transitions between the magnetic sublevels is proportional to the relative probabilities @xmath37 according to @xcite : @xmath38 where @xmath39 for @xmath40 polarized light and @xmath41 for @xmath42 , and the matrices are 6-@xmath43 and 3-@xmath43 symbols , respectively .
this relation should be taken into account if multiple magnetic sublevels are populated and the corresponding spectral components are not well resolved . in our case
the ground state is the 5p@xmath2 level in rb and the excited state is the 5d level , which are coupled by 776 nm laser radiation .
the experimental values for scalar and tensor polarizabilities of the 5p@xmath2 level are equal to @xmath44 and @xmath45 @xcite .
the atomic unit of the polarizability is the cube of the bohr radius @xmath46 , but in the experiment the units of @xmath47 are more practical .
the conversion is given by @xmath48=2.482\times10^{-4 } \alpha[a_0 ^ 3]$ ] .
to measure the stark shift of the 5d level in rubidium we used two - stage laser excitation @xmath49 in an external dc electric field on the order of 1kv / cm .
rb-87 atoms were laser cooled in a regular six - beam magneto - optical trap ( mot ) with an axial magnetic field gradient of up to 20g / cm .
the cloud of 300@xmath50 m in diameter contains about @xmath51 atoms at a temperature of 300@xmath50k . the mot configuration and the excitation scheme are similar to one described in ref @xcite .
compared to ref.@xcite , the atomic cloud was formed in the center of a plane capacitor consisting of two metallic meshes , as shown in fig.[fig1 ] .
the capacitor was placed inside a vacuum glass cell ( 3@xmath523@xmath5212 cm ) providing easy optical access .
[ t ! ] the mesh consists of non - magnetic stainless steel wires with a diameter of 25@xmath50 m and has an optical transparency of 80% . to manufacture one of the capacitor plates the mesh was made taut and then glued to a flat glass plate with a hole of 1 cm in diameter .
we saw to it that glue did not penetrate through the mesh to the front surface .
two plates were glued together using four glass posts of calibrated length , thus forming a plane capacitor with rectangular plates of @xmath53 cm and the separation of 1 cm .
the distance between the plates and the hole size provides clearance for the laser cooling beams .
one pair of laser cooling beams were sent through the mesh as shown in fig.[fig1 ] , which did not significantly change the cloud shape or the number of atoms .
although all glass components were manufactured in the laboratory of optics , p.n .
lebedev physics institute , and have superior flatness and well - defined sizes within a few @xmath50 m , the glue can influence the distance @xmath54 between the meshes . to reach an uncertainty of 0.5% in polarizability one should know the electric field to 0.2% , which corresponds to 20@xmath50 m uncertainty in the distance .
the parallelism of the glass plates was checked in the air using a micrometer and found to be parallel to within @xmath55 .
this angle was taken into account in calculations of the electric field .
the distance @xmath54 between the meshes was measured optically in vacuum using a high na lens assembly ( @xmath56 ) which imaged a free standing mesh on a ccd camera .
the lens and the camera were rigidly placed on a three - coordinate translation stage outside the vacuum chamber .
the translation stage axis was aligned with @xmath57 accuracy perpendicular to the capacitor plate ( along the @xmath58-axis , fig.[fig1 ] ) .
the focus position for one of the meshes was determined from a number of shots using a gradient filter method and then by fitting the position of the translation stage to the highest image sharpness .
this method provides a statistical uncertainty of 5@xmath50 m . moving the translation stage in the @xmath59-@xmath60 direction we performed similar measurements in the region within @xmath613 mm of the hole center .
the distance remained constant within 20@xmath50 m , the scatter can be explained by the mesh thickness . the final result including the averaged mesh thickness gives @xmath62
. the refractive index of air contributes to the result on a negligible level .
the field distribution in our capacitor was simulated using a finite element analysis , the result for the @xmath59-@xmath60 plane as well as the zoom in of the central volume is shown in fig.[fig2](a ) .
the position of the atomic cloud was controlled within @xmath63 mm with respect to the center of the capacitor by two ccd cameras . as follows from fig.[fig2 ] , the field variation within this volume is less than 0.05% .
the potential difference between the plates could be varied from zero to 2.5kv using a high - voltage power supply ( stanford research systems ps350 ) .
the accuracy of the device was studied with a high - precision voltmeter and corresponded to 0.1% variation .
measurement of the stark shift of each of the @xmath64 transitions ( fig.[fig3 ] ) was performed in a pulsed regime , the pulse sequence was repeated every 20@xmath50s and is depicted in fig.[fig4 ] .
first , atoms were laser cooled using the 780 nm transition coupling @xmath65 and @xmath66 levels . since the mot is formed close to the zero of the quadrupole magnetic field , all magnetic sublevels of the @xmath66 level become nearly equally populated , which does not allow for determination of polarizabilities ( [ eq1 ] ) , so we prepare the atoms into a particular magnetic state using optical pumping at 780 nm . after switching off the mot
beams we waited for 200ns and applied a circularly @xmath67 polarized pump pulse along the @xmath58-axis ( fig.[fig1 ] ) to transfer atoms to the @xmath68 magnetic state , as shown in fig.[fig3 ] .
the pulse had an intensity of 100 mw/@xmath69 and a duration of 500ns , which is much longer than the reverse pumping rate .
experimental results presented in the next section verify that nearly all atoms addressed by the probe 5p@xmath705d radiation were pumped into the @xmath71 sublevel .
the probe beam at 776 nm was tuned close to resonance with one of the fine structure sublevels 5d@xmath72 , which are separated by 89ghz .
the probe beam was directed at an angle of 11@xmath73 to the @xmath58-axis , allowing for independent control of its polarization .
polarization of the probe beam was changed from @xmath67 to @xmath74 with respect to its wave vector @xmath75 . in this case
we can address different magnetic sublevels of the 5d multiplet and thus derive polarizabilities using ( [ eq1 ] ) . due to the set angle to the quantization axis , probe radiation always contains an admixture of linear polarized light ( with respect to the @xmath58-axis ) , which is taken into account in our analysis .
thus , with the pump beam we couple the sublevel @xmath68 either with one of the sublevels from @xmath76 multiplet or with one from the @xmath77 as shown in fig.[fig3 ] .
the hyperfine structure of the 5d level is about 100mhz and is well resolvable .
the probe pulse with an intensity of 100 mw/@xmath69 was switched on right after the pump beam was switched off . the time delay between the two pulses
was chosen to be 10ns to avoid overlap between pulses .
the strong pump beam causes an ac stark shift of the 5p@xmath2 level , which influences the results of our measurement .
the 5p@xmath2 level lifetime equals 30ns and most of the atoms excited to the 5p@xmath2 remain there when the probe beam is on .
the probe pulse duration was 50ns , which is much shorter than the 5d level lifetime ( 300ns ) .
this prevents optical pumping back to the 5p@xmath2 level and re - distribution of the population between magnetic components .
approximately 30% of atoms in the 5d state decay to the ground state _ via _ the 6p level , emitting 410 nm photons . in our experiment ,
`` blue '' photons were collected onto a photomultiplier tube equipped with a narrow - band 410 nm filter .
photons were counted in a time window of 1 @xmath50s , as shown in fig.[fig4 ] .
the probe laser was scanned over the resonance with an acousto - optical modulator ( aom ) .
for each of the frequency steps ( typically 50 per line ) the signal was accumulated for 0.1s .
a typical count rate at the resonance position was @xmath78cps .
the mot magnetic field gradient was continuously switched on .
the influence of magnetic field on our results is discussed below .
measurement of the stark shift for each of the lines @xmath64 was performed at different voltages on the capacitor plates .
for each of the selected voltages three spectral lines were recorded , namely , with positive voltage polarity , grounded electrodes and negative polarity , to prevent any influence of charging .
the zero - voltage data was also used in the data analysis .
typical spectra of @xmath79 and @xmath80 transitions at different electric field strengths and probe beam polarizations are shown in figs.[fig5],[fig6 ] , respectively . from these spectra
we derive the frequency shifts of the corresponding transitions , estimate the quality of initial state preparation and estimate the excitation probabilities to different magnetic sublevels of the 5d level .
the spectra were fitted by three different models : ( a ) sum of two gaussian functions , ( b ) sum of two lorentzian functions and ( c ) sum of two lorentzian functions convoluted with the probe pulse spectral profile .
the pulse shape was measured in @xcite and its fourier spectrum width equals 20 mhz for 50ns duration .
the model used 6 fit parameters : two amplitudes , two central frequencies and two widths .
the different fit functions were used to test the model dependency of our fitting procedure and to evaluate the corresponding uncertainty .
the fit gave us the frequency of each of the hyperfine components at different values of the electric field .
to derive the polarizability of the @xmath17 level we used transitions from the @xmath81 level to two different hyperfine sublevels excited at different polarizations of the probe beam : @xmath82 ( fig.[fig5 ] , top ) and @xmath83 ( fig.[fig5 ] , bottom ) .
for the @xmath16 level we use , respectively , @xmath84 ( fig.[fig6 ] , top ) and @xmath85 ( fig.[fig6 ] , bottom ) .
since the probe beam radiation always contains a fraction of linearly polarized light if projected on the @xmath58-axis ( tilted by 11@xmath73 with respect to @xmath75 ) , all mentioned hyperfine components ( except @xmath84 ) contain two different magnetic sublevels , as follows from fig.[fig3 ] . after projecting on the @xmath58-axis ,
the circularly polarized pump beam will consist of 96% circular polarization ( the same sign ) and 4% linear polarization .
fig.[fig6 ] shows an example of the data for the @xmath82 and @xmath83 hyperfine sublelevs obtained at different electric field strengths .
the difference in the two curves is due to the tensor polarizability .
fitting the data with parabolic dependence , @xmath86 , according to ( [ eq3],[eq4 ] ) allows us to derive the sensitivity @xmath87 of each transition to the electric field .
the results are : @xmath88 where the entry in the parentheses after the level notation denotes the polarization of the probe beam . the spectrum shown in fig.[fig6](top )
contains only one hyperfine component , @xmath84 .
the @xmath85 one is not distinguishable from the noise .
this component can be excited by the linearly polarized fraction of the probe beam from the @xmath89 sublevel . knowing the fraction of linear polarization in our probe beam
, we can set a limit to the @xmath89 sublevel population ( see table[table2 ] ) .
table [ table2 ] also shows the relative probabilities to excite different magnetic sublevels of the final @xmath90 state , which are calculated using ( [ eq4 ] ) .
[ t ! ] c|c|c initial state & @xmath91 & population + @xmath66&3 , 3&@xmath92 + @xmath66&3 , 2&@xmath93 + c|c|c|c final state & @xmath91 & probe polarization & excitation + & & ( with respect to @xmath75 ) & probability + @xmath17&4 , 4&@xmath67 & 0.64 + @xmath17&4 , 3&@xmath67&0.01 + @xmath17&3 , 3&@xmath74 & 0.004 + @xmath17&3 , 2&@xmath74&0.032 + @xmath16&3 , 3&@xmath67&0.024 + @xmath16&3 , 2&@xmath74 & 0.192 + @xmath16&2 , 2&@xmath74 & 0.192 + using the results ( [ eq7 ] ) , table[table2 ] , and relation ( [ eq4 ] ) we get a system of linear equations for different magnetic and hyperfine sublevels from which we derive scalar and tensor polarizabilities .
polarizabilities for the 5p level are taken from @xcite , their uncertainties negligibly contribute to our error budget .
although the rb cloud resides in the minimum of the mot magnetic field , we expect a residual magnetic field on the order of 0.1 g due to the final cloud size and adjustment imperfections .
switching off the magnetic field in the experiment takes about 500@xmath50s , which would drastically reduce the duty cycle and the count rate .
therefore we decided to take the magnetic field into account instead of switching it off .
the influence of the magnetic field can be considered as additional degradation of the efficiency of the optical pumping .
we found that magnetic field reduces relative population of the @xmath94 level in our experiment from 98% to 97% .
more information about optical pumping in the magnetic field can be found in appendix i. to verify that the field does not significantly influence the results of our measurements , we measured the dependencies similar to that shown in fig.[fig7 ] at axial magnetic field gradients of 10g / cm and 5g / cm .
we did not observe any significant difference within the statistical uncertainty . since
our regular measurements were performed at a gradient of 10g / cm , we conservatively add a systematic uncertainty to the @xmath87 values ( [ eq7 ] ) of 0.1% due to the influence of the magnetic field . the different fitting procedures ( a , b , c ) described above influence the results at the level of 0.03% , so we add this value to the uncertainty of coefficients @xmath87 coming from the line shape model .
calculation shows that the residual population of the 5p@xmath95 level ( see table[table2 ] ) can influence the coefficients @xmath87 on the level of 0.07% .
imperfection of the probe beam polarization and error in determination of the angle between the @xmath58-axis and @xmath75 also result in an uncertainty of 0.1% .
[ t ! ] l|c effect & uncertainty , % + statistical uncertainty & 0.2 + electric field determination&0.3 + residual magnetic field&0.1 + line shape model&0.03 + optical pumping&0.07 + probe beam polarization&0.1 + ac stark shift&0.1 + * sum*&0.41 + the ac stark shift of the 5p@xmath2 level caused by a strong pump beam may influence the result of the measurement if the pump and probe pulses overlap .
experimental study of this effect shows that for the time delay used in the experiment ( 50ns ) , the residual pump beam perturbs the result on the level of 0.1% .
all mentioned uncertainties , including the uncertainty of electric field determination , are summarized in table[table3 ] .
adding up all contributing uncertainties quadratically , we get 0.4% for the coefficient @xmath87 .
this uncertainty directly converts in the uncertainty of scalar and tensor polarizability uncertainties as follows from ( [ eq4 ] ) .
since the main contribution to @xmath87 values comes from the scalar polarizability @xmath0 , its relative uncertainty is similar to the relative uncertainty of @xmath87 .
tensor polarizability @xmath1 is more than ten times smaller compared to @xmath0 , which means that its relative uncertainty is larger .
scalar and tensor polarizabilities for the 5@xmath96 levels measured in our experiment and corresponding uncertainties are summarized in table[table4 ] .
] .results for the tensor and scalar polarizability measurements of the 5@xmath96 levels in rb-87 in atomic units . here
@xmath97 is a standard deviation . [
cols="<,^,^",options="header " , ] in conclusion , we determined the scalar and tensor polarizabilities of the 5d level in rb .
using laser cooled atoms placed in a constant electric field and two - step laser excitation we demonstrated a relative uncertainty of 0.4% for the scalar polarizability and 4% for the tensor polarizability .
the demonstrated uncertainty for the scalar polarizability is comparable to accurate measurements in ground state alkali atoms .
our result is close to the theoretical prediction @xcite where the model potential approach was implemented .
we are grateful to i.veinstein and v.ovsiannikov for discussions and acknowledge support from rfbr grants # 12 - 02 - 00867-a and # 11 - 02 - 00987-a .
consider an atom placed in the pumping beam propagating along the @xmath58 axis ( fig.[fig8 ] ) .
a magnetic field directed at some angle to the @xmath58 axis will cause precession of the magnetic moment of the atom and therefore changes in the populations of magnetic sublevels will occur .
the maximum influence of the magnetic field will take place when it is perpendicular to the @xmath58 axis .
changes in populations of magnetic sublevels due to optical pumping can be found by solving the master equation for the density matrix .
influence of the magnetic field can be calculated by projecting the initial wavefunction onto the basis with quantization axis along the direction of the magnetic field .
the initial wavefunction is @xmath98 where @xmath99 describes populations of the magnetic sublevels with projection of the magnetic moment onto the @xmath58 axis equal to @xmath100 . in the new basis
this wavefunction will have a form @xmath101 where @xmath102 is a state with projection of the moment onto the direction of the magnetic field equal to @xmath103 and the coefficients @xmath104 can be calculated according to the equation : here the angles @xmath106 and @xmath107 can be expressed over @xmath108,@xmath109 and @xmath110 , where @xmath110 is an angle between the original quantization axis and the magnetic field .
if the magnetic field is perpendicular to the @xmath58 axis we have @xmath111 and @xmath112 , @xmath113 . taking into account evolution of the wavefunction in time we obtain : we accomplished numerical calculations of the population dynamics in the magnetic field perpendicular to the @xmath58 axis with 500ns of optical pumping and further 100ns evolution of the atom in the magnetic field without pumping .
the results of the simulations for different values of the applied magnetic field are shown in fig.[fig9 ] .
we can see that optical pumping remains efficient with magnetic fields up to 2 g .
larger fields leads to rapid mixing of states even during the pumping .
imperfection in the optical pumping in the experiment was about 3% , corresponding to a magnetic field less than 0.3 g. 99 l. d. landau and e.m .
lifshitz , quantum mechanics , pergamon press , new york , ( 1977 ) .
j. mitroy , m.s .
safronova , and charles w. clark , topical review : theory and applications of atomic and ionic polarizabilities , j. phys .
b 43 , 202001 ( 2010 ) .
h. katori , m. takamoto , vg palchikov , vd ovsiannikov , phys .
91 , 173005 ( 2003 ) .
k. beloy , u.i .
safronova , a. derevianko , phys .
97 , 040801 ( 2006 ) .
thomas middelmann , stephan falke , christian lisdat , and uwe sterr , phys .
109 , 263004 ( 2012 ) .
sherman , n.d .
lemke , n .
hinkley , m. pizzocaro , r.w .
fox , a.d .
ludlow , and c.w .
oates , phys .
108 , 153002 ( 2012 ) .
s. ulzega , a. hofer , p. moroshkin , a. weis , eur .
76 , 1074 ( 2006 ) .
a. derevianko , w. r. johnson , m. s. safronova , and j. f. babb , phys .
82 , 3589 ( 1999 ) .
axel grabowski , rolf heidemann , robert lw , jrgen stuhler , and tilman pfau , fortschritte der physik - progress of physic , 54 765 - 775 ( 2006 ) .
m. s. osullivan and b. p. stoicheff , phys .
a 33 1640 ( 1986 ) .
jianming zhao , hao zhang , zhigang feng , xingbo zhu , linjie zhang , changyong li , and suotang jia , journal of the physical society of japan 80 034303 ( 2011 ) .
w. hogervorst , and s. svanberg , physica scripta . vol .
12 , 67 , ( 1975 ) .
d. touahri , o. acef , a. clairon , j .- j .
zondy , r. felder , l. hilico , b. de beauvoir , f. biraben , and f. nez , opt . commun .
133 , pp . 471 , ( 1997 ) .
j. e. bernard , a. a. madej , k. j. siemsen , l. marmet , c. latrasse , d. touahri , m. poulin , m. allard , and m. ttu , opt . commun . 173 , 357 , ( 2000 ) .
quinn , practical realisation of the definition of the metre , including recommended radiations of other optical frequency standards ( 2001 ) .
metrologia , 40:103133 , n. l. manakov , v. d. ovsiannikov , j. phys .
b - atomic molecular and optical physics 10 , 569 ( 1977 ) .
kamenski , v.d .
ovsiannikov , j. phys .
39 , 2247 ( 2006 )
kondratev , i.l .
beigman , l.a .
vainshtein , bulletin of the lebedev physics institute 35 , 12 , 355 ( 2008 ) .
l. a. vainshtein , v. p. shevelko , program `` atom '' , preprint no .
43 of lebedev phys .
moscow , ( 1996 ) .
ovsiannikov , private communications .
daniel a. steck , `` rubidium 87 d line data , '' available online at http://steck.us/alkalidata ( revision 2.1.4 , 23 december 2010 ) . c. krenn , w. scherf , o. khait , m. musso , and l. windholz , zeitschrift fur physik d 41 , 229 ( 1997 ) .
snigirev , a.a .
golovizin , g.a .
vishnyakova , a.v .
akimov , v.n .
sorokin , n.n .
kolachevskii , quantum electron . , 2012 , 42 ( 8) , 71472 | we report on accurate measurements of the scalar @xmath0 and tensor @xmath1 polarizabilities of the 5d fine structure levels 5d@xmath2 and 5d@xmath3 in rb .
the measured values ( in atomic units ) @xmath4 , @xmath5 , @xmath6 and @xmath7 show reasonable correspondence to previously published theoretical predictions , but are more accurate .
we implemented laser excitation of the 5d level in a laser cooled cloud of optically polarized rb-87 atoms placed in a constant electric field .
@xmath8e - mail : [email protected] |
the spontaneous chiral symmetry breaking ( @xmath0sb ) is one of the most important features in low - energy qcd .
this is considered to be the origin of several hadron masses , such as the lightest nucleon mass . however , there is a possibility that only a part of the lightest nucleon mass is generated by the spontaneous @xmath0sb and the remaining part is the chiral invariant mass .
this structure is nicely expressed in so called parity doublet models ( see , e.g. refs .
@xcite ) .
it is an interesting question to ask how much amount of the nucleon mass is generated by the spontaneous @xmath0sb , or to investigate the origin of nucleon mass . studying dense baryonic matter
would give some clues to understand the origin of our mass , since a partial restoration of chiral symmetry will occur at high density region .
we expect that the mass generated by the spontaneous @xmath0sb will become small near the chiral phase transition point .
it is not so an easy task to study the dense baryonic matter from the first principle , namely starting from qcd itself : it may not be good to use the perturbative analysis , and the lattice qcd is not applicable due to the sign problem at this moment .
then , instead of the analysis from the first principle , it may be useful to make an analysis based on effective models , especially for qualitative understanding .
holographic qcd ( hqcd ) models ( see , for reviews , e.g. refs .
@xcite and references therein . ) are constructed based on the ads / cft correspondence @xcite and powerful tools to study the low - energy hadron physics . there
exist several ways to apply hqcd models for dense baryonic matter ( see e.g. refs .
recently the holographic mean field theory approach was proposed to study dense baryonic matter in ref .
this approach allows us to predict the equation of state between the chemical potential and the baryon number density . in ref .
@xcite , this approach was applied to a top - down model of hqcd @xcite including the baryon fields in the framework of the sakai - sugimoto ( ss ) model @xcite .
it is known @xcite that the ss model provides the repulsive force mediated by iso - singlet mesons such as @xmath1 meson among nucleons , while the attractive force mediated by the scalar mesons are not generated . as a result ref .
@xcite shows that the chemical potential increases monotonically with the baryon number density . on the other hand , when the attraction mediated by the scalar meson is appropriately included , the chemical potential is expect to decrease up until the normal nuclear matter density , and
then turn to increase ( see e.g. ref .
thus , it is interesting to study whether the chemical potential decreases with increasing density when the scalar degree of freedom is included . in this paper , for studying this , we adopt a bottom - up model given in ref .
@xcite which includes five - dimensional baryon field included in the model proposed in refs .
there the five dimensional scalar field @xmath2 is explicitly included to express the chiral symmetry breaking by its vacuum expectation value ( vev ) .
yet another interest appears in a hqcd model of ref .
since there is no chirality in five dimension , the hqcd model includes two baryon fields ; one transforms linearly under u(2)@xmath3 and another under u(2)@xmath4 .
the existence of two baryon fields naturally generates the parity doublet structure mentioned above . in ref .
@xcite , the boundary condition is adopted in such a way that all of the nucleon mass is generated by the chiral symmetry breaking . in the present analysis
, we will show that we can adjust the amount of nucleon mass coming from the chiral symmetry breaking by changing the boundary value of the five - dimensional baryon fields : the percentages of the chiral invariant mass in the nucleon mass is controlled by changing the boundary value .
we study how the equation of state in the dense baryonic matter depends on the percentage of the nucleon mass originated from the spontaneous @xmath0sb in the holographic mean field theory approach .
our result shows that , larger the percentage of the mass coming from the spontaneous @xmath0sb is , more rapidly the effective nucleon mass , which is extracted from the equation of state by comparing it with the one obtained in a walecka type model given in ref .
@xcite , with increasing baryon number density .
this paper is organized as follows : in section [ sec : parity ] , we first review the model proposed in ref .
@xcite , and then show the parity doubling structure .
we study the equation of state at non - zero baryon density in the model in section [ holographic_mft ] .
we also discuss the interpretation of our results in terms of a walecka - type model .
finally , we give a summary and discussions in section [ sec : sd ] .
we summarize several intricate formulas needed in this paper in appendix [ parity ] .
in this subsection we briefly review the holographic qcd model including baryons given in ref .
@xcite .
the fields relevant to the present analysis are the scalar meson field @xmath2 and two baryon fields @xmath5 and @xmath6 , as well as the 5-dimensional gauge fields @xmath7 and @xmath8 , which transform under the 5-dimensional chiral symmetry as @xmath9 where @xmath10 denote the transformation matrix of chiral symmetry , and @xmath11 with @xmath12 . by using these fields ,
the bulk action is given as @xmath13 where @xmath14 with @xmath15 and @xmath16 being the bulk masses for baryons and mesons , @xmath17 the scalar - baryon coupling constant , @xmath18 the gauge coupling constant .
the vielbein @xmath19 appearing in eqs .
( [ action_n1 ] ) and ( [ action_n2 ] ) satisfies @xmath20 where @xmath21 labels the general space - time coordinate and @xmath22 labels the local lorentz space - time , with @xmath23 . by fixing the gauge for the lorentz transformation ,
we take the vielbein as @xmath24 the dirac matrices @xmath25 are defined as @xmath26 and @xmath27 which satisfy the anti - commutation relation @xmath28 the covariant derivatives for baryon and scalar meson are defined as @xmath29 where @xmath30/(2i)$ ] .
@xmath31 is the spin connection given by @xmath32 in this subsection , we study the parity doubling structure of baryons in the model described in the previous subsection .
note that the analysis in this subsection is done for zero chemical potential , so that only the scalar field @xmath2 has a mean field part , or 4-dimensional vacuum expectation value ( vev ) , expressed by @xmath33 .
the equation of motion ( eom ) for @xmath33 is read from the action @xmath34 in eq .
( [ s x ] ) as @xmath35 the solution for this eom is obtained as @xcite @xmath36 where @xmath21 is the current quark mass and @xmath37 is the quark condensate @xmath38 . by using this vacuum solution , the eoms for @xmath5 and @xmath6 are given by @xmath39 as done in ref .
@xcite , we decompose the bulk fields @xmath5 and @xmath6 as @xmath40 where @xmath41 the mode expansions of @xmath42 and @xmath43 are performed as @xmath44 it is convenient to introduce @xmath45 and @xmath46 as @xmath47 which satisfy @xmath48 and @xmath49 with @xmath50 corresponding to mass eigenvalues .
it should be noticed that eq .
( [ eomf- ] ) is rewritten as @xmath51 which is the same form as in eq .
( [ eomf+ ] ) .
this implies that the solutions of eq .
( [ eomf+ ] ) and those of eq .
( [ eomf- ] ) are not independent with each other . for example , a solution of eq .
( [ eomf+ ] ) with negative energy eigenvalue is actually a solution of eq .
( [ eomf- ] ) with positive energy eigenvalue , which is the reflection of the charge conjugation invariance at zero density . for solving eq .
( [ eomf+ ] ) we need to fix the boundary conditions for @xmath52 and @xmath53 : at the uv boundary ( @xmath54 ) , @xmath52 and @xmath53 should be zero required by the normalizability
. the value of @xmath55 at the ir boundary can be set @xmath56 without loss of generality since the coupled differential equations in eq .
( [ eomf+ ] ) are homogeneous equations . in ref .
@xcite , the value of @xmath57 at the ir boundary was taken as @xmath58 in such a way that all of the mass of ground state baryon is generated by the chiral symmetry breaking expressed by the vev of @xmath33 . in the present analysis , we regard the ir value of @xmath57 , i.e. @xmath59 , as a parameter , which turns out to control the percentages of the chiral invariant mass included in the nucleon mass .
we summarize the boundary condition in table [ tab:1 ] for a convenience .
.boundary conditions for baryon fields [ cols="^,^,^",options="header " , ] we solve the equations of motion in eq .
( [ eomn+ ] ) for given values of @xmath60 and @xmath61 , with regarding @xmath62 in eq .
( [ def : mu ] ) as an eigenvalue . using the solutions for the baryonic mean fields @xmath63 and @xmath64 we calculate the baryon number density from eq .
( [ numberdensity_b ] ) .
we first study the density dependence of the chiral condensate for checking the partial chiral restoration . here
we define the in - medium condensate through the holographic mean field @xmath65 as @xmath66 we plot the density dependence of the @xmath37 normalized by the vacuum value @xmath67 in fig .
[ fignd_sigma ] . for several choices of @xmath60 . ]
this shows that the quark condensate @xmath37 decreases with the increasing number density , which can be regarded as a sign of the partial chiral symmetry restoration .
when the value of @xmath60 is decreased , the corresponding value of @xmath17 becomes larger ( see fig .
[ figb_g1 ] ) to reproduce the nucleon mass . since the larger @xmath17 implies the larger correction to the scalar from the nucleon matter , the smaller @xmath60 we choose , the more rapidly the condensate @xmath37 decreases .
the degreasing property of the chiral condensate is similar to the one obtained in ref .
@xcite .
we next show the resultant equation of state , a relation between the chemical potential and the baryon number density in fig .
[ fignd_mu ] . , and the vertical axis does the chemical potential by the nucleon mass of @xmath68gev .
the dashed line shows the eos for @xmath69 , the solid line for @xmath70 and the dotted line for @xmath71 . ]
this figure shows that the chemical potential increases with the increasing baryon number density .
this does not agree with the nature , in which the chemical potential decreases against the density in the low density region below the normal nuclear matter density .
this decreasing property is achieved by the subtle cancellation between the repulsive and attractive forces .
so this increasing property indicates that , in the present model , the repulsive force mediated by the u(1 ) gauge field is stronger than the attractive force mediated by the scalar degree included in x field . for studying the attractive force mediated by the scalar fields , we extract the density dependence of the effective nucleon mass using the walecka type model ( see e.g. refs .
@xcite ) , in which the chemical potential @xmath62 is expressed as @xmath72 where @xmath73 is the baryon number density , @xmath74 is the coupling for @xmath75th eigenstate of the omega mesons , @xmath76 is its mass , @xmath77 is the fermi momentum , and @xmath78 is the effective nucleon mass .
note that , in the free fermi gas , @xmath77 is related to @xmath73 as @xmath79 , which leads to @xmath80 .
in the present hqcd model , the @xmath81 coupling is calculated in vacuum as @xmath82 depending on the value of @xmath60 .
using these couplings together with the masses of @xmath83 , we convert the density dependence of @xmath62 obtained above into the one of the effective nucleon mass @xmath78 through eq .
( [ walecka ] ) .
we plot the density dependence of the effective mass @xmath78 in fig .
[ fig : effective mass ] . .
] this shows that the effective mass decreases with increasing density .
the decreasing rate is larger than the one obtained in ref .
@xcite , which is the reflection of the iterative corrections included through the holographic mean field theory .
it should be noted that the decreasing of @xmath78 is more rapid for smaller value of @xmath60 . in other word , the larger the percentage of the mass coming from the chiral symmetry breaking is , more rapidly the effective mass @xmath78 decreases with density . in fig .
[ fig : distribution ] , we plot the baryon charge distribution @xmath84 defined in eq .
( [ numberdensity_b ] ) for @xmath85 , @xmath86 and @xmath87 with @xmath88 fixed . . ]
this figure shows that the distribution is broader for larger value of @xmath89 .
this indicates that the distribution becomes more important for larger density , as shown in ref . @xcite .
we develope the holographic mean field approach in a bottom - up holographic qcd model proposed in ref .
@xcite which includes five - dimensional baryon field in the model proposed in refs .
we first study the mass spectrum of baryons with paying attention to the chiral invariant mass @xmath90 , which were formulated in parity doublet models ( see , e.g. refs .
we found the parameter ( @xmath60 ) , which is one boundary value of two baryon fields , controls the percentage of the chiral invariant mass : for @xmath69 all of the mass of the ground - state nucleon is generated by the spontaneous chiral symmetry breaking , while for @xmath91 , more than half of the nucleon mass is actually the chiral invariant mass .
we studied the density dependence of the chiral condensate .
our result shows that the quark condensate @xmath37 decreases with the increasing number density , which is consistent with the analysis done in ref .
furthermore , we found that the @xmath37 decreases more rapidly for smaller value of @xmath60 .
this is because the sigma coupling to the nucleon is larger for smaller @xmath60 .
we next calculated the equation of state between the baryon chemical potential and the baryon number density using the holographic mean field approach proposed in ref .
the resultant equation of state shows that the chemical potential increases with the increasing baryon number density .
this indicates that , in the present model , the repulsive force mediated by the u(1 ) gauge field is stronger than the attractive force mediated by the scalar degree included in x field . for studying the attractive force mediated by the scalar fields , we extract the density dependence of the effective nucleon mass using a walecka type model .
our result shows that the effective mass decreases with increasing density .
furthermore , the decreasing rate is more rapid for smaller value of @xmath60 .
this is consistent with the fact that the percentage of the chiral invariant mass is larger for larger value of @xmath92 .
in other word , the larger the percentage of the mass coming from the spontaneous @xmath0sb is , more rapidly the effective nucleon mass decreases with increasing baryon number density .
we also studied the baryon number distribution in the holographic direction .
our results show that the distribution is concentrated near the ir boundary for smaller @xmath89 .
this indicates that the distribution becomes more important for larger density . in the present analysis
, we made an analysis only at the mean field level .
so a natural extension is to consider the fluctuations on the top of the mean field obtained here .
it is also interesting to study the relation between the isospin chemical potential and the isospin density based on the approach developed in this paper , since the relation has a relevance to the symmetry energy .
we leave these works to the future project .
m.h . would like to thank useful discussions with youngman kim and chang - hwan lee .
we are also grateful to yong - liang ma for helpful discussions .
this work was supported in part by grant - in - aid for scientific research on innovative areas ( no .
2104 ) `` quest on new hadrons with variety of flavors '' from mext , and by the jsps grant - in - aid for scientific research ( s ) no .
22224003 , ( c ) no .
br.h . would like to thank the nagoya university program for leading graduate schools `` leadership development program for space exploration and research '' for the financial support .
in this appendix , we consider the parity transformation properties of the 5d fields . as in the 4-dimension , a parity transformation should flip the sign of normal three spatial coordinates .
but the 5th coordinate @xmath93 , as it is defined in the range @xmath94 , does not participate in the parity transformation .
@xmath95 using these conventions , we obtain the parity transformation of 5d fields as @xmath96 for the 5d spinors , their parity transformation properties are express as @xmath97 where @xmath98 and @xmath99 are arbitrary phases . c. e. detar and t. kunihiro , phys
. rev .
d * 39 * , 2805 ( 1989 ) .
d. jido , m. oka and a. hosaka , prog .
phys . * 106 * , 873 ( 2001 ) [ hep - ph/0110005 ] . c. sasaki and i. mishustin , phys .
c * 82 * , 035204 ( 2010 ) [ arxiv:1005.4811 [ hep - ph ] ] .
w. -g .
paeng , h. k. lee , m. rho and c. sasaki , phys .
d * 85 * , 054022 ( 2012 ) [ arxiv:1109.5431 [ hep - ph ] ] .
j. erdmenger , n. evans , i. kirsch and e. threlfall , eur .
phys . j. a * 35 * , 81 ( 2008 ) [ arxiv:0711.4467 [ hep - th ] ] .
y. kim , i. j. shin and t. tsukioka , prog .
part . nucl .
phys . * 68 * , 55 ( 2013 ) [ arxiv:1205.4852 [ hep - ph ] ] . j. m. maldacena , adv .
* 2 * , 231 ( 1998 ) [ int .
j. theor .
* 38 * , 1113 ( 1999 ) ] .
k. -y .
kim , s. -j .
sin and i. zahed , hep - th/0608046 ; jhep * 0801 * , 002 ( 2008 ) [ arxiv:0708.1469 [ hep - th ] ] . n. horigome and y. tanii , jhep * 0701 * , 072 ( 2007 ) [ hep - th/0608198 ] . s. kobayashi , d. mateos , s. matsuura , r. c. myers and r. m. thomson , jhep * 0702 * , 016 ( 2007 ) [ hep - th/0611099 ] .
y. kim , c. -h .
lee and h. -u .
yee , phys .
d * 77 * , 085030 ( 2008 ) [ arxiv:0707.2637 [ hep - ph ] ] ; k - i .
kim , s. h. lee and y. kim , in _ proceedings of workshop on hadron and nuclear physics , hnp09 , osaka , japan , 2009 _
( world scientific , hackensack , 2010 ) .
o. bergman , g. lifschytz and m. lippert , jhep * 0711 * , 056 ( 2007 ) [ arxiv:0708.0326 [ hep - th ] ] .
y. seo and s. -j .
sin , jhep * 0804 * , 010 ( 2008 ) [ arxiv:0802.0568 [ hep - th ] ] .
s. nakamura , y. seo , s. -j .
sin and k. p. yogendran , j. korean phys .
soc . * 52 * , 1734 ( 2008 ) [ hep - th/0611021 ] ; prog .
phys . * 120 * , 51 ( 2008 ) [ arxiv:0708.2818 [ hep - th ] ] .
m. rozali , h. -h .
shieh , m. van raamsdon and j. wu , jhep * 0801 * , 053 ( 2008 ) [ arxiv:0708.1322 [ hep - th ] ]
. m. harada , s. nakamura and s. takemoto , phys .
d * 86 * , 021901 ( 2012 ) [ arxiv:1112.2114 [ hep - th ] ] .
d. k. hong , m. rho , h. -u .
yee and p.
yi , jhep * 0709 * , 063 ( 2007 ) [ arxiv:0705.2632 [ hep - th ] ] . t. sakai and s. sugimoto , prog .
113 * , 843 ( 2005 ) ; prog .
phys . * 114 * , 1083 ( 2005 ) .
a. dymarsky , d. melnikov and j. sonnenschein , jhep * 1106 * , 145 ( 2011 ) [ arxiv:1012.1616 [ hep - th ] ] . , , and , * * , ( ) , . | we develope the holographic mean field approach in a bottom - up holographic qcd model including baryons and scalar mesons in addition to vector mesons and pions .
we study the effect of parity doubling structure of baryons at non - zero density to the equation of state between the chemical potential and the baryon number density .
we first show that we can adjust the amount of nucleon mass coming from the chiral symmetry breaking by changing the boundary value of the five - dimensional baryon fields .
then , introducing the mean field for the baryon fields , we calculate the equation of state between the baryon number density and its corresponding chemical potential .
then , comparing the predicted equation of state with the one obtained in a walecka type model , we extract the density dependence of the effective nucleon mass .
the result shows that the effective mass decreases with increasing density , and that the rate of decreasing is more rapid for larger percentage of the mass from the chiral symmetry breaking . |
the precise quantum signatures of different qualitative types of classical motion and the very definition of chaos in quantum mechanics are still the issues of an unsettled discussion ( see e.g. @xcite ) . due to unitarity of quantum dynamics
, quantum chaos can not be defined in the same way as the classical chaos @xcite , namely through the exponential sensitivity on the variation of initial conditions .
however , peres @xcite proposed an alternative concept which can be used in classical as well as in quantum mechanics : one can study the stability of quantum motion with respect to a small variation in the hamiltonian , or more generally , a variation of the unitary evolution operator . clearly , in classical mechanics this concept , when applied to individual trajectories ( or to phase space distribution functions as we show below in sect . 4 ) , is equivalent to the sensitivity to initial conditions : integrable systems with regular orbits are stable against small variation in the hamiltonian ( the statement of kam theorem ) , where s for chaotic orbits varying the hamiltonian has a similar effect as varying the initial condition : exponential divergence of two orbits for two nearby chaotic hamiltonians .
this paper is devoted to a systematic theoretical study of the stability of a unitary time evolution with respect to small static variations of the unitary propagator .
it will be primarily applied to the schr " odinger propagator in quantum dynamics ( sect . 2 and 3 ) ,
however an alternative application to the stability of classical unitary perron - frobenius evolution will be considered for comparison ( sect . 4 ) .
the quantity of the central interest here is the _ fidelity _ of unitary ( quantum ) motion .
let us consider a unitary operator @xmath6 being either ( i ) a short - time propagator @xmath7 , or ( ii ) a floquet map @xmath8 of ( periodically time - dependent ) hamiltonian @xmath9 ( @xmath10 ) , or ( iii ) a quantum poincar ' e map or any other quantized symplectic map . in any case a general small perturbation of the unitary operator can be written in the following form @xmath11 where @xmath12 is some self - adjoint operator , @xmath2 is a perturbation strength and @xmath3 is an effective planck constant which is taken out explicitly so that the observable @xmath12 has a well defined classical limit ( e.g. the corresponding weyl symbol ) .
the influence of a small perturbation to the unitary evolution is quantitatively described by the overlap @xmath13 measuring the hilbert space distance between the exact and the perturbed time evolution from the same initial pure state @xmath14 , @xmath15 , where _ integer _
@xmath16 is a discrete time ( in units of the period @xmath17 ) .
this defines the _ fidelity _ @xmath18 where @xmath19 gives the expectation value in the initial state @xmath20 .
more generally , it may be useful to statistically average the fidelity over an ensemble of _ different pure initial states _ @xmath21 appearing with _ probabilities _ @xmath22 .
thus we will write the fidelity in terms of a _ statistical density operator _
@xmath23 , namely as eq .
( [ eq : ft ] ) using the conventional statistical expectation value @xmath24 .
the theoretical discussion in this paper is fully general , however , we will later apply our theory in the two extreme situations , namely for the ( coherent ) pure initial state @xmath25 , and for the full hilbert space average @xmath26 which is also equivalent to considering a _ random _ pure initial state of maximal information entropy .
integer @xmath27 denotes a dimension of the relevant hilbert space which can be written semiclassically by the thomas - fermi rule @xmath28 where @xmath29 is the classical phase - space ( or energy shell ) volume and @xmath30 is the number of ( relevant ) degrees of freedom . the quantity @xmath31 , or its equivalent definitions , have already been discussed in several different contexts in quantum physics .
we name just a few , namely those which provided direct motivation for the present work : first , @xmath31 has been proposed by peres @xcite as a measure of the stability of quantum motion .
second , @xmath32 is the _ loschmidt echo _ characterizing the _ dynamical irreversibility _ , which has been used e.g. in spin - echo experiments @xcite where one is interested in the overlap between the initial state @xmath20 and an _ echo _ state @xmath33 obtained by composing forward time evolution , imperfect time inversion with a residual ( non - invertible ) interaction described by the operator @xmath34 , and backward time evolution .
third , the fidelity has become a standard measure characterizing the loss of phase coherence in the quantum computation @xcite .
fourth , it was used to characterize the `` hypersensitivity to perturbations '' in related studies of information theoretic characterization of quantum chaos @xcite , though in a different context of a stochastically time - dependent perturbation . the main result of this paper is a relation of the fidelity to the ergodic properties of quantum dynamics , more precisely to the time autocorrelation function of the generator of perturbation @xmath12 .
quantum dynamics of finite and bound systems has always a _ discrete spectrum _ since the effective hilbert space dimension @xmath27 is finite , hence it is _ non - ergodic _ and _ non - mixing _
@xcite : time correlation functions have fluctuating tails of order @xmath35 . in order to reach a genuinely complex quantum motion with a continuous spectrum one
has to enforce @xmath36 by considering one of the following two limits : semi - classical limit of effective planck s constant @xmath4 , or thermodynamic limit of number of particles , or number of freedoms @xmath37 .
our result is surprising in the sense that it predicts the fidelity to decay slower if the integrated time correlation function is smaller , i.e. if the relaxation process is faster
. in particular , for ergodic and sufficiently mixing dynamics ( such that the time integrated autocorrelation function of the perturbation @xmath12 is finite ) the fidelity is found to decay _ exponentially _ on a time - scale @xmath38 , whereas for a ` more regular ' , non - ergodic dynamics with non - vanishing time - averaged correlation functions , the fidelity decay is found to be qualitatively faster with a characteristic time - scale @xmath39 . however , this surprising and apparently counterintuitive result is correctly reconciled with the expected corresponding behavior in the classical limit due to non - trivial non - interchangability of the limits @xmath5 and @xmath4 . in all cases ,
the relevant time - scale for the decay of fidelity in the semi - classical regime can be explicitly computed in terms of the classical quantities only ( e.g. classical correlation functions ) .
the main theoretical predictions are clearly demonstrated on the numerical examples of a kicked top and a pair of coupled kicked tops .
furthermore , our theory on fidelity is general and can be applied to any perturbed unitary evolution ; as an example we consider the stability of the classical unitary perron - frobenius ( liouville ) evolution for maps , where we show how ` classical fidelity ' behaves in a qualitatively different way than the quantum fidelity and conforms to the corresponding linearly stable and exponentially unstable behaviors in the respective limiting cases of regular and chaotic dynamics .
two short announcements of our theory with applications in the contexts of many - body quantum dynamics and quantum computing have already been reported @xcite . in section 2
, we present a theoretical derivation of the general relation between the fidelity of unitary motion and the correlation decay , discuss qualitatively different regimes of stability in the semiclassical range of small effective @xmath3 and outline the corresponding time - scales . in section 3 , our theoretical predictions are demonstrated in detail by numerical experiments on a quantized kicked top .
in addition , we consider also a pair of coupled kicked tops in order to illustrate some dimensionality dependent aspects . in section 4 , we demonstrate the conceptual difference between the stability of classical and quantum unitary evolution by applying our theory to the case of classical ( liouvillian ) unitary evolution of phase space densities . in section 5 ,
we conclude by pointing to some potentially important applications and implications of our results .
we start by rewriting the fidelity ( [ eq : ft ] ) in terms of the heisenberg evolution of the perturbation @xmath40 .
let us recursively insert the unity @xmath41 in the definition ( [ eq : ft ] ) for @xmath31 and observe @xmath42 for @xmath43 running from @xmath44 downto @xmath45 , yielding @xmath46 the obvious next step is to expand the product ( [ eq : fprod ] ) into a power - series in @xmath2 @xmath47 where the operator @xmath48 denotes a left - to - right time ordering .
such a perturbative expansion converges absolutely for any @xmath2 provided only that the perturbation @xmath12 is a bounded operator .
therefore , the fidelity may be computed by an arbitrary truncation of the above expansion ( [ eq : fsum ] ) where its order @xmath49 ( @xmath50 ) depends only on the desired accuracy @xmath51 .
we can see that the fidelity @xmath31 can be expressed entirely in terms of time correlation functions of the generator @xmath12 of perturbation . before proceeding
further we note that the first order term ( @xmath52 ) can be trivially eliminated , namely by transforming the perturbations as @xmath53 with @xmath54 ( i.e. the perturbation @xmath12 has to be _ traceless _ if @xmath26 ) .
this does not affect the absolute value @xmath55 since a shift by a multiple of unity reflects in a simple complex rotation of fidelity @xmath56 .
so , to the lowest non - trivial order ( @xmath57 ) we find @xmath58 where @xmath59 is a 2-point time correlation function of the quantum observable @xmath12 . in the following
, the regime of sufficiently small @xmath2 , such that the fidelity remains close to @xmath60 ( i.e. @xmath61 ) and the above formula ( [ eq : f2nd ] ) is accurate , will be referred to as the regime of _ linear response_. although we have a general convergent formula for systematic higher order corrections ( [ eq : fsum ] ) we note that all the qualitative and in most cases even quantitative results of this paper can be explained by this simple linear - response kubo - like formula ( [ eq : f2nd ] ) .
an interesting and somewhat counterintuitive conclusion can be drawn from the expression ( [ eq : f2nd ] ) , namely the _ smaller _ the time - integrated correlation function the _ higher _ the fidelity . in the semiclassical regime of approaching the classical limit @xmath4
the quantum correlation function @xmath62 goes over to the corresponding classical correlation function provided a state @xmath63 converges to some classical phase space distribution , like e.g. for @xmath26 .
therefore , the fidelity for classically chaotic systems will decay with the rate which is _ inversely proportional _ to their rate of mixing , and furthermore for classically non - ergodic , i.e. regular or integrable motion , the correlation functions will generally not decay to zero and the fidelity will therefore decay much faster . of course , this surprising general conclusions can only be true if the initial state @xmath20 ( or the statistical density matrix @xmath63 ) is _ sufficiently random _
, i.e. if it is not a small linear combination of the eigenstates of @xmath6 ( or @xmath64 ) as will be discussed in more detail later .
now let us have a closer look at fidelity in the two qualitatively different regimes of correlation decay .
here we assume that the system is ( classically ) ergodic and mixing such that the correlation function of the perturbation @xmath12 decays sufficiently fast ; this typically ( but not neccessarily ! ) corresponds to globally chaotic classical motion .
due to _ ergodicity _ we will safely assume @xmath26 , i.e. average over the full hilbert space @xmath65 . for any other initial state
( e.g. in the worst case for the minimal wavepacket coherent state ) one obtains identical results on @xmath31 for sufficiently long times , i.e. longer than the _
ehrenfest ergodic _
time @xmath66 ( for a classically chaotic system with maximal lyapunov exponent @xmath67 ) needed for a minimal wavepacket to spread effectively over the accessible phase space .
the state averaged quantum correlation function is homogeneous in time @xmath68 so we simplify the second order linear response
formula for the fidelity @xmath69 and note again that we have assumed @xmath12 to be traceless @xmath70 .
if the decay of correlation function @xmath71 is sufficiently fast , namely it is sufficient that the sums @xmath72 exist for @xmath73 and that a certain characteristic _ mixing time _ exists , e.g. @xmath74 , then the above formula can be further simplified . for times
@xmath75 we can neglect the second term under the summation in ( [ eq : fc ] ) and obtain a linear decay in time @xmath16 in the linear response regime @xmath76 which is a transport coefficient given by the integral of correlation function .
we can make a stronger statement in a non - linear - response regime if we make an additional assumption on the factorization of higher order time - correlations
@xmath77point mixing .
this implies that @xmath78-point correlation @xmath79 is appreciably different from zero for @xmath80 only if all the ( ordered ) time indices @xmath81 are _ paired _ with the time differences within each pair @xmath82 being of the order or less than @xmath83 .
then we can make a further reduction , namely if @xmath84 @xmath85 so that we obtain a global exponential decay @xmath86 with a time - scale @xmath87 .
we should stress again that the above result ( [ eq : fmix ] ) has been derived by the assumption of true quantum mixing which can be justified only in the limit @xmath88 , e.g. either in semiclassical or thermodynamic limt .
however , precise study of finite size effects and the relevant time and perturbation scales will be given in subsection [ sec : timepert ] .
the opposite situation of a non - mixing and non - ergodic quantum dynamics , which typically corresponds to integrable , quasi - integrable ( kam ) , or mixed classical dynamics , is characterized by a non - vanishing time - average of the correlation function @xmath89 here , due to non - ergodicity , time - average @xmath90 depends on the choice of the initial state @xmath20 or , more generally , on the density matrix @xmath63 .
we assume that a characteristic averaging time - scale @xmath91 exists , namely it is an effective time @xmath92 at which the limiting process ( [ eq : cinfty ] ) converges .
therefore , for sufficiently large times @xmath93 the linear - response formula ( [ eq : f2nd ] ) gives , in contrast to ( [ eq : sigma ] ) , a simple _ quadratic decay _ in time @xmath94 with time - scale @xmath95 .
one should observe that the non - ergodic time - scale @xmath96 can be much smaller than the ergodic - mixing time - scale @xmath97 ( [ eq : fmix ] ) provided @xmath3 is fixed , or the limit @xmath5 is taken prior to the limit @xmath4 .
again , we can make much stronger general statement using an additional assumption .
namely , if we assume that @xmath98 then we can re - write the @xmath99-tuple sums in the series ( [ eq : fsum ] ) in terms of a _ time average perturbation operator _
@xmath100 namely @xmath101 note that @xmath102 is by construction an _ integral of motion _
@xcite , @xmath103\equiv 0 $ ] , and reduces to a trivial multiple of identity ( in fact , it vanishes with our choice of @xmath70 ) in the ergodic case of previous subsection . whereas in an ergodic and mixing case , @xmath104th order term of ( [ eq : fsum ] ) grows with time only as @xmath105 ( for even @xmath99 ) since it is dominated by pair time correlations , here in a non - ergodic case , the non - trivial time average operator @xmath102 already gives the dominant effect , namely @xmath106 for @xmath104th order term of ( [ eq : fsum ] ) , so the effect of pair time correlations can safely be neglected for sufficiently long times ( @xmath98 ) .
observe also that time averaged correlation is just a second moment of the time average operator @xmath107 in the semiclassical regime , @xmath90 goes to a purely classical , @xmath3-independent quantity @xmath108 _ only _ for such initial states @xmath20 , corresponding to @xmath3-independent distribution of initial conditions in the classical phase space , e.g. for random states or @xmath26 .
formula ( [ eq : favg ] ) expresses the fidelity in terms of a sequence of moments @xmath109 .
it has been shown @xcite that for a general class of one - dimensional spin chains , the moments in _ thermodynamic limit _
@xmath37 of long chains tend to a _
behavior @xmath110 , @xmath111 , so fidelity exhibits a gaussian decay @xmath112 although good numerical agreement with this formula has been observed in several related contexts , e.g. in quantum computing @xcite , it can not be generally valid , in particular not in the semiclassical regime with a small number of freedoms @xmath30 , as we show below . in general , we can rewrite the formula ( [ eq : favg ] ) in terms of ( average ) _ local density of states _ ( ldos ) of the time average @xmath102 which is defined ( for finite @xmath27 ) as @xmath113 where @xmath114 are the eigenvalues and @xmath115 the eigenstates of self - adjont operator @xmath102 .
namely @xmath116 so the fidelity @xmath117 of non - ergodic systems is just a fourier transformation of ldos @xmath118 of the time average perturbation @xmath102 ( which becomes the simple ( global ) density of states in case of an uniform average @xmath26 ) .
thus only a gaussian ldos would yield a gaussian fidelity decay ( [ eq : fgauss ] ) , however we often find good finite time gaussian behavior which goes beyond the second moment @xmath119 , even if the asymptotic time dependence is completely different , see e.g. fig .
[ fig : analit ] for an example of kicked top .
let us now try to give some semi - classical estimates on the fidelity decay in non - ergodic regime by expressing the asymptotics @xmath4 in terms of classical information only .
here we assume that our ( classical ) map is integrable so that there exist @xmath30 canonical constants of motion the action variables @xmath120 , which are quantized using ebk rule @xmath121 where integers @xmath122 are _ quantum numbers _ and integers @xmath123 maslov indices . since the time average operator @xmath102 commutes with @xmath6 and with the actions @xmath124 , it is diagonal in ( generically non - degenerate ) basis of eigenstates of @xmath124 ( quantized tori ) @xmath125 , namely in the leading semiclassical order @xmath126 where @xmath127 is the corresponding classical time - averaged observable in action space .
the fidelity ( [ eq : favg ] ) can therefore be written as @xmath128 now , provided the diagonal elements of the density matrix can be written in terms of some smooth function @xmath129 , and replacing the sum ( [ eq : fscs ] ) by an integral over the action space , which is justified for small @xmath3 up to classically long time @xmath130 , we obtain @xmath131 the obvious next step is to compute this integral by a method of stationary phase . however , the result depends on the precise form of the function @xmath132 ( which may also explicitly depend on @xmath3 ) so we only work out the details for two important special cases .
let us first assume uniform averaging over ( random ) initial states @xmath26 , so @xmath133 .
then , for large @xmath134 the above integral ( [ eq : fsemi ] ) can be written as a sum of contributions stemming from , say @xmath17 points , @xmath135 where the phase is stationary , @xmath136 , yielding @xmath137 where @xmath138 is a matrix of second derivatives at the stationary point @xmath139 , and @xmath140 where @xmath141 are the numbers of positive / negative eigenvalues of the matrix @xmath142 .
we repeat that the stationary phase formula ( [ eq : fsqrt ] ) is expected to be correct in the range @xmath143 .
most interesting to note is the asymptotic power - law time and perturbation dependence @xmath144 , which allows for a possible crossover to a gaussian decay ( [ eq : fgauss ] ) when approaching the thermodynamic limit @xmath37 .
now let us consider a single @xmath30-dimensional general coherent state centered at @xmath145 in action - angle space , @xmath146 , @xmath147 where @xmath148 is a positive symmetric @xmath149 matrix of squeezing parameters , giving @xmath150 and @xmath151 using the assumption @xmath152 , we see that a unique stationary point @xmath153 of the exponent approaches @xmath154 as @xmath5 , @xmath155 so we may explicitly evaluate ( [ eq : fcohst ] ) by the method of stationary phase without any lower bound on the range of time @xmath16 , @xmath156 note that the fidelity decay for a coherent initial state with regular classical motion has a time - scale @xmath157 which is consistent with ( [ eq : fr2 ] ) with @xmath158 and is typically ( for small @xmath3 ) larger than the time - scale @xmath159 for a random initial state .
however , surprisingly enough , for sufficiently small @xmath2 , fidelity decay for coherent state with regular classical dynamics can still be faster than the fidelity decay for ergodic and mixing dynamics .
it should be noted that the above derivation for fidelity decay of a coherent state ( [ eq : cs]-[eq : taunecoh ] ) remains valid in a quasi - integrable ( kam ) situation of mixed classical phase space , provided that the initial wave packet is launched in a _ regular region _ of phase space where ( local ) action - angle variables exist . from the derivation of gaussian behavior ( [ eq : fcoh ] ) we can also begin to understand the reason for a faster _ gaussian _ decay of fidelity for a non - mixing system in contrast to the _ exponential _ decay for a mixing case .
namely , in the non - mixing case , the displacement between centers of the wave packets of two nearby regular evolutions grows _ ballistically _ and thus produces destructive interference ( i.e. decay of fidelity ) much faster than in the case of mixing dynamics where the destructive interference builds up in a _ diffusive _ , random way .
the theoretical relations of the previous subsections are strictly justified in the asymptotics @xmath36 . however , in any experimentally or practically relevant situation the dimensionality of the relevant hilbert ( sub)space @xmath27 is finite , so we have to discuss in detail the limitations and additional time and perturbation scales associated with this .
first of all , for finite @xmath27 , fidelity @xmath31 can not decay indefinitely but starts to fluctuate for long times due to discreteness of the spectrum of the evolution operator @xmath6 .
let us write the eigenphases of @xmath6 and @xmath64 , and the corresponding eigenvectors , respectively , as @xmath160 , @xmath161 , and @xmath162 , @xmath163 , @xmath164 , satisfying @xmath165 , @xmath166 .
let us define a unitary operator @xmath167 which maps the eigenbasis of @xmath6 to the eigenbasis of @xmath64 , namely @xmath168 for all @xmath169 , with matrix elements @xmath170 , and write the matrix elements of the statistical averaging operator as @xmath171 .
note that the matrix @xmath172 is real orthogonal , @xmath173 , if @xmath6 and @xmath64 possess a common anti - unitary symmetry ( e.g. time - reversal ) .
now it is straightforward to rewrite the fidelity ( [ eq : ft ] ) as @xmath174 where @xmath175 is a matrix element of the operator product @xmath176 . at this point
we are interested in the long time fluctuations so we compute the time average fidelity fluctuation @xmath177 in the process of averaging over the time we have assumed that the eigenphases are non - degenerate so @xmath178 .
so we see that fidelity fluctuation @xmath179 depends on the orthogonal / unitary matrix @xmath172 and the statistical matrix @xmath180 .
clearly , @xmath172 converges to an identity matrix for sufficiently _ small _ @xmath2 , while for _ large _ enough @xmath2 and sufficiently complex perturbation @xmath12 one may assume that it becomes a random orthogonal / unitary matrix . of course , the critical perturbation strength @xmath181 for a transition between the two regimes is very much system dependent and can not be discussed in general .
let us now discuss three different limiting cases of the statistical matrix @xmath180 : + + * ( i ) * first , consider the simplest case where the initial state is a pure eigenstate of @xmath6 say , @xmath182 .
then @xmath183 and ( [ eq : frms ] ) rewrites to @xmath184 now , for _ weak perturbation _
@xmath185 fidelity does not decay at all @xmath186 , while for _ strong perturbation _ @xmath187 we may estimate ( [ eq : frmsa ] ) using the fact that @xmath172 become real / complex gaussian random variables with variance @xmath188 for the orthogonal(@xmath189)/unitary(@xmath190 ) case , namely @xmath191^{1/2}$ ] .
this result can be generalized to the case where initial state can be written as a finite superposition of a small number , say @xmath192 , of eigenstates of @xmath6 : then for a weak perturbation @xmath193 . with this simple observation
we can easily explain the numerical result of peres @xcite where no - decay of fidelity was found for a coherent initial state sitting in the center of elliptic island , thus being a superposition of a very small number of eigenstates ( it is almost an eigenstate ) , while the behavior in generic cases may be drastically different as described in the present work .
+ + * ( ii ) * second , consider the case of a _ random pure _ initial state @xmath194 , giving @xmath195 . the principle of maximal entropy forces the coefficients @xmath196 to be independent random complex gaussian variables with variance @xmath188 .
thus , averaging ( [ eq : frms ] ) over @xmath196 s yields @xmath197 here , for weak perturbation @xmath185 we find @xmath198 while for strong perturbation @xmath187 we have @xmath199 + + * ( iii ) * third , in the limiting case of a uniform average over all initial states @xmath26 , @xmath200 , we have @xmath201 again , for weak perturbation @xmath185 , @xmath202 , and for strong perturbation @xmath187 , @xmath203 . + + note that the formulae ( [ eq : frmsa]-[eq : frmsc ] ) state simply that fidelity fluctuation is an _ inverse participation ratio _ ( ipr ) of the _ perturbed _
eigenstate(s ) ( single one in case ( i ) or an average over all in case ( iii ) ) in terms of the _ unperturbed _ eigenstates , and is thus directly related to the _ localization _ properties of eigenstates of @xmath64 in terms of eigenstates of @xmath6
. however , except for the pathological case of initial state being a small combination of eigenstates of @xmath6 with weak perturbation , the fidelity fluctuation is always between the limiting values @xmath204 , and @xmath205 . for an illustrations of transitions between weak and strong perturbations in case ( iii ) of a kicked top , see fig .
[ fig : frms ] . ) in the mixing regime ( @xmath206 ) and with the whole hilbert space average ( [ eq : frmsc ] ) is shown as a function of the perturbation strength @xmath2 at four different values of @xmath207 ( curves from right to left , respectively ) .
data are scaled in such a way @xmath208 ( note that @xmath209 ) that the weak perturbation regime corresponds to value @xmath60 while the horizontal lines give the corresponding values in the strong perturbation regime @xmath210 .
note that the crossover perturbation scales as @xmath211 . ] therefore , fidelity will decay according to one of the asymptotic formulae ( [ eq : fmix],[eq : fgauss],[eq : fsqrt],[eq : fcoh ] ) until it reaches the value of finite size fluctuation .
this condition determines the time - scale @xmath212 , @xmath213 in non - ergodic and non - mixing regime this is the only relevant finite size time - scale , while in the mixing regime the situation is more complicated . here , combining ( [ eq : tstar ] ) with an exponential decay ( [ eq : fmix ] ) we get @xmath214 where @xmath215 is a parameter which typically lies in the range @xmath216 ( except if the initial state is non - random _ and _ the dynamics is non - ergodic ( e.g. case ( i ) ) _ and _ the perturbation is small @xmath217 ) and depends on ipr and the statistical operator @xmath63 as discussed above , and @xmath218 is the classical limit of the transport coefficient ( [ eq : sigma ] ) . the second new time - scale is related the to asymptotic non - decay of time correlations for finite @xmath27 quantum dynamics , namely even if the system is classically mixing the quantum correlation function will have a small non - vanishing ( @xmath3-dependent ) time average @xmath219 where @xmath220 .
however , since the classical system is ergodic and mixing , we will use a version of _ quantum chaos conjecture _ saying that @xcite then @xmath221 are independent gaussian random variables with a variance given by the fourier transformation @xmath222 of the corresponding classical correlation function @xmath223 at frequency @xmath224 . on the diagonal we have @xmath225 and additional factor of @xmath226 due to random matrix measure on the diagonal ( see e.g. @xcite ) , so we can write @xmath227 where @xmath228 is the classical limit of ( [ eq : sigma ] ) . due to ergodicity , for large @xmath27 , this does not depend on the statistical operator @xmath63 ; neither in cases ( ii , iii ) , nor in case ( i ) if one assumes additional ensemble averaging . the decay of fidelity ( [ eq : fc ] ) will start to be dominated by the average plateau ( [ eq : finitesize ] ) at time @xmath16 , when @xmath229 , i.e. for times @xmath16 greater than @xmath230 @xmath231 which is just the _ heisenberg time _ associated to the inverse density of states .
now , depending on the interrelation among four ( or five ) time - scales @xmath232 , @xmath233 , @xmath234 , @xmath235 , ( and @xmath236 if we are considering coherent initial states , like e.g. @xcite ) we can have four ( or five ) different regimes depending on three main scaling parameters : perturbation strength @xmath2 , planck s constant @xmath3 , and dimensionality @xmath30 .
note that we always have @xmath237 .
all different regimes can be reached by changing only @xmath2 while keeping @xmath3 and @xmath30 fixed ( see fig . [
fig : delta ] ) : + + ( * a * ) for sufficiently small perturbation @xmath2 we will have @xmath238 .
this means that @xmath239 and we will have initially quadratic decay ( [ eq : fr2 ] ) with @xmath90 given by an average finite size plateau ( [ eq : finitesize ] ) .
this will occur for @xmath240 where @xmath241 in fact , in this regime , also referred to @xcite as _ perturbative _ , one may use a first order stationary perturbation theory on the eigenstates of @xmath64 , yielding @xmath242 , @xmath243 , and rewrite ( following @xcite ) the finite size fidelity ( [ eq : ftfin ] ) in terms of a fourier transform of a probability distribution @xmath244 of diagonal matrix elements @xmath245 , @xmath246 .
since @xmath244 is conjectured to be gaussian for classically ergodic and mixing system @xcite , it follows that @xmath247 is also a gaussian with a semiclassically long time - scale @xmath248 @xmath249 ( * b * ) if @xmath250 we will have a crossover from initial exponential decay of fidelity ( [ eq : fmix ] ) to a gaussian decay ( [ eq : fp ] ) at @xmath251 , which will terminate and go over to fluctuating behavior when @xmath252 .
note that this will happen before time @xmath253 which is estimated ( [ eq : tstar ] ) based on a slower exponential decay ( [ eq : fmix ] ) .
this regime will exist in perturbation range @xmath254 with an upper border @xmath255 determined by the condition @xmath256 @xmath257 ( * c * ) if we still further increase @xmath2 , we have the most interesting , fully nonperturbative regime , when @xmath258 and we will have a full exponential decay ( [ eq : fmix ] ) , up to time @xmath253 when the fidelity reaches finite size fluctuations .
this regime continues for @xmath259 where the border @xmath260 is determined by the condition @xmath261 which is a point where the arguments leading to the factorization ( [ eq : factor ] ) and exponential decay ( [ eq : fmix ] ) are no longer valid .
we note that the relative size of this window range @xmath262 increases both , in the semiclassical and in the thermodynamic limit .
this regime also corresponds to ` fermi golden rule decay ' discussed in @xcite .
+ + ( * d * ) further increasing @xmath263 , the estimated fidelity decay time eventually becomes smaller than the classical mixing time @xmath83 , @xmath264 .
in this regime , the perturbation is simply too strong so that the fidelity effectively decays within the shortest observable time - scale ( @xmath83 ) .
however , if we consider a non - random , e.g. _ coherent initial state _ then the quantum correlation function relaxes on a slightly longer , namely eherenfest time - scale @xmath265 so the regime ( c ) should terminate already at a little smaller upper border @xmath266 which is naturally determined by the condition @xmath267 @xmath268^{1/2 } } \sim \frac{\delta_{\rm mix}}{[\ln(1/\hbar)]^{1/2}}.\ ] ] for coherent initial states one thus obtains an extra but very narrow regime @xmath269 ( describing the time - range @xmath270 ) where the fidelity decay can be computed in terms of classical lyapunov exponents @xcite .
in this regime things are simpler , as we do not have to worry about the average plateau in the correlation function due to a finite @xmath27 because we already have a higher average time correlation @xmath271 ( [ eq : cinfty ] ) due to non - mixing nature of dynamics .
thus we have here only two relevant time - scales , namely @xmath96 giving initial quadratic decay ( [ eq : fr2 ] ) , and the finite size fluctuation time - scale @xmath272 ( [ eq : tstar ] ) which depends on the properties of the initial state ( power law ( [ eq : fsqrt ] ) for a random initial state , versus gaussian ( [ eq : fcoh ] ) for a coherent initial state ) .
we conjecture , based on a rigorous result in spin @xmath273 chains @xcite , that the fidelity decay in the thermodynamic limit @xmath37 will generically approach a gaussian ( [ eq : fgauss ] ) , which is consistent with increasing power @xmath274 of power - law fidelity decay ( [ eq : fsqrt ] ) for finite dimension @xmath30 .
+ + we can summarize our findings by the following question : can we have a regime with @xmath275 , so that the fidelity will decay faster for a `` regular '' ( non - ergodic ) than for a `` chaotic '' ( ergodic and mixing ) dynamics ? indeed , if @xmath276 where @xmath218 and @xmath108 are computed for a mixing and non - ergodic classical dynamics , respectively , then the decay in a non - ergodic case will be faster than in a mixing case .
we can see that the condition ( [ eq : delta_con ] ) can be generally satisfied for a random initial state , while for a coherent initial state it can be satisfied above the ( second ) perturbative border ( see fig .
[ fig : delta ] ) , @xmath277 , only in more than one dimension @xmath278 .
we note that our result is not contradicting any of the existing findings on quantum - classical correspondence .
for example , a growth of quantum dynamical entropies @xcite persists only up to logarithmically short ehrenfest time @xmath265 , which is the upper bound for the validity of another semiclassical approach to the fidelity decay @xcite and within which one would always find @xmath279 above the perturbative border @xmath280 , whereas our theory reveals new nontrivial quantum phenomena with a semiclassical prediction ( but not correspondence ! ) much beyond that time . if we let @xmath4 first , and then @xmath281 , i.e. we keep @xmath282 , then we recover a result supported by a classical intuition , namely that the regular ( non - ergodic ) dynamics is more stable than the chaotic ( ergodic and mixing ) dynamics . on the other hand , if we let @xmath5 first , and only after that @xmath4 , i.e. satisfying ( [ eq : delta_con ] ) , we find somewhat counterintuitive results saying that chaotic ( mixing ) dynamics is more stable than the regular one .
we can conclude the section by saying that we have _ three non - commuting limits _ , namely the _ semiclassical limit _
@xmath283 the _ perturbation strength _
limit @xmath281 , and the _ thermodynamic limit _
@xmath284 , such that no pair of these limits commutes .
here we wish to verify and demonstrate the results of the previous section by numerical experiments . for this purpose
we choose the haake s quantized kicked top @xcite since this model served as a model example for many related studies @xcite .
the unitary propagator reads @xmath285 where @xmath286 ( @xmath287 ) are quantum angular momentum operators , @xmath288=i \epsilon_{klr } j_r$ ] .
( half)integer @xmath289 determines the size of the hilbert space @xmath290 and the value of effective planck s constant @xmath291 .
the perturbation is defined by perturbing the parameter @xmath292 , @xmath293 , so that the generator @xmath12 is @xmath294 physically , the system ( [ eq : ukt ] ) represents a twist around @xmath295-axis followed by a rotation for an angle @xmath296 around @xmath297-axis .
the classical limit is obtained by letting @xmath298 and writing the classical angular momentum in terms of a unit vector on a sphere @xmath299 .
the heisenberg equation for the su(2 ) operators @xmath300 , @xmath301 , reduces to the classical area preserving map of a sphere @xmath302 note that in the classical limit the perturbation generator is @xmath303 for @xmath304 the system is integrable , while with increasing @xmath292 there is a transition to chaotic motion .
the second parameter @xmath296 is usually set to @xmath305 , however in our numerical simulation we will use two different values exhibiting qualitatively different correlation decay ( for large @xmath292 ) : the standard case @xmath306 where @xmath223 decays in oscillatory way and the case @xmath307 where @xmath223 decays monotonically ( see fig .
[ fig : class30 ] ) . in the case of
@xmath306 we have two discrete symmetries .
the evolution @xmath6 commutes with @xmath308 and @xmath309 , the rotations for @xmath310 around @xmath311 and @xmath297 axes , respectively .
the hilbert space is therefore reducible into three invariant subspaces ( using notation of peres s book @xcite with the basis @xmath312 of eigenstates of @xmath313 and assuming @xmath289 to be an _ even integer _ ) : ee of dimension @xmath314 with the basis states @xmath315 and @xmath316 ; oo of dimension @xmath317 with the basis @xmath318 ; oe of dimension @xmath289 with the basis @xmath319 and @xmath320 with @xmath99 in all three cases running through @xmath321 . for @xmath322 the spaces oo and ee coalesce as @xmath309 is the only discrete symmetry left . in numerical experiments we always choose oe
subspace so that the dimension of the hilbert space is @xmath209 .
we will compute the fidelity using three different statistical operators @xmath63 : ( 1 ) @xmath26 corresponding to full hilbert space average , ( 2 ) pure random initial state @xmath25 ( components @xmath323 being independent gaussian pseudo - random numbers ) giving the same results as ( 1 ) however with a higher finite size fluctuation plateau @xmath324 ( as discussed in subsect .
[ sec : timepert ] ) , and ( 3 ) pure minimal wavepacket initial state @xmath325 , namely su(2 ) coherent state @xmath326 centered at the point @xmath327 on a unit sphere @xmath328 in this regime we will demonstrate two main different decays of fidelity as discussed in subsect . [ sec : timepert ] , namely the exponential decay @xmath329 ( [ eq : fmix ] ) for @xmath330 and a finite size ( perturbative ) gaussian decay @xmath247 ( [ eq : fp ] ) for a small perturbation @xmath331 .
we choose large @xmath206 to ensure fast mixing .
this value seems to be excessively large , but for smaller @xmath292 we still have `` sticky '' structures in classical phase space which in turn cause slow algebraic tails in the classical correlation function .
we must remember that the derivation of an exponential decay required well defined mixing time - scale @xmath83 which is not the case if we have a slow power - law decay of correlations . exponential decay time @xmath97 ( [ eq : fmix ] ) is determined by an integral / sum of time correlation function ( [ eq : sigma ] ) , which can be calculated in the semiclassical regime ( @xmath218 ) by means of the correlation function of the classical map @xmath332 @xmath333 is a `` traceless '' perturbation where averaging over the sphere gives @xmath334 .
( [ eq : cclass ] ) for @xmath206 , and @xmath307 ( top , full curve ) and @xmath306 ( bottom , broken curve ) . finite time
integrated correlation function is shown in the inset , converging to @xmath335 and @xmath336 , for @xmath306 and @xmath337 , respectively .
averaging over @xmath338 random initial conditions on a sphere is performed . ]
the classical correlation functions calculated by using the classical map ( [ eq : ktclass ] ) are shown in fig .
[ fig : class30 ] . for @xmath306
the correlation function is oscillating with an exponential envelope hence the transport coefficient @xmath335 is quite small . for @xmath307
the correlation decay is monotonic and exponential with @xmath339 .
the decay of quantum fidelity ( [ eq : fmix ] ) can now be obtained by using the _ classical _ limit @xmath340 : @xmath341 this formula has been compared with the exact numerical calculation of fidelity ( [ eq : ft ] ) where averaging over the whole hilbert space has been employed , i.e. @xmath26 , and , as expected due to ergodicity , there was _ no difference _ for sufficiently large @xmath289 when we have chosen a fixed coherent initial state . as the finite size fidelity fluctuation level @xmath342 ( [ eq : frmsc ] ) decreases with increasing hilbert space dimension
, we chose large @xmath343 in order to be able to check exponential decay ( [ eq : fmclass ] ) over as many orders of magnitude as possible . ,
@xmath343 and full trace average .
top figure is for @xmath306 and for @xmath344 ( solid , dashed , dotted curves , respectively ) bottom figure is for @xmath307 and @xmath345 , ( solid , dashed , dotted curves , respectively ) .
chain line in both cases gives the theory ( [ eq : fmclass ] ) with classically computed @xmath218 .
note that the largest @xmath346 case in the top figure ( dotted curve ) corresponds to @xmath347 , so it is already over the upper border of the regime ( *
c * , subsect .
[ sec : timepert ] ) @xmath263 but the agreement with the theory ( [ eq : fmclass ] ) is still quite good , appart from oscillations .
this is due to the oscillatory nature of time - correlations making the factorization assumption ( [ eq : factor ] ) justified ( on average ) even for much smaller time @xmath16 as required . ] , @xmath343 and full trace average .
top figure is for @xmath306 and for @xmath344 ( solid , dashed , dotted curves , respectively ) bottom figure is for @xmath307 and @xmath345 , ( solid , dashed , dotted curves , respectively ) .
chain line in both cases gives the theory ( [ eq : fmclass ] ) with classically computed @xmath218 .
note that the largest @xmath346 case in the top figure ( dotted curve ) corresponds to @xmath347 , so it is already over the upper border of the regime ( * c * , subsect .
[ sec : timepert ] ) @xmath263 but the agreement with the theory ( [ eq : fmclass ] ) is still quite good , appart from oscillations .
this is due to the oscillatory nature of time - correlations making the factorization assumption ( [ eq : factor ] ) justified ( on average ) even for much smaller time @xmath16 as required . ]
the results are shown in fig .
[ fig : j40k ] . the smallest and the largest @xmath2 shown roughly correspond to borders @xmath255 and @xmath348 , respectively . as we can see , the agreement with an exponential decay is excellent , at least over four decades in probability @xmath32 .
note that for @xmath306 and the largest @xmath346 the time - scale of the decay of fidelity is comparable to the classical time - scale @xmath83 so the factorization assumption ( [ eq : factor ] ) is strictly not applicable any more .
however , due to oscillatory correlation decay , overall agreement with the theory ( [ eq : fmclass ] ) is still rather good but the oscillations in the correlation decay reflect in the oscillations of the fidelity decay ( around the theoretical exponential curve ) .
of course , one does not need such a large @xmath289 in order to have an exponential decay , but for smaller @xmath289 the fluctuation level @xmath179 will be higher so the exponential decay ( [ eq : fmclass ] ) will persist for correspondingly smaller time , namely up to @xmath349 ( [ eq : tstar ] ) .
( solid curves ) and the corresponding classical sum @xmath350 ( dashed curves saturating at @xmath218 and ending at @xmath351 ) for @xmath352 .
upper curves are for @xmath307 while lower curves are for @xmath306 .
chain lines are best fitting asymptotic linear functions corresponding to @xmath353 , @xmath354 for @xmath306 and @xmath355 for @xmath307 . ]
then we focus on the so - called perturbative regime @xmath331 where the fidelity decay will be dictated by a finite size correlation average ( [ eq : finitesize ] ) , so according to eq .
( [ eq : fp ] ) @xmath356 we numerically computed @xmath90 ( [ eq : cinfty ] ) for @xmath357 , @xmath206 in order to show that it is given by the theoretical value ( [ eq : finitesize ] ) .
quantum correlation function has been computed @xmath358 by means of a traceless perturbation @xmath359\mathbbm{1}$ ] . in fig .
[ fig : cinf ] we show a finite time correlation sum @xmath360 which exhibits a crossover , at the heisenberg time @xmath361 , from the plateau given by @xmath218 to a linear increase @xmath353 due to finite size correlation average ( [ eq : finitesize ] ) @xmath362 . for @xmath206 , @xmath357 , and @xmath306 and @xmath337 , calculated as a full trace hilbert space average as a function of the scaled variable @xmath363 . for @xmath306 data for @xmath364 ( solid curve ) and @xmath365 ( dotted curve ) are shown . for @xmath307 , @xmath366 ( solid ) and @xmath367 (
dotted ) are shown .
note that for both @xmath296 the curves for both @xmath2 practically overlap .
the chain curves are theoretical predictions ( [ eq : fpert ] ) with classically computed @xmath218 . ]
the excellent agreement between prediction ( [ eq : fpert ] ) and full numerical calculation of fidelity is shown in fig .
[ fig : j15kpert ] . in view of our findings
this so - called @xcite perturbative regime can be understood as a simple consequence of a finite hilbert space dimension .
for times larger than the heisenberg time @xmath230 every quantum system behaves effectively as an integrable one , e.g. with a finite time average correlation plateau . in a kicked top
this regime is realized for sufficiently small value of @xmath292 .
if the classical phase space has a mixed ( kam ) structure , non - mixing regime of fidelity decay may be obtained by choosing a localized initial state ( e.g. coherent state ) located in a regular part of the phase space .
however , such a situation may easily lead to the opposite conclusion ( as compared to generic situation ) for insufficiently large dimension @xmath27 . as discussed in subsect.[sec : timepert ] , the fidelity fluctuation plateau is determined by the number of constituent propagator eigenstates @xmath162 which are effectively needed to expand the initial state . for a coherent state sitting inside a ( not too large )
regular ( kam ) island this number can be fairly small for numerically realizable hilbert space dimensions , thus prohibiting any significant fidelity decay as observed in @xcite .
nevertheless , we would still see the initial quadratic decay in the linear response regime but we would not be able to verify higher orders in the long - time expansion of fidelity . therefore , in order to make a situation numerically as clean as possible , we choose a small value of parameter @xmath368 , such that the classical dynamics is almost integrable and that the majority of phase space corresponds to regular motion so that the number of constituent eigenstates for coherent states is as large as possible ( on average ) . here
we focus only on the case @xmath306 . for small @xmath292 ,
the quantum and classical evolution is a ( slightly perturbed ) rotation around @xmath297 axis and the time averaged perturbation can be computed analytically ( in the leading order in @xmath292 ) @xmath369 we will now use ( approximate ) analytical results for @xmath370 to compare with numerics for @xmath368 .
it should be noted that our leading order analytical approximations could easily be systematically improved using a classical perturbation theory ( treating @xmath292 as a perturbing parameter ) .
however , since the agreement , as shown below , is almost perfect in all cases , we see no need for refinement at this level . first , we consider the full trace average , @xmath371 , and starting from the expression ( [ eq : favg ] ) , @xmath372 , write the fidelity as a sum over all eigenvalues of @xmath373 , namely @xmath374 , for @xmath321 ( in oe subspace ) , @xmath375 for large @xmath289 we can replace the sum with an integral and get @xmath376 , @xmath306 , @xmath377 , @xmath378 , and @xmath371 , in oe subspace .
solid curve gives the result of a numerical simulation .
isolated dots denote differences between numerical calculation and the analytic formula ( [ eq : ferfi ] ) for @xmath370 @xmath379 .
the dotted line gives a predicted asymptotic decay @xmath380 , and the dashed / chain curves are the predicted fidelity decays at small times , namely the second order expansion @xmath381 , and improved by the gaussian ( [ eq : fgauss ] ) . ] where @xmath382 is a complex error function with a limit @xmath383 to which it approaches by oscillating around @xmath60 .
we therefore have an analytic expression for the fidelity ( [ eq : ferfi ] ) in the case of an uniform average over whole hilbert space or , equivalently , over one random initial state .
its asymptotic decay is @xmath384 which agrees with the general semiclassical asymptotics ( [ eq : fsqrt ] ) and we expect initial quadratic decay ( [ eq : fr2 ] ) for small times @xmath385 decay rate @xmath96 is determined by the time averaged correlation @xmath90 ( [ eq : cinfty ] ) which can be calculated explicitly in the limit @xmath370 where the classical correlation function @xmath223 alternates for even / odd times as @xmath386 , @xmath387 , giving @xmath388 and the fidelity is expected to decay as ( [ eq : fr2 ] ) with @xmath389 , for short times , @xmath385 .
the short - time formula ( [ eq : fr2 ] ) and the full analytic expression ( [ eq : ferfi ] ) are compared with the numerical simulation in fig .
[ fig : analit ] .
the agreement is very good and , surprisingly enough , the gaussian approximation ( [ eq : fgauss ] ) for small times is observed to be valid considerably beyond the second order expansion ( [ eq : fr2 ] ) .
quite interesting is the regime where the decay time @xmath390 for a `` regular '' dynamics will be smaller than a decay time @xmath391 ( [ eq : fmclass ] ) for a `` chaotic '' dynamics .
this will happen for @xmath392 ( [ eq : delta_con ] ) .
this border has the same scaling with @xmath289 as @xmath348 ( [ eq : deltam ] ) . ,
@xmath393 and @xmath378 on oe subspace .
the dashed curve is a simulation for @xmath206 ( mixing regime , full trace average ) .
the pluses are for a pure coherent initial state ( see text for details ) at @xmath368 ( non - mixing regime ) . the chain and solid curves are , respectively , the theoretical exponential ( [ eq : fmclass ] ) and gaussian ( [ eq : fktcoh ] ) decays . ]
second , considering su(2 ) coherent initial state ( [ eq : su2coh ] ) , we could proceed along the same line , namely by an analytic calculation . but rather than that , we will illustrate the usefulness of a semiclassical formula for @xmath394 ( [ eq : fcoh ] ) .
this is a more general approach , as the explicit analytical calculation is usually not possible .
let us by @xmath395 denote the spherical angular coordinates measured with respect to the y - axis .
then @xmath396 represent canonical action - angle coordinates for the integrable case @xmath370 .
furthermore , the coherent state ( [ eq : su2coh ] ) acquires a semiclassical gaussian form ( [ eq : cs ] ) in the ebk basis @xmath397 , @xmath398 , namely @xmath399 the squeezing parameter @xmath148 reads @xmath400 in order to apply the general formula ( [ eq : fcoh ] ) we need to express the classical time average ( [ eq : ave ] ) in terms of a canonical action , @xmath401 , and evaluate the derivative , @xmath402 , giving @xmath403 . rewriting this expression in terms of original spherical angles @xmath404 and @xmath405 measured with respect to @xmath295-axis
, we obtain @xmath406 here the same interesting question , namely when do we have @xmath407 , results in a condition ( [ eq : delta_con ] ) @xmath408 ( we did not write the factor involving trigonometric functions ( [ eq : fktcoh ] ) which is of order @xmath60 ) .
this condition generally can not be meet if we also require @xmath409 ( in order to see full exponential decay in the mixing regime ) , except if @xmath410 .
this is for instance satisfied for @xmath378 and @xmath306 , for which @xmath335 .
we checked this by a numerical simulation and the result for a coherent state centered at @xmath411 , for which @xmath412 , is shown in fig .
[ fig : parad1 ] .
we can see that for @xmath413 the fidelity in a non - mixing regime ( @xmath414 ) is lower than the fidelity in a mixing regime ( @xmath415 ) . for larger times , @xmath416 ,
non - mixing decay @xmath417 displays revivals of fidelity .
finally , we want to visualize the phenomenon of faster decay of fidelity in a regular , non - mixing regime @xmath368 in comparison with the chaotic , mixing regime @xmath206 by using the phase space representation of wave functions .
a convenient and popular choice is a husimi function @xmath418 of a state @xmath20 defined as @xmath419 we have chosen a pure random initial state @xmath20 and propagated it , in the first case with the propagator @xmath6 for @xmath368 , and in the second case for @xmath206 , as well as with the perturbed propagators @xmath64 in both cases .
the state and all the other parameters ( e.g. perturbation strength @xmath2 ) were identical in both cases .
then we compared the differences between the husimi functions @xmath420 of the perturbed and the unperturbed time evolution .
the results for @xmath368 are shown in fig .
[ fig : hus1 ] and for @xmath206 in fig .
[ fig : hus2 ] . for chosen @xmath421
the fidelity decay for the regular dynamics is much faster than for the chaotic one and this effect can also be observed in husimi functions by comparing the right columns of both figures .
but one should stress that by considering the difference of husimi functions @xmath422 , we loose information about the relative phases of the perturbed and unperturbed wave functions in the coherent - state basis which are more important for the fidelity decay in the regular case ( @xmath368 ) .
however , results shown in figs . [ fig : hus1],[fig : hus2 ] suggest that not only the quantum phases but also the amplitudes ( in some classically motivated , e.g. coherent state basis ) exhibit larger susceptibility to system perturbations for the regular ( non - mixing ) as compared to the chaotic ( mixing ) dynamics . for a regular evolution @xmath368 , @xmath306 and three different times @xmath423 and @xmath424 ( left column ) . gaussian random initial state
is used and @xmath425 , @xmath421 . in the right column
we show pictures of absolute difference between husimi functions of the unperturbed and perturbed time evolution , @xmath426 .
fidelity at the times shown is @xmath427 and @xmath428 .
top picture in the right column gives the classical phase space portrait . ] except for a chaotic ( mixing ) dynamics , @xmath206 .
numerical values of the fidelity are now @xmath429 and @xmath430 . ] as we have already remarked at the end of sect .
[ sec : timepert ] , for a one dimensional @xmath431 systems , the ` surprising ' behavior @xmath407 is for coherent initial states possible only around the border ( [ eq : deltas ] ) @xmath255 ( unless @xmath218 is very small like in the example of fig .
[ fig : parad1 ] ) where the exponential decay in the mixing regime goes over to a gaussian due to finite size @xmath27 .
however , for more than one degree of freedom , @xmath432 , such behavior is generally possible well above the finite size perturbative border @xmath255 . in order to illustrate this phenomenon we will now briefly consider a numerical example of a pair of coupled kicked tops @xcite where @xmath433 . and @xmath425 .
the upper curves are for @xmath434 ( mixing regime ) , solid curve for a coherent initial state ( [ eq : ket ] ) and dashed curve for a random initial state , and the lower dotted curve is for @xmath435 ( non - mixing regime ) with a coherent initial state ( [ eq : ket ] ) .
the exponential and gaussian chain curves give , respectively , the expected theoretical decays ( [ eq : fmix ] ) and ( [ eq : fcoh ] , here decay time is determined by the best fit ) . ]
we consider coupled kicked tops with a unitary propagator ( a simplified version of that of ref .
@xcite ) @xmath436 where @xmath437 and @xmath438 are two independent quantum angular momentum vectors .
perturbed propagator is obtained by perturbing the parameter @xmath439 , so that @xmath440 .
the perturbation generator is therefore @xmath441 with @xmath291 .
a pair of coupled kicked tops possesses the same pair of discrete symmetries as the single top for @xmath306 @xcite , namely @xmath308 and @xmath309 , and in addition it is unvariant under the permutation of the identical tops .
however in our simple - minded numerical experiment reported here we have used the propagator ( [ eq:2kt ] ) over the full @xmath442 dimensional hilbert space thus making the appropriate average over all symmetry classes .
the classical limit is obtained by @xmath443 and writing the classical angular momentum vectors in terms of two unit vectors on the sphere @xmath444 . in component notation
we get the following equations of motion @xmath445 we have chosen two regimes , namely non - ergodic ( kam ) regime for @xmath435 where the vast majority of classical orbits are stable , and the mixing regime for @xmath434 where no significant traces of stable classical orbits were found and where the correlation sum was to a very good accuracy given by the first term only @xmath446 our motivation here was to compare the non - ergodic and mixing fidelity decays for the coherent initial state which is here the dyadic product of su(2 ) coherent states ( [ eq : su2coh ] ) @xmath447 in fig.[fig : parad2 ] we show the fidelity decay at @xmath425 and @xmath448 in non - ergodic and mixing cases started from the coherent state ( [ eq : ket ] ) with @xmath449 .
we find excellent agreement between the theortical predictions ( [ eq : fmix ] ) and ( [ eq : fcoh ] ) and the numerics .
note that we are here already in the regime @xmath450 where the fidelity decay in ergodic - mixing regime is slower @xmath451 than in non - ergodic regime @xmath452 . in the ergodic - mixing regime ( @xmath434 )
we show for comparison also the fidelity decay for a random initial state which is ( due to ergodicity ) almost identical to the case of coherent initial state .
we note that overall fidelity decay here is similar as in a one - dimensional case , however , the scaling of various time and perturbation scales on @xmath291 is different as discussed in subsect.[sec : timepert ] .
we want to conclude this section with an interesting experimental observation , namely that in the non - ergodic regime the revivals of fidelity ( quantum reccurrences ) beyond the time scale @xmath253 are much less pronounced in @xmath453 than in @xmath454 , e.g. compare figs.[fig : parad1 ] and [ fig : parad2 ] .
in this section we wish to contrast the surprising findings on the stability of quantum motion with an application of the general idea ( section 2 ) to a fully classical concept , namely to ( unitary ) perron - frobenius propagator of classical phase space densities .
let us consider a classical dynamical system being given by an invertible measure ( volume ) preserving map @xmath455 on a @xmath456 @xmath457 dimensional phase space @xmath458 .
no additional properties of the map will be assumed except of being at least piece - wise differentiable so that the stability matrix @xmath459 exists almost everywhere .
unitary perron - frobenius propagator @xmath460 is defined on a phase space density from @xmath461 as @xmath462 the next step is to perturb the map in a most general way by composing it @xmath463 with a near identity map @xmath464 which is generated by an arbitrary ( smooth ) vector field @xmath465 thru the @xmath466flow @xmath467 vector field @xmath465 should be divergence free @xmath468 to keep the perturbed map @xmath469 volume preserving ( or hamiltonian for symplectic maps ) .
now we can write the perturbed unitary perron - frobenius propagator @xmath470 in the standardized general form ( [ eq : u_d ] ) with the self - adjoint perturbation operator @xmath471 @xmath472 we define the classical fidelity as the inner product between the phase space densities propagated by two slightly different perron - frobenius propagators from the same fixed initial density @xmath473 @xmath474 note that @xmath475 follows the corresponding quantum fidelity @xmath31 ( 2 ) for very short times @xmath476 , since the latter can be alternatively written as @xmath477 in terms of wigner functions @xmath478 and @xmath479 of the states @xmath480 and @xmath481 , respectively , corresponding to the liouville densities @xmath482 and @xmath483 in ( 74 ) if @xmath484 .
@xmath475 can be given a probabilistic interpretation , namely if @xmath473 is a characteristic function on a set @xmath485 , @xmath486 or @xmath45 , for @xmath487 or @xmath488 , respectively , with the volume @xmath489 then @xmath490 is a _ probability _ that a _
perturbed _ system with the initial condition chosen at random from the set @xmath491 is found , after time @xmath16 , in the image of the same set @xmath491 propagated for time @xmath16 with the _ unperturbed _ system . in order to proceed formally along eqs .
( [ eq : fprod]-[eq : f2nd ] ) we define and straightforwardly compute the ` classical heisenberg operators ' @xmath492 finally , plugging this into @xmath466expansion ( [ eq : fsum ] ) , observing that the first order always vanishes since @xmath493 , and simplifying the second order by integrating by parts , we obtain @xmath494 note that now the rhs does not look like a correlation function at all !
the reason for this lies in an essentially different form of the generator of perturbation ( [ eq : acl ] ) which in classical case involves phase space derivatives invoking the stability matrix , while in quantum mechanics it is a simple quantization of the observable @xmath465 .
now , observe that the classical linear response formula ( [ eq : fcl2 ] ) gives precisely the same result as one intuitively expects based on the stability of individual orbits . indeed , in the case of chaotic dynamics , the integrand ( for sufficiently long times @xmath43 ) becomes dominated by the leading eigenvalue of the stability matrix @xmath495 , namely the largest lyapunov exponent @xmath67 , and ( [ eq : fcl2 ] ) can be written as @xmath496 where the unspecified constant depends on the initial distribution @xmath473 .
this means the classical fidelity for chaotic dynamic decays on logarithhmically short time - scale @xmath497 . on the other hand , in the case of regular dynamics , all eigenvalues of the stability matrices
@xmath495 have modulus @xmath60 so the integrand remains bounded by some constant and the fidelity can be estimated as @xmath498 note that the fidelity decay for regular dynamics has the same @xmath2 and @xmath16 dependence as in the quantum chase ( [ eq : fr2 ] ) with the decay time - scale @xmath499 .
in this paper we have presented a simple theory on intrinsic decoherence of deterministic ( isolated ) quantum dynamical systems due to small static perturbations of the evolution operator .
the central object of study is the fidelity @xmath31 of quantum motion computed either with a single ( pure ) initial state or averaged over an ensemble of initial states described by the statistical density matrix .
the main result of the paper is a simple linear response ( or kubo - like ) formula ( [ eq : f2nd ] ) which relates the fidelity decay to the total sum ( or integral ) of two point time autocorrelation function of the generator of perturbation . in the limit of infinite hilbert space dimension we found _ exponential _ fidelity decay on a time - scale @xmath500 , for _ quantum ergodic and mixing _ systems , whereas for _ non - ergodic _
systems we have found much faster decay ( in the sufficiently ` quantum ' regime where @xmath501 ) on a time - scale @xmath502 for random initial states or @xmath503 for coherent initial states ( minimal uncertainty wavepackets ) where the fidelity @xmath31 is given by a _ fourier transformation of the local density of states of the time averaged perturbation operator_. a special emphasis was given to the semiclassical theory of fidelity of small but finite values of @xmath3 , where different regimes and the corresponding time and perturbation scales are carefully discussed , and where fidelity decay may asymptotically ( as @xmath4 ) be evaluated in terms of classical quantities only .
interestingly , finite size fluctuations of fidelity ( for very long times at a finite hilbert space dimension ) have been shown to be given by the _
inverse participation ratio _ of the eigenstates of the perturbed evolution operator in the eigenbasis of the unperturbed propagator .
the surprising aspects of our relations are mainly due to non - interchangability of the limits @xmath5 and @xmath4 as the relevant decay time - scales are only functions of the ratio @xmath504 .
therefore , a different and intuitively expected behaviour , namely faster fidelity decay for mixing than regular dynamics , is obtained in the ` classical ' regime where @xmath505 ( or making the limit @xmath4 prior to @xmath5 ) . a similar
, reassuring result has been found by applying our formalism to inspect the analogous classical fidelity for the unitary perron - frobenius evolution of volume ( area ) preserving maps : there the classical fidelity for regular dynamics has been found to decay on a time - scale @xmath38 , while for a chaotic dynamics , the fidelity decay is governed by a maximal lyapunov exponent @xmath67 on a short time - scale @xmath506 .
we note that our findings should be of primary importance for understanding the problem of stability of quantum computation @xcite where different regimes and time - scales @xcite should now be explained in terms of intrinsic dynamics of a particular quantum algorithm .
actually , our relation between fidelity and correlation decay may prove very useful in design of any new technology which is aimed at manipulating individual quantum states .
on the other hand , our results may also shed new light on the relation between decoherence and dynamics @xcite .
in particular , due to quite special role of coherent initial states on the short ehrenfest time - scale one may expect the results to be quite different for a random initial state and/or for longer time - scales ( see @xcite ) .
we thank t. h. seligman for very fruitful discussions and for many encouraging and useful remarks .
we also acknowledge discussions with g. usaj , s. tomsovic and g. veble . the work has been supported by the ministry of education , science and sport of slovenia .
1 pastawski h m , levstein p r and usaj g 1995 _ quantum dynamical echoes in the spin - diffusion in mesoscopic systems _ phys . rev .
lett.*75 * 43104313 ; levstein p r , usaj g and pastawski h m 1998 _ attenuation of polarization echoes in nuclear magnetic resonance : a study of the emergence of dynamical irreversibility in many - body quantum system _ j. chem .
phys . * 108 * 27182724 .
narnhofer h and thirring w 1989 _ mixing property of quantum systems _ j. stat . phys . * 57 * 811 - 25 ; jona - lasinio g and presilla c 1996 _ chaotic properties of quantum many - body systems in the thermodynamic limit _ phys . rev . lett .
* 77 * 432225 .
feingold m and peres a 1986 _ distribution of matrix - elements of chaotic systems _ phys .
a * 34 * 59195 ; feingold m , leitner d m and piro o 1989 _ semiclassical structure of hamiltonians _ phys . rev .
a * 39 * 650714 ; prosen t and robnik m 1993 _ distribution and fluctuation properties of transition probabilities in a system between integrability and chaos _ j. phys
. a : math .
* 26 * l31926 ; prosen t 1994 _ statistical properties of matrix elements in a hamilton system between integrability and chaos _ ann .
( ny ) * 235 * 11564 song p h and shepelyansky d l 2001 _ quantum computing of quantum chaos and imperfection effects _ phys .
* 86 * 216265 ; georgeot b and shepelyansky d l 2001 _ stable quantum computation of unstable classical chaos _ phys .
86 * 539396 ( also quant - ph/0101004 ) and _ efficient quantum computing insensitive to phase errors _ preprint quant - ph/0102082 | we derive a simple and general relation between the fidelity of quantum motion , characterizing the stability of quantum dynamics with respect to arbitrary static perturbation of the unitary evolution propagator , and the integrated time auto - correlation function of the generator of perturbation .
surprisingly , this relation predicts the slower decay of fidelity the faster decay of correlations is .
in particular , for non - ergodic and non - mixing dynamics , where asymptotic decay of correlations is absent , a qualitatively different and faster decay of fidelity is predicted on a time scale @xmath0 as opposed to mixing dynamics where the fidelity is found to decay exponentially on a time - scale @xmath1 , where @xmath2 is a strength of perturbation . a detailed discussion of a semi - classical regime of small effective values of planck constant @xmath3
is given where classical correlation functions can be used to predict quantum fidelity decay .
note that the correct and intuitively expected classical stability behavior is recovered in the classical limit @xmath4 , as the two limits @xmath5 and @xmath4 do not commute .
in addition we also discuss non - trivial dependence on the number of degrees of freedom .
all the theoretical results are clearly demonstrated numerically on a celebrated example of a quantized kicked top . to appear in ( march 2002 ) |
consider lz transitions in a qubit coupled to a bath of @xmath0 quantum harmonic oscillators , as described by the hamiltonian @xmath1 with the qubit - oscillator coupling @xmath2(b_{j}+b_{j}^{\dag}).\ ] ] the energy difference between the diabatic qubit states changes linearly in time as @xmath3 ( with level - crossing speed @xmath4 and their intrinsic interaction amplitude is @xmath5 .
the @xmath6 are pauli operators .
the first two terms of ( [ e.ham1 ] ) define the standard landau - zener problem for an isolated two - level system .
the @xmath0 harmonic oscillators in ( [ e.ham1 ] ) can have different frequencies @xmath7 , qubit - oscillator couplings @xmath8 , and interaction angles @xmath9 .
the oscillators affect the qubit ( i ) by changing its energies via the diagonal coupling @xmath10 , and ( ii ) by inducing transitions between its levels via the transverse coupling @xmath11 . in order to calculate the probability that the qubit state flips due to the lz sweep
, we will work in an interaction picture and split the hamiltonian ( [ e.ham1 ] ) into an interaction @xmath12 $ ] involving bit flips as described by @xmath13 , and the ( bit - flip-)free hamiltonian @xmath14 .
a polaron transformation @xcite diagonalizes @xmath14 in terms of qubit - state dependent shifted oscillators , giving @xmath14 the form @xmath15 , with creation and annihilation operators for the shifted oscillators @xmath16 with @xmath17 ; the @xmath18 corresponds to the qubit state @xmath19 and the @xmath20 to @xmath21 . with bath oscillators shifting in this way , the reorganization energy @xcite gained by the system has the same value @xmath22 for both qubit states .
eigenstates of the shifted oscillators are labelled as @xmath23 , where the @xmath0 components @xmath24 of the vector @xmath25 are single - oscillator excitation numbers .
the free time - evolution operator @xmath26 can be written as @xmath27 , where the inner product of the vectors @xmath25 and @xmath28 shows up .
we next define the interaction - picture hamiltonian as @xmath29 . in order to bring @xmath30 into a useful form
, we write the oscillator operators @xmath31 in @xmath32 as @xmath33 .
we choose to associate the `` @xmath18''-oscillators with the term @xmath34 of @xmath35 , and the `` @xmath20''-operators with the other term .
we then write @xmath36 , where @xmath37 is the unit vector with @xmath38 component equal to @xmath39 , and likewise @xmath40 . in this form ,
the ` bra s ' of shifted oscillator states in @xmath32 correspond to the ` bra s ' of the qubit , while for the ` kets ' this must still be arranged by using the completeness relation @xmath41 .
the interaction - picture hamiltonian then becomes @xmath42 involving the two infinite - dimensional matrices @xmath43 with @xmath44
we focus on the situation that at time @xmath45 the system starts in its ground state @xmath46 .
we are now interested in the survival probability @xmath47 of the initial state @xmath19 of the qubit .
@xmath47 equals the square of the norm of the projected final oscillator state @xmath48 , which can be written as @xmath49 , where @xmath50 $ ] is the time evolution operator in the interaction picture . in a time - ordered expansion of @xmath51
, only the even powers of @xmath52 will contribute to @xmath47 .
as is well known , the @xmath53-order term in the expansion involves a @xmath54-fold time integral with variables @xmath55 in the interval ( @xmath56 ) .
it is advantageous to make the variable transformation @xcite @xmath57 for @xmath58 , @xmath59 , and @xmath60 .
we label the @xmath0-oscillator state after @xmath61 interactions as @xmath62 . for brevity
, we define the frequencies @xmath63 and @xmath64 .
the perturbation series for @xmath48 then becomes @xmath65 |{{\bf n}^{(2k)}_{+}}\rangle .
\end{aligned}\ ] ] it would be convenient to symmetrize the @xmath66-integrals at this point , but in general the integrand is not symmetric under permutation of the @xmath66 .
when transforming the @xmath66 to new variables @xmath67 and @xmath68 for @xmath69 , one finds that the @xmath70-integral yields the delta - function @xmath71 .
now , since the initial state is @xmath72 , the sum over the @xmath73 in this delta function equals @xmath74 , which evidently is @xmath75 .
likewise , the variables @xmath76 are all positive or zero .
therefore , the delta - function can only `` click '' in the subspace @xmath77 and will do so only if the vector @xmath78 vanishes .
the physical meaning of the latter statement is discussed below . performing the other @xmath79-integrals
is cumbersome , so we quickly return to the integrals in eq .
( [ e.integralsxy ] ) , but this time armed with the knowledge that only the subspace @xmath80 contributes .
since within this subspace the integrand _ is _ symmetric in the variables @xmath66 , it is correct to symmetrize the @xmath66-integrals in eq .
( [ e.integralsxy ] ) , i.e. we can replace them by @xmath81 .
after performing these standard integrals , the @xmath76-integrals can be evaluated as well .
for example , the @xmath82-integral @xmath83 vanishes unless @xmath84 vanishes , in which case it is indeed the `` subspace '' defined by @xmath85 that makes this integral equal to @xmath86 . from the time
integrals we find the following _ selection rule _ : when starting in the ground state @xmath72 , the only @xmath53-order processes contributing to the survival probability @xmath87 are those with @xmath88 .
hence the oscillators will end up in their initial state @xmath89 in case the qubit ends up in @xmath19 .
this striking result agrees with the so - called no - go theorem @xcite , which we extended here to spin - boson problems .
the time integrals do not forbid occupation of the states @xmath90 at intermediate times , nor do they restrict the intermediate oscillator states @xmath91 , but further restrictions may originate from vanishing matrix elements @xmath92 , see eq .
( [ wpm ] ) . only when the qubit ends up in @xmath21 can the lz driving dissipate energy into the bath .
hence qubit and bath end up entangled . in line with the selection rule
, we find that @xmath93 simplifies into @xmath94 , where the parameter @xmath95 is still to be determined by using eq .
( [ wpm ] ) .
we finally obtain our central result : the exact landau - zener transition probability for a qubit arbitrarily coupled to an oscillator bath at @xmath96 is @xmath97 where the parameter @xmath98 is given by @xmath99 by introducing the spectral density @xmath100 and assuming that oscillators with equal frequencies @xmath101 have equal coupling angles @xmath102 , eq .
( [ w2final ] ) becomes @xmath103\cos[\theta(\omega)]j(\omega)/\omega \big)^2 \nonumber \\ & & + \frac{\hbar^{2}}{4\pi}\int_{0}^{\infty}\mbox{d}\omega\,\sin^{2}[\theta(\omega)]j(\omega).\end{aligned}\ ] ] no specific form of @xmath104 or @xmath102 is presupposed , but for the limit @xmath105 describing a continuum of oscillators , the two integrals in eq .
( [ w2finalcontinuum ] ) have to be finite .
we first focus on purely transverse bath coupling and obtain @xmath107 , where @xmath108 is the integrated spectral density .
the @xmath38 oscillator reduces the final survival probability by a factor @xmath109 $ ] , independent of the other oscillators or of the value of @xmath110 . for @xmath111 ,
the bath ends in @xmath89 when the qubit ends in @xmath19 , whereas the bath contains an odd number of bosons when the qubit ends in @xmath19 .
qubit and bath therefore end up fully entangled , in the sense that the qubit coherence vanishes after tracing out the bath .
nonzero final qubit coherence requires @xmath112 .
notice also the interesting phenomenon of _ bath - assisted adiabatic following _ : for large bath coupling , i.e. @xmath113 , the qubit ends up in the initially unpopulated state , even in the absence of an intrinsic interaction @xmath5 .
in solid - state environments , decoherence often occurs much faster than the bath - induced relaxation so that the latter is neglected . hence the `` standard model '' for lz transitions in dissipative environments
@xcite is the single qubit diagonally coupled to an oscillator bath , which is obtained from our hamiltonian ( [ e.ham1 ] ) by setting all @xmath114 . by doing the same in eq .
( [ w2final ] ) , we find that for a qubit diagonally coupled to a bath at @xmath96 , the landau - zener transition probability @xmath115 $ ] , which is the well - known result for an _ isolated _ qubit ! thus , although the bath coupling does not commute with the qubit hamiltonian , there is no bath dependence in the transition probability , neither via qubit - bath couplings @xmath8 nor via the oscillator frequencies @xmath110 , no matter how fast the lz sweep is performed .
we discuss this below .
next assume that all oscillators couple to the qubit with the same angle @xmath116 .
this may follow from microscopic considerations or may be engineered . for example , when all oscillators couple diagonally to the qubit in a basis described by pauli matrices @xmath117 , then an experiment where the qubit is driven via @xmath118 while @xmath119 is kept constant , is described by our hamiltonian ( [ e.ham1 ] ) with all angles @xmath116 .
the reorganization energy then becomes @xmath120 , where @xmath121 occurs for diagonal coupling . from eq .
( [ w2final ] ) we deduce @xmath122 unlike for transverse coupling , the oscillators do not affect @xmath123 independently .
relaxation @xmath124 , dephasing @xmath125 , and intrinsic interaction act together . in the continuum limit @xmath126 , eq .
( [ w2finaltheta ] ) holds for spectral densities @xmath104 that give rise to finite reorganization energies @xmath127 and finite integrated spectral densities @xmath128 . as a main application of our calculations
, we propose to gauge the dissipative environment of a qubit via lz transitions . by gauging we
mean in short : the measurement of @xmath127 and @xmath128 and the subsequent parameter fixing of appropriate model spectral densities .
more in detail , we propose to perform lz sweeps and to determine transition probabilities for several fixed values of the tunable intrinsic interaction @xmath5 .
notice that @xmath129 as a function of @xmath5 is a parabola in eq .
( [ w2finaltheta ] ) .
figure [ fig : gauging ] as a function of intrinsic interaction @xmath5 , for several values of the coupling angle @xmath130 .
parameters : @xmath131 and @xmath132 . ]
depicts corresponding final transition probabilities for several coupling angles .
diagonal ( @xmath133 ) and transverse ( @xmath106 ) coupling are limiting cases which have in common that @xmath129 and hence @xmath134 assume their minima for @xmath135 . neither limiting case is suitable for gauging @xmath127 . since for intermediate cases @xmath136 we find a minimal transition probability for a _ nonzero _ internal interaction @xmath137 , one can determine the reorganization energy @xmath127 by measuring @xmath138 .
moreover , the integrated spectral density @xmath128 can then be identified as @xmath139 . as a consistency test
, one can check whether @xmath140 holds .
consistency can be further tested by changing the basis in which the lz sweep is performed : the values of @xmath127 and @xmath128 should come out independent of the angle @xmath130 . if not , then @xmath102 is not constant and eq .
( [ w2finalcontinuum ] ) applies . with the values of @xmath127 and @xmath128 thus determined
, one can fix parameters in appropriate model spectral densities . for example , suppose @xmath104 has the known form @xmath141 with a power @xmath142 given by the physical nature of the environment @xcite , but with unkown strength @xmath143 and cutoff frequency @xmath144 .
( @xmath145 is the ohmic case and for @xmath146 , @xmath127 would diverge . )
then @xmath147 , in terms of the gamma function , and @xmath148 .
hence lz gauging fixes @xmath149 and @xmath143
. other spectral densities may depend on more parameters .
lz gauging fixes two of them .
the efficient use of our gauging scenario requires the following : for @xmath134 to change considerably , one chooses @xmath150 such that @xmath151 and varies @xmath5 on the scale of @xmath152 . then @xmath127 can be measured accurately if @xmath153 is not much smaller than @xmath154 , or @xmath155 not much smaller than unity for an ohmic bath .
so @xmath134 is robust under dephasing , even if dephasing is much faster than relaxation ( @xmath156 ) , unless the qubit - bath coupling is strong .
this robustness has found its use in experiments on molecular nanomagnets @xcite .
we presented in eqs .
( [ transitionsexact ] ) and ( [ w2final ] ) exact landau - zener transition probabilities for a qubit with arbitrary linear coupling to a zero - temperature bath .
we found that qubit and bath end up entangled .
our results apply to experiments where the initial and final qubit energies are off - resonant with relevant bath frequencies , for example in circuit qed @xcite , where qubit energies can be varied over a broad range and spectral densities are peaked @xcite .
indeed , our predictions for transverse coupling generalize our detailed study of lz sweeps in circuit qed @xcite to more realistic situations where peaked spectral densities have nonzero widths .
other applications include tunable atoms in optical cavities and in photonic crystals . for diagonal coupling
we find that the transition probability does not depend on the bath at all .
our exact result settles a long - standing discussion @xcite , at least for zero temperature .
it corroborates and interpolates between what ao and rammer @xcite found for the fast - passage limit @xmath157 and for the opposite adiabatic limit @xmath157 , which both have been confirmed numerically @xcite .
for fast lz sweeps , the bath clearly has no time to affect the transition , but the absence of any bath influence also for slower sweeps is a highly nontrivial property of this standard model . a qubit undergoing a lz sweep measures ` global ' properties of the bath , namely the frequency - integrated spectral density , and for strong coupling the reorganization energy .
sample - dependent spikes in spectral densities are therefore averaged out , even in a single sweep , so that model parameters can be determined .
we therefore propose to employ lz transitions for a valuable gauging of the dissipative environment of a tunable qubit . | we calculate the _ exact _ landau - zener transitions probabilities for a qubit with arbitrary linear coupling to a bath at zero temperature . the final quantum state exhibits a peculiar entanglement between the qubit and the bath . in the special case of a diagonal coupling , the bath does not influence the transition probability , whatever the speed of the landau - zener sweep . it is proposed to use landau - zener transitions to determine both the reorganization energy and the integrated spectral density of the bath .
possible applications include circuit qed and molecular nanomagnets .
quite a number of quantum two - state systems are presently tested as candidate qubits , the units of quantum information .
good qubits are well isolated from their environment , but easy to manipulate .
this somewhat conflicting requirement has spurred renewed interest in the dynamics of qubits coupled to an environment or heat bath @xcite .
qubits can be seen as bath detectors .
for example , static qubits probe via their decay rates the bath spectral density at their transition frequencies . in solid - state environments
these rates can be strongly frequency- and sample - dependent .
we discuss an in this respect superior ` bath detection mode ' of the qubit .
one way of changing the state of a two - level system involves the forced crossing of its diabatic energies . for constant level - crossing speed
this is known as the landau - zener ( lz ) problem @xcite , which for a two - level system can be solved exactly @xcite .
this is no longer the case when taking its environment into account @xcite that may cause thermal excitation and quantum tunneling . in the low - temperature tunneling regime , analytical estimates for transition probabilities exist only for very fast and very slow sweeps , and the literature is not unanimous about the latter limit @xcite . in another line of research , incited by the paper @xcite
, it was recently proven that lz transition probabilities can be calculated exactly for some many - level systems as well , although only for some initial states @xcite . in this letter
, we extend the analysis to quantum dissipative systems and study lz transitions in spin - boson problems in a new way .
first we calculate zero - temperature lz transition probabilities exactly .
second , by making use of this exact dependence on the bath parameters , we propose to gauge the dissipative environment of a qubit by performing lz sweeps .
one advantage of this bath detection mode of the qubit is that effects of spiky bath spectral densities are averaged out in every sweep . |
the research on the striking universality properties of slow relaxation dynamics has been driving great efforts in the last decades .
a large and growing literature can be found in which non - exponential behaviour has been observed for correlation functions : supercooled colloidal systems @xcite , glasses and granular material @xcite , liquid crystal polymers @xcite , hydrated proteins @xcite , growth @xcite , plasmas @xcite and disordered vortex lattices in superconductors @xcite are just a few examples .
those systems present features which are similar to those found in systems with anomalous diffusion .
the attempt to obtain response functions which are able to explain such relaxation processes is a subject more than a hundred years old .
rudolph kohlrausch used stretched exponentials @xmath0 $ ] with @xmath1 to describe charge relaxation in a leyden gas @xcite .
later on , his son , friedrich kohlrausch @xcite observed two distinct universalities : the stretched exponential with @xmath2 , and the power law behaviour . the former behaviour is now known as the kohlrausch - williams - watts ( kww ) stretched exponential .
there are now two main methods to describe those relaxations : a fractional derivatives approach @xcite , and a recurrence method applied to the mori equation @xcite .
however , new methods appear every day and we should mention here the attempt of chamberlin @xcite to explain this universal behaviour , the diffusive method of perez - madrid @xcite , and the new method recently proposed by schwartz and edwards @xcite for non - linear fields . in this work
, we present an alternative way to obtain the correlation function and a result which is general as long as the diffusive processes can be described by a generalized langevin equation ( gle ) .
we derive a correlation function which is an even function of time .
the asymptotic behaviour is similar to those obtained using fractional derivatives .
however , for short times , our method can lead to richer behaviours .
the paper is organized as follows : first , we give the outline of the problem of a system governed by a gle and show the connection to diffusion ; we then continue by defining in a clear manner the noise , memory and correlation functions .
we construct the memory function from the fluctuation - dissipation theorem using a coloured noise , from a given density of states .
after that , we discuss normal and anomalous diffusions . finally , we analyze the difference between markovian and non - markovian behaviours .
we shall start by writing the gle for an operator @xmath3 @xmath4 where @xmath5 is a stochastic noise subject to the conditions @xmath6 , @xmath7 and @xmath8 here , the angular brackets denote an ensemble average .
some correlation functions depend on two times , being consequently non - stationary @xcite .
equation ( [ fdt ] ) is the kubo fluctuation - dissipation theorem ( fdt ) @xcite .
the fdt can be violated in many slow relaxation processes @xcite . before continuing ,
it is important to stress the connection between the generalized langevin equation for a given operator @xmath3 and diffusion . to do this
, we define the variable @xmath9 now , we can study the asymptotic behaviour of its second moment @xmath10 to characterize the type of diffusion presented by the system : for @xmath11 , we have normal diffusion , and for @xmath12 , subdiffusion ( superdiffusion ) .
the memory kernel @xmath13 indicates that the motion is non - markovian ; when @xmath14 , eq .
( [ gle ] ) becomes the usual langevin equation .
our main interest is to discuss the behaviour of the correlation function@xmath15 from which we can describe most of the processes of interest , including relaxation .
we use the above conditions for the noise to obtain a self - consistent equation for @xmath16 @xmath17 if we then apply the laplace transform ( denoted by a tilde ) to eq .
( [ self_consistent ] ) , we get @xmath18 from the analysis of this equation , it is possible to obtain plenty of information concerning the asymptotic behaviour of the system . in order to make progress ,
we need to make some assumptions about the origin of the memory . the direct way is to connect the random force , @xmath5 , to a thermal bath composed of harmonic oscillators @xcite .
therefore , for a system in contact with a heat reservoir ( canonical ) , the memory becomes @xmath19 where @xmath20 is the noise density of states .
the memory is clearly even for any noise distribution .
we shall use a coloured noise given by a generalization of the debye spectrum@xmath21 with @xmath22 as a debye cutoff frequency .
the motivation for considering such cases arises from previous studies @xcite in which it was proved that if @xmath23 as @xmath24 , then the diffusion exponent is given by @xmath25
in order to obtain @xmath16 , we state that its derivative must vanish at the origin , due to eq .
( [ self_consistent ] ) ; therefore , the correlation function can be neither an exponential nor a stretched exponential .
the analytical continuation of the laplace transform of an even function is an odd function , and vice - versa . from eq .
( [ laplace_r ] ) , we see that @xmath26 is odd , because @xmath27 is even ( see eq .
( [ memory ] ) ) . following the same line of thought
, we arrive at the conclusion that @xmath16 is even .
lee @xcite has also shown that both the memory and the correlation function must be even functions for any hamiltonian system .
consequently , we can write @xmath28 with @xmath29 and , @xmath30 we replace those , eqs .
( [ memory ] ) and ( [ noise_dos ] ) in eq .
( [ self_consistent ] ) to obtain the following recurrence relation @xmath31!\,\omega _ { s}^{2l}}{(2l+1+\nu ) } a_{n-1-l},\ ] ] which displays a complex behaviour , where every order depends on all previous ones .
this is not a surprise for a non - markovian system and it is very useful to have it made explicit in a power series .
the power series defined by eq .
( [ r2 ] ) is convergent : it is straightforward to prove by induction that @xmath32 . besides that , it is easy to show that the absolute values of @xmath33 decrease for @xmath34 , for a certain @xmath35 . consequently , since eq .
( [ r2 ] ) is an alternating series , and @xmath36 , it is possible to conclude that it converges @xcite for every value of @xmath37 . by playing with the parameters @xmath38 , @xmath22 , and @xmath39 , it is possible to generate a large number of functions .
before we analyze anomalous diffusion , it is very useful to give a more detailed description of normal diffusion . if one takes eq .
( [ noise_dos ] ) with @xmath40 and @xmath22 finite , diffusion is normal , despite the noise being coloured . using this noise in eq .
( [ memory ] ) , we get @xmath41 . applying the laplace
transform yields @xmath42 .
it is interesting to note that this memory function is realized in electron gas models @xcite .
we now show that it is possible to find at least three time scales for this memory .
the first one is the short time scale , given by @xmath43 . in the short time scale , given by @xmath44 , the first term in the expansion , eq .
( [ r2 ] ) , gives @xmath45 where @xmath46 .
for this particular case , @xmath47 introduces the intermediate time scale @xmath48 . in the long time scale , @xmath49 , or , equivalently , @xmath50 , then @xmath51 . therefore , from eq .
( [ laplace_r ] ) , we obtain @xmath52 , with @xmath53 . this result is the same as the one obtained from the normal langevin equation , which can be obtained from the limiting case @xmath54 . for normal diffusion ( @xmath55 ) . here , @xmath56 and @xmath57 and @xmath58 , for curves @xmath59 and @xmath60 , respectively .
curve @xmath61 is the plot of @xmath62 .
note that curves @xmath60 and @xmath61 are indistinguishable , except near the origin ( see inset ) .
inset : plot of curves @xmath60 , @xmath61 and @xmath63 , near the origin.,width=245 ] in fig . [ fig1 ] we plot the correlation function @xmath64 as a function of @xmath65 for normal diffusion .
we use @xmath66 , and @xmath67 and @xmath58 in curves @xmath68 , and @xmath69 , respectively .
curve @xmath70 is the plot of the function @xmath62 .
it is clear that curve @xmath68 is not an exponential ; however , in many practical situations it is fine to approximate curve @xmath69 by an exponential for times larger than @xmath71 . for short times , @xmath72 , eq .
( [ r3 ] ) is the solution and we disregard the exponential as the solution in both cases .
the inset shows the correlation function @xmath16 near the origin for @xmath73 .
together with curve @xmath69 and @xmath70 , we plot @xmath74 . in short ,
`` good exponentials '' , such as curve @xmath70 can not exist as a solution in this case .
in many experimental situations , the ratio @xmath75 can be very large .
nevertheless , we propose that there will always be a short time behaviour which permits distinction from an exponential .
even if the distinction is not accessible to the experimental probe , it is very important to bear in mind that the correlation function is even .
for @xmath76 and @xmath77 , @xmath78 , and @xmath79 from top to bottom .
notice the transition from a stretched exponential behaviour to a power law.,width=245 ] now let us return to the general memory . by substituting eq .
( [ noise_dos ] ) into eq .
( [ memory ] ) with @xmath80 and calculating the laplace transform , we get @xmath81 here , @xmath82 where @xmath83 is the lerch transcendent @xcite . this function has been found most useful in certain quantum statistics @xcite .
this memory function has an asymptotic behaviour of the following form @xmath84 where @xmath85 this cutoff shows that the exponent @xmath86 , eq .
( [ diff_exponent ] ) , can not be larger than @xmath87 ( ballistic motion ) .
recent results @xcite confirm that cutoff ; they obtain @xmath88 when @xmath89 . by inverse transforming eq .
( [ laplace_r ] ) , using eq .
( [ limga ] ) , we get @xmath90 which is a useful approximation for @xmath91 . here
@xmath92 is the mittag - leffler function @xcite .
this function behaves as a stretched exponential for short times and as an inverse power law in the long time regime ( see fig .
[ fig2 ] ) . due to the asymptotic power - law behaviour @xmath93^{-1},\ ] ] for @xmath94
, we obtain an infinite diffusion constant @xmath95 , which is a sign of superdiffusion @xcite .
the power law exponent @xmath96 is in the interval @xmath97 as observed long ago by friedrich kohlrausch . for @xmath98 ,
the mittag - leffler function is the usual exponential and @xmath99 is the relaxation time .
the limiting cases correspond to @xmath88 and @xmath100 ; for these values , we have @xmath101 and @xmath102 , respectively .
this last case corresponds to the limit of subdiffusion , _
i.e. _ , @xmath103 .
since the argument of the function is negative , the mittag - leffler function is in this case a cosine . for short times , @xmath104 the first term in the expansion gives a result identical to eq .
( [ r3 ] ) , with @xmath105 . in fig .
[ fig2 ] , we display the correlation function @xmath16 as a function of t for anomalous diffusion , obtained from the numerical integration of eq .
( [ self_consistent ] ) .
we select broadband noise with @xmath76 . from top to bottom , @xmath106 and @xmath79 .
we notice the transition from stretched exponential to power law behaviour .
note as well that as the behaviour becomes more anomalous , _
i.e. _ as @xmath38 increases , the relaxation becomes slower .
recently some experimental situations have been found where non - markovian processes display markovian behaviour after a long time . in particular , we call attention to the experiments of merikoski _ et al . _ about flame evolution @xcite .
we show now that this transition is possible for normal diffusion . for large @xmath65 , eq .
( [ self_consistent ] ) can be written as @xmath107 we can justify this approximation by the following arguments : if @xmath108 , then @xmath109 and for those values @xmath110 gives the major contribution .
as @xmath111 increases , say @xmath112 , both @xmath113 and @xmath114 are small . finally , for @xmath115 , @xmath116 , but @xmath110 is very small .
notice that this decoupling happens in many situations in physics , otherwise all correlated systems would present aging . for normal diffusion , the convergence of the integral of the memory function to a constant value @xmath117 makes the decoupling acceptable . in that case , for broadband noise @xmath118 , as expected ( see fig .
[ fig1 ] ) . consequently ,
as long as the diffusion is normal , @xmath119 is finite independently of the character of @xmath27 ; after a time @xmath120 all the processes behave as those governed by the usual langevin equation . for anomalous diffusion , the dependence on time of the integral of the memory in eq .
( [ approx_corr ] ) makes the decoupling fail . in short , the exponential behavior in normal diffusion
is associated with time invariance while the non - exponential behaviour is associated with an aging process which is in essence non - markovian .
we discussed the relaxation properties of a dynamical variable @xmath121 and we proved that neither exponentials nor stretched exponentials can give a full description of the relaxation process .
however , these functions can be the asymptotic behaviour found in many experimental and theoretical works and we found an even function that can approach an exponential as we broaden the noise ( increase the ratio @xmath122 ) . for broad noise , the correlation function for anomalous diffusion will be a stretched exponential and finally a power - law .
fractional derivatives have been successfully applied to the study of subdiffusive behaviour @xcite .
those yield @xmath123 , for @xmath124 .
the gle formalism allows @xmath38 to assume the values in the range @xcite @xmath125 , which encompass both subdiffusion and superdiffusion .
there is a great advantage in our formulation , since the noise density of states exists not only in systems governed by a gle , but in most physical systems .
also , we avoid the levy function which has infinite mean square deviation .
as we have already mentioned , glassy systems seem to be a rich field for studying these phenomena
. it would be very helpful if the relaxation process could be studied and exponent @xmath126 determined for those diffusive processes .
other related phenomena are chaos synchronization @xcite and anomalous reaction rates@xcite , which we expect to discuss soon . | we study the relaxation process in normal and anomalous diffusion regimes for systems described by a generalized langevin equation ( gle ) .
we demonstrate the existence of a very general correlation function which describes the relaxation phenomena .
such function is even ; therefore , it can not be an exponential or a stretched exponential .
however , for a proper choice of the parameters , those functions can be reproduced within certain intervals with good precision .
we also show the passage from the non - markovian to the markovian behaviour in the normal diffusion regime . for times longer than the relaxation time ,
the correlation function for anomalous diffusion becomes a power law for broadband noise . |
one of the outstanding issues in @xmath35 meson physics is the semileptonic branching fraction puzzle .
experimentally @xmath36 is measured to be ( @xmath37)% @xcite , whereas theoretical calculations have difficulties accommodating a branching fraction below @xmath38 @xcite .
one way to reduce the theoretical expectations is through a two - fold enhancement in the assumed @xmath39 rate @xcite , which is estimated to be @xmath40 from the measured inclusive rates for @xmath41 and @xmath42 .
recently , buchalla _
et al._@xcite and blok _
et al._@xcite have suggested that a significant fraction of the @xmath39 transition hadronizes into @xmath43 .
this is supported by cleo s
@xcite observation of `` wrong - sign '' @xmath44 mesons from @xmath35 decays , @xmath45 , where the @xmath44 comes from the virtual @xmath46 .
the aleph @xcite and delphi @xcite collaborations have also observed sizeable @xmath47 decay rates .
exclusive @xmath35 decays involving wrong - sign @xmath44 mesons can result from ( 1 ) resonant @xmath48 decays , where the @xmath46 hadronizes to an excited @xmath5 meson that decays into @xmath49 ; and ( 2 ) non - resonant @xmath50 decays .
this paper explores one possibility in the first case , namely , the decays @xmath51 where @xmath52 is the narrow p - wave @xmath5 meson with @xmath53 .
the `` upper - vertex '' production of @xmath52 from @xmath46 hadronization is shown in figure [ fig : feynman](a ) .
in addition , @xmath52 mesons can be produced from `` lower - vertex '' decays @xmath54 with the creation of an @xmath55 quark pair , as shown in figure [ fig : feynman](b ) .
this produces right - sign @xmath44 mesons ; however , the decay rate is expected to be small . throughout this paper charge
conjugate states are implied .
continuum @xmath52 production has been thoroughly studied @xcite .
the @xmath52 is just above the @xmath56 mass threshold and decays dominantly into @xmath57 and @xmath58 .
other possible decay channels are negligible : @xmath59 due to isospin conservation , @xmath60 due to ozi suppression @xcite , @xmath61 or @xmath62 due to angular momentum and parity conservation , and @xmath63 due to the small radiative decay rate .
the data used in this analysis were selected from hadronic events collected by the cleo ii detector at the cornell electron storage ring ( cesr ) .
the cleo ii detector @xcite is a large solenoidal detector with 67 tracking layers and a csi electromagnetic calorimeter that provides efficient @xmath64 reconstruction .
the data consist of an integrated luminosity of 3.11 fb@xmath65 at the @xmath66 resonance , corresponding to @xmath67 @xmath68 events . to evaluate non-@xmath68 backgrounds we also collected 1.61 fb@xmath65 of `` continuum '' data 60 mev below the @xmath66 resonance .
the inclusive @xmath69 decay is studied by reconstructing the decay channels @xmath70 and @xmath71 using the decay modes @xmath72 and @xmath73 .
the @xmath74 is reconstructed using the decay modes @xmath75 and @xmath76 .
hadronic events are required to satisfy the ratio of fox - wolfram moments @xcite @xmath77 to reduce the background from continuum events .
charged tracks , except pions from @xmath78 decays , are required to be consistent with coming from the primary interaction point . charged kaon and pion candidates are identified using specific ionization ( @xmath79 ) and , when available , time - of - flight ( tof ) information . for kaon identification , we consider the relative probability for a charged track to be a kaon , @xmath80 , where @xmath81 is the @xmath82 probability for a given particle hypothesis .
the requirement on @xmath83 depends on the decay mode of interest .
pion candidates are identified by requiring the @xmath79 and , when available , tof information to be within 3 standard deviations ( @xmath84 ) of that expected for pions .
we select @xmath78 candidates through the decay to @xmath85 by requiring a decay vertex displaced from the primary interaction point and a @xmath78 invariant mass within 10 mev / c@xmath86 of its nominal value .
we reconstruct @xmath64 candidates through the decay to @xmath87 by requiring candidates to have an invariant mass within 2.5 standard deviations ( @xmath88 mev / c@xmath86 ) of the nominal @xmath64 mass .
the @xmath89 and @xmath76 combinations are required to have a kaon identification of @xmath90 and @xmath91 , respectively , and an invariant mass within 15 and 25 mev / c@xmath86 ( @xmath92 ) of the nominal @xmath74 mass , respectively .
in addition , we select regions of the @xmath93 dalitz plot to take advantage of the known resonant substructure @xcite . for the @xmath70 mode ,
the dalitz cut reduces the signal efficiency by 40% and the background by 80% .
we relax the dalitz cut for the @xmath71 mode since the combinatoric background is substantially lower .
the @xmath73 candidates are required to have a mass difference @xmath94 within 1.5 mev / c@xmath86 ( @xmath92 ) of the nominal value of 145.4 mev / c@xmath86 , where @xmath95 is the reconstructed invariant mass of @xmath96 .
similarly , the @xmath72 candidates are required to have a mass difference @xmath97 within 1.5 mev / c@xmath86 ( @xmath92 ) of the nominal value of 142.1 mev / c@xmath86 . to form @xmath52 candidates charged kaons
are combined with @xmath98 candidates and @xmath78 s are combined with @xmath99 candidates .
since the primary kaons from @xmath70 decays have low momentum , we can impose a stringent @xmath100 requirement on the @xmath101 with negligible loss of efficiency .
the @xmath52 candidates are required to have a scaled momentum @xmath102 , which is the kinematic limit for @xmath69 decays .
( we ignore the negligible contributions from @xmath103 decays . )
upper - vertex @xmath52 production results in a maximum @xmath104 of 0.35 , and this requirement is imposed when determining the @xmath52 decay constant . the @xmath52 decay channels with @xmath64 s in the final state often have multiple @xmath52 candidates per event .
we select the candidate with the highest @xmath82 probability of being a @xmath52 , which is derived from the invariant masses of the reconstructed @xmath64 , @xmath74 and @xmath105 mesons .
the @xmath52 signal is identified using the @xmath56 mass difference , @xmath106 and @xmath107 , where @xmath108 and @xmath109 are the known masses @xcite .
the @xmath56 mass difference signal has a resolution that is two to four times smaller than the corresponding signal in the reconstructed @xmath56 invariant mass distribution .
the @xmath110 and @xmath111 distributions are shown in figure [ fig : mass diff ] , where the @xmath75 and @xmath76 modes have been added together .
the data is fit with a gaussian signal and a threshold background function .
the gaussian width is fixed to that expected from a geant - based monte carlo simulation @xcite ( @xmath112 mev / c@xmath86 , depending on the mode ) and the mean is fixed to the measured @xmath52 mass difference from continuum data ( @xmath113 mev / c@xmath86 and @xmath114 mev / c@xmath86 . )
we observe @xmath115 signal events in the @xmath57 mode and @xmath116 events in the @xmath71 mode .
however , when the @xmath57 candidates are further subdivided into the @xmath75 and @xmath76 decay channels there is a discrepancy in the @xmath52 yields . as shown in figure [ fig:2 d0 modes ] , we observe @xmath117 signal events in the @xmath110 distribution for the @xmath75 channel and @xmath118 @xmath52 signal events for the @xmath93 channel . after accounting for branching fractions and efficiencies , discussed below , this results in a @xmath119 discrepancy in the @xmath57 rates between the two @xmath74 modes .
we can not rule out the fact that background sources may be contributing a false @xmath52 signal in the @xmath93 channel , but not in the @xmath75 channel .
however , no such mechanism has been uncovered . to be conservative
, we choose to quote only an upper limit for the decay @xmath69 .
since the @xmath52 reconstruction efficiency increases rapidly with @xmath104 and the @xmath52 momentum distribution from @xmath35 decays is not known , we compute the inclusive @xmath69 branching fraction by dividing the data into four equal regions of @xmath104 from 0.05 to 0.45 and summing the efficiency corrected yields .
the @xmath70 and @xmath58 branching fractions are equal according to isospin , and their ratio has been measured to be within 30% of unity @xcite .
we measure the branching fraction @xmath69 to be @xmath120 from the @xmath57 mode and @xmath121 from the @xmath71 mode , where the error is statistical only .
the two measurements are statistically consistent . the @xmath104 distribution for our @xmath52 candidates is shown in figure [ fig : xp dist ] .
several cross - checks , shown in figure [ fig : cross checks ] , were performed to corroborate the validity of the @xmath52 signal . the scaled continuum background from data after satisfying all selection cuts is negligible , and there is no excess in the @xmath110 signal region ( @xmath122 events ) . the uncertainty in the continuum @xmath52 contribution is included in the systematic error .
there is also no evidence of peaking in the @xmath110 signal region for wrong - sign @xmath123 combinations ( @xmath124 events ) , @xmath74 mass sidebands ( @xmath125 events ) , and @xmath98 mass sidebands ( @xmath126 events ) .
we have also searched for the @xmath74 signal from @xmath127 candidates in the @xmath110 signal region , @xmath128 mev / c@xmath129 mev / c@xmath86 , by relaxing the @xmath74 mass cut and histogramming the invariant mass of all @xmath89 and @xmath76 combinations that satisfy the remaining selection criteria . in events with multiple candidates per @xmath74 decay mode
we select the candidate with the highest @xmath82 probability , which is derived from the reconstructed @xmath64 and @xmath52 masses . we observe @xmath130 @xmath74 events . however , there are also real @xmath74 s in the random @xmath57 combinations under the @xmath52 peak ; after a @xmath110 sideband subtraction the @xmath74 invariant mass spectrum yields @xmath131 events ( see figure [ fig : d0 yield](a ) ) .
this is consistent with our @xmath70 yield in figure [ fig : mass diff ] .
similarly , we have studied the @xmath98 signal from @xmath127 candidates in the @xmath110 signal region .
we observe @xmath132 @xmath74 events . as in the @xmath74 case
there are also real @xmath98 s in the random @xmath57 combinations under the @xmath52 peak . after a @xmath110 sideband subtraction the @xmath98 mass difference spectrum yields @xmath133 events ( see figure [ fig : d0 yield](b ) ) , consistent with our @xmath70 yield .
finally , we have studied the @xmath52 production from continuum @xmath134 events .
the selection criteria is similar to that used to find @xmath52 from @xmath35 decays , but since continuum charm production has a hard fragmentation , we require @xmath135 .
in addition , we remove the @xmath136 cut , relax the charged kaon identification to @xmath137 , and remove the dalitz cut for @xmath93 .
the mass difference distribution for @xmath57 and @xmath71 combinations are shown in figure [ fig : continuum ] , where the @xmath75 and @xmath76 modes have been added together .
we extract the @xmath52 signal by fitting the data with a gaussian signal and a threshold background function .
the gaussian width is fixed to the value predicted by monte carlo ( 2.1 mev / c@xmath86 ) , and the mean is allowed to float .
we observe @xmath138 events in the @xmath70 mode with a mass difference of @xmath139 mev / c@xmath86 ( statistical error only ) , and @xmath140 events in the @xmath141 mode with a mass difference of @xmath142 mev / c@xmath86 .
the results are consistent with the previous cleo analysis @xcite .
there are several sources of systematic error .
we assign a systematic error of 16% to account for the @xmath119 discrepancy between the @xmath57 rates for the @xmath75 and @xmath76 modes .
this accomodates different methods of computing the weighted average of the @xmath69 branching fraction from the four separate decay chains .
uncertainties due to reconstruction efficiencies include 1.5% per charged track , 5% per @xmath64 , 5% for slow pions from @xmath105 , and 5% for @xmath78 .
we also include systematic errors of 7% for monte carlo statistics , 5% for kaon identification and the dalitz decay cut efficiency , 4% for uncertainties in the yield for @xmath143 , and 8% for uncertainties in the continuum @xmath52 contribution that passes our selection criteria .
the total systematic error is 24% . averaging the @xmath57 and @xmath71 modes together
, we obtain @xmath144 . since the @xmath52 signal is observed largely in only one decay mode @xmath70 with @xmath93 , and since there is a discrepancy between this mode and the corresponding mode involving @xmath75 , we instead prefer to quote an upper limit on the branching fraction to be @xmath145 at the 90% c.l .
@xcite this decay rate limit is small relative to the total rate expected for @xmath2 of about @xmath146 from the wrong - sign @xmath44 meson yield in @xmath35 decays @xcite .
this is not surprising considering the @xmath147 system has appreciable phase space beyond the @xmath52 mass @xcite .
also , cleo s @xcite recent observation of exclusive @xmath50 decays shows that the @xmath148 invariant mass distribution lies mostly above the @xmath52 mass .
measurement of the @xmath69 decay rate also provides an estimate of the @xmath52 decay constant , @xmath149 , assuming that the @xmath52 comes dominantly from upper - vertex decays .
the inclusive decay rate for @xmath35 mesons into ground state or excited @xmath5 mesons can be calculated assuming factorization @xcite , @xmath150 where @xmath151 is the bsw @xcite parameter for the effective charged current , and @xmath152 is a kinematic factor with @xmath153 and @xmath154 . for scalar or pseudoscalar @xmath155 mesons , @xmath156 , and for vector or axial - vector @xmath155 mesons , @xmath157 .
we have tightened the @xmath104 requirement to @xmath158 since this is the kinematic limit for upper - vertex @xmath159 decays .
the production of ground state and excited @xmath5 mesons from lower - vertex decays such as @xmath160 is expected to be suppressed .
this is certainly true for @xmath161 decays where the fraction of @xmath5 produced at the lower - vertex is measured to be @xmath162 @xcite .
moreover , there is no evidence of @xmath52 production in the region @xmath163 where lower - vertex production is likely to occur ( see figure [ fig : xp dist ] . ) with the assumption @xmath164 we can extract @xmath149 from the ratio of inclusive rates , @xmath165 many systematic errors cancel in the ratio . when computing the @xmath52 decay constant from the above equation , we use @xmath166 of the measured @xmath69 branching fraction to account for uncertainties in the upper and lower vertex contributions to @xmath52 .
this accomodates the excess of @xmath69 candidates observed at low @xmath167 as seen in figure [ fig : xp dist ] . from our upper limit on @xmath69 and
cleo s @xcite measurement of @xmath168 , we derive @xmath169 at the 90% c.l .
the central value is @xmath170 , where the first error is due to the total error in the inclusive @xmath171 and @xmath69 branching fractions , and the second is the uncertainty in the non - factorizable and lower - vertex contributions to the @xmath69 decay rate . using the measured value of @xmath172 mev
@xcite gives @xmath173 mev which corresponds to an upper limit of @xmath174 mev .
this limit accomodates the prediction of @xmath175 mev by veseli and dunietz @xcite .
in summary , we have searched for @xmath35 mesons decaying into the p - wave @xmath176 meson . the upper limit of @xmath177 at the 90% c.l .
accounts for at most only a fraction of the total wrong - sign @xmath178 rate .
assuming factorization , the decay constant @xmath149 is at least a factor of 2.5 times smaller than the decay constant for the pseudoscalar @xmath5 .
we gratefully acknowledge the effort of the cesr staff in providing us with excellent luminosity and running conditions .
, j.r.p . , and i.p.j.s . thank the nyi program of the nsf , m.s .
thanks the pff program of the nsf , g.e . thanks
the heisenberg foundation , k.k.g .
, t.s . , and h.y .
thank the oji program of doe , j.r.p .
, m.s . and v.s .
thank the a.p .
sloan foundation , r.w .
thanks the alexander von humboldt stiftung , m.s . thanks
research corporation , and s.d .
thanks the swiss national science foundation for support .
this work was supported by the national science foundation , the u.s .
department of energy , and the natural sciences and engineering research council of canada .
kuhn , s. nussinov , and r. ruckl , z. phys .
c * 5 * , 117 ( 1980 ) .
m. bauer , b. stech , and m. wirbel , z. phys .
c * 29 * , 637 ( 1985 ) .
cleo collaboration , x. fu _ et al .
_ , cleo conf-95 - 11 , eps0169 , contributed to the 1995 international europhysics conference on high energy physics , brussels , belgium . | we have searched for the decay @xmath0 and measured an upper limit for the inclusive branching fraction of @xmath1 at the 90% confidence level .
this limit is small compared with the total expected @xmath2 rate .
assuming factorization , the @xmath3 decay constant is constrained to be @xmath4 mev at the 90% confidence level , at least 2.5 times smaller than that of @xmath5 . 6.5 in 9.0 in -0.50 in 0.00 in 0.00 in m. bishai,@xmath6 j. fast,@xmath6 j. w. hinson,@xmath6 n. menon,@xmath6 d. h. miller,@xmath6 e. i. shibata,@xmath6 i. p. j. shipsey,@xmath6 m. yurko,@xmath6 s. glenn,@xmath7 s. d. johnson,@xmath7 y. kwon,@xmath8 s. roberts,@xmath7 e. h. thorndike,@xmath7 c. p. jessop,@xmath9 k. lingel,@xmath9 h. marsiske,@xmath9 m. l. perl,@xmath9 v. savinov,@xmath9 d. ugolini,@xmath9 r. wang,@xmath9 x. zhou,@xmath9 t. e. coan,@xmath10 v. fadeyev,@xmath10 i. korolkov,@xmath10 y. maravin,@xmath10 i. narsky,@xmath10 v. shelkov,@xmath10 j. staeck,@xmath10 r. stroynowski,@xmath10 i. volobouev,@xmath10 j. ye,@xmath10 m. artuso,@xmath11 f. azfar,@xmath11 a. efimov,@xmath11 m. goldberg,@xmath11 d. he,@xmath11 s. kopp,@xmath11 g. c. moneti,@xmath11 r. mountain,@xmath11 s. schuh,@xmath11 t. skwarnicki,@xmath11 s. stone,@xmath11 g. viehhauser,@xmath11 x. xing,@xmath11 j. bartelt,@xmath12 s. e. csorna,@xmath12 v. jain,@xmath13 k. w. mclean,@xmath12 s. marka,@xmath12 r. godang,@xmath14 k. kinoshita,@xmath14 i. c. lai,@xmath14 p. pomianowski,@xmath14 s. schrenk,@xmath14 g. bonvicini,@xmath15 d. cinabro,@xmath15 r. greene,@xmath15 l. p. perera,@xmath15 g. j. zhou,@xmath15 b. barish,@xmath16 m. chadha,@xmath16 s. chan,@xmath16 g. eigen,@xmath16 j. s. miller,@xmath16 c. ogrady,@xmath16 m. schmidtler,@xmath16 j. urheim,@xmath16 a. j. weinstein,@xmath16 f. wrthwein,@xmath16 d. w. bliss,@xmath17 g. masek,@xmath17 h. p. paar,@xmath17 s. prell,@xmath17 v. sharma,@xmath17 d. m. asner,@xmath18 j. gronberg,@xmath18 t. s. hill,@xmath18 d. j. lange,@xmath18 r. j. morrison,@xmath18 h. n. nelson,@xmath18 t. k. nelson,@xmath18 j. d. richman,@xmath18 d. roberts,@xmath18 a. ryd,@xmath18 m. s. witherell,@xmath18 r. balest,@xmath19 b. h. behrens,@xmath19 w. t. ford,@xmath19 h. park,@xmath19 j. roy,@xmath19 j. g. smith,@xmath19 j. p. alexander,@xmath20 c. bebek,@xmath20 b. e. berger,@xmath20 k. berkelman,@xmath20 k. bloom,@xmath20 v. boisvert,@xmath20 d. g. cassel,@xmath20 h. a. cho,@xmath20 d. s. crowcroft,@xmath20 m. dickson,@xmath20 s. von dombrowski,@xmath20 p. s. drell,@xmath20 k. m. ecklund,@xmath20 r. ehrlich,@xmath20 a. d. foland,@xmath20 p. gaidarev,@xmath20 l. gibbons,@xmath20 b. gittelman,@xmath20 s. w. gray,@xmath20 d. l. hartill,@xmath20 b. k. heltsley,@xmath20 p.
i. hopman,@xmath20 j. kandaswamy,@xmath20 p. c. kim,@xmath20 d. l. kreinick,@xmath20 t. lee,@xmath20 y. liu,@xmath20 n. b. mistry,@xmath20 c. r. ng,@xmath20 e. nordberg,@xmath20 m. ogg,@xmath21 j. r. patterson,@xmath20 d. peterson,@xmath20 d. riley,@xmath20 a. soffer,@xmath20 b. valant - spaight,@xmath20 c. ward,@xmath20 m. athanas,@xmath22 p. avery,@xmath22 c. d. jones,@xmath22 m. lohner,@xmath22 c. prescott,@xmath22 j. yelton,@xmath22 j. zheng,@xmath22 g. brandenburg,@xmath23 r. a. briere,@xmath23 a. ershov,@xmath23 y. s. gao,@xmath23 d. y .- j .
kim,@xmath23 r. wilson,@xmath23 h. yamamoto,@xmath23 t. e. browder,@xmath24 y. li,@xmath24 j. l. rodriguez,@xmath24 t. bergfeld,@xmath25 b. i. eisenstein,@xmath25 j. ernst,@xmath25 g. e. gladding,@xmath25 g. d. gollin,@xmath25 r. m. hans,@xmath25 e. johnson,@xmath25 i. karliner,@xmath25 m. a. marsh,@xmath25 m. palmer,@xmath25 m. selen,@xmath25 j. j. thaler,@xmath25 k. w. edwards,@xmath26 a. bellerive,@xmath27 r. janicek,@xmath27 d. b. macfarlane,@xmath27 p. m. patel,@xmath27 a. j. sadoff,@xmath28 r. ammar,@xmath29 p. baringer,@xmath29 a. bean,@xmath29 d. besson,@xmath29 d. coppage,@xmath29 c. darling,@xmath29 r. davis,@xmath29 s. kotov,@xmath29 i. kravchenko,@xmath29 n. kwak,@xmath29 l. zhou,@xmath29 s. anderson,@xmath30 y. kubota,@xmath30 s. j. lee,@xmath30 j. j. oneill,@xmath30 s. patton,@xmath30 r. poling,@xmath30 t. riehle,@xmath30 a. smith,@xmath30 m. s. alam,@xmath31 s. b. athar,@xmath31 z. ling,@xmath31 a. h. mahmood,@xmath31 h. severini,@xmath31 s. timm,@xmath31 f. wappler,@xmath31 a. anastassov,@xmath32 j. e. duboscq,@xmath32 d. fujino,@xmath33 k. k. gan,@xmath32 t. hart,@xmath32 k. honscheid,@xmath32 h. kagan,@xmath32 r. kass,@xmath32 j. lee,@xmath32 m. b. spencer,@xmath32 m. sung,@xmath32 a. undrus,@xmath33 r. wanke,@xmath32 a. wolf,@xmath32 m. m. zoeller,@xmath32 b. nemati,@xmath34 s. j. richichi,@xmath34 w. r. ross,@xmath34 and p. skubic@xmath34 @xmath6purdue university , west lafayette , indiana 47907 + @xmath7university of rochester , rochester , new york 14627 + @xmath9stanford linear accelerator center , stanford university , stanford , california 94309 + @xmath10southern methodist university , dallas , texas 75275 + @xmath11syracuse university , syracuse , new york 13244 + @xmath12vanderbilt university , nashville , tennessee 37235 + @xmath14virginia polytechnic institute and state university , blacksburg , virginia 24061 + @xmath15wayne state university , detroit , michigan 48202 + @xmath16california institute of technology , pasadena , california 91125 + @xmath17university of california , san diego , la jolla , california 92093 + @xmath18university of california , santa barbara , california 93106 + @xmath19university of colorado , boulder , colorado 80309 - 0390 + @xmath20cornell university , ithaca , new york 14853 + @xmath22university of florida , gainesville , florida 32611 + @xmath23harvard university , cambridge , massachusetts 02138 + @xmath24university of hawaii at manoa , honolulu , hawaii 96822 + @xmath25university of illinois , urbana - champaign , illinois 61801 + @xmath26carleton university , ottawa , ontario , canada k1s
5b6 + and the institute of particle physics , canada + @xmath27mcgill university , montral , qubec , canada h3a
2t8 + and the institute of particle physics , canada + @xmath28ithaca college , ithaca , new york 14850 + @xmath29university of kansas , lawrence , kansas 66045 + @xmath30university of minnesota , minneapolis , minnesota 55455 + @xmath31state university of new york at albany , albany , new york 12222 + @xmath32ohio state university , columbus , ohio 43210 + @xmath34university of oklahoma , norman , oklahoma 73019 |
rayleigh - taylor ( rt ) instability occurs when a heavy fluid is supported by a lighter fluid in a gravitational field , or , equivalently , when a heavy fluid is accelerated by a lighter fluid .
rt instability and the related processes have found applications in various astronomical systems , such as the expansion of supernova remnants ( e.g. , ribeyre et al .
2004 ) ( where inertial acceleration plays the role of the gravitational field ) , the interiors of red giants ( e.g. , chairborne and lagard 2010 ) , the radio bubbles in galaxy clusters ( pizzolato and soker 2006 ) .
the evolution of the rt instability is influenced by many different factors .
for example , viscosity tends to reduce the growth rate and to stabilize the system ( e.g. , chandrasekhar 1961 ) .
growth rate of the short - wavelength unstable perturbations decreases because of the compressibility ( e.g. , shivamoggi 2008 ) .
a dynamically important radiation field affects rt instability as well ( jacquet and krumholz 2011 ) . however , the most important effects in the astrophysical context are probably those due to the presence of the magnetic fields .
one can decompose the magnetic field lines into a component perpendicular to the interface and a component parallel to it .
we will deal only with the effect of a tangential magnetic field .
incompressible rt instability in a plane parallel to a uniform tangential magnetic field in both fluids has been studied analytically by chandrasekhar ( 1961 ) .
the linear stability theory shows that a tangential magnetic field slows down the growth rate of the rt instability .
the growth rate @xmath0 for the modes with wavenumber @xmath1 parallel to the magnetic field lines is given by @xmath2 here , we use cartesian coordinates and denote the quantities of the plasma below the discontinuity ( @xmath3 ) with a subscript 1 and those in above the discontinuity ( @xmath4 ) with a subscript 2 . the magnetic field permeating
the plasma is uniform and tangent to the discontinuity , so @xmath5 , while gravity is perpendicular to it , so @xmath6 . if we set @xmath7 , the classical dispersion relation for rt instability is obtained , i.e. @xmath8 there are a number of astrophysical systems in which the magnetic rt instability is expected to be important , among them are the accretion onto the magnetized compact objects ( wang and nepveu 1983 ) , buoyant bubbles generated by the radio jets in clusters of galaxies ( robinson et al .
2005 ) , and the thin shell of ejecta swept up by a pulsar wind ( bucciantini et al .
but one should note magnetic rt growth rate ( i.e. , equation ( 1 ) ) is based on the ideal mhd approximation , in which the multifluid nature is neglected for simplicity .
however , a partially ionized plasma represents a state which often exists .
thus , we are interested to know how the growth rate of rt instability is modified in a partially ionized medium , in particular when the coupling between the ionized and the neutral components is not complete .
there are a few studies related to this issue .
for example , chhajlani ( 1998 ) studied magnetic rt instability considering the surface tension and finite larmor radius correction ( flr ) in the absence of gravity and the pressure gradient for the neutral particles .
it was found that an increase in the collision frequency causes a decrease in the growth rate of the system .
just recently , diaz , soler & ballester ( 2012 ) ( hereafter dsb ) studied rt instability in a partially ionized compressible ( and also incomessible ) plasma .
their purpose was to study the stability thresholds and the linear growth rate of the rt unstable modes in a two - fluid plasma consisting of ions and neutrals .
they also studied the effects of compressibility and the collision between the particles .
they calculated the growth rate as a function of the wavelength of the perturbations for different values of the gravity and concluded that collisions are not able to fully suppress the instability .
but in the highly collisional regime the growth rate is significantly lowered by an order of magnitude in comparison to the classical result , specially for low values of gravity .
also , the linear growth rate is significantly lowered by compressibility and ion - neutral collisions compared to the incompressible collisionless case .
then , as an astrophysical implication , dsb applied their results to the solar prominences . in this article , although we follow a similar problem to dsb , not only our presentation of the results are different from that study , but also we apply the results to another astronomical object which is subject to rt instability . more specifically , we study rt instability in a two - fluid magnetized medium consisting of the ions and the neutrals and obtain the growth rate of the unstable modes for different wavenumbers ( not different gravity like dsb ) and compare them with the fully ionized or neutral cases .
we also determine the most unstable mode and its relation to the wavenumber .
then , possible effects of ionization and the collisions on the growth rates are analyzed .
more important , we apply our results to an astronomical object different from dsb system ,
i.e. the interaction zone between loop1 and local bubble , which seems to be subject to rt instability and the maximum growth rate and the corresponding wavenumber are obtained .
breitschwerdt , freyberg and egger ( 2000 ) showed the local clouds surrounding the solar system have been formed as a result of the growing magnetic rt instability in the interaction zone between the loop1 and local bubble .
we now know that these clouds are partially ionized ( slavin 2008 , welsh 2009 ) .
thus , we can apply our results to the interaction region between the loop i and the local bubble . in the second section
, we will present the basic equations and the assumptions of the model . in the third section
, we will analyze the growth rate of the unstable perturbations and finally in forth section we perform an application to interaction region between the loop i and the local bubble .
we follow the analysis of chandrasekhar ( 1961 ) , but including neutrals and ions which are coupled via collisions .
our basic equations are similar to the other related two - fluid studies ( e.g. , shadmehri and downes 2007 ) , but here there is a gravitational acceleration in the equations of motion for both the ions and the neutrals . in order to proceed analytically
, it is assumed that the system is incompressible and the non - ideal dissipative process related to the evolution of the magnetic field are neglected .
so , the convective term in the induction equation dominates the resistive one . we suppose the neutral and ion components have not velocity in the unperturbed state .
the magnetic field is assumed to be parallel to the interface , i.e. @xmath9 .
finally , all the unperturbed physics quantities are assumed to be spatially uniform in each medium .
the basic equations are @xmath10 @xmath11 @xmath12 @xmath13 @xmath14 @xmath15 where @xmath16 is the collision rate coefficients per unit mass so that @xmath17 is the neutral - ion collision frequency .
the collision frequency determines the coupling between each component and the magnetic field . here
, @xmath18 is a uniform vertical gravitational acceleration .
now , we perturb the physical variables as @xmath19 .
thus , linearized equations for the neutrals become @xmath20 @xmath21 @xmath22 where @xmath23 and @xmath24 are the @xmath25 and the @xmath26 components of the perturbed velocity of the neutrals , respectively . also , the linearized equations for the ions are @xmath27 @xmath28 @xmath29 @xmath30 where @xmath31 and @xmath32 are the @xmath25 and the @xmath26 components of the perturbed velocity of the ions , respectively .
after some mathematical manipulations , we can reduce the above differential equations to a set of two differential equations for @xmath24 and @xmath32 , i.e. @xmath33 @xmath34 where @xmath35 .
up to this point , our basic linearized equations are similar to shadmehri and downes ( 2007 ) who studied two - fluid kelvin - helmholtz instability . however , the boundary conditions for rt instability are different from kelvin - helmholtz instability . having solutions of the above equations and by imposing appropriate boundary conditions , we can obtain a dispersion relation for rt instability .
behavior of the flow at the upper and the lower layers is determined by the general solutions of the linear differential equations ( [ 8 ] ) and ( [ 9 ] ) .
one can easily show that the general solution of the equations is a linear superposition of two independent solutions @xmath36 and @xmath37 .
now , we must apply the following proper boundary conditions to obtain a physical solution : ( 1 ) the perturbations tends to zero as @xmath26 goes to the infinity ; ( 2 ) the @xmath26-component of the velocity is continuous at the interface ; ( 3 ) the total pressure is also continuous at the interface .
thus , the general solutions become @xmath38 @xmath39 where @xmath40 , @xmath41 , @xmath42 and @xmath43 are constants to be determined from the above boundary conditions .
continuity of the vertical displacement at @xmath44 ( the second boundary condition ) gives the following relations @xmath45 also , based on the continuity of the ions and neutrals pressures at the interface @xmath44 ( the third boundary condition ) , we have @xmath46 @xmath47 where @xmath48 is the perturbed magnetic pressure .
having solutions ( [ 10 ] ) and ( [ 11 ] ) , we can simply obtain perturbed pressures and substitute them into the above equations .
therefore , we obtain @xmath49 @xmath50 and after lengthy ( but straightforward ) mathematical manipulations , we then obtain @xmath51 @xmath52x^2+\big[2y^2\lambda(\alpha_i\alpha_n+1)-\end{aligned}\ ] ] @xmath53x -2y^3(\alpha_n-1)+\\\end{aligned}\ ] ] @xmath54 where @xmath55 @xmath56 where @xmath57 is alfven velocity for @xmath58 .
dispersion relation ( [ eq : main ] ) is the main equation of our stability analysis .
obviously , if we neglect collision between ions and neutrals ( i.e. , @xmath59 ) , equation ( [ eq : main ] ) simply reduces to a dispersion relation for the ions which are tied to the magnetic field lines and another dispersion relation for the neutral component , i.e. @xmath60 if we set the first parenthesis equal to zero , magnetic criterion for rt instability is obtained .
also , the second parenthesis gives the classical condition of non - magnetic rt instability .
although dispersion relation for rt instability within one fluid approximation gives analytical solutions , it is very unlikely to obtain roots of equation ( [ eq : main ] ) analytically .
so , we follow the problem numerically by assuming some numerical values for the input parameters .
we assume @xmath61 and the dispersion relation ( [ eq : main ] ) is numerically solved for different values for the parameters @xmath62 , @xmath63 and @xmath64 .
obviously , we would have an unstable mode if @xmath25 has a positive real part .
figure 1 shows non - dimensional growth rate @xmath65 of the unstable modes versus the non - dimensional wavenumber @xmath66 .
parameter @xmath63 denotes the ratio of densities of the neutrals and ions in layer 1 . in figure 1 , we assume @xmath67 and each curve is labeled by the corresponding non - dimensional collision rate @xmath62 .
figure 2 shows growth rate of the perturbations versus the wavenumber but for @xmath68 . in this case , the ions are stable and the neutrals determine growth rate of the perturbations . also , figure 3 and 4 show the growth rate of the neutrals and the ions for @xmath69 , respectively .
again , we can see the stabilizing role of the collision between the ions and the neutrals .
note that when there is no collision between the ions and neutrals ( i.e. , @xmath70 ) , each component of our two - fluid system behaves independently .
but when collision between ions and neutrals is considered , we found two unstable modes up to a certain wavenumber that are related on the neutral and ion fluids separately , with the neutral ones having a tendency with much larger growth rate .
curves with the same color are corresponding to the same value of @xmath62 . since magnetic field
has a stabilizing role , we can see that unstable mode corresponding to the ions has a smaller growth rate in comparison to the neutrals unstable mode because of the coupling to the magnetic field lines .
dsb obtained plots of dispersion relation of the unstable modes versus the non - dimensional gravity but we plot unstable modes versus wavenumber .
therefor , we can conclude : + ( 1 ) the neutral mode is unstable for all wavenumbers but for ions we can find a critical wavenumber for which the instability becomes ineffective .
this result is valid irrespective of the ionization fraction and the collision rate . +
( 2 ) growth rate of the unstable perturbations for the ions tends to become zero as the collision rate increases and thereby , behavior of the system is determined by the growth rate of the neutrals .
moreover , in this case , the profile of the growth rate for the neutrals is similar to the ions without collision .
+ ( 3 ) however , as the collision frequency increases , not only growth rate of ions reduces but the unstable perturbations for the neutrals are significantly reduced in particular at short wavelengths . + in the next section we apply our results to an astronomical object .
our solar system is embedded in an ionized cloud named local cloud . in vicinity of local clouds
there are also other cloudlets of comparable size .
winds and supernovae events that are associated with clusters of massive early - type stars have a profound effect on the surrounding interstellar medium ( ism ) , including the creation of large cavities .
these cavities , which are often referred to as `` interstellar bubbles '' , are typically @xmath71 in diameter and have low neutral gas densities of @xmath72 ( weaver et al .
the local clouds are inside a local x - ray emitting cavity which called the local bubble .
breitschwerdt et .
al ( 2000 ) presented observational evidences based rosat pspc data that manifest existence of an interaction shell between our local interstellar bubble and the adjacent loop i superbubble .
they showed that due to the overpressure in loop i , a rayleigh - taylor instability would operate , even in the presence of tangential magnetic field .
their calculations showed that the most unstable mode has a growth time about @xmath73 years ( depending on magnetic field strength ) which was in agreement with the interaction time between the two bubbles .
moreover , the wavelength of the fastest growing mode was about 2.2 pc which was comparable to or less than the thickness of the interaction zone . also , breitschwerdt et al .
( 2006 ) have performed 3d high resolution hydrodynamic simulations of the local bubble ( lb ) and the neighboring loop i ( l1 ) and reproduced the observed sizes of the local and loop i superbubbles , the generation of blobs like the local cloud as a consequence of a dynamical instability ( breitschwerdt et al .
nevertheless , reports suggest the local clouds are partially ionized ( e.g. , slavin 2008 , welsh et al .
since the clouds are formed due to rt instability ( breitschwerdt et al . 2000 and 2006 ) , we conclude the interaction zone between loop i and local bubble is partially ionized and because loop i is hot , the interaction shell must be ionized .
now we are interested to know what effect does the ionization may have on the growth rate of system .
+ we begin with breitschwerdt and slavin s assumptions for our system and obtain a dispersion equation from equations ( [ eq:13])and ( [ eq:14 ] ) for a case where @xmath74 and @xmath75 .
thus , @xmath76 @xmath77 where @xmath78 @xmath79 where @xmath80 is alfven velocity for @xmath81 .
we have the following input numbers ( breitschwerdt et al .
2000 ) @xmath82 now , we can estimate @xmath63 and @xmath62 . slavin ( 2008 )
found the ionization fraction is around @xmath83 for hydrogen and around @xmath84 for helium .
also , welsh et al . (
2009 ) presented an amount of 0.1 for ionized fraction .
if we suppose the ionized fraction is 0.1 , then value of @xmath63 becomes 9 .
we can write the collision frequency ( shadmehri et al .
2008 ) , @xmath85 where @xmath86 .
we neglected ionization of helium .
thus , @xmath87 having the above input numbers , we can solve equation ( @xmath88 ) to find growth time of the fastest growing mode .
figure 5 shows the results of such a calculation for the specified parameters .
then , the the most unstable mode of system is @xmath89 we found shorter growth time in comparison to the classical magnetic rt instability ( around @xmath90yrs ) .
the wavenumber of the fastest growing mode is @xmath91 thus , the size of structures formed by rt instability reduces in comparison to the classical magnetic rt instability that is @xmath92 . when the nondimensional collision frequency lambda is large , the ionization fraction has also a vital role .
ionized particles are coupled to the magnetic field lines , but their coupling to the neutral particles is determined via collisions ( i.e. , @xmath62 ) .
now , we can consider two identical systems with the same input parameters except their ionization fractions .
if both systems have a large collision rate , the system with a larger ionization fraction is more affected by the magnetic field lines in comparison to the system with a smaller ionization fraction .
in other words , although the ionized and the neutral particles are coupled to the same level , but when the ionization fraction is larger the system is more under influence of the magnetic field lines .
figure 5 clearly shows this effect .
each curve is labeled by its ionization fraction . here , the non - dimensional collision frequency is @xmath93 .
we can see that as the ionization fraction increases , the growth rate of the unstable mode decreases simply because more particles are affected by the magnetic field lines .
it is because of the two - fluid nature of the system .
the effect is more significant when the collision frequency decreases and the coupling between the ionized and the neutral particles is not complete .
thus , it seems that one - fluid approach is not adequate even when the collision frequency is large but the ionization fraction is not large enough .
however , it is difficult to determine a critical value for lambda so that beyond which the system tends to mhd case . because such a transition depends on the ionization fraction among the other input parameters .
moreover , when two - fluid approach is adopted the maximum growth rate is modified .
but compressibility does not lead to such an effect .
in fact , compressibility becomes less effective when the density contrast of the layers increases .
so , the compressibility correction in a two - fluid system subject to rt instability depends on the density contrast of the layers . and various non - dimensional collision rate @xmath62 .
curves with the same color are corresponding to the same value of @xmath62.,width=264,height=226 ] .,width=264,height=226 ] .,width=264,height=226 ] and different ionization fractions.,width=264,height=226 ]
in this study , we investigated the magnetic rt instability in a two - fluid medium consisting of the neutrals and the ions . a general dispersion relation is obtained . by analyzing the unstable roots of the dispersion equation
, two unstable modes are found that are related to those of the neutral and ion fluids separately , with the neutral ones having a much higher growing rate .
for each ionization fraction and collision rate , the stability of the system only depends on the behavior of the neutrals . for lower values of collision rate
the curves are similar to the collisionless case .
we found that the growth rate of the unstable perturbations decreases when the collision rate increases .
also , the instability of the system strongly depends on the ionization fraction . when the ionization fraction increases , for a given collision rate , the growth rate of the perturbations decreases .
finally , we apply our results for interaction zone between the loop i and the local bubble that is caused form local clouds .
we obtained a shorter value for the growth time and size of the clouds .
although classical magnetic rt instability has been applied to this system for explaining some of the observed structures , our analysis shows that rt instability may operates less effectively if the two - fluid nature of the system is considered .
but we note that our results are valid within linear regime and non - linear numerical simulations are needed to confirm the linear results .
we note that magnetic effects only suppress the linear growth rate for perturbations aligned with the magnetic field . those perpendicular to it are unaffected .
indeed , as shown by stone & gardiner ( 2007 ) , in 3d the net effect of magnetic fields is actually to enhance the non - linear growth rate by suppressing secondary kh instabilities .
thus in the real world , it seems like magentic rt instability never occur .
the hydrodynamic modes perpendicular to the field always end up taking over .
we also think that the hall effect is an interesting problem , but it is beyond the scope of the present study .
our analysis is restricted to a two - fluid case , i.e. a system consisting of ion and neutral particles .
we could also start from the one fluid mhd equations , but considering modified induction equation with resistivity , hall and ambipolar terms . in that framework
we could study possible effects of non - ideal terms ( including hall term ) .
but we think it deserves a separate analysis independent of the present study . | we study linear theory of the magnetized rayleigh - taylor instability in a system consisting of ions and neutrals .
both components are affected by a uniform vertical gravitational field .
we consider ions and neutrals as two separate fluid systems where they can exchange momentum through collisions .
however , ions have direct interaction with the magnetic field lines but neutrals are not affected by the field directly .
the equations of our two - fluid model are linearized and by applying a set of proper boundary conditions , a general dispersion relation is derived for our two superposed fluids separated by a horizontal boundary .
we found two unstable modes for a range of the wavenumbers .
it seems that one of the unstable modes corresponds to the ions and the other one is for the neutrals .
both modes are reduced with increasing the collision rate of the particles and the ionization fraction .
we show that if the two - fluid nature is considered , rt instability would not be suppressed and also show that the growth time of the perturbations increases . as an example
, we apply our analysis to the local clouds which seems to have arisen because of the rt instability . assuming that the clouds are partially ionized , we find that the growth rate of these clouds increases in comparison to a fully ionized case . |
understanding the origin of mass is the main open puzzle in particle physics today . in the standard model
all masses arise as a result of the spontaneous breaking of the 1 gauge symmetry .
this implies the existence of an elementary higgs boson , not yet found .
stabilizing the mass of the higgs most likely requires new physics and supersymmetry has so far been the leading contender .
another aspect of this problem is the smallness of neutrino masses . despite the tremendous effort that has led to the discovery of neutrino mass @xcite the mechanism of neutrino mass generation will remain open for years to come ( a detailed analysis of the three
neutrino oscillation parameters can be found in @xcite ) .
the most popular mechanism to generate neutrino masses is the seesaw mechanism @xcite .
although the seesaw fits naturally in so(10 ) unification models , we currently have no clear hints that uniquely point towards any unification scheme .
therefore it may well be that neutrino masses arise from garden variety physics having nothing to do with unification , such as certain seesaw variants @xcite , and models with radiative generation @xcite .
in such models the physics of neutrino mass would then be characterized by much lower scales @xcite , potentially affecting the decay properties of the higgs boson .
this is especially so if neutrino masses arise due to the spontaneous violation of ungauged lepton number . in this broad class of models
the higgs boson will have an important decay channel into the singlet goldstone boson ( called majoron ) associated to lepton number violation @xcite , @xmath1 here we focus on the specific case of low - energy supersymmetry with spontaneous violation of r parity , as the origin of neutrino mass .
r parity is defined as @xmath2 with @xmath3 , @xmath4 , @xmath5 denoting spin , baryon and lepton numbers , respectively @xcite . in this model r
parity violation takes place `` a la higgs '' , i.e. , spontaneously , due to non - zero sneutrino vacuum expectation values ( vevs ) @xcite . in this case one of the neutral cp - odd scalars is identified with the majoron .
in contrast with the seesaw majoron , ours is characterized by a small scale ( tev - like ) and carries only one unit of lepton number .
this scheme leads to the bilinear r
parity violation model , the simplest effective description of r - parity violation @xcite ( for calculations including also trilinear terms see , for example @xcite ) .
the model not only accounts for the observed pattern of neutrino masses and mixing @xcite , but also makes predictions for the decay branching ratios of the lightest supersymmetric particle @xcite from the current measurements of neutrino mixing angles @xcite . in previous studies
@xcite it was noted that the spontaneously broken r parity ( sbrp ) model leads to the possibility of invisibly decaying higgs bosons , provided there is an 1 singlet superfield @xmath6 coupling to the electroweak doublet higgses , the same that appears in the nmssm .
in this paper we reanalyse this issue taking into account the small masses indicated by current neutrino oscillation data @xcite ev . ] .
we focus on the lowest - lying neutral cp - even scalar boson of the model .
we show explicitly that the presence of the 1 singlet superfield @xmath6 plays a triple role : ( i ) it gives a model where neutrino masses are obtained from first principles without any type of fine - tuning , even when radiative corrections are negligible , ( ii ) it solves the @xmath0-problem `` a la nmssm '' @xmath7-symmetry on the non - renormalizable operators @xcite , or that they are simply inflated away . ] , and ( iii ) it makes the invisible higgs boson decay in eq .
( [ eq : hjj ] ) potentially the most important mode of higgs boson decay .
the latter is remarkable , given the smallness of neutrino masses required to fit current neutrino oscillation data .
we also verify that the production of such higgs boson in @xmath8 annihilation can be as large as that characterizing the standard case , and that therefore this situation should be taken as part of the agenda of future accelerators probing the mechanism of mass generation .
the most general superpotential terms involving the minimal supersymmetric standard model ( mssm ) superfields in the presence of the 1 singlet superfields @xmath9 carrying a conserved lepton number assigned as @xmath10 , respectively is given as @xcite @xmath11 the first three terms together with the @xmath12 term define the r - parity conserving mssm , the terms in the last row only involve the 1 singlet superfields @xmath13 has been included in the first row as it is relevant in electroweak breaking .
] , while the remaining terms couple the singlets to the mssm fields .
we stress the importance of the dirac - yukawa term which connects the right - handed neutrino superfields to the lepton doublet superfields , thus fixing lepton number .
the presence of singlets in the model is essential in order to drive the spontaneous violation of r parity and electroweak symmetries in a phenomenologically consistent way .
like all other yukawa couplings @xmath14 we assume that @xmath15 is an arbitrary non - symmetric complex matrix in generation space . for technical simplicity
we take the simplest case with just one pair of lepton number carrying 1 singlet superfields , @xmath16 and @xmath17 , in order to avoid inessential complication .
this in turn implies , @xmath18 and @xmath19 .
the full scalar potential along neutral directions is given by @xmath20\\\nonumber & & + \sum_{\alpha } \tilde{m}_{\alpha}^2 |z_{\alpha}|^2 + \frac{1}{8 } ( g^2 + { g'}^2 ) \big ( |h_u|^2 - |h_d|^2 - \sum_{i=1}^3 |\tilde{\nu}_i|^2\big)^2,\end{aligned}\ ] ] where @xmath21 denotes any neutral scalar field in the theory .
the pattern of spontaneous symmetry breaking of both electroweak and r parity symmetries works in a very simple way . the spontaneous breaking of r parity is driven by nonzero vevs for the scalar neutrinos .
the scale characterizing r parity breaking is set by the isosinglet vevs @xmath22 and @xmath23 we also have very small left - handed sneutrino vacuum expectation values @xmath24 the spontaneous breaking of r parity also entails the spontaneous violation of total lepton number .
this implies that one of the neutral cp
odd scalars , which we call majoron , and which is given by the imaginary part of @xmath25 remains massless , as it is the nambu - goldstone boson associated to the breaking of lepton number .
note that this majoron is quite different from the one that emerges in the seesaw majoron model , as it is characterized by a different lepton number ( one unit instead of two ) and by a different scale , determined by the combination @xmath26 tev .
note that eq .
( [ eq : rpv ] ) is the origin of lepton number violation in this model and plays a crucial role in determining the neutrino masses . on the other hand ,
electroweak breaking is driven by the isodoublet vevs @xmath27 and @xmath28 , with the combination @xmath29 fixed by the w mass @xmath30 while the ratio of isodoublet vevs yields @xmath31 this basically recovers the standard tree level spontaneous breaking of the electroweak symmetry in the mssm @xcite .
since neutrino masses are so much smaller than all other fermion mass terms in the model , once can find the effective neutrino mass matrix in a seesaw type approximation . from the full neutral fermion mass matrix ,
( [ eq : mass ] ) , one calculates the effective @xmath32 neutrino mass matrix @xmath33 as @xmath34 where @xmath35 is the @xmath36 matrix of all other neutral fermion states , see eq .
( [ eq : mass ] ) , and the @xmath37 matrix @xmath38 is given as @xmath39 where the matrices @xmath38 and @xmath40 are given in eqs .
( [ eq : mrpv ] ) and ( [ eq : md ] ) .
the inverse of @xmath35 is too long to be given explicitly here .
after some algebraic manipulation , the effective neutrino mass matrix can be cast into a very simple form @xmath41 where one can define the effective bilinear r parity violating parameters @xmath42 and @xmath43 as @xmath44 and @xmath45 here the parameter @xmath0 is @xmath46 while the coefficients appearing in eq .
( [ eq : eff ] ) are given by @xmath47 @xmath48 @xmath49 and @xmath50 is given as @xmath51 note that @xmath52 and @xmath53 above are defined as @xmath54 the `` photino '' mass parameter is defined as @xmath55 .
( [ eq : eff ] ) resembles very closely the corresponding expression for the explicit bilinear r - parity breaking model @xcite , once the dominant 1-loop corrections are taken into account .
note that the tree - level result of the explicit bilinear model can be recovered in the limit @xmath56 . in this limit the coefficients @xmath57 and @xmath58 go to zero , while @xmath59 in this limit only one non - zero neutrino mass remains . whether the 1-loop corrections or the contribution from the singlet fields
are more important in determining the neutrino masses depends essentially on the relative size of the coefficient @xmath58 in eq .
( [ eq : eff ] ) compared to the corresponding 1-loop coefficient .
both extremes can be realized in our model .
we note , however , that as discussed below large branching ratios of the higgs into invisible final states require sizeable values of @xmath60 and @xmath61 ( as well as singlets not being too heavy ) . for such choices of parameters
we have found that the `` singlino '' contribution to eq .
( [ eq : eff ] ) is usually much more important than the 1-loop corrections to the neutrino masses .
note also that the model does not predict whether the atmospheric ( solar ) mass scale is mainly due to the first ( third ) term in eq .
( [ eq : eff ] ) or vice versa . we have checked numerically that both possibilities can be realized and `` good '' points ( in the sense of being appropriate for neutrino physics ) can be found easily in either case . with the above choices and definitions we can obtain the neutral scalar boson mass matrices as in ref .
@xcite by evaluating the second derivatives of the scalar potential in eq .
( [ scalarpot ] ) at the minimum .
this results in @xmath62 mass matrices for the real and imaginary parts of the neutral scalars . we have checked , in particular , that in the cp - odd sector we find both the goldstone `` eaten '' by the @xmath63 as well as the goldstone boson corresponding to the spontaneous breaking of r - parity , namely the majoron , eq .
( [ maj ] ) . in the basis @xmath64 these fields
are given as , @xmath65 j&=&n_4 ( -n_1 v_d , n_1 v_u , n_2 v_{l1 } , n_2 v_{l2 } , n_2 v_{l3},0 , n_3 v_s ,- n_3 v_r)\end{aligned}\ ] ] where the normalization constants @xmath66 are given as @xmath67 n_1&=&v_{l1}^2 + v_{l2}^2 + v_{l3}^2\nonumber \\[+2 mm ] n_2&=&v_d^2 + v_u^2\nonumber \\[+2 mm ] n_3&=&n_1 + n_2 \nonumber \\[+2 mm ] n_4&=&\frac{1}{\sqrt{n_1 ^ 2 n_2 + n_2 ^ 2 n_1 + n_3 ^ 2(v_r^2+v_s^2)}}\end{aligned}\ ] ] and can easily be checked to be orthogonal , i. e. they satisfy @xmath68 . in order to study the phenomenology of the scalar sector we need some information about the parameters of the sbrp model .
broadly speaking there are four types of parameters that are to a large extent undetermined .
first there are yukawa couplings , like @xmath60 , @xmath61 and @xmath69 .
in contrast to @xmath70 , @xmath71 and @xmath72 these are not fixed by fermion masses .
then there are mssm parameters such as @xmath73 , the effective higgsino mixing parameter @xmath0 , the supersymmetry breaking scalar mass parameters @xmath74 and @xmath75 .
these are partially restricted by negative collider searches for supersymmetric particles @xcite .
then there are singlet sector mass parameters , such as @xmath76 , @xmath77 and @xmath78 .
finally there is the important yukawa coupling @xmath79 , which determines the strength of effective r parity breaking parameters , through eq .
( [ eq : eps ] ) .
this is constrained by neutrino oscillation data . in section [ sec : numerical - results ] we will discuss our strategy to choose the parameters in such a way that the results can be easily interpreted .
we will also show there that a fully cubic superpotential , without any mass scale parameter such as the @xmath80 term , also leads to a realistic model @xcite consistent with neutrino oscillation data . before that , however , we consider the corresponding higgs boson phenomenology , focusing on higgs boson production and decays , and stressing the potentially large invisible decay branching ratio .
supersymmetric higgs bosons can be produced at the @xmath81 collider through their couplings to @xmath82 , via the so called bjorken process . in our sbrp model
there are 8 neutral cp even states @xmath83 and 6 neutral cp
odd higgs bosons @xmath84 , in addition to the majoron @xmath85 .
one must diagonalize the scalar boson mass matrix in order to find the coupling of the massive scalars to the @xmath82 .
the lagrangean is @xmath86 with each @xmath87 given as a weighted combination of the five 1 doublet scalars , @xmath88 where @xmath89 is the @xmath62 rotation matrix for the cp - even scalars .
note that we leave the discussion of the cp
odd scalars for elsewhere .
moreover , here we focus mainly on the production of the lightest cp even supersymmetric higgs boson @xmath90 .
the main difference between the production of this state and the lightest cp even higgs boson of the mssm is the fact that ours contains an admixture of the 1 singlet scalar fields @xmath91 and @xmath92 , and its coupling to the @xmath82 is correspondingly reduced by a factor @xmath93 in comparison with the standard model case in the usual notation . ] . when the lightest cp even higgs boson is mainly singlet its production cross section in @xmath81 annihilation will be suppressed .
we now turn to the lightest higgs boson decays . given that other mssm decay modes are less important ,
we are particularly interested here in the ratio @xmath94 of the invisible decay to the standard model decay into b - jets . for this
we have to look separately at the decay widths , @xmath95 and @xmath96^{3/2}\ ] ] from these expressions we see that @xmath97 will be small if the component of the lightest higgs boson along @xmath98 is small .
on the other hand the magnitude of @xmath99 will depend on the @xmath100 coupling .
this is in general given by a complicated expression , but for the situation that we are considering here with @xmath101 we have to a very good approximation @xmath102 where @xmath103 .
under this approximation we can write the coupling @xmath104 for the vertex @xmath105 of the majoron with the _ unrotated _ higgs boson @xmath106 , in the following form @xmath107 g'_2&=&h h_0 v_d \frac{v_s v_r}{v^2 } -\frac{2 v_u}{v^2}\ , \sum_{j=1}^3 \epsilon_j^2 \nonumber \\[+1 mm ] g'_i&=&-\frac{2 \epsilon_{i-2}}{v^2}\sum_{j=1}^3 \epsilon_j v_{lj } \quad ( i=3,4,5 ) \nonumber \\[+1 mm ] g'_6&= & -\sqrt{2}\ , h \left(a_h + \widehat{m}_{\phi } \right)\ , \frac{v_s v_r}{v^2 } -\sqrt{2}\ , h\ , \widehat{m}_r \nonumber \\[+1 mm ] g'_7&=&-h^2 \frac{v_s v_r^2}{v^2}\nonumber \\[+1 mm ] g'_8&=&-h^2 \frac{v_s^2 v_r}{v^2}\end{aligned}\ ] ] where @xmath52 and @xmath53 have been defined in eq .
( [ eq : effmr ] ) . from these expressions
we conclude that @xmath100 can be large in two situations .
the first is , of course , if the lightest higgs boson is mainly a combination of the @xmath108 and @xmath92 fields . in this case not only @xmath100 will be large , but also @xmath109 will be small suppressing @xmath110 .
unfortunately the production would be suppressed , as singlets do not couple to the @xmath82 .
the phenomenologically novel and interesting situation is when @xmath60 and @xmath61 are large . in this case
the higgs boson behaves as the lightest mssm higgs boson ( with moderately reduced production cross section ) but with a large branching to the invisible channel @xmath111 .
the sensitivities of lep experiments to the invisible channel @xmath112 have been discussed since long ago @xcite and the current status has been presented in ref .
in order to evaluate the experimental sensitivities to the parameters of the model we must take into account both the production as well as higgs decays .
in this section we discuss the numerical results on the invisible decay of the higgs boson in our model .
we start with a brief discussion of the sbrp parameters .
unknown parameters of the spontaneous r - parity breaking model fall into three different groups .
first , there are the mssm parameters , mainly the unknown soft susy breaking terms .
the second group of parameters are the @xmath113 and left - handed sneutrino vevs @xmath114 .
we trade the latter for the parameters @xmath43 using eq .
( [ eq : deflam0 ] ) .
these six parameters occur also in the explicit bilinear model . and , finally , there are the parameters of the singlet sector , namely singlet vevs @xmath115 , @xmath116 and @xmath117 , yukawa couplings @xmath60 , @xmath61 and @xmath69 and the singlet mass terms @xmath76 , @xmath118 , @xmath78 , as well as the corresponding soft terms .
we have checked by a rather generous scan that the results presented below qualitatively do not depend on the choice of mssm parameters , as expected .
thus , for definiteness we will fix the mssm parameters in the following to the sps1a benchmark point @xcite , defined by @xmath119 we have run down this set of parameters to the electro - weak scale using the program package spheno @xcite .
we stress again that different choices of mssm parameters will not lead to qualitatively different results .
we first consider the general model defined by the superpotential in eq .
( [ eq : wsuppot ] ) reduced to one generation of @xmath120 and @xmath3 fields . for the singlet parameters we choose as a starting point @xmath121 gev and @xmath122 gev , as well as @xmath123 , @xmath124 and @xmath125 .
we have tried other values of parameters and obtained qualitatively similar results to the ones discussed below .
the explicit bilinear parameters are then fixed approximately such that neutrino masses and mixing angles @xcite are in agreement with experimental data @xcite .
slightly different values of parameters are found , depending on whether the first or the third term in eq .
( [ eq : eff ] ) is responsible for the atmospheric neutrino mass scale .
both possibilities lead to very similar results for the invisible decay of the higgs .
this can be understood quite easily .
the ratio of the atmospheric and solar neutrino mass scale is only of the order of ( @xmath126 ) are approximately equal to the larger mass . ] and the changes in parameters @xmath127 and @xmath128 are only of the order of the square root of this number .
such a small change can always be compensated by a slight adjustment of other parameters , leading to the same ( or very similar ) final result .
after having defined our `` preferred '' choice of parameters in the following we will vary one unknown parameter at a time .
we now turn to a discussion of the results . in fig .
( [ fig : plot54 ] ) we show the ratio @xmath129 as a function of @xmath130 for different choices of @xmath60 ( left ) and for different choices of @xmath115 ( right ) and all other parameters fixed .
larger values of @xmath129 are found for smaller values of @xmath131 , as expected .
however , one sees explicitly that even for values of @xmath132 , @xmath129 can be larger than @xmath133 .
this means that the lightest higgs can decay mainly invisibly , even when the cross section for its production is essentially equal to the usual ( mssm ) doublet higgs boson cross section .
this is the main result of this work . ,
defined in eq .
( [ eq : ratio ] ) , as function of @xmath130 .
a ) to the left , for different values of the parameter @xmath60 , from top to bottom : @xmath134 .
b ) to the right , for different values of the parameter @xmath135 : @xmath136 gev .
the plots show explicitly that @xmath137 is possible even for @xmath132 .
this is the main result of the current paper.,title="fig:",width=283,height=188 ] , defined in eq .
( [ eq : ratio ] ) , as function of @xmath130 .
a ) to the left , for different values of the parameter @xmath60 , from top to bottom : @xmath134 .
b ) to the right , for different values of the parameter @xmath135 : @xmath136 gev .
the plots show explicitly that @xmath137 is possible even for @xmath132 .
this is the main result of the current paper.,title="fig:",width=283,height=188 ] in fig .
( [ fig : plot59 ] ) we show @xmath129 as function of @xmath138 ( left ) and as function of @xmath60 ( to the right ) .
the figure shows that large values of @xmath129 are obtained for small values of @xmath139 and for large values of @xmath60 .
the decreasing of @xmath129 with increasing values of @xmath139 can be easily understood , since in the limit @xmath140 the majoron should obviously decouple . ,
defined in eq .
( [ eq : ratio ] ) , as function of @xmath139 ( left ) and as function of @xmath60 ( to the right ) .
small ( large ) values of @xmath139 ( @xmath60 ) lead to large values of @xmath129.,title="fig:",width=283,height=188 ] , defined in eq .
( [ eq : ratio ] ) , as function of @xmath139 ( left ) and as function of @xmath60 ( to the right ) . small ( large ) values of @xmath139 ( @xmath60 ) lead to large values of @xmath129.,title="fig:",width=283,height=188 ] other singlet - sector parameters also can have an important impact on @xmath129 , as demonstrated in fig .
( [ fig : plot60 ] ) .
as shown in the left panel of this figure , larger values of @xmath61 lead to larger values of @xmath129 . for values of @xmath60
smaller than about @xmath141 ( for our specific choice of the other parameters ) the order of the lines is exchanged .
this is due to a level - crossing in the eigenvalues . below this value ,
the lightest higgs is mainly a singlet and thus even though it decays dominantly invisibly its production cross section is very much reduced .
on the other hand , the right panel of figure ( [ fig : plot60 ] ) shows that the value of @xmath117 is normally somewhat less important than the value of @xmath139 in determining @xmath129 . again
this can be qualitatively understood since @xmath139 is the parameter whose magnitude determines the breaking of lepton number ( indeed , with the help of the approximate couplings @xmath104 in eq .
( [ eq:16 ] ) one can see that the parameters @xmath60 , @xmath61 , @xmath115 and @xmath116 should be the most important ones ) . ,
defined in eq .
( [ eq : ratio ] ) , as a ) left figure : function of @xmath142 for @xmath143 ( on the right part of the plot from top to bottom ) .
the right panel b ) gives @xmath129 as function of @xmath144 for different values of the parameter @xmath135 for @xmath145 gev.,title="fig:",width=283,height=188 ] , defined in eq .
( [ eq : ratio ] ) , as a ) left figure : function of @xmath142 for @xmath143 ( on the right part of the plot from top to bottom ) .
the right panel b ) gives @xmath129 as function of @xmath144 for different values of the parameter @xmath135 for @xmath145 gev.,title="fig:",width=283,height=188 ] as a summary of this section we conclude that large branching ratios of the doublet like higgs boson into invisible final states are possible in the sbrp model , despite the smallness of the neutrino masses indicated by oscillation data .
large values of @xmath129 occur for large values of the yukawa couplings and for small values of @xmath115 .
the presence of the field @xmath6 plays a crucial role in getting the invisible higgs boson decays that are not suppressed by the small neutrino masses . before concluding we illustrate the results we have obtained for the case of a restricted sbrp model described by the superpotential in eq .
( [ eq : wsuppot ] ) containing only cubic terms @xcite .
the restricted model provides a potential `` solution '' to the @xmath0 problem in the context of spontaneous r parity violation .
we give results for the same parameter choices as above , except that no mass parameters are now present in the basic superpotential .
, defined in eq .
( [ eq : ratio ] ) , as function of @xmath130 , a ) to the left , for different values of the parameter @xmath60 and b ) to the right , for different values of the parameter @xmath135 . as in the general case ( fig .
[ fig : plot54 ] ) , large values of @xmath129 can be found even for @xmath146 also in the cubic - only case.,title="fig:",width=283,height=188 ] , defined in eq .
( [ eq : ratio ] ) , as function of @xmath130 , a ) to the left , for different values of the parameter @xmath60 and b ) to the right , for different values of the parameter @xmath135 . as in the general case ( fig .
[ fig : plot54 ] ) , large values of @xmath129 can be found even for @xmath146 also in the cubic - only case.,title="fig:",width=283,height=188 ] , defined in eq .
( [ eq : ratio ] ) , as function of the parameter @xmath139 ( left ) and as function of @xmath60 ( right ) .
the qualitative behaviour is similar to the general case , compare to figure [ fig : plot59].,title="fig:",width=283,height=188 ] , defined in eq .
( [ eq : ratio ] ) , as function of the parameter @xmath139 ( left ) and as function of @xmath60 ( right ) .
the qualitative behaviour is similar to the general case , compare to figure [ fig : plot59].,title="fig:",width=283,height=188 ] even though acceptable physical solutions consistent with experiment ( supersymmetric particle searches as well as neutrino oscillation data ) are somewhat harder to find , they exist .
( [ fig : plot143 ] ) and ( [ fig : plot144 ] ) show @xmath129 as function of @xmath130 and as function of @xmath60 and @xmath139 for the cubic - only case , compare to figs .
( [ fig : plot54 ] ) and ( [ fig : plot59 ] ) for the general case .
as can be seen , the qualitative behaviour is very similar in all cases , although the parameters for which acceptable solutions are found are usually restricted to narrower ranges in the cubic - only case .
these figures demonstrate that also in the cubic - only case large production cross section and large invisible branching ratios for the lightest higgs decay can occur at the same time .
we have discussed the possibility of an invisibly decaying higgs boson in the context of the spontaneously broken r parity model .
one of the neutral cp - odd scalars in this model corresponds to the 1 singlet nambu - goldstone boson associated to the breaking of lepton number .
in contrast to the mssm , where the higgs boson can decay invisibly only to supersymmetric states ( in regions of parameters where the higgs is heavier than twice the lightest neutralino mass ) in our case the higgs can decay mainly due to eq .
( [ eq : hjj ] ) , instead of the standard model channels , over large regions of parameters , given that there is no kinematical barrier for this decay .
we have reanalysed this striking suggestion in view of the recent data on neutrino oscillations that indicate non - zero but small neutrino masses .
we have explicitly shown that ( i ) despite the smallness of neutrino masses , invisible higgs boson decay may indeed provide the most important mode of higgs boson decays and ( ii ) its production cross section need not be suppressed with respect to that characterizing the standard mssm case . as a result , our analysis indicates that invisibly decaying higgs bosons should be an important topic in the agenda of future accelerators , such as the large hadron collider and the next linear collider .
in fact the interest on this possibility goes beyond the model we have taken as framework , it is much more general .
however the sbrp model provides an attractive explanation for the origin of the neutrino masses that can also be probed at future collider experiments through the predicted pattern of the lightest supersymmetric particle ( lsp ) decays which directly traces the experimentally observed neutrino mixing angles .
this work was supported by spanish grant bfm2002 - 00345 , by the european commission human potential program rtn network hprn - ct-2000 - 00148 and by the european science foundation network grant n.86 .
is supported by a mcyt ramon y cajal contract .
a. v. m. was supported by a phd fellowship from generalitat valenciana .
jcr was supported by the portuguese _ fundao para a cincia e a tecnologia _ under the contract cfif - plurianual and grant pocti / fnu/4989/2002 .
we thank werner porod for useful discussions .
in the basis @xmath147 the mass matrix of the neutral fermions following from eq . ( [ eq : wsuppot ] ) can be written as @xmath148 where the matrix @xmath149 is the mssm neutralino mass matrix : @xmath150 here , @xmath151 .
@xmath152 is the r - parity violating neutrino - neutralino mixing part , which also appears in explicit bilinear r - parity breaking models : @xmath153 -\frac{1}{2}g'v_{l \mu } & \frac{1}{2}gv_{l \mu } & 0 & \epsilon_{\mu}\\[2 mm ] -\frac{1}{2}g'v_{l \tau } & \frac{1}{2}gv_{l \tau } & 0 & \epsilon_{\tau } \end{array } \right ) , \label{eq : mrpv}\ ] ] where @xmath154 are the vevs of the left - sneutrinos , @xmath113 are defined by @xmath155 , and @xmath115 is the vev of the right - sneutrino . here
@xmath156 is given as @xmath157 and @xmath158 is @xmath159 the `` dirac '' mass matrix is defined in the usual way : @xmath160 the @xmath120 and @xmath3 states are coupled by @xmath161 @xmath162 and @xmath163 are @xmath164 @xmath165 here , @xmath166 and @xmath167 . finally @xmath168 is @xmath169 we briefly comment on the case of three generations of neutral fermions in the singlet sector . for three copies of @xmath120 and @xmath3 fields the mass matrix of the neutral fermions can be written in exactly the same form as given in eq .
( [ eq : mass ] ) with some rather straight - forward generalizations of the above definitions .
these changes are : @xmath60 and @xmath170 become @xmath171 matrices @xmath172 and @xmath173 . in eq .
( [ eq : mchinuc ] ) the matrix becomes a @xmath174 matrix , @xmath76 is a symmetric @xmath171 matrix and eqs .
( [ eq : mnucp ] ) and ( [ eq : sp ] ) have to be replaced by @xmath175 @xmath176 where @xmath177 and @xmath178 .
notice that even with three generations of @xmath120 and @xmath3 fields one neutrino mass is zero at the tree - level .
the @xmath179 scalar mass matrix is a symmetric matrix that in the basis of the real part of @xmath180 can be written in the form , @xmath181 m^{s^2}_{h\widetilde
l}\!{}^t & m^{s^2}_{\widetilde l \widetilde l } & m^{s^2}_{\widetilde l s}\\[+2 mm ] m^{s^2}_{hs}\!{}^t & m^{s^2}_{\widetilde l s}\!{}^t & m^{s^2}_{ss } \end{array } \right]\end{aligned}\ ] ] where @xmath182 is a symmetric @xmath183 matrix , @xmath184 and @xmath185 are symmetric @xmath186 matrices , while @xmath187 and @xmath188 are @xmath189 matrices and finally @xmath190 is ( a non - symmetric ) @xmath191 matrix . in this notation @xmath192 denotes the sneutrinos and @xmath3 the singlet fields . we can write the mass matrix by giving the components of the various blocks .
we get , @xmath194 m^{s^2}_{hh_{12}}&=&-\frac{1}{4}(g^2+g'^2 ) v_d v_u -\omega + h_0 ^ 2 v_u v_d\\[+2 mm ] m^{s^2}_{hh_{22}}&=&\frac{1}{4}(g^2+g'^2 ) v_u^2 + \omega \cot\beta - \frac{\sqrt{2}}{2 } \frac{v_r}{v_u } \sum_{i=1}^3 a_{h_{\nu } } h_{\nu}^{i}\ , v_{li } -\frac{\sqrt{2}}{2}\ , \widehat{m}_r \frac{v_s}{v_u } \sum_{i=1}^3 h_{\nu}^{i } v_{li}\end{aligned}\ ] ] where , @xmath195 and @xmath0 , @xmath52 and @xmath53 are defined in eqs .
( [ eq : defmu ] ) and ( [ eq : effmr ] ) . @xmath197 @xmath199 m^{s^2}_{\widetilde l s_{i2}}&= & \frac{\sqrt{2}}{2 } \widehat{m}_r\ , v_u h_{\nu}^{i } \\[+2 mm ] m^{s^2}_{\widetilde l s_{i3}}&= & \frac{\sqrt{2}}{2}\ , v_u a_{h_{\nu } } h_{\nu}^{i } - \frac{\sqrt{2}}{2 } h_{\nu}^{i } \mu v_d + h_{\nu}^{i } v_r \sum_{k=1}^3 h_{\nu}^{k } v_{lk}\end{aligned}\ ] ] @xmath201 m^{s^2}_{h \widetilde l_{2i } } & = & -\frac{1}{4}(g^2+g'^2 ) v_u v_{li } + \frac{\sqrt{2}}{2}\ , v_r\ , a_{h_{\nu } } h_{\nu}^{i } + \frac{\sqrt{2}}{2 } \widehat{m}_r\ , v_s h_{\nu}^{i } + v_u\ , h_{\nu}^{i } \sum_{k=1}^3 h_{\nu}^{k } v_{lk}\end{aligned}\ ] ] @xmath203 m^{s^2}_{h s_{12}}&=&-\frac{1}{2 } h h_0\ , v_r v_u \\[+2 mm ] m^{s^2}_{h s_{13}}&=&-\frac{1}{2 } h h_0\ , v_s v_u - \frac{\sqrt{2}}{2 } \mu \sum_{k=1}^3 h_{\nu}^{k } v_{lk}\\[+2 mm ] m^{s^2}_{h s_{21}}&=&\sqrt{2 } h_0 \mu v_u
-\frac{\sqrt{2}}{2}h_0 \left(a_{h_0}+ \widehat{m}_{\phi}\right ) v_d + \frac{1}{2 } h\ , v_s \sum_{k=1}^3 h_{\nu}^{k } v_{lk } \\[+2 mm ] m^{s^2}_{h s_{22}}&=&-\frac{1}{2 } h h_0\ , v_r v_d + \frac{\sqrt{2}}{2 } \widehat{m}_r \sum_{k=1}^3 h_{\nu}^{k } v_{lk } \\[+2 mm ] m^{s^2}_{h s_{23}}&=&-\frac{1}{2 } h h_0\ , v_s v_d + v_u v_r \sum_{k=1}^3 h_{\nu}^{k } h_{\nu}^{k } + \frac{\sqrt{2}}{2 } \sum_{k=1}^3 a_{h_{\nu } } h_{\nu}^{k } v_{lk}\end{aligned}\ ] ] @xmath205 m^{s^2}_{s s_{12}}&\hskip-3mm=\hskip -3 mm & \frac{\sqrt{2}}{2 } h \left ( a_h + \widehat{m}_{\phi}\right ) v_r + \sqrt{2 } h\ , \widehat{m}_r v_s + \frac{1}{2 } h\ , v_u \sum_{k=1}^3 h_{\nu}^{k } v_{lk } \\[+2 mm ] m^{s^2}_{s s_{13}}&\hskip-3mm=\hskip -3 mm & \frac{\sqrt{2}}{2 } h \left ( a_h + \widehat{m}_{\phi}\right ) v_s -\frac{1}{2 } h_0 v_d \sum_{k=1}^3 h_{\nu}^{k } v_{lk } + \sqrt{2}\ , h\ , \widehat{m}_r v_r \\[+2 mm ] m^{s^2}_{s s_{22}}&\hskip-3mm=\hskip -3 mm & -\gamma \frac{v_r}{v_s } - \frac{\sqrt{2}}{2 } \frac{v_u}{v_s } \widehat{m}_r \sum_{k=1}^3 h_{\nu}^{k } v_{lk } \\[+2 mm ] m^{s^2}_{s s_{23}}&\hskip-3mm=\hskip -3 mm & \gamma + h^2 v_r v_s\\[+2 mm ] m^{s^2}_{s s_{33}}&\hskip-3mm=\hskip -3 mm & -\gamma \frac{v_s}{v_r } + \frac{\sqrt{2}}{2 } \frac{\mu v_d}{v_r } \sum_{k=1}^3 h_{\nu}^{k } v_{lk } - \frac{\sqrt{2}}{2 } \frac{v_u}{v_r } \sum_{k=1}^3 a_{h_{\nu } } h_{\nu}^{k } v_{lk } \end{aligned}\ ] ] where @xmath206
the @xmath179 pseudoscalar mass matrix is a symmetric matrix that can be written in the form , @xmath207 m^{p^2}_{h\widetilde l}\!{}^t & m^{p^2}_{\widetilde l \widetilde l } & m^{p^2}_{\widetilde l s}\\[+2 mm ] m^{p^2}_{hs}\!{}^t & m^{p^2}_{\widetilde l s}\!{}^t & m^{p^2}_{ss } \end{array } \right]\end{aligned}\ ] ] where the blocks have the same structure as before .
we can write the mass matrix by giving the components of the various blocks .
we get , @xmath209 m^{p^2}_{hh_{12}}&=&\omega \\[+2 mm ] m^{p^2}_{hh_{22}}&=&\omega \cot\beta - \frac{\sqrt{2}}{2 } \frac{v_r}{v_u } \sum_{i=1}^3 a_{h_{\nu } } h_{\nu}^{i}\ , v_{li } -\frac{\sqrt{2}}{2}\ , \widehat{m}_r \frac{v_s}{v_u } \sum_{k=1}^3 h_{\nu}^{k } v_{lk } \end{aligned}\ ] ] @xmath214 m^{p^2}_{\widetilde l s_{i2}}&= & \frac{\sqrt{2}}{2 } \widehat{m}_r\ , v_u\ , h_{\nu}^{i } \\[+2 mm ] m^{p^2}_{\widetilde l s_{i3}}&= & -\frac{\sqrt{2}}{2}\ , v_u a_{h_{\nu } } h_{\nu}^{i } + \frac{\sqrt{2}}{2 } h_{\nu}^{i } \mu v_d \end{aligned}\ ] ] @xmath218 m^{p^2}_{h s_{12}}&=&-\frac{1}{2 } h h_0\ , v_r v_u \\[+2 mm ] m^{p^2}_{h s_{13}}&=&-\frac{1}{2 } h h_0\ , v_s v_u - \frac{\sqrt{2}}{2 } \mu \sum_{k=1}^3 h_{\nu}^{k } v_{lk}\\[+2 mm ] m^{p^2}_{h s_{21}}&= & \frac{\sqrt{2}}{2}h_0 \left(a_{h_0}- \widehat{m}_{\phi}\right ) v_d + \frac{1}{2}\ , h\ , v_s \sum_{k=1}^3 h_{\nu}^{k } v_{lk } \\[+2 mm ] m^{p^2}_{h s_{22}}&=&-\frac{1}{2 } h h_0\ , v_r v_d + \frac{\sqrt{2}}{2}\widehat{m}_r \sum_{k=1}^3 h_{\nu}^{k } v_{lk } \\[+2 mm ] m^{p^2}_{h s_{23}}&=&-\frac{1}{2 } h h_0\ , v_s v_d - \frac{\sqrt{2}}{2 } \sum_{k=1}^3 a_{h_{\nu } } h_{\nu}^{k } v_{lk}\end{aligned}\ ] ] @xmath220 m^{p^2}_{s s_{12}}&\hskip-3mm=\hskip -3 mm & -\frac{\sqrt{2}}{2 } h \left ( a_h - \widehat{m}_{\phi}\right ) v_r -\frac{1}{2 } h\ , v_u \sum_{k=1}^3 h_{\nu}^{k } v_{lk } \\[+2 mm ] m^{p^2}_{s s_{13}}&\hskip-3mm=\hskip -3 mm & - \frac{\sqrt{2}}{2 } h \left ( a_h - \widehat{m}_{\phi}\right ) v_s -\frac{1}{2 } h_0 v_d \sum_{k=1}^3 h_{\nu}^{k } v_{lk}\\[+2 mm ] m^{p^2}_{s s_{22}}&\hskip-3mm=\hskip -3 mm & -\gamma \frac{v_r}{v_s } -\frac{\sqrt{2}}{2 } \widehat{m}_r \frac{v_u}{v_s } \sum_{k=1}^3 h_{\nu}^{k } v_{lk } \\[+2 mm ] m^{p^2}_{s s_{23}}&\hskip-3mm=\hskip -3 mm & -\gamma \\[+2 mm ] m^{p^2}_{s s_{33}}&\hskip-3mm=\hskip -3 mm & -\gamma \frac{v_s}{v_r } + \frac{\sqrt{2}}{2 } \frac{\mu v_d}{v_r } \sum_{k=1}^3 h_{\nu}^{k } v_{lk } - \frac{\sqrt{2}}{2 } \frac{v_u}{v_r } \sum_{k=1}^3 a_{h_{\nu } } h_{\nu}^{k } v_{lk } \end{aligned}\ ] ] super - kamiokande collaboration , y. fukuda _ et al .
* 81 * , 1562 ( 1998 ) , [ hep - ex/9807003 ] .
sno collaboration , q. r. ahmad _ et al .
* 89 * , 011301 ( 2002 ) , [ nucl - ex/0204008 ] .
t. araki et al . , [ kamland collaboration ] , arxiv : hep - ex/0406035 .
talk by g. gratta at neutrino 2004 , 14 19 june 2004 , paris , http://neutrino2004.in2p3.fr/ ; k. eguchi _
* 90 * , 021802 ( 2003 ) , [ hep - ex/0212021 ] .
m. maltoni , t. schwetz , m. a. tortola and j. w. f. valle , new j. phys .
* 6 * , 122 ( 2004 ) , [ hep - ph/0405172 ] .
see also m. maltoni et al , phys .
rev . * d68 * , 113010 ( 2003 ) , [ hep - ph/0309130 ] , and references therein .
m. gell - mann , p. ramond and r. slansky , ( 1979 ) , print-80 - 0576 ( cern ) .
r. n. mohapatra and g. senjanovic , phys . rev .
* d23 * , 165 ( 1981 ) . y. chikashige , r. n. mohapatra and r. d. peccei , phys .
* b98 * , 265 ( 1981 ) . j. schechter and j. w. f. valle , phys
* d22 * , 2227 ( 1980 ) .
phys . rev .
* d25 * , 774 ( 1982 ) .
r. n. mohapatra and j. w. f. valle , phys . rev .
* d34 * , 1642 ( 1986 ) .
a. zee , phys . lett . *
b93 * , 389 ( 1980 ) .
k. s. babu , phys
b203 * , 132 ( 1988 ) .
j. w. f. valle , prog .
nucl . phys .
* 26 * , 91 ( 1991 ) .
a. s. joshipura and j. w. f. valle , nucl .
b397 * , 105 ( 1993 ) . c. s. aulakh and r. n. mohapatra , phys .
b * 119 * , 136 ( 1982 ) .
g. g. ross and j. w. f. valle , phys .
b * 151 * , 375 ( 1985 ) .
j. r. ellis _ et al .
_ , phys .
b * 150 * , 142 ( 1985 ) .
l. j. hall and m. suzuki , nucl .
b * 231 * , 419 ( 1984 ) .
a. masiero and j. w. f. valle , phys
b251 * , 273 ( 1990 ) . j. c. romao , c. a. santos and j. w. f. valle , phys . lett . *
b288 * , 311 ( 1992 ) .
m. shiraishi , i. umemura and k. yamamoto , phys . lett . *
b313 * , 89 ( 1993 ) .
m. a. diaz , j. c. romao and j. w. f. valle , nucl .
b524 * , 23 ( 1998 ) , [ hep - ph/9706315 ] .
e. j. chun and s. k. kang , phys
* d61 * , 075012 ( 2000 ) , [ hep - ph/9909429 ] .
a. abada , s. davidson and m. losada , phys . rev . *
d65 * , 075010 ( 2002 ) , [ hep - ph/0111332 ] . m. a. diaz _
et al . _ ,
phys . rev . * d68 * ,
013009 ( 2003 ) , [ hep - ph/0302021 ] . m. hirsch _
_ , phys . rev . *
d62 * , 113008 ( 2000 ) , [ hep - ph/0004115 ] , err - ibid .
* d65*:119901,2002 .
j. c. romao _
_ , phys . rev . *
d61 * , 071703 ( 2000 ) , [ hep - ph/9907499 ] .
m. hirsch and j. w. f. valle , new j. phys .
6 76 ( 2004 ) , hep - ph/0405015 .
m. hirsch _
et al . _ ,
phys . rev . * d66 * , 095006 ( 2002 ) , [ hep - ph/0207334 ] .
w. porod _ et al .
_ , phys . rev . *
d63 * , 115004 ( 2001 ) , [ hep - ph/0011248 ] .
d. restrepo , w. porod and j. w. f. valle , phys .
* d64 * , 055011 ( 2001 ) , [ hep - ph/0104040 ] .
m. hirsch and w. porod , phys . rev .
* d68 * , 115007 ( 2003 ) , [ hep - ph/0307364 ] .
f. de campos _ et al .
_ , nucl . phys . *
b451 * , 3 ( 1995 ) , [ hep - ph/9502237 ] . c. panagiotakopoulos and k. tamvakis , phys
b446 * , 224 ( 1999 ) , [ hep - ph/9809475 ] .
j. c. romao , f. de campos and j. w. f. valle , phys .
b292 * , 329 ( 1992 ) , [ hep - ph/9207269 ] .
r. barbieri , s. ferrara and c. a. savoy , phys .
b119 * , 343 ( 1982 ) .
delphi collaboration , p. abreu _ et al .
b502 * , 24 ( 2001 ) , [ hep - ex/0102045 ] .
j. c. romao , a. ioannisian and j. w. f. valle , phys .
d * 55 * ( 1997 ) 427 [ arxiv : hep - ph/9607401 ] .
a. lopez - fernandez _
et al . _ ,
. lett . * b312 * , 240 ( 1993 ) , [ hep - ph/9304255 ] .
f. de campos _ et al .
_ , phys . rev . * d55 * , 1316 ( 1997 ) , [ hep - ph/9601269 ] . j. abdallah _ et al .
_ [ delphi collaboration ] , eur .
j. c * 32 * ( 2004 ) 475 [ arxiv : hep - ex/0401022 ] .
b. c. allanach _
et al . _ ,
j. * c25 * , 113 ( 2002 ) , [ hep - ph/0202233 ] .
w. porod , comput .
commun . 153
( 2003 ) 275 [ arxiv : hep - ph/0301101 | the higgs boson may decay mainly to an invisible mode characterized by missing energy , instead of the standard model channels .
this is a generic feature of many models where neutrino masses arise from the spontaneous breaking of ungauged lepton number at relatively low scales , such as spontaneously broken r - parity models . taking these models as framework , we reanalyze this striking suggestion in view of the recent data on neutrino oscillations that indicate non - zero neutrino masses .
we show that , despite the smallness of neutrino masses , the higgs boson can decay mainly to the invisible goldstone boson associated to the spontaneous breaking of lepton number .
this requires a gauge singlet superfield coupling to the electroweak doublet higgses , as in the next to minimal supersymmetric standard model ( nmssm ) scenario for solving the @xmath0-problem .
the search for invisibly decaying higgs bosons should be taken into account in the planning of future accelerators , such as the large hadron collider and the next linear collider . |
the theoretical description of the electronic properties of transition metal compounds with partially filled @xmath1 and @xmath2-shells and strong coulomb interaction between the electrons remains a challenging , fundamental problem in condensed matter physics .
@xcite the interplay between electronic and lattice degrees of freedom in such materials results in their diverse physical properties and rich phase diagrams making these compounds particularly attractive for technological applications .
@xcite moreover , orbital degeneracy is an important and often inevitable cause of this complexity .
together with the hund s exchange interaction , it has important implications for the electronic and magnetic properties of correlated materials , leading to formation of local moments and complicated multiplet structures .
the electronic properties of correlated materials can be understood by employing the so - called lda+dmft approach , @xcite a combination of _ ab initio _ local density approximation ( lda ) of the density functional theory and dynamical mean - field theory ( dmft ) .
nowadays , the lda+dmft technique has become a state - of - the - art method for realistic description of correlated electron materials from first principles .
this approach provides a systematic many - body treatment of the effect of local electronic correlations by taking into account temporal fluctuations while spatial fluctuations are neglected .
applications of lda+dmft for correlated electron compounds such as transition metals and their oxides have provided important insights into our understanding of the electronic and magnetic properties of these materials . in particular , by employing the lda+dmft technique it has become possible to obtain a good quantitative description of localized as well as delocalized electron states .
in addition , the approach allows one to determine the electronic and magnetic properties of correlated compounds in both paramagnetic and magnetically ordered states . nevertheless , there are two important limitations of conventional implementations of the lda+dmft method .
the first originates from the single - site ( local ) nature of dmft .
in particular , the key assumption of the theory is the limit of infinite spatial dimension , which allows one to perform an exact mapping of a complex lattice model ( such as the hubbard model ) to a quantum impurity with an energy - dependent external bath , resulting in @xmath3-independent self - energy .
however , in some cases the non - local spatial correlations can be essential to provide a correct description of the properties of correlated materials .
@xcite for instance , the standard lda+dmft calculations are not able to capture the reduction of magnetic transition temperature due to long - wavelength spin waves . to resolve this problem several methods
have been recently proposed , @xcite which we leave beyond the scope of our paper .
the second limitation concerns the spin rotational symmetry of the hund s exchange interaction .
since correlated materials often have several bands at the fermi level , it requires specific treatment of the local coulomb interaction , which in a cubic environment consists of the intra- and inter - orbital coulomb interactions @xmath4 and @xmath5 , hund s exchange @xmath6 , and the pair - hopping coupling @xmath7 .
these interactions obey spin and orbital rotational symmetry , thereby @xmath8 ensures the rotational invariance in the orbital space and @xmath9 can be assumed whenever the spin - orbital coupling is negligible .
unfortunately , it is difficult to handle all these multiband interactions including hund s exchange coupling and the pair hopping term with the quantum monte carlo ( qmc ) method .
in particular , a straightforward implementation leads to a severe sign problem making such simulations unfeasible .
therefore , at present , the most material - specific calculations employ the approximate form of the coulomb repulsion restricted to the ising - type exchange interaction .
these calculations often provide a good quantitative description of the electronic , magnetic , and structural properties of correlated materials as a function of the reduced temperature @xmath10 , where @xmath11 is the calculated temperature of magnetic ordering .
@xcite however , the correct symmetry of the exchange interaction turns out to be essential for quantitative description of the electronic and magnetic properties of correlated systems .
@xcite this problem can be overcome by using some quantum impurity solvers such as numerical renormalization group ( nrg ) , @xcite exact diagonalization ( ed ) , @xcite continuous - time quantum monte carlo ( ct - qmc ) , @xcite and others , @xcite which allow one to treat the coulomb interaction in its general form with preserved spin rotational symmetry .
these calculations performed for the two- and three - band hubbard models on the infinite - dimensional bethe @xcite and hypercubic @xcite lattices show a substantial overestimation of the magnetic transition temperature for the approximate ising - type form of the exchange coulomb interaction with respect to the rotationally invariant one . in accordance with this
, recent lda+dmft calculations of correlated compounds also indicate that the magnetic transition temperatures appear to be significantly overestimated by using the density - density type of coulomb interaction .
@xcite however , applications of these techniques so far have been mostly limited to simple model systems and only a few realistic calculations for 3@xmath1 compounds have been recently presented .
@xcite this is mostly because of high computational costs ( exponential with the number of orbitals ) of these methods implemented with the full rotationally invariant coulomb interaction which makes such calculations for @xmath12 and @xmath13 materials extremely expensive . obviously , the lda+dmft investigations of correlated materials with the coulomb interaction in its general form with preserved spin rotational symmetry remain problematic and pose a great theoretical challenge . in this paper , we present an implementation of the lda+dmft approach which allows us to take into account rotational symmetry of the exchange coulomb interaction .
this approach is formulated in terms of the hirsch - fye quantum monte carlo algorithm @xcite where the spin rotational invariance of hund s exchange is approximated by averaging over all possible directions of the spin quantization axis .
it provides a robust and computationally efficient method which allows us to simulate the five - orbital systems at high temperatures . using this technique we perform benchmark calculations for the two- and three - band hubbard models on the infinite - dimensional bethe lattice .
in addition , we employ the proposed approach to calculate the electronic and magnetic properties of paramagnetic @xmath0 iron and nickel . to outline the importance of rotational symmetry of the exchange coulomb interaction we compare our results with those obtained by using the density - density approximation .
this paper is organized as follows . in sec .
[ sec : method ] we present a detailed formulation of the proposed approach which allows one to treat rotational invariance of the exchange interaction . in sec .
[ sec : results ] we employ this technique to compute the electronic and magnetic properties of the two- and three - band models on the bethe lattice , paramagnetic @xmath0 iron , and nickel .
the obtained results are compared with those of previous calculations and experimental data .
finally , conclusions are presented in sec .
[ sec : conclusions ] .
the multiband hamiltonian with full rotationally invariant on - site coulomb interaction can be written in the following form : @xcite @xmath14 where @xmath15 ( @xmath16 ) denotes the creation ( annihilation ) operator of an electron with spin @xmath17 ( @xmath18 ) at orbital @xmath19 , @xmath20 , @xmath4 is the screened coulomb interaction parameter , and @xmath6 is the hund s exchange coupling .
the first three terms in hamiltonian ( [ full_ham ] ) correspond to the density - density part of coulomb interaction and contain the exchange interaction in the ising - type form .
the remaining part consists of spin - flip ( 4th ) and pair hopping ( 5th ) terms . using the @xmath21-projection of the spin operator , @xmath22 , and the orbital occupancy operator , @xmath23 ,
the density - density part can be rewritten as @xmath24 where @xmath25 is the average value of the coulomb interaction .
the spin - flip term in eq . can be expressed via operators @xmath26 and @xmath27 as @xmath28 the pair hopping term acts only on high energy states with two electrons on the same orbital and thereby can be neglected . however , taking into account the spin - flip term in eq .
( [ full_ham ] ) is crucial for the correct description of spin dynamics .
therefore , the coulomb interaction hamiltonian can be written as @xmath29 hamiltonian ( 4 ) with exchange term taken as a vector product , @xmath30 , is invariant with respect to the spin quantization axis rotations while the density - density counterpart with the ising - type exchange term , @xmath31 , is not . to restore the spin rotational symmetry of hamiltonian ( [ dd_ham ] )
we here employ the method originally proposed by hubbard .
@xcite the coulomb interaction in ref .
was considered in the following form : @xmath32 where @xmath33 , @xmath34 , @xmath35 , @xmath36 is the total spin of the atom , and @xmath37 is an arbitrary unit vector that can be interpreted as a quantization axis direction . in order to restore the spin rotational symmetry in eq .
( [ from_hubbard ] ) , averaging over all possible directions of @xmath37 was introduced using the functional integral technique .
@xcite here we implement this method with the hirsch - fye quantum monte carlo algorithm ( ) .
@xcite the is based on the discrete hubbard - stratonovich transformation which employs the identity @xmath38 \nonumber \\ & & = \dfrac{1}{2 } \sum_{s_{\mu\nu}=\pm 1 } \textrm{exp } \ { \lambda_{\mu\nu } s_{\mu\nu } ( \hat{n}_\mu - \hat{n}_\nu ) \},\end{aligned}\ ] ] where @xmath39 and @xmath40 are combined spin - orbital indices
, @xmath41 is an ising - like variable taking the values @xmath42 , @xmath43 stands for the matrix element of the coulomb interaction operator , and @xmath44}$ ] . the imaginary time interval @xmath45 $ ] is divided into @xmath46 slices of length @xmath47 , so that @xmath48 , where @xmath49 , and @xmath50 denotes the inverse temperature . using the trotter decomposition
, the partition function of the system can be approximated as @xmath51 where @xmath52 is the non - interacting ( quadratic in fermion operators ) part of hamiltonian ( 1 ) and @xmath53 describes the coulomb interaction .
therefore , the partition function can be written as a sum over all auxiliary field configurations : @xmath54 where @xmath55 is the partition function for a particular configuration of auxiliary fields , @xmath56 denotes the set of all auxiliary fields , and @xmath57 is the number of auxiliary fields for @xmath58 orbitals . in the case of the density - density form of the local coulomb interaction ,
the single - electron dynamical potential has only the @xmath21 component and can be expressed as @xmath59 for the quantization axis defined by polar angle @xmath60 and azimuthal angle @xmath61 , the single - electron dynamical potential can be written as @xmath62 where @xmath63 is the potential calculated by eq .
( [ p1 ] ) for the quantization axis chosen to coincide with the @xmath21 axis , and @xmath64 stands for a transformation matrix in the spin variables and reads as @xmath65 by integrating over all possible directions of the quantization axis , we obtain the partition function of the system with preserved spin rotational symmetry : @xmath66 here , @xmath67 is the partition function for a particular configuration of auxiliary fields with the quantization axis defined by angles @xmath60 and @xmath61 . similarly to the original hf - qmc algorithm , it can be demonstrated that @xmath68,\end{aligned}\ ] ] where @xmath69 is the green s function for a particular configuration of auxiliary fields , angles @xmath60 and @xmath61 .
the resulting green s function has the following form @xmath70 where @xmath71 . to integrate over all auxiliary field configurations and all possible directions of the quantization axis in eq .
( [ new_gf ] ) , we employ the quantum monte carlo technique in which @xmath72}$ ] is interpreted as a stochastic weight .
similarly to the original algorithm , the green s functions of two configurations with potentials @xmath73 and @xmath74 are related to each other as @xmath75 where @xmath76 denotes the unit matrix . however , in contrast to the hf - qmc , eq .
( [ new_gf2 ] ) now contains off - diagonal in the spin indices elements . in the case of a single auxiliary spin - flip
, the fast matrix update algorithm has the same form as in the hf - qmc method .
note , however , that all equations have the matrix form in the spin indices .
the ratio of stochastic weights for two configurations which differ by the quantization axis direction can be calculated as @xmath77}{\textrm{det}[g'(s,\theta',\phi ' ) ] } = \textrm{det}[a].\end{aligned}\ ] ] if the new quantization axis direction defined by angles @xmath78 and @xmath79 is accepted , the corresponding green s function @xmath80 is calculated via the full update procedure using eq .
( [ new_gf2 ] ) .
the physical meaning of the proposed rotationally invariant algorithm can be expressed as averaging over all possible directions of fluctuating spin polarization in the 3@xmath1 shell , in contrast to polarization along the @xmath21-axis only in the density - density hamiltonian .
thereby this technique allows one to take into account not only the longitudinal spin fluctuations but also the transverse ones .
in this section , we first present results of our model calculations in comparison with previous studies . in particular , we employ the proposed approach to compute the two- and three - band hubbard models on the infinite - dimensional bethe lattice .
we benchmark these calculations with the previously published results of the ct - qmc computations .
@xcite to study the role of symmetry of the exchange interaction we compare our results with those obtained by employing the density - density form of the local coulomb interaction .
in particular , we compute the uniform magnetic susceptibility @xmath81 by calculating magnetization @xmath82 induced by the external magnetic field @xmath83 applied along the @xmath21 axis ( these calculations include the polarization of the impurity weiss field ) . in our calculations
, we adopted a few magnetic fields in the range from 0.01 to 0.04 ev to ensure linearity of response .
the temperature dependence of magnetic susceptibility @xmath84 is fitted to the curie - weiss law @xmath85 , where @xmath86 is a material - specific constant and @xmath87 is the curie temperature .
next we investigate the electronic and magnetic properties of paramagnetic iron and nickel .
we outline the importance of the correct spin rotational symmetry of the exchange interaction to describe the properties of these materials .
these results are presented in secs .
iii b and iii c. in recent years the properties of the two- and three - orbital hubbard models have been extensively investigated by using dynamical mean - field approach .
@xcite here we would like to refer to ref . where the two- and three - orbital hubbard models on the infinite - dimensional bethe lattice were studied by means of the continuous - time quantum monte carlo method with full rotationally invariant coulomb interaction .
in particular , it was established that the effect of spin - flip interactions considerably depends on a band filling .
it was shown that the mott - hubbard physics dominates due to the strong effective coulomb interaction @xmath4 in the particle - hole symmetric case , while away from half - filling formation of local magnetic moments is more plausible due to the hund s exchange . in agreement with previous studies,@xcite
it was also shown that using the density - density form of the coulomb interaction results in an overestimation of the magnetic transition temperature .
for the benchmark purposes we here reproduce these calculations by employing the proposed rotationally invariant method . in fig .
[ fig : model_susc ] ( upper panel ) we present our results for the inverse uniform magnetic susceptibility calculated for the two - band hubbard model on the infinite - dimensional bethe lattice .
we have chosen the same set of parameters as in ref . .
namely , we consider the two - band model at half - filling with @xmath88 for coulomb interaction and @xmath89 for hund s exchange , where @xmath90 is a half of the non - interacting bandwidth .
the calculated curie temperatures are 0.49@xmath90 and 0.39@xmath90 for the conventional and rotationally invariant hf - qmc , respectively .
these findings are in good quantitative agreement with the results of the ct - qmc calculations which give @xmath91 and @xmath92 for the density - density and rotationally invariant interactions .
similar calculations with @xmath93 ( see table [ tab : model ] ) give @xmath91 and @xmath94 for the density - density and rotationally invariant hf - qmc methods , respectively .
we notice that the effective local moments calculated by the conventional and rotationally invariant hf - qmc methods agree well with those obtained by the ct - qmc . .
the ct - qmc results were taken from ref . .
[ fig : model_susc],scaledwidth=46.0% ] to proceed further , we investigate the properties of the three - band hubbard model with @xmath93 and @xmath89 at one - third electron filling ( two - electron occupancy ) . in fig .
[ fig : model_susc ] ( lower panel ) we present our results for the inverse uniform magnetic susceptibility . our results for @xmath95 and half - filling are summarized in table [ tab : model ] . for both sets of parameters ,
the broken rotational symmetry of the coulomb interaction leads to an overestimation of the curie temperature .
however , in agreement with ref . , this overestimation is found to be more pronounced at the @xmath96 filling than at the half - filling .
this is due to the hund s exchange interaction which plays a dominating role away from half - filling .
the curie temperatures calculated by the rotationally invariant hf - qmc algorithm are @xmath97 and @xmath98 for the @xmath99 and @xmath96 electron filling , respectively .
these findings are in good quantitative agreement with the results of rotationally invariant ct - qmc calculations which give @xmath100 and @xmath101 for the half - filling and one - third filling , respectively .
our findings clearly indicate that the curie temperature is overestimated by employing the density - density form of coulomb interaction .
hence , the retaining of spin rotational symmetry is crucial for the correct description of the magnetic transition temperature .
we find good quantitative agreement between the results obtained by rotationally invariant hf - qmc and ct - qmc methods .
this demonstrates the validity of our method for accurate description of the magnetic properties of correlated electron systems .
we note that transverse spin fluctuations can be regarded as an important source of magnetic response softening .
it is expected that the proper treatment of the spin rotational symmetry of coulomb interaction is even more important in the five - band case .
.curie temperatures ( in units of @xmath90 ) and effective local magnetic moments ( in @xmath102 ) as obtained by dmft for the two- and three - band hubbard models on the infinite - dimensional bethe lattice .
the results corresponding to the density - density interaction are denoted as hf - qmc and .
the calculations were carried out with @xmath103 .
the results were taken from ref . .
[ tab : model ] [ cols="^,^,^,^,^,^",options="header " , ] elemental iron is one of the oldest and experimentally best studied itinerant ferromagnets .
various properties of iron can be understood on the basis of band - structure calculations .
@xcite in particular , these calculations provide a good description of the low - temperature ferromagnetic phase of fe .
however , applications of conventional band - structure techniques to describe the properties of paramagnetic iron , in particular , close to the @xmath104 phase transition , do not lead to satisfactory results .
this is mainly due to the presence of local magnetic moments above the curie temperature which are important for quantitative description of paramagnetic state . in this respect ,
the lda+dmft approach provides the best formalism which allows one to unify the localized and itinerant electron behavior in metallic magnets .
( black ) and @xmath105 ( red ) densities of states for paramagnetic @xmath0 iron as obtained by the rotationally invariant hf - qmc within lda+dmft in comparison with the non - magnetic lda results .
the fermi level is indicated by the vertical ( gray ) line at zero energy .
inset : imaginary parts of the obtained self - energies .
[ fig : fe_dos_1spin],scaledwidth=46.0% ] applications of lda+dmft have shown to provide a good quantitative description of the electronic , magnetic , and structural properties of iron .
@xcite however , an agreement was achieved only in terms of the reduced temperature @xmath106 , while the calculated curie temperature @xmath11 was found to be almost twice larger than the experimental value of @xmath107 k ( ref . ) .
recently the properties of iron have been investigated by means of @xmath108-qmc method , which uses the static approximation for the charge degrees of freedom and treats the exchange coulomb interaction in the rotationally invariant form .
@xcite these calculations indicate that substantial part of the curie temperature overestimation comes from the approximate ( density - density ) treatment of the exchange coulomb interaction .
we now calculate the electronic structure and magnetic properties of paramagnetic bcc iron by employing the lda+dmft implemented with the rotationally invariant hf - qmc method .
we first calculate the non - magnetic lda electronic structure of @xmath0 iron using the tight - binding linear muffin - tin orbital ( tb - lmto ) approach .
@xcite for these calculations we adopt lattice constant @xmath109 .
we construct an effective low - energy hamiltonian in the basis of fe @xmath110 wannier orbitals using the @xmath111th - order muffin - tin ( @xmath111mto ) method .
@xcite here we adopt common definitions for the screened coulomb interaction and hund s exchange parameters in the @xmath12 shell , namely , @xmath112 and @xmath113 , where @xmath114 , @xmath115 , and @xmath116 are the slater integrals .
we take @xmath117 ev and @xmath118 ev in accordance with the previous estimations @xcite and solve the five - orbital impurity problem within dmft .
iron as obtained by the lda+dmft .
the straight lines depict the least - squares fit to the curie - weiss law .
the experimental value @xmath119 k is denoted by the ( black ) arrow .
the experimental value of the effective local magnetic moment is @xmath120 ( ref . ) .
[ fig : fe_susc],scaledwidth=46.0% ] in fig .
[ fig : fe_dos_1spin ] we present the partial density of states and the corresponding imaginary parts of the self - energies obtained by the rotationally invariant hf - qmc method at @xmath121 k in comparison with the non - magnetic lda results .
our calculations reproduce the splitting in the density of states of the @xmath105 orbitals near the fermi level , which is absent in the lda calculations .
this result agrees well with the previous calculations @xcite as well as with the experimental data and is one of the characteristic features of @xmath0 iron .
the calculated self - energy for the @xmath122 states exhibits a fermi - liquid - like behavior , whereas the @xmath105 self - energy diverges at low frequencies .
our calculations with @xmath123 recover the fermi - liquid - like behavior for the @xmath105 states resulting in the suppression of the splitting .
this indicates that the splitting can be attributed to the exchange coulomb interaction .
@xcite we found no evidence for the hubbard subbands formation in the calculated quasiparticle spectrum .
thereby we conclude that in @xmath0 iron the correlation effects are mainly affected by the strength of the hund s coupling @xmath6 rather than by the coulomb interaction @xmath4 which allow us to refer @xmath0 iron as a hund s metal .
we note that importance of the hund s exchange in multiorbital systems has been recently studied in ref . .
the temperature dependence of the inverse uniform magnetic susceptibility calculated by the lda+dmft shows a linear behavior at high temperatures ( fig .
[ fig : fe_susc ] ) in accordance with the curie - weiss law .
it is clearly seen that the limited to the ising - type exchange interaction substantially overestimates the curie temperature value and yields @xmath124 k. the rotationally invariant method gives @xmath125 k , which is in good agreement with the experimental value of @xmath107 k ( ref . ) .
the calculated values of the effective local moment extracted from the uniform magnetic susceptibility are @xmath126 and @xmath127 for the hf - qmc with density - density and rotationally invariant interactions , respectively .
our estimates are in reasonable agreement with the experimental value of @xmath128 ( ref . ) , which appears to be underestimated by both methods .
we note that preserving the spin rotational symmetry turns out to be crucial for quantitative description of magnetic properties of correlated materials .
in particular , for @xmath0 iron this leads to substantial improvement of the magnetic transition temperature value .
elemental nickel is another example of itinerant electron ferromagnets which together with iron serves as a benchmark material for electronic structure methods .
various low temperature properties of nickel can be understood by employing standard band - structure approaches .
@xcite nevertheless , these techniques generally fail to reproduce many characteristic features of nickel , such as an existence of satellite structure @xcite at @xmath129 ev , @xmath12 electron bandwidth , @xcite and the value of exchange splitting .
@xcite applications of lda+dmft to study the electronic and magnetic properties of nickel have given a good quantitative description of many of these phenomena .
@xcite the calculated magnetic properties of nickel are shown to be in good agreement with experiment .
in contrast to iron , the overestimation of the magnetic transition temperature for nickel by lda+dmft with the ising - type exchange interaction is not so significant .
@xcite however , as demonstrated below , preserving the spin rotational symmetry leads to the underestimation of the curie temperature .
we now compute the electronic structure and magnetic properties of paramagnetic nickel by employing the lda+dmft implemented with rotationally invariant hf - qmc method . to obtain the non - magnetic lda electronic structure of fcc nickel we employ the tight - binding linear muffin - tin orbital approach .
the calculations were performed for the lattice constant @xmath130 . using these results
we construct an effective low - energy hamiltonian in the basis of ni @xmath110 wannier orbitals by employing the @xmath111mto method . in accordance with the previous estimations ,
@xcite we take @xmath117 ev and @xmath131 ev for the screened coulomb interaction and hund s exchange , respectively . , scaledwidth=46.0% ] in fig
[ fig : ni_dos_1spin ] we present the partial densities of states and the imaginary parts of the self - energies obtained by the rotationally invariant algorithm at @xmath132 k in comparison with the non - magnetic lda results .
the inclusion of electronic correlations results in a small reduction of the bandwidth of nickel with respect to the non - magnetic lda result .
in addition , a satellite - like structure emerges at about @xmath133 ev .
in contrast to iron , the obtained self - energies for both the @xmath122 and @xmath105 orbitals exhibit the fermi - liquid - like behavior .
this can also be seen from the calculated amplitudes of the effective damping @xmath134 which are @xmath135 ev and @xmath136 ev for the @xmath122 and @xmath105 states of nickel , respectively , while @xmath137 ev for the @xmath122 states of @xmath0 iron .
this indicates a more coherent nature of the electronic properties of nickel in comparison with iron . to quantify this qualitative difference @xcite we calculate spin - spin correlation functions for iron and nickel . in fig .
[ fig : sisj ] we present the impurity spin - spin correlation functions on the real and imaginary energy axes calculated for @xmath0 iron and nickel at @xmath138 ( @xmath87 refers here to the corresponding calculated value of the curie temperature ) .
the height of peak on the real energy axis can be interpreted as a value of the local magnetic moment .
the pronounced peak for iron indicates the presence of well localized magnetic moments above @xmath87 , while magnetism of nickel is more itinerant . iron and nickel calculated by lda+dmft at @xmath139 .
[ fig : sisj],scaledwidth=46.0% ] in fig .
[ fig : ni_susc ] we show the temperature dependence of the inverse uniform magnetic susceptibility calculated by lda+dmft . from these data
we estimate the values of the curie temperature . by employing the density - density approximation
we find @xmath140 k , whereas the inclusion of the spin rotational symmetry leads to an almost twice smaller value of about 400 k. both results are in reasonable agreement with the experimental value of 631 k. to proceed further , we compute the effective local magnetic moments by the hf - qmc and rotationally invariant methods which give @xmath141 and @xmath142 , respectively .
these findings are in good agreement with the experimental value of @xmath143 ( ref . ) . the fact that in @xmath0 iron the lda+dmft with proper spin rotational symmetry results in an overestimated value of @xmath11 , while it appears to be underestimated in nickel , can be dealt with non - local effects which are neglected in dmft or with the more itinerant nature of magnetism in nickel than in @xmath0 iron .
k is denoted by the ( black ) arrow .
the experimental value of the effective local magnetic moment is @xmath144 ( ref . ) .
[ fig : ni_susc],scaledwidth=46.0% ]
in conclusion , we have presented an implementation of the lda+dmft approach which allows one to take into account the spin rotational symmetry of the exchange coulomb interaction .
the computational scheme is based on extension of the hirsch - fye quantum monte carlo algorithm in which the spin rotational invariance of hund s exchange is approximated by averaging over all possible directions of the spin quantization axis .
the proposed approach provides a robust and computationally efficient method which allows us to compute high temperature electronic properties of the five - orbital systems .
we have used this approach to perform benchmark calculations for the two- and three - band hubbard models on the infinite - dimensional bethe lattice .
our results agree quantitatively well with those obtained using the continuous - time quantum monte carlo technique .
the proposed method is employed to compute the electronic and magnetic properties of paramagnetic @xmath0 iron and nickel .
the obtained curie temperatures agree well with experiment .
our results indicate that the density - density approximation for the coulomb interaction leads to a substantial overestimation of the magnetic transition temperature .
the authors thank t. costi , e. dagotto , a. liebsch , a. poteryaev , and s. sakai for valuable comments .
the authors are also grateful to a. antipov for providing the ct - qmc data .
this work was supported by the russian foundation for basic research ( projects nos .
10 - 02 - 00046a , 12 - 02 - 91371-ct@xmath145a , 12 - 02 - 31207 ) , the ministry of education and science of russian federation through projects 14.a18.21.0076 , 12.740.11.0026 , the fund of the president of the russian federation for the support of scientific schools nsh-6172.2012.2 , the dynasty foundation , program of the russian academy of science presidium `` quantum microphysics of condensed matter '' 12-p-2 - 1017 , 12-m-23 - 2020 .
support by the deutsche forschergemeinschaft through trr 80 and for 1346 is gratefully acknowledged .
w. metzner and d. vollhardt , phys .
lett . * 62 * , 324 ( 1989 ) ; g. kotliar and d. vollhardt , phys . today *
57*(3 ) , 53 ( 2004 ) ; a. georges , g. kotliar , w. krauth , and m. j. rozenberg , rev .
phys . * 68 * , 13 ( 1996 ) .
k. held , i. a. nekrasov , g. keller , v. eyert , n. blmer , a. k. mcmahan , r. t. scalettar , th .
pruschke , v. i. anisimov , and d. vollhardt , phys .
status solidi b * 243 * , 2599 ( 2006 ) ; k. held , adv .
phys . * 56 * , 829 ( 2007 ) ; j. kunes , i. leonov , m. kollar , k. byczuk , v. i. anisimov , and d. vollhardt , eur .
j. special topics * 180 * , 5 ( 2010 ) ; v. anisimov and y. izyumov , _ electronic structure of strongly correlated materials _
( springer , berlin , 2010 ) .
s. biermann , a. poteryaev , a. i. lichtenstein , a. georges , phys .
lett . * 94 * , 026404 ( 2005 ) ; a. kutepov , k. haule , s. y. savrasov , and g. kotliar , phys . rev .
b * 82 * , 045105 ( 2010 ) ; b. lazarovits , k. kim , k. haule , and g. kotliar , _ ibid . _ * 81 * , 115117 ( 2010 ) ; a. s. belozerov , m. a. korotin , v. i. anisimov , a. i. poteryaev , _ ibid . _
* 85 * , 045109 ( 2012 ) .
a. i. lichtenstein and m. i. katsnelson , phys .
b * 62 * , r9283 ( 2000 ) ; g. kotliar , s. y. savrasov , g. palsson , and g. biroli , phys . rev . lett .
* 87 * , 186401 ( 2001 ) ; p. sun and g. kotliar , phys .
b * 66 * , 085120 ( 2002 ) ; s. biermann , f. aryasetiawan , and a. georges , phys . rev . lett . * 90 * , 086402 ( 2003 ) ; m. potthoff , m. aichhorn , and c. dahnken , _ ibid . _ * 91 * , 206402 ( 2003 ) ; a. toschi , a. a. katanin , and k. held , phys .
b * 75 * , 045118 ( 2007 ) ; a. n. rubtsov , m. i. katsnelson , and a. i. lichtenstein , _ ibid . _ * 77 * , 033101 ( 2008 ) .
a. n. rubtsov , v. v. savkin , and a. i. lichtenstein , phys .
b * 72 * , 035122 ( 2005 ) ; p. werner , a. comanac , l. de medici , m. troyer , and a. j. millis , phys . rev
97 * , 076405 ( 2006 ) ; k. haule , phys . rev .
b * 75 * , 155113 ( 2007 ) ; a. m. lauchli and p. werner , _ ibid . _ * 80 * , 235117 ( 2009 ) ; e. gull _
et al . _ , rev .
phys . * 83 * , 349 ( 2011 ) .
t. a. costi
_ et al . _ ,
102 * , 056802 ( 2009 ) ; e. gorelov , t. o. wehling , a. n. rubtsov , m. i. katsnelson , and a. i. lichtenstein , phys . rev .
b * 80 * , 155132 ( 2009 ) ; v. v. mazurenko , s. n. iskakov , a. n. rudenko , v. i. anisimov , and a. i. lichtenstein , _ ibid . _
* 82 * , 193403 ( 2010 ) ; m. aichhorn , s. biermann , t. miyake , a. georges , and m. imada , _ ibid . _ * 82 * , 064504 ( 2010 ) ; z. p. yin , k. haule , and g. kotliar , nature mat .
* 10 * , 932 ( 2011 ) ; b. surer , m. troyer , p. werner , t. o. wehling , a. m. lauchli , a. wilhelm , and a. i. lichtenstein , phys . rev .
b * 85 * , 085114 ( 2012 ) . strictly speaking , this parametrization of the hamiltonian is appropriate for the two- and three - orbital systems .
the five - band calculations require three effective slater integrals ( @xmath114 , @xmath115 , and @xmath116 ) which can be linked to the @xmath4 and @xmath6 parameters .
p. werner and a. j. millis , phys .
lett . * 99 * , 126405 ( 2007 ) ; p. werner , e. gull , m. troyer , and a. j. millis , _ ibid . _
* 101 * , 166405 ( 2008 ) ; p. werner , e. gull , and a. j. millis , phys .
b * 79 * , 115119 ( 2009 ) .
o. k. andersen and o. jepsen , phys .
lett . * 53 * , 2571 ( 1984 ) ; o. k. andersen , z. pawlowska , and o. jepsen , phys .
b * 34 * , 5253 ( 1986 ) .
the exchange - correlation potential was considered in the form of von barth and hedin [ u. von barth and l. hedin , j. phys .
c * 5 * , 1629 ( 1972 ) ] with the langreth - mehl - hu non - local exchange correlation correction [ c. d. hu and d. c. langreth , phys .
scr . * 32 * , 391 ( 1985 ) ] .
j. minar , l. chioncel , a. perlov , h. ebert , m. i. katsnelson , and a. i. lichtenstein , phys .
b * 72 * , 045125 ( 2005 ) ; j. braun , j. minar , h. ebert , m. i. katsnelson , and a. i. lichtenstein , phys . rev
. lett . * 97 * , 227601 ( 2006 ) ; i. di marco , j. minar , s. chadov , m. i. katsnelson , h. ebert , and a. i. lichtenstein , phys . rev .
b * 79 * , 115111 ( 2009 ) ; j. minar , j. phys .
matter * 23 * , 253201 ( 2011 ) ; j. kolorenc , a. i. poteryaev , and a. i. lichtenstein , phys . rev .
b * 85 * , 235136 ( 2012 ) . | we present a rotationally invariant hirsch - fye quantum monte carlo algorithm in which the spin rotational invariance of hund s exchange is approximated by averaging over all possible directions of the spin quantization axis .
we employ this technique to perform benchmark calculations for the two- and three - band hubbard models on the infinite - dimensional bethe lattice .
our results agree quantitatively well with those obtained using the continuous - time quantum monte carlo method with rotationally invariant coulomb interaction .
the proposed approach is employed to compute the electronic and magnetic properties of paramagnetic @xmath0 iron and nickel .
the obtained curie temperatures agree well with experiment .
our results indicate that the magnetic transition temperature is significantly overestimated by using the density - density type of coulomb interaction . |
the comet c/1917 f1 mellish , formerly designated as @xmath0 or @xmath1 , was discovered by j.e .
mellish on march 20 , 1917 and was observed for 96 days @xcite from many places on the earth . in the southern hemisphere ,
the comet reached up to + 1 magnitude .
astronomer j.f .
skjellerup noted that the brightness of the cometary head was about + 3 magnitude , with the diffuse coma and narrow tail about 10@xmath2 long on april 19 , 1917 @xcite .
the comet is a halley - type comet , with a relatively low inclination , and has one of the smallest perihelion distances .
it was observed only at one apparition . despite the relatively long observational arc
, the precision of the orbital elements is questionable .
@xcite published a slightly modified orbit of the comet and noted that the orbit is given with a period of @xmath3yr @xcite .
the nominal orbital elements , according to jpl solar system dynamics database @xcite , are presented in table 1 .
although the ascending and descending node of the nominal orbit are currently far away from the orbit of the earth , we noted a small difference in the eccentricity ( e.g. @xmath4 @xmath5 ) would change the orbit into an earth - crossing orbit as the heliocentric distance of the ascending node would become equal to 1au .
the orbit has a notably small perihelion distance ( @xmath6au ) .
several authors @xcite determined that the comet
earth distance is close enough to observe a meteor shower and predicted the radiant positions and activity of the shower on dec .
20 respectively ) and the geocentric velocity of meteors @xmath7km / s .
the first few meteors associated with the comet c/1917 f1 mellish were obtained by the harvard super schmidt photographic survey @xcite .
several candidates of this meteor stream , later designated as the december monocerotids ( mon ) , were also detected and distinguished by radar surveys @xcite .
another study connecting the december monocerotids with the comet c/1917 f1 mellish was made by @xcite .
surprisingly , the radar data published by @xcite and @xcite revealed that the meteors having similar radiant positions , activity and geocentric velocities appear to have 10@xmath2 lower inclinations .
@xcite noted that december monocerotids seem to have 2 components .
the author also speculates that the stream may have a common origin with the geminid meteor stream .
moreover , harvard radio data revealed a possible meteor stream with low inclined orbits but almost the same orbital elements as the december monocerotids , active between november 27 - december 7 @xcite .
the possible genetic connection between the comet and the december monocerotids was studied by @xcite .
various photographic searches confirmed the existence of a weak stream at @xmath8@xmath2 , @xmath9@xmath2 on november 27 , with @xmath10km / s @xcite .
the stream was named as @xmath11-orionids ( xi - orionids , @xmath12orionids ) , currently recognized as established meteor shower november orionids ( noo ) within the iau meteor data center ( iau mdc ) catalogue @xcite and later by @xcite . moreover
, other photographic december monocerotids were published @xcite and complex analysis of december monocerotids and @xmath11-orionids done by @xcite .
even some historical records of fireballs might confirm that december monocerotids were active in past centuries @xcite . in @xcite , @xcite published his telescopic meteor observation from dec .
11 , 1964 and assigned 5 meteors to the new stream called 11 canis minorids .
@xcite computed that these meteors have parabolic orbits and much higher inclinations ( over 100@xmath2 ) .
a year later , the author suggested a connection between the shower and comet c/1917 f1 mellish @xcite and determined the activity during december 9 - 14 .
@xcite revealed that 9 meteors that create the second component of the december monocerotids might be 11 canis minorids activity within december 4 - 15 .
their inclination was determined as @xmath13@xmath2 and the perihelion distance as @xmath14 , which is closer to the sun compared to the nominal orbit of the comet c/1917 f1 mellish .
the radiants might look like an extension of the mon activity , however , the shower might be active during the mon activity .
the geocentric velocity is similar @xmath15km / s .
the maximum activity of 11 canis minorids is expected at @xmath16@xmath2 ( december 3 ) .
now , the december monocetorids and november orionids are weak ( few meteors per hour at maximum ) but annual established meteor showers . the shower 11 canis minorids , ( december canis minorid according to iau nomenclature )
is classified as a `` working '' shower . despite several investigations , past publications analyzed only a small number of orbits and provided
disperse data on the mean orbit , the position of radiant , the activity and precision of orbits , and did not reveal the orbital evolution of the meteoroid particles released from the parent comet , in order to explain the current state of these meteor showers .
a significant number of the analyzed orbits are hyperbolic or parabolic .
this work uses recent and precise video multi - station orbits , obtained by the sonotaco video network in japan , which provide the highest number of relatively precise meteor orbits detected continually between the years @xmath17 .
the network operates with over 25 similar video - optical systems and uses the same software for meteor detection and orbit analysis ( ufocapture , ufoanalyser and ufoorbit ; @xcite ) .
our goal is to provide a more complex description of meteor showers related to the comet c/1917 f1 mellish and reveal some of their obscured characteristics .
the eventual objective is the investigation of the orbital evolution of meteors and the parent comet , and determination of ejection velocities near the perihelion .
[ f1 ] [ f3 ] .orbital elements of the comet c/1917 f1 mellish in equinox j2000 reference frame ( jpl ) . [ cols="^,^",options="header " , ]
we demonstrated that the december monocerotids and november orionids obtained from the sonotaco database of 3 year observations ( 2007 - 2009 ) have most likely a common origin and come from the comet c/1917 f1 mellish .
the common origin is supported by their similar orbital characteristics , the activity , physical properties assumed from the beginning , and the terminal heights of the meteors , the descending nodes of both showers as a function of the solar longitude , and the narrow southworth - hawkins d - criterion for most december monocerotids and november orionids with respect to the parent comet ( @xmath18 ) .
direct modeling of the stream was not an option , while the orbit uncertainty of the comet avoids the selection of reliable and real starting point of the numerical integration from the past to the current date .
therefore , we studied the orbital evolution of the nominal orbit of the comet , cometary clones within the known orbital uncertainties , and the orbital evolution of precise meteor orbits . the dispersion of the orbital elements , the radiants and nominal orbit of the parent comet , currently beyond the orbital elements , suggest that the november orionids is an older stream than december monocerotids .
the orbital evolution of both streams , the nominal orbit of the comet and its clones imply that the orbital evolution is causing november orionids to have 10@xmath2 lower inclinations than december monocerotids .
there is a non - zero chance of a close encounter of the comet with venus or mercury which could cause a sudden change in the inclination of the parent comet .
furthermore , a close encounter with a planet might cause the tidal breakup of the comet and create a significant release of matter .
the scenario of the cometary core disintegration might also be supported by the extremely low perihelion distance of the parent comet and both meteor showers .
another option is a gradual shift in the inclination , demonstrated in the simulation . but a change in the inclination of more than @xmath19 of the parent comet would be solved only through a longer orbital evolution .
this option is also obscured while we do not observe any orbits between relatively well defined clumps of the december monocerotids and november orionids in the @xmath20 phase space and radiant sky - plane distributions .
eventually there is a wide and massive stream of the meteoroids but only some of them have ascending nodes close to the earth s orbit ; and due to selectional effects , we may observe two distinguished streams and only the distant edge of the stream .
the nodal distance of the comet is currently more than 0.2au from the earth s orbit and it retreated in the past increasingly , as well as the ascending nodes of the observed meteors .
the observed shower meteors might have left the cometary nucleus a few centuries ago but , due to the stable orbit of the comet , both streams might be replenished regularly and weak shower activity might be observed each year .
the semimajor axes of both meteor streams are much lower than the nominal orbit of the comet or its clones evolution 5000yr to the past .
almost no change of the semimajor axes of meteoroids within the orbital evolution suggests that these particles were injected directly into these orbits right after ejection from the cometary nucleus .
we determined that transversal component of the ejection velocity would be about @xmath21 if the ejection occurred at the perihelion .
further precise orbits and physical data of the december monocerotids and november orionids are needed for additional research .
the key question is the accuracy of the c/1917 f1 mellish orbit .
a new measurements of the photographic plates of the comet might reveal a more precise orbit and bring new light onto the orbital evolution of the comet and its meteors .
our work did not confirm any december canis minorids meteors in the sonotaco database ; however , 6 candidates were selected .
their connection to the comet c/1917 f1 mellish is uncertain .
this work was supported by slovak grant agency vega , no . 1/0636/09 and by a grant of comenius uviversity , no . | december monocerotids and november orionids are weak but established annual meteor showers active throughout november and december .
analysis of a high quality orbits subset of the sonotaco video meteor database shows that the distribution of orbital elements , geocentric velocity and also the orbital evolution of the meteors and potential parent body may imply a common origin for these meteors coming from the parent comet c/1917 f1 mellish .
this is also confirmed by the physical properties and activity of these shower meteors .
an assumed release of meteoroids at the perihelion of the comet in the past and the sky - plane radiant distribution reveal that the december monocerotid stream might be younger than the november orionids .
a meteoroid transversal component of ejection velocity at the perihelion must be larger than 100 m/s .
a few authors have also associated december canis minorids with the comet c/1917 f1 mellish .
however , we did not find any connection .
[ firstpage ] comets , individual : c/1917 f1 mellish meteors , meteoroids celestial mechanics catalogues |
many optimization problems of interest can be phrased as picking a maximum - weight independent subset from a ground set of elements , for a suitable definition of independence .
a well - known example is the ( maximum - weight ) independent set problem on graphs , where we wish to find a set of vertices , no two of which are adjacent .
a more tractable problem in this setting is the maximum - weight matching problem , in which we wish to find a set of edges such that no two edges share an endpoint .
this notion of independence can be naturally extended to hypergraphs , where a set of hyperedges is considered independent if no two hyperedges share a vertex . in the previous examples ,
independent sets are characterized by forbidding certain _ pairs _ of elements from the ground set .
a somewhat related , but different notion of independence comes from the independent sets of a matroid .
for example , in the uniform matroid of rank @xmath9 , any set of at most @xmath9 elements is independent . for graphic matroids ,
a set of edges in an undirected graph is independent if and only if it does not contain a cycle ; the optimization goal is to find a maximum - weight acyclic subgraph of a graph @xmath0 . in transversal matroids ,
a set of left - vertices of a bipartite graph is independent if and only if there is a matching that matches each vertex in this set to some right - vertex . in many applications ,
the elements of the ground set and their weights are not known in advance , but arrive online one at a time .
when an item arrives , we must immediately decide to either irrevocably accept it into the final solution , or reject it and never be able to go back to it again .
we will be interested in competitive analysis , that is , comparing the performance of an online algorithm to an optimal offline algorithm which is given the whole input in advance . in this setting , even simple problems like selecting a maximum - weight element become difficult , because we do not know if elements that come in the future will have weight significantly higher or lower than the element currently under consideration .
if we make no assumptions about the input , any algorithm can be fooled into performing arbitrarily poorly by ofering it a medium - weight item , followed by a high - weight item if it accepts , and a low - weight item if it rejects . to solve such problems , which frequently arise in practice ,
various assumptions are made .
for instance , one might assume that weights are all drawn from a known distribution , or ( if independent sets may contain several elements ) that the weight of any single element is small compared to the weight of the best independent set .
one useful assumption that can be made is that the elements of the ground set appear in a random order .
the basic problem in which the goal is to select the maximum - weight element is well known as the _ secretary problem_. it was first published by martin gardner in @xcite , though it appears to have arisen as folklore a decade previously @xcite .
an optimal solution is to observe the first @xmath10 elements , and select the first element from the rest with weight greater than the heaviest element seen in the first set ; this algorithm gives a @xmath11 probability of finding the heaviest element , and has been attributed to several authors ( see @xcite ) .
motivated by this simple observation , several results have appeared for more complex problems in this random permutation model ; these are often called secretary - type problems .
typically , given a random permutation of elements appearing in an online fashion , the goal is to find a maximum - weight independent set .
for example , kleinberg @xcite gives a @xmath12-competitive algorithm for the problem of selecting at most @xmath9 elements from the set to maximize their sum .
et al . _
@xcite give a constant - competitive algorithm for the more general knapsack secretary problem , in which each element has a size and weight , and the goal is to find a maximum - weight set of elements whose total size is at most a given integer @xmath13 .
babaioff _ et al . _
@xcite had earlier introduced the so - called _ matroid secretary problem _ , and gave an @xmath14-competitive algorithm to find the max - weight independent set of elements , where @xmath9 is the rank of the underlying matroid .
a @xmath15-competitive algorithm was also given in @xcite for the special case of graphic matroids ; this was based on their @xmath16-competitive algorithm algorithm for the important case of _ transversal matroids _ , where @xmath17 is the maximum degree of any left - vertex .
recently , dimitrov and plaxton @xcite improved the latter to a ratio of @xmath15 for all transversal matroids .
a significant open question is whether there exists a @xmath18-competitive algorithm for general matroids , or for other secretary problems with non - matroid constraints .
these secretary - type problems arise in many practical situations where decisions must be made in real - time without knowledge of the future , or with very limited knowledge .
for example , a factory needs to decide which orders to fulfil , without knowing whether more valuable orders will be placed later .
buyers and sellers of houses must decide whether to go through with a transaction , though they may receive a better offer in a week or a month .
below , we give an example from online advertising systems , which we use as a recurring motivation through the paper .
internet - based systems are now being used to sell advertising space in other media , such as newspapers , radio and television broadcasts , etc .
advertisers in these media typically plan advertising campaigns and reserve slots well in advance to coincide with product launches , peak shopping seasons , or other events . in such situations ,
it is unreasonable to run an auction immediately before the event to determine which ads are shown , as is done for sponsored search and other online advertising .
consider an automatic advertising reservation system , in which the seller controls a number of _ slots _ , each representing a position in which an advertisement ( hereafter _ ad _ ) can be published .
advertisers / bidders appear periodically , and report which slots they would like to place an ad in , and how much they are willing to pay for each slot
. when an advertiser reports a bid , the system must immediately decide whether or not to accept it ; if a bid is accepted , the ad _ must _ be placed in the corresponding slot , and if not , the ad is permanently rejected .
note that in disallowing the removal of an accepted ad , our model differs significantly from that of @xcite , in which the seller can subsequently remove an accepted ad if he makes a compensatory payment to the advertiser .
we model this system as as an online edge - weighted matching problem on a bipartite graph @xmath19 : the vertices of set @xmath3 correspond to the set of slots , and those of set @xmath4 to the ads .
for each vertex @xmath7 , its neighbors in @xmath3 correspond to the slots in which ad @xmath6 can appear , and the weight of edge @xmath20 is the amount the advertiser is willing to pay if @xmath6 appears in slot @xmath21 .
initially , the seller knows the set of slots @xmath3 ; vertices of @xmath4 appear sequentially in a random order , as advertisers bid on slots .
when a vertex @xmath7 is seen , all the edges from @xmath6 to @xmath3 are revealed , together with their weights ; the seller must immediately decide whether to accept ad @xmath6 , and if so , which of the relevant slots to place it in .
the seller s goal , obviously , is to maximize his revenue .
subsequently , we refer to this problem as bipartite vertex - at - a - time matching ( bvm ) .
we describe our results for bvm and other problems below .
recall that the elements of a transversal matroid are one partite set @xmath4 ( subsequently referred to as the _ left vertices _ ) of a bipartite graph , and a set of vertices @xmath22 is independent if the graph constains a perfect matching from @xmath23 to the other partite set .
that is , the transversal matroid secretary problem is equivalent to the special case of bvm in which all edges incident to each @xmath7 have the same weight .
( equivalently , the weights are on vertices of @xmath4 instead of edges . ) in section [ sec : bvm ] , we give a simpler and tighter analysis for an algorithm essentially similar to that of dimitrov and plaxton @xcite for transversal matroids ; this allows us to improve the competitive ratio from 16 to 8 , even for the more general bvm problem .
in addition to an improved ratio , our methods are of interest as they appear robust to changes in the model and can be naturally applied to more general problems .
we illustrate this in section [ sec : hypergraphs ] by extending our algorithms to hypergraph problems , with applications to more complex advertising systems in which advertisers desire _ bundles _ of slots , as opposed to a single slot .
in particular , we obtain constant - competitive algorithms for finding independent edge sets in hypergraphs of constant edge - size .
we also introduce secretary problems with _ groups _ , to model applications in which we do not see a truly random permutation of elements .
we assume that an adversary can group the elements arbitrarily , but once the groups are constructed , they appear in random order .
when a group appears , the algorithm can see all the elements in the group .
we discuss this idea further in section [ sec : groups ] .
finally , in section [ sec : graphic ] , we obtain a simple @xmath8-competitive algorithm for the problem of finding independent edge - sets in graphic matroids , improving the ratio of 16 from @xcite .
the majority of our algorithms follow the `` sample - and - price '' method common to many solutions to secretary problems .
that is , we look at a random sample of elements containing a constant fraction of the input , and use the values observed to determine _ prices _ or thresholds . in the second half ,
we accept an element if its weight / value is above the given price .
for instance , in the optimal solution to the original secretary problem , the price is set to be the highest value seen in the first @xmath11 fraction of the input , and we accept any element from the remaining set with value greater than this price .
recall that in the bvm problem , the algorithm is initially given one partite set @xmath3 of a bipartite graph @xmath19 , together with the size of the other partite set @xmath4 .
the algorithm sees the vertices of @xmath4 sequentially , in a random order .
when a vertex @xmath7 is seen , all edges incident to @xmath6 are revealed , together with their weights
. the algorithm must immediately either match @xmath6 to an available vertex of @xmath3 , or decide that @xmath6 will remain permanently unmatched . in this section , we show that an algorithm based on that of @xcite gives a competitive ratio of 8 for this problem . before presenting the algorithm for bvm
, we describe a closely related algorithm simulate that is easier to analyze , and then show that our final algorithm does at least as well as simulate .
let greedy denote the following greedy algorithm for the offline edge - weighted bipartite matching problem : : + sort edges of @xmath24 in decreasing order of weight .
+ matching @xmath25 + for each edge @xmath26 , in sorted order + if @xmath27 is a matching : + @xmath28 + return @xmath29 .
let @xmath30 denote the weight of a set of edges @xmath31 , and @xmath32 denote the weight of an optimum ( max - weight ) matching on @xmath0 .
it is easy to see the following proposition , that greedy is a 2-approximation .
@xmath33 .
we now describe the algorithm simulate , which we use purely to analyze our final algorithm for bvm . : + sort edges of @xmath34 in decreasing order of weight .
+ @xmath35 + mark each vertex @xmath7 as unassigned .
+ for each edge @xmath36 , in sorted order + if @xmath6 is unassigned * and * @xmath37 is a matching : + mark @xmath6 as assigned + flip a coin with probability @xmath38 of heads + if heads , @xmath39 + else @xmath40 + @xmath41 + for each vertex @xmath42 + if @xmath21 has degree @xmath43 in @xmath44 + delete all edges incident to @xmath21 from @xmath44 .
say that an edge @xmath45 is _ considered _ by simulate if we flip a coin and assign @xmath45 to either @xmath46 or @xmath47 .
we make two observations about simulate : once any edge incident to a vertex @xmath7 has been considered , no other edge incident to @xmath6 will be considered later .
second , once an edge incident to @xmath42 has been added to @xmath46 , no subsequent edge incident to @xmath21 will be considered .
( note that multiple edges incident to @xmath21 might be considered until one of these edges is added to @xmath46 . )
observe that from our description of simulate , @xmath46 is a matching , but @xmath47 may not be , as a vertex @xmath42 may be incident to multiple edges of @xmath47 .
hence , we have a final pruning step in case there are multiple edges incident to the same vertex of @xmath3 ; this gives us a matching @xmath44 .
we now prove three statements about simulate , and later show that the matching returned by our online algorithm is at least as good as @xmath44 .
[ prop : wm1 ] @xmath48 } \ge p { \text{\sc opt}}/2 $ ] .
simulate tosses a coin ( at most ) once for each vertex in @xmath4 ; @xmath46 is precisely the matching one would obtain from running greedy on @xmath49 , where @xmath50 denotes the vertices which came up heads .
( if the coin for a vertex comes up tails , this vertex has no effect on @xmath46 . )
if @xmath51 denotes the weight of an optimum matching on @xmath52 , it is easy to see that
@xmath53 } \ge p { \text{\sc opt}}$ ] , and hence that @xmath48 } \ge p { \text{\sc opt}}/2 $ ] . [ lem : wm2 ] @xmath54 } \ge ( 1-p ) { \text{\sc opt}}/2 $ ] .
consider any history of coin tosses in which an arbitrary edge @xmath45 is being considered , and we are about to flip a coin to determine whether @xmath45 is added to @xmath46 or @xmath47 .
its expected contribution to @xmath46 is @xmath55 , and to @xmath47 , is @xmath56 .
this holds for each edge @xmath45 and any history in which @xmath45 can contribute to the weight of @xmath46 or @xmath47 ; hence @xmath54 } = \frac{(1-p)}{p } { \mathbb{e}[w(m_1)]}$ ] , completing the proof .
[ lem : wm3 ] @xmath57 } \ge \frac{p^2(1-p)}{2 } { \text{\sc opt}}$ ] . for each vertex @xmath58
, let @xmath59 denote the revenue earned by vertex @xmath60 in @xmath47 , which we define as the sum of the weights of edges in @xmath47 incident to @xmath60 .
( hence , @xmath61 . ) for each edge @xmath45 incident to @xmath60 , let @xmath62}$ ] denote the expected revenue earned by @xmath60 in @xmath47 , conditioned on the fact that @xmath45 is the first edge incident to @xmath60 selected by simulate for @xmath47 .
it is easy to see that @xmath62 } \le w(e)/p$ ] , by considering how @xmath60 can earn revenue : if the next edge incident to @xmath60 considered by simulate is added to @xmath46 ( which happens with probability @xmath38 ) , then @xmath60 earns precisely @xmath63 , as no later edge incident to @xmath60 can ever be considered . in general , if @xmath60 is incident to @xmath64 edges in @xmath47 , the revenue it earns is at most @xmath65 , and the probability of this event is at most @xmath66 ; this is because the next @xmath67 edges incident to @xmath60 that are considered must be added to @xmath47 , and the @xmath64th edge is added to @xmath46 .
therefore , @xmath68 } \le w(e ) \sum_{i=1}^\infty i \cdot p ( 1-p)^{i-1 } = w(e)/p$ ] .
similarly , for each vertex @xmath58 , let @xmath69 denote the revenue earned by vertex @xmath60 in @xmath44 , which is the weight of the ( at most one ) edge incident to @xmath60 in @xmath44 .
let @xmath70}$ ] denote the expected revenue earned by @xmath60 , conditioned on @xmath45 being the first edge incident to @xmath60 added to @xmath47 . with probability @xmath38 , the next considered edge incident to @xmath60
is added to @xmath46 , and hence @xmath60 has degree 1 in @xmath47 .
@xmath70 } \ge p w(e)$ ] , and so @xmath70 } \ge p^2 { \mathbb{e}[{revenue}_2(v)|e]}$ ] ; it follows that @xmath57 } \ge p^2 { \mathbb{e}[w(m_2 ) ] } = \frac{p^2 ( 1-p)}{2 } { \text{\sc opt}}$ ] . before describing our final algorithm for ebp ,
we show that the matching returned by an intermediate algorithm sampleandpermute is at least as good as @xmath44 , which implies that we have a @xmath71-competitive algorithm : setting @xmath72 , we get a @xmath73-competitive algorithm
. however , our pruning step allows us to take an edge for @xmath44 only if its right endpoint has degree 1 ; a more careful pruning step allows more edges in the matching .
we use this fact to give a tighter analysis for the next algorithm , obtaining a competitive ratio of 8 . : + @xmath74 + for each @xmath7 : + with probability @xmath38 , @xmath75 + @xmath76 greedy(@xmath77 $ ] ) .
+ for each @xmath42 : + set @xmath78 to be the weight of the edge incident to @xmath21 in @xmath46 .
+ @xmath79 + for each @xmath80 , in random order : + let @xmath81 be the highest - weight edge such that @xmath82 + add @xmath45 to @xmath47 .
+ if @xmath27 is a matching , add @xmath45 to @xmath29 .
note that the matching @xmath46 in sampleandpermute is precisely the same as @xmath46 from simulate ; intuitively , in the former , we toss all the coins at once and run greedy , while in the latter , we toss coins while constructing the greedy matching .
( more precisely , the two algorithms to generate the matchings are equivalent . )
similarly , the `` matching '' @xmath47 in this algorithm is essentially @xmath47 from simulate .
the difference between the two algorithms is in the pruning step : to construct @xmath44 in simulate , we delete all edges incident to any vertex @xmath42 with degree greater than 1 ; in sampleandpermute , we add to @xmath29 the first such edge seen in our permutation of @xmath83 .
it follows immediately from lemma [ lem : wm3 ] that @xmath84 } \ge p^2(1-p ) { \text{\sc opt}}/2 $ ] , but accounting for the difference in pruning allows the following tighter statement , which we prove in the appendix .
[ lem : tighterbound ] @xmath84 } \ge \frac{p ( 1-p)}{2 } { \text{\sc opt}}$ ] .
we now present our final algorithm , a trivial modification of sampleandpermute for the online bvm problem .
+ @xmath85 + let @xmath50 be the first @xmath9 vertices of @xmath4 .
+ @xmath76 greedy(@xmath77 $ ] ) .
+ for each @xmath42 : + set @xmath78 to be the weight of the edge incident to @xmath21 in @xmath46 .
+ @xmath25 + for each subsequent @xmath80 , : + let @xmath81 be the highest - weight edge such that @xmath82 + if @xmath27 is a matching , accept @xmath45 for @xmath29 .
as the input to sampleandprice is a random permutation , @xmath50 is a subset of @xmath4 in which each vertex of @xmath4 is selected with probability @xmath38 ; it is easy to see that this algorithm is equivalent to sampleandpermute .
therefore , @xmath84 } \ge \frac{p ( 1-p)}{2 } { \text{\sc opt}}$ ] ; setting @xmath86 implies that the expected competitive ratio is 8 .
in the hypergraph edge - at - a - time matching ( hem ) problem , we are initally given the vertex set of a hypergraph ; subsequently , hyperedges appear in a random order . when an edge ( together with its weight ) is revealed , the algorithm must immediately decide whether or not to accept it ; as before , the goal is for the algorithm to select a maximum - weight set of disjoint edges . for arbitrary hypergraphs , one can observe that even the offline version of this problem is np - complete ( and also hard to approximate ) via an easy reduction from the independent set problem .
however , the difficulty is related to the size of the hyperedges ; if all edges contain only 2 vertices , for instance , then we are simply trying to find a matching in a ( possibly non - bipartite ) graph .
( even in this special case , the problem is of interest in an online setting . )
let @xmath17 denote the maximum size of an edge in the hypergraph .
we provide an @xmath87-competitive algorithm for the hem problem by solving the more general hypergraph vertex - at - a - time matching ( hvm ) problem , described as follows : we are initally given a subset @xmath3 of the vertex set of a hypergraph .
the remaining vertices @xmath4 arrive online ; each edge of the hypergraph is constrained to contain exactly one vertex of @xmath4 , together with some vertices of @xmath3 .
the vertices of @xmath4 appear online in a random order ; when @xmath7 is revealed , the algorithm also sees all edges incident to @xmath6 , together with their weight . at this point , the algorithm must immediately decide whether or not to accept some edge containing @xmath6 , and if so , which edge ; again , the goal is for the algorithm to select a maximum - weight set of disjoint edges . here , let @xmath17 denote the maximum number of vertices of @xmath3 contained in a single edge ( so the largest edge has @xmath88 vertices ) .
first , we observe that the hem problem with edge size @xmath17 reduces to the hvm problem with edge size @xmath88 : let @xmath3 be the vertex set of the original hypergraph , and add one vertex to @xmath4 for each original edge .
an edge of the new hypergraph consists of an old edge , together with the corresponding vertex of @xmath4 .
clearly , observing a random permutation of @xmath4 together with the incident edges is equivalent to a random permutation of the edge set of the original hypergraph .
also , notice that the the bvm problem of section [ sec : bvm ] is simply the special case of hvm when @xmath89 .
( see figure 1 at the end of this section . )
these hypergraph problems capture the notion of demand _
bundles_. for instance , in ad reservation systems , advertisers rarely make reservations for a single ad at a time ; they are more likely to plan advertising campaigns involving multiple individual ads . in many campaigns ,
advertisers create various ads which are related to and complement or reinforce each other ; these advertisers might be interested in acquiring a bundle or set of slots for this campaign .
they submit to the reservation system the bundles they are interested in , together with the price they are willing to pay ; the system must either accept a request for an entire bundle or reject it , as it does not receive revenue for providing the advertiser with a part of the bundle .
if each advertiser submits a request for a single bundle , we obtain the hem problem with vertex set corresponding to the set of slots . more generally , an advertiser may submit a request for _ one of _ a set of bundles , together with a price for each bundle .
( for example , an advertiser might want an ad to appear in any three out of four local newspapers . )
this leads to the hvm problem , with vertex set @xmath4 corresponding to the set of advertisers , and set @xmath3 to the set of slots : we receive a random permutation of advertisers , and each advertiser informs us of the bundles she is interested in , together with a price for each bundle .
let greedy denote the offline algorithm for hvm that sorts edges in decreasing order of weight , and selects an edge if it is disjoint from all previously selected edges .
for ease of exposition , we subsequently assume that the hypergraph is @xmath90-uniform ; that is , that each edge contains exactly @xmath17 vertices of @xmath3 together with one vertex of @xmath4 .
greedy returns a @xmath90-approximation to the maximum - weight disjoint edge set .
we again define an algorithm simulate , as in section [ sec : bvm ] : sort edges of @xmath24 in decreasing order of weight .
+ mark each vertex @xmath7 as unassigned .
+ @xmath35 + for each edge @xmath26 in sorted order : + let @xmath6 be the vertex of @xmath4 in @xmath45 + if @xmath6 is unassigned * and * @xmath45 is disjoint from @xmath46 : + mark @xmath6 as assigned .
+ flip a coin with probability @xmath38 of heads + if heads , add @xmath45 to @xmath46 + if tails , add @xmath45 to @xmath47 + @xmath91 + for each @xmath92 : + add @xmath45 to @xmath44 if @xmath45 is disjoint from the rest of @xmath47 . as before , we let @xmath30 denote the weight of an edge set @xmath31 .
the proofs of the following two propositions are exactly analogous to proposition [ prop : wm1 ] and lemma [ lem : wm2 ] .
@xmath48 } \ge p \cdot { \text{\sc opt}}/(d+1)$ ] .
@xmath54 } \ge ( 1-p ) { \text{\sc opt}}/(d+1)$ ] .
it is now slightly more complex to bound the weight of @xmath44 than it was for the bvm problem ; for bvm , the set of edges in @xmath47 incident to @xmath58 interfere only with each other , but in the hypergraph version , edges @xmath93 and @xmath94 might not intersect , though they may both intersect @xmath95 , and hence all of @xmath96 will have to be deleted .
however , we can use a similar intuition : in bvm , we charge all edges of @xmath47 incident to @xmath60 to the heaviest such edge ; in expectation , each edge is charged a constant number of times . for the hvm problem
, we charge all the edges in a `` connected component '' to the heaviest edge in the component , and argue that ( with a suitable choice of @xmath38 ) the average size of the components is small .
more formally , we prove the following lemma : [ lem : hvm ] setting @xmath97 , @xmath57 } \ge \frac{{\text{\sc opt}}}{12 d(d+1)}$ ] .
construct an auxiliary directed graph @xmath31 as follows : for each @xmath98 , add a corresponding vertex @xmath99 to @xmath31 . if @xmath100 is the heaviest edge in @xmath47 that intersects @xmath45 , add a directed arc from @xmath99 to @xmath101 to @xmath31 .
( if @xmath45 itself is this heaviest edge , @xmath99 has no out - neighbors . )
note that the graph @xmath31 is obviously a forest . for each @xmath92 , if @xmath99 is not the root of its tree in @xmath31 , we define @xmath102 to be 0 , and if it is the root , we set @xmath102 to be the weight of all edges of @xmath47 in the tree . clearly , @xmath103 .
we define @xmath104 to be equal to the weight of @xmath45 iff @xmath45 is an edge in @xmath47 that does not intersect any other such edge .
( in which case , it follows that @xmath99 is the root of its tree . )
we prove that @xmath105 } \ge \frac{{\mathbb{e}[{revenue}_2(e)]}}{6}$ ] , which proves the lemma , since @xmath106 .
first , note that the probability that any edge @xmath45 added to @xmath47 intersects an edge added later is at most @xmath107 : for each vertex @xmath108 of @xmath3 contained in @xmath45 , the probability that @xmath45 intersects a later edge because of @xmath108 is at most @xmath109 , as with probability @xmath110 , the next edge containing @xmath108 considered by simulate will be added to @xmath46 .
as @xmath45 contains only @xmath17 vertices in @xmath3 , the desired probability is at most @xmath107 .
( every vertex of @xmath4 is incident to at most one edge in @xmath47 , and so @xmath45 can not intersect any other edge through its vertex in @xmath4 . )
it follows that the probability that any @xmath111 has a child is at most @xmath107 .
we also count the expected number of children of @xmath99 ; the edge corresponding to each child of @xmath99 must share some vertex with @xmath45 , and the expected number of children through a particular vertex is at most @xmath112 . as @xmath45 contains @xmath17 vertices of @xmath3 ,
the expected number of children of @xmath99 is at most @xmath113 ; since @xmath114 , the expected number of children is at most @xmath115 .
it follows that the expected size of a subtree rooted at @xmath99 is at most @xmath116 .
note that @xmath102 and @xmath104 are both @xmath117 if @xmath99 is not the root of its tree in @xmath31 .
conditioned on @xmath99 being a root , @xmath118 } \le 3 w(e)$ ] , as @xmath45 is the heaviest edge in its tree , and the expected size of the tree is at most @xmath116 .
@xmath105}$ ] is at least @xmath119 , as @xmath45 intersects no previously added edges , and with probability at least @xmath107 , it intersects no edge added to @xmath47 later .
therefore , the ratio of these expectations is at most @xmath120 , completing the proof .
now , we define our final algorithm sampleandprice for the hvm problem : + @xmath121 + let @xmath50 be the first @xmath9 elements of @xmath4 .
+ @xmath76 greedy(@xmath122 ) .
+ for each @xmath123 : + set @xmath124 to be the weight of the edge incident to @xmath60 in @xmath46 .
+ @xmath25 + for each subsequent @xmath80 : + let @xmath45 be the highest - weight edge containing @xmath6 such that for each @xmath125 , @xmath126 + if @xmath45 is disjoint from @xmath29 , add @xmath45 to @xmath29 . as before , since the input is a random permutation of @xmath4 , @xmath50 is a subset of @xmath4 in which every vertex is selected independently with probability @xmath127 , and the matching @xmath29 is at least as good as @xmath44 from simulate .
therefore , we have proved the following theorem : [ thm : hvm ] sampleandprice is an @xmath87-competitive algorithm for the hvm secretary problem .
note that @xmath29 may also contain extra edges that occur earlier in the permutation than edges they intersect ; for the bvm problem , this was the difference between lemma [ lem : wm3 ] and the stronger bound [ lem : tighterbound ] .
we do not provide a tighter analysis similar to lemma [ lem : tighterbound ] for the hvm problem in this extended abstract , nor make an attempt to optimize the constants of lemma [ lem : hvm ] .
in particular , for the hem problem with @xmath128 ( finding an online matching in a non - bipartite graph @xmath129 , given a random permutatation of @xmath24 ) , we have a constant bound on the competitive ratio ; a smaller constant can easily be obtained .
( hvm ) at ( 2,3 ) hvm(@xmath88 ) ; ( hem ) at ( 0,1.5 ) hem(@xmath17 ) ; ( bvm ) at ( 4,1.5 ) bvm ; ( bem ) at ( 0,0 ) edge - at - a - time matchings ; at ( 0,-0.5 ) ( in arbitrary graphs ) ; ( tvs ) at ( 4,0 ) transversal matroid secretary ; ( hvm ) ( hem ) ; ( hvm ) ( bvm ) ; ( hem ) ( bem ) ; ( bvm ) ( tvs ) ;
consider a secretary - type problem in which , instead of receiving a random permutation of the elements , elements can be grouped by an adversary .
the algorithm receives the number of groups in advance , instead of the number of elements .
however , once the groups have been constructed , they arrive in random order ; when a group arrives , the algorithm can see all its elements at once .
note that the groups are fixed in advance ; the adversary can not construct groups in response to the algorithm s choices or the set of groups seen so far .
the effect of such grouping on the difficulty of the problem is not immediately clear : the adversary can ensure that some permutations of the element set never occur , which might make the problem more difficult . on the other hand , as the algorithm is allowed to see several elements at once , it may be easier to compute a good solution .
for instance , consider the classical secretary problem with groups .
an optimal algorithm will never hire any but the best secretary from a group , and it is easy to obtain an @xmath45-competitive algorithm : ignore all but the best secretary from each group , and run the standard secretary algorithm on these .
that is , observe a constant ( @xmath11 ) fraction of the groups , and note the value / price of the best secretary seen so far . from the rest of the input ,
hire the best secretary from the first group with a secretary to beat this price .
perhaps a reason this problem is as easy as the original version is that only one element is to be selected . by way of contrast , consider the following matching problem , even restricted to bipartite graphs : the algorithm is initially given the vertex set of a bipartite graph , and an adversary groups the edges arbitrarily .
the groups arrive in random order ; when a group arrives , the algorithm sees the weights of all edges it contains .
the goal is to find a maximum - weight matching ; note that as a special case of hem with @xmath128 , we have an @xmath18-competitive algorithm for this problem without edge grouping .
a natural sample - and - price algorithm for this problem is as follows : look at a constant fraction of the input , and construct a matching with these edges ( either the optimal matching , or the greedy matchings we used in the previous sections ) . use the weights of edges in the matching to set vertex prices , and in the remainder of the input , select an edge if its weight is at least the price of each of its endpoints , and if it does not conflict with edges already selected .
unfortunately , this algorithm does not work : consider a bipartite graph @xmath130 , with @xmath131 and @xmath132 .
we have two groups of edges : @xmath133 , with @xmath134 , and @xmath135 , with @xmath136 .
assuming @xmath137 , @xmath138 corresponds to an optimal matching , with weight @xmath139 .
if @xmath138 arrives first , the price of each @xmath140 is @xmath141 .
subsequently , when @xmath142 arrives , @xmath143 = @xmath144 , and hence no edge of @xmath142 beats the price of its right endpoint .
if @xmath142 arrives first , the price of each @xmath145 is @xmath146 .
subsequently , when @xmath138 arrives , @xmath147 , and so no edge except @xmath148 beats the price of its left endpoint , for a total revenue of @xmath149 .
we believe , therefore , that the introduction of groups affects these secretary - type problems in non - trivial ways , and these problems are likely to be of theoretical interest ; in addition , they have applications to problems where groups occur naturally , and we do not receive a random permutation of the entire element set . to take another example from the advertising world ,
when a merchant plans a campaign , she may submit to the reservation system multiple ads , together with the slots in which each ad can be placed , and a price for each ad - slot combination .
even if the merchants arrive in a random order , this does not correspond to a random permutation of ads , and hence our previous analysis is not directly applicable .
we model this ( as in bvm ) as an edge - weighted matching problem on a bipartite graph @xmath19 in which vertices of @xmath4 may be grouped ; here , the groups correspond to the set of ads for a given advertiser .
the algorithm initially receives @xmath3 ( the set of slots ) , and the number of advertisers / groups ; the adversary can construct groups from @xmath4 arbitrarily .
once the groups have been fixed , a random permutation of the groups is seen , and when a group arrives , the algorithm must decide which ads to accept , and where to place them ; as always , decisions are irrevocable .
we refer to this as the bvm problem with groups .
[ thm : bvmlogn ] there is an @xmath150-competitive algorithm for the bvm problem with groups .
it is easy to prove this theorem using standard techniques : sample the first half of the vertices , and let @xmath151 denote the weight of the heaviest edge seen so far .
pick an integer @xmath152 uniformly at random in @xmath153 $ ] , and set a threshold of @xmath154 . in the second half , greedily construct a matching using edges with weight above the threshold .
( see , for instance , theorem 3.2 of @xcite for analysis of an essentially similar algorithm . ) for completeness , we give a proof of theorem [ thm : bvmlogn ] in section [ subsec : otherproofs ] of the appendix .
a natural question is whether one can find a constant - competitive algorithm for bvm with groups .
note that one must be careful about using sample - and - price algorithms : first , as the example above shows , the natural algorithm with groups of edges instead of vertices does not work .
second , one might sample a constant fraction of groups , construct a matching @xmath46 on the sampled groups , and then use @xmath46 to set prices
. however , once prices have been set in this way , the edges assigned to a group @xmath155 may not be the same as the edges that would have been assigned to @xmath155 in @xmath46 if @xmath155 had been sampled .
this was not the case for the basic bvm problem : if an edge @xmath20 is in @xmath47 , then by construction fixing all other
coin flips
if the coin for @xmath6 had come up heads instead of tails , @xmath20 would be in @xmath46 .
as the example in figure [ fig:2 ] shows , this desirable property no longer holds once groups are introduced .
= [ circle , draw , inner sep=0pt , minimum size=6 mm ] ; \(a ) at ( 0,2 ) [ vertex ] @xmath156 ; ( b ) at ( 0,1 ) [ vertex ] @xmath13 ; ( c ) at ( 0,0 ) [ vertex ] @xmath157 ; \(x ) at ( 4,1.5 ) [ vertex ] @xmath158 ; ( y ) at ( 4,0.5 ) [ vertex ] @xmath159 ; \(a ) ( x ) ; at ( 2,1.9 ) 4 ; ( b ) ( x ) ; at ( 2,1.4 ) 3 ; ( b ) ( y ) ; at ( 2,0.9 ) 2 ; ( c ) ( y ) ; at ( 2,0.4 ) 1 ; we conjecture that the following algorithm samplewithgroups is constant - competitive for bvm with groups . here
, @xmath160 denotes the set of groups : + sample each group with probability @xmath38 .
+ construct a greedy matching @xmath46 on the set of sampled groups @xmath161 .
+ @xmath162 . + for each group @xmath155 in @xmath163 : + let @xmath164 denote the edges assigned to vertices of @xmath155 in the greedy matching on @xmath165 .
+ @xmath166 .
+ @xmath41 + for each @xmath42 : +
if @xmath21 has degree @xmath43 in @xmath44 : + delete all edges incident to @xmath21 from @xmath44 .
it is easy to see that @xmath48 } \ge p { \text{\sc opt}}/2 $ ] . by construction ,
the edges assigned to @xmath155 in @xmath47 are precisely those that would have been assigned to @xmath155 in @xmath46 if @xmath155 had been sampled .
( hence , this algorithm differs from the natural sampleandprice . )
therefore , it follows that the probability an edge contributes to @xmath47 is @xmath167 times the probability it contributes to @xmath46 .
if @xmath86 , it follows that @xmath54 } = { \mathbb{e}[w(m_1)]}$ ] , and further , that the expected degree of @xmath42 in @xmath47 is equal to its expected degree in @xmath46 , which is at most 1 since @xmath46 is a matching .
this does not suffice to give a lower bound on the expected weight of @xmath44 , but we conjecture that the expected weight of @xmath44 is at most a constant factor lower than that of @xmath47 .
samplewithgroups is constant - competitive for the bvm problem with groups .
in this section , we describe a @xmath8-competitive algorithm for the graphic matroid secretary problem . here , we are initially given the set of vertices @xmath168 of an undirected edge - weighted graph @xmath169 together with the size of its edge set @xmath170 .
the edges of the graph appear in a random order , and the goal is to accept a maximum - weight subset of edges @xmath31 that does not contain any cycles . as always , the decision to accept an edge must be made upon its arrival , and can not be revoked .
this problem is equivalent to finding the maximum - weight spanning tree ( assuming @xmath0 is connected ) and is also equivalent to finding the maximum - weight independent set in the graphic matroid defined by the graph @xmath0 .
babaioff et al .
@xcite give a 16-competitive algorithm for the secretary version of this problem based on a related algorithm for transversal matroids .
we give a simple reduction to the classical secretary problem , losing a factor of @xmath171 in the reduction . in this way
, we obtain a @xmath172-competitive algorithm for the graphic matroid secretary problem .
fix an ordering @xmath173 on the vertices of @xmath0 .
consider two directed graphs : graph @xmath174 is obtained by orienting every edge of @xmath0 from higher numbered to lower numbered vertex , and graph @xmath175 by orienting every edge from lower to higher numbered vertex .
our online algorithm initially flips a fair coin @xmath176 . for each vertex
@xmath60 independently , it runs a secretary algorithm to find the maximum - weight edge leaving @xmath60 in @xmath177 .
the output of the algorithm is @xmath178 , the union of all edges accepted by the individual secretary algorithms .
since the graph @xmath177 is acyclic and each vertex has at most one outgoing edge , the set of edges @xmath178 must be acyclic even in the undirected sense .
it remains to show a lower bound on the weight of @xmath178 . for each vertex @xmath60 , let @xmath179 be the heaviest edge leaving vertex @xmath60 in @xmath177 .
let @xmath180 .
let @xmath181 be a maximum - weight acyclic subgraph of @xmath0 .
[ prop : dir - tree ] @xmath182 .
conditioned on the coin flip @xmath158 , each secretary algorithm recovers at least @xmath11 fraction of the weight of the heaviest edge leaving its vertex . hence @xmath183
= \frac{1}{e } w(f_x)$ ] for @xmath184 . using proposition [ prop : dir - tree ] , @xmath185 = \frac{1}{e } \left ( \frac12 e[w(f ' ) ~|~ x=0 ] + \frac12 e[w(f ' ) ~|~ x=1 ]
\right ) \ge \frac{1}{2e}w(f^*)$ ] . therefore
, we obtain the following theorem : there is a @xmath8-competitive algorithm for the graphic matroid secretary problem .
we list several problems that remain to be solved : * an improved understanding of groups and their contribution to the difficulty of secretary - type problems is likely to be of interest .
in particular , it may be possible to find a constant - competitive algorithm for the bvm problem with groups . *
few lower bounds for these problems are known beyond @xmath11 for the original secretary problem ; obtaining such bounds may require new techniques . * in the basic bvm problem , we lose a factor of 2 by constructing greedy matchings . if , instead , we modified our algorithm to set prices using an optimal matching @xmath46 on the sampled vertices , is the resulting algorithm 4-competitive ?
is it even @xmath18-competitive ? * finally , obtaining an @xmath18-competitive algorithm for the general matroid secretary problem is still open , though the competitive ratios for important special cases such as transversal and graphic matroids have been reduced to small constants . * acknowledgments : * we would like to thank florin constantin , jon feldman , and s. muthukrishnan for helpful discussions on bvm and related problems . 99 m. babaioff , n. immorlica and r. kleinberg .
matroids , secretary problems , and online mechanisms . , 434443 , 2007 .
m. babaioff , n. immorlica , d. kempe and r. kleinberg .
a knapsack secretary problem with applications .
, 1628 , 2007 .
f. constantin , j. feldman , s. muthukrishnan , m. pl .
online ad slotting with cancellations .
submitted to _
soda 2009_. preliminary version in _ workshop on ad auctions _
, 2008 . arxiv:0805.1213v1 .
n. b. dimitrov and c. g. plaxton .
competitive weighted matching in transversal matroids . in _
icalp 2008 , to appear_. t. s. ferguson . who solved the secretary problem ?
vol . 4 , 282289 , 1989 . m. gardner .
mathematical games column .
feb . , mar .
, 35 , 1960 .
r. kleinberg .
a multiple - choice secretary problem with applications to online auctions .
, 630631 , 2005 .
for each @xmath58 , we let @xmath59 be the revenue earned by @xmath60 in @xmath47 , which is the total weight of edges in @xmath47 incident to @xmath60 . similarly , @xmath69 denotes the weight of the ( at most one ) edge of @xmath44 incident to @xmath60 .
let @xmath186 be the probability that @xmath60 is incident to @xmath64 edges in @xmath47 .
finally , we let @xmath187}$ ] and @xmath188}$ ] be the expected revenue earned by @xmath60 in @xmath47 and @xmath44 respectively , conditioned on @xmath60 being incident to @xmath64 edges in @xmath47 .
first , we note that @xmath188 } = \frac{{\mathbb{e}[{revenue}_2(v)|i]}}{i}$ ] , as for each set of coin flips in which @xmath60 has degree @xmath64 in @xmath47 , we may see any of the @xmath64 edges incident to @xmath60 first in the random permutation ; on average , then , we receive a @xmath189 fraction of @xmath59 .
we then have the following equations : @xmath190 } & = & \sum_{i=1}^{\infty } p_i \cdot { \mathbb{e}[{revenue}_2(v)|i]}.\\ { \mathbb{e}[{revenue}_3(v ) ] } & = & \sum_{i=1}^{\infty } p_i \cdot \frac{{\mathbb{e}[{revenue}_2(v)|i]}}{i}.\end{aligned}\ ] ] for ease of notation below , we use @xmath191 to denote @xmath187}$ ] .
we wish to bound @xmath192}$ ] in terms of @xmath193}$ ] , and we do this as follows : first , we show that @xmath194 , and @xmath195 .
next , we prove that subject to these constraints , the worst - case ratio of these two expectations occurs when all the constraints hold with equality .
we can then evaluate the sums , and show that @xmath192 } \ge p { \mathbb{e}[{revenue}_2(v)]}$ ] , completing our proof .
it is easy to see that @xmath195 ; consider any partial history of simulate in which @xmath67 edges incident to @xmath60 have been added to @xmath47 so far ; as we process edges in decreasing order of weight , the @xmath64th edge must be the lightest of those seen so far .
as this is true for each ( partial ) history , it holds in expectation , and so @xmath196 .
similarly , to see that @xmath197 , consider a partial history until the @xmath198st edge has just been added : @xmath47 will have @xmath67 edges incident to @xmath60 if the coin for the next edge incident to @xmath60 considered by simulate comes up `` heads '' , with probability @xmath38 .
@xmath47 will have @xmath64 edges incident to @xmath60 if the coin for the next edge incident to @xmath60 comes up `` tails '' , and that for the following edge comes up heads , with probability @xmath199 .
, in which case @xmath60 will have @xmath64 edges with probability @xmath200
. however , this only helps the analysis .
alternatively , one can assume the existence of a large number of `` zero - weight '' edges incident to @xmath60 .
] again , as this holds for each history , we have @xmath201 . to see that the worst - case ratio occurs
when all these constraints hold with equality , notice that the ratio between successive terms of equations ( 1 ) and ( 2 ) is increasing : the ratio between the @xmath64th terms is simply @xmath64 .
let @xmath202 denote the worst - case ratio of the expectations ; from lemma [ lem : wm3 ] , we already know that @xmath203 . if @xmath204 , for @xmath205 , the ratio between the @xmath64th term of the two sums is at most @xmath202 , while for @xmath206 , the ratio is greater than @xmath202 . consider a choice of @xmath191 s and @xmath186 s such that the ratio between ( 1 ) and ( 2 ) be as large as possible , and suppose the constraints on @xmath186 and @xmath191 do not all hold with equality .
let @xmath9 be an index such that @xmath207 or @xmath208 .
if @xmath209 , then by increasing @xmath210 or @xmath211 , we do not violate any constraint , and the increase in ( 1 ) is greater than @xmath202 times the increase in ( 2 ) .
similarly , if @xmath212 , by decreasing @xmath213 or @xmath214 to achieve equality , and also decreasing @xmath215 or @xmath216 to maintain feasibility , the decrease in ( 1 ) is less than @xmath202 times the decrease in ( 2 ) . in either of these situations , we increase the ratio between the two sums , contradicting our initial setting of @xmath217 .
finally , we can now evaluate this worst case ratio . setting @xmath218 and @xmath219
, we find : @xmath190 } & = & \sum_{i=1}^{\infty } i w_1 p_1 ( 1-p)^{i-1 } = w_1 p_1 / p^2 . \\ { \mathbb{e}[{revenue}_3(v ) ] } & = & \sum_{i=1}^{\infty } w_1 p_1 ( 1-p)^{i-1 } = w_1 p_1 / p = p { \mathbb{e}[{revenue}_2(v)]}\\\end{aligned}\ ] ] as @xmath220 } = { \mathbb{e}[w(m_2 ) ] } \ge ( 1-p ) { \text{\sc opt}}/2 $ ] , we have @xmath57 } \ge p ( 1-p ) { \text{\sc opt}}/2 $ ] , completing the proof of lemma [ lem : tighterbound ] .
theorem [ thm : bvmlogn ] we show that the algorithm of theorem [ thm : bvmlogn ] is @xmath221-competitive for bvm with groups , closely following the analysis of @xcite for an @xmath14-competitive algorithm for general matroids .
recall that the algorithm observes the first half of the vertices , and picks a random integer @xmath222 $ ] .
if @xmath151 is the weight of the heaviest edge seen so far , the algorithm sets a threshold of @xmath154 , and in the second half , greedily constructs a matching using edges of weight greater than this threshold .
let @xmath32 be an optimal matching ; we also abuse notation and use @xmath32 to refer to the weight of this matching , though the meaning will be clear from context .
let @xmath223 denote the weights of edges in @xmath32 , such that @xmath224 for @xmath225 .
let @xmath226 denote the largest index in @xmath227 $ ] such that @xmath228 . clearly , @xmath229 , as the remaining edges all have weight less than @xmath230 , and there are fewer than @xmath231 of them .
for any set of edges @xmath31 , we use @xmath232 to denote the number of edges in @xmath31 with weight at least @xmath191 , and @xmath233 to denote the number of edges in @xmath31 with weight at least @xmath234 .
now , we have : @xmath235 in order to obtain an @xmath150-competitive algorithm , it suffices to show that for each @xmath237 , @xmath238 } \ge n_i({\text{\sc opt}})/o(\log n)$ ] .
first , consider the case of @xmath239 : @xmath240 , and we argue that @xmath241 } \ge 1/4 ( { \lceil \log n \rceil } + 1)$ ] . with probability @xmath242 , the vertex @xmath60 incident to the heaviest edge appears in the second half , and the heaviest edge not incident to any vertex of @xmath60 s group appears in the first half . if this occurs , and the algorithm picks @xmath243 ( which happens with probability @xmath244 ) , then the only edges with weight above the threshold are those incident to vertices in @xmath60 s group .
thereore , the greedy algorithm will select the heaviest edge with probability @xmath245 , and hence @xmath241 } \ge 1/4 ( { \lceil \log n \rceil } + 1)$ ] .
we now complete the argument for each @xmath246 .
let @xmath60 be the vertex incident to the heaviest edge .
we consider two cases : first , that at least half the edges of @xmath32 with weight at least @xmath191 are incident to vertices not in the same group as @xmath60 , and second , that more than half these edges are incident to vertices of @xmath60 s group .
in the former case , suppose that @xmath60 is seen in the first half .
let @xmath151 be the weight of this heaviest edge , and let @xmath247 be the smallest integer in @xmath248 $ ] such that @xmath249
. may be greater than @xmath250 , as the heaviest edge may not be in @xmath251 .
however , it is easy to see that @xmath252 , and since @xmath253 , there always exists such an index @xmath247 . ] with probability @xmath254 , the algorithm picks @xmath255 , and the threshold is set to be @xmath256 .
let @xmath158 denote the event that the threshold is set to be @xmath257 ; as we have seen , @xmath258 \ge 1/2 ( { \lceil \log n \rceil } + 1)$ ] .
we show that conditioned on @xmath158 , @xmath238}$ ] is sufficiently large .
recall that @xmath32 contains a matching of size @xmath64 using edges of weight at least @xmath191 ; it follows that in expectation , using edges of this weight , there is a matching in the second half of size at least @xmath259 .
( this is because at least half of these @xmath64 edges are in other groups ; even conditioned on @xmath60 appearing in the first half , each of the remaining @xmath260 edges could appear in either half . )
since we construct a greedy matching using edges of weight at least @xmath234 , the expected size of this matching is at least @xmath261 .
hence , with probability at least @xmath262 , @xmath238 } \ge i/8 $ ] .
that is , @xmath238 } \ge i/16
( { \lceil \log n \rceil } + 1)$ ] .
we now consider the second case , when more than half the edges of @xmath32 with weight at least @xmath191 are in the same group as @xmath60 .
let @xmath108 be the vertex outside this group incident to the heaviest - weight edge .
suppose @xmath108 s group appears in the first half , and @xmath60 s group in the second .
let @xmath151 be the weight of the heaviest edge incident to @xmath108 ; if @xmath263 and we pick @xmath264 , the _ only _ edges above the threshold will be vertices in @xmath60 s group . since we construct the greedy matching using only the group of @xmath60 , and there exists a matching in this group with more than @xmath265 edges of weight @xmath191 , the matching we construct has at least @xmath259 edges of weight at least @xmath191 . if @xmath266 , then with probability @xmath267 , we pick an index @xmath152 such that @xmath268 .
again , we will find a matching in which at least @xmath259 edges have weight at least @xmath234 .
therefore , with probability at least @xmath269 , we find a matching of size at least @xmath259 .
therefore , @xmath238 } \ge i/16 ( { \lceil \log n \rceil } + 1)$ ] .
proposition [ prop : dir - tree ] let @xmath270 denote the heaviest edge incident to @xmath60 ; clearly @xmath271 .
it remains to show that this latter sum is at least @xmath272 . to see this ,
consider the tree @xmath181 , and root it arbitrarily .
for each edge @xmath273 , the weight of @xmath45 is at most @xmath270 , where @xmath60 is the vertex further from the root .
each vertex @xmath60 is charged by at most one edge , and so @xmath274 . | we examine several online matching problems , with applications to internet advertising reservation systems .
consider an edge - weighted bipartite graph @xmath0 , with partite sets @xmath1 .
we develop an @xmath2-competitive algorithm for the following secretary problem : initially given @xmath3 , and the size of @xmath4 , the algorithm receives the vertices of @xmath4 sequentially , in a random order .
when a vertex @xmath5 is seen , all edges incident to @xmath6 are revealed , together with their weights .
the algorithm must immediately either match @xmath6 to an available vertex of @xmath3 , or decide that @xmath6 will remain unmatched . in @xcite ,
the authors show a 16-competitive algorithm for the transversal matroid secretary problem , which is the special case with weights on vertices , not edges .
( equivalently , one may assume that for each @xmath7 , the weights on all edges incident to @xmath6 are identical . )
we use a similar algorithm , but simplify and improve the analysis to obtain a better competitive ratio for the more general problem .
perhaps of more interest is the fact that our analysis is easily extended to obtain competitive algorithms for similar problems , such as to find disjoint sets of edges in hypergraphs where edges arrive online .
we also introduce secretary problems with adversarially chosen _
groups_. finally , we give a @xmath8-competitive algorithm for the secretary problem on graphic matroids , where , with edges appearing online , the goal is to find a maximum - weight acyclic subgraph of a given graph . |
over the years compass models have proved especially useful in the modeling of materials with multi - orbital degrees of freedom@xcite ; these hamiltonians describe fictitious spin degrees of freedom that have fully anisotropic spatial interactions@xcite .
currently they are very prominent in both theoretical and experimental studies of the physics of iridium - oxide materials@xcite , the @xmath0rucl@xmath1 compound@xcite and finally the purported application of the kitaev model@xcite to quantum computing .
neutron scattering@xcite and nmr experiments@xcite have provided convincing evidence that kitaev model physics dominates the effective hamiltonian describing @xmath0rucl@xmath1 , although the inclusion of further interactions , especially heisenberg exchange , seem to be necessary for a consistent description of the experiments .
additionally , thermal conductivity experiments could shed more light on the relevant interactions in these exotic quantum magnets .
the standard theoretical approach to the spin dynamics analysis of the kitaev model is the elegant and powerful majorana fermion solution@xcite .
so far most studies have focused on the zero temperature limit@xcite . however , at finite temperatures monte carlo methods have been used@xcite to sample the auxiliary majorana fields and there are exact diagonalization studies on restricted lattices@xcite . in the high temperature limit ,
a long standing approach to spin dynamics has been the moment method@xcite . in this work ,
we point out that the peculiar structure of compass model hamiltonians maps the moment enumeration to a branching model , though there are certain extra initial constraints specific to each model . in the following ,
we discuss as examples the compass model on the two dimensional honeycomb lattice ( the so called kitaev model ) and the two - dimensional compass model on a square lattice .
our main focus is to compare the qualitative features obtained from different model branching processes to the exact spin autocorrelation results obtained by diagonalization of the hamiltonian on a finite size lattice .
the validity of the mapping rests so far on the qualitative agreement with frequency spectra obtained by ed .
it is amusing that such simple combinatorial models reproduce the essential features of these quantum many - body models , providing another instance of correspondence between quantum mechanical and statistical mechanics problems .
it would be interesting also to consider the inverse procedure , where statistical mechanics problems can be solved by corresponding compass models .
primarily we are interested in the evaluation of the spin autocorrelation function @xmath2 where @xmath3 are pauli spin-1/2 operators , and @xmath4 , @xmath5
. we will also briefly discuss further spatial correlations , @xmath6 @xmath7 denotes a thermal average at temperature @xmath8 ( with @xmath9 ) . in the infinite temperature limit , @xmath10
, the autocorrelation function reduces to a trace over all the hilbert space , @xmath11 @xmath12 being the number of spins on the lattice .
expanding in powers of time , @xmath13 the autocorrelation function analysis reduces to the evaluation of the moments @xmath14@xcite , @xmath15 where @xmath16=ha - ah$ ] is the liouville operator .
the time fourier transform of @xmath17 , @xmath18 can alternatively be evaluated by an extension in complex frequencies @xmath19 , @xmath20 @xmath21.\ ] ] @xmath22 is then conveniently expressed as a continued fraction expansion , @xmath23 with the coefficients @xmath24 related the moments @xmath14 by recursion relations@xcite . in the following
we will present an approximate evaluation of the moments @xmath14 and the corresponding structure of @xmath24 by mapping them to combinatorial branching models .
the kitaev model on a honeycomb lattice is given by the hamiltonian , @xmath25 where @xmath26 denotes the nearest neighbor bonds in the three directions on the lattice , with the convention indicated in fig.[ktv](0 ) .
the central operator @xmath27 ( that subsequently is abbreviated just by @xmath28 ) represents the operator @xmath29 at the initial time @xmath30 .
we find that repeatedly applying the liouville operator , @xmath31 , as in ( [ liouville ] ) creates strings of operators as depicted in the subsequent parts of fig.[ktv ] . because of the 3-fold geometry of the honeycomb lattice ,
the string operators are organized into three branches , each one starting with an @xmath32 , @xmath33 or @xmath34 bond .
fig.[ktv]@xmath35 indicate the two possible operator strings created by one application of @xmath31 on @xmath36 .
the third sting , along the z - bond , vanishes at this order . in ( @xmath37 )
the four possible ( non - vanishing ) strings created by the application of @xmath31 on ( @xmath38 ) are shown ( note that there are an additional four , symmetrically related , strings for ( @xmath39 ) which are not shown ) .
then , the application of the trace in ( [ liouville ] ) gives zero for all but the first diagram shown in fig.[ktv]@xmath40 : the zero - length string which is just the operator @xmath41 .
consequently there are only two contributions to the second moment , @xmath42 : this one and the analogous diagram arising from the application of @xmath31 on ( @xmath39).the problem of applying @xmath31 on a operator string is therefore mapped to a branching process in three directions , where at each iteration the tail operator in a branch either disappears or a new one is created in one of the three possible directions ( although with some specific restrictions at the origin which we will discus later ) . in the high temperature limit , the trace of any string of non - zero length vanishes .
thus , evaluating the moments reduces to counting all of the possible branching processes that culminate in the complete annihilation of the strings ( i.e. they return to just a single @xmath41 operator ) .
each branching configuration contributes a factor of @xmath43 from each application of @xmath31 to the value of the contribution to a moment .
this arises from the spin commutation involving the bond-@xmath44 ; the factor for annihilating a particular link turns out to always have the opposite sign to that corresponding to its creation . as an example , the third order diagram shown in fig.[ktv]@xmath45 represents the operator string : @xmath46 henceforth we will only consider @xmath47 , although generalising to the anisotropic case is feasible . in summary , it therefore seems that the minimal model is branching in three independent directions .
however , as we will now discuss , this is not quite adequate .
there are some extra initial constraints and also some `` higher order '' processes that need consideration .
we find that the branching model must obey the following ` rules ' : \(i ) side - branching is not possible and the operator at a vertex is given by the direction of the missing bond ; \(ii ) applying @xmath31 in the middle of a branch annihilates the entire operator string ; \(iii ) independent branching along all three directions is not possible : there are initial restrictions
. a branch along the @xmath19-direction can not be created at first order .
instead , it can only be initiated when either an @xmath48-direction or a @xmath49-direction branch already exists , as depicted in fig.[ktv](@xmath37 ) .
this initial restriction effects the lowest order moments and thus the high frequency behavior of the spectral function .
\(iv ) branches must be annihilated in the reverse order to which they were created or the contribution from the entire string vanishes .
\(v ) possible higher order processes are not taken into account where the branching structure is modified e.g. at the origin and then eventually reconstructed and annihilated .
\(vi ) finally , as shown in fig.[ktv]@xmath50 , the branching process is self - avoiding . values . ]
[ ktvspectrum ] in the inset of fig.[ktv ] we show the values of @xmath24 evaluated from the moments @xmath14 for the two models , the branching process without ( shown in red ) and with ( shown in black ) the initial constraints .
it is clear that for large values of @xmath51 the @xmath24 s asymptotically coincide . in these branching model calculations , however , we have omitted the self - avoidance constraint . to estimate the applicability of the branching model we compare in fig.[ktv ] the spectra obrained by the branching models with ( c ) and without ( un ) initial constraints
also shown are results from an exact diagonalization study ( ed ) on a lattice of 4 by 6 spins . with periodic boundary conditions . here
we calculated in the high temperature limit using the microcanonical lanczos method@xcite .
100 lanczos iterations were used to converge to an infinite temperature state and 400 further iterations to obtain the continuous fraction expansion .
the @xmath52function peak at zero frequency is a finite size effect due to an excess of degenerate states in small lattices .
the agreement is fair with the model when including initial constraints ( despite omitting the self - avoidance requirement ) and can be fitted by a form @xmath53 .
next , we will study the 2d - compass model , using the same approach as previously .
the hamiltonian is given by , @xmath54 with the bond - labeling convention used depicted in fig.[compass](0 ) .
analogously to our discussion of the kitaev model , repeated applications of the liouville operator creates strings of operators which can propagate with up to four branches : see fig.[compass ] . the same procedure for obtaining the moments @xmath14 and corresponding @xmath24 of this combinatorial model can then be applied , but with the same reservations on possibly omitted high - order processes .
this time there are a different set of restrictions specific to the 2d compass model .
firstly , we find that the operator strings must propagate alternatively along @xmath48-links and @xmath49-links . trying to extend the string along two consecutive @xmath48-links ( or @xmath49-links )
causes the entire string to vanish .
this also means that the internal spin operators along the length of the string are all @xmath27 s .
secondly , we note that , as there was in the kitaev model , there are initial constraints on the order in which the four different branches can be created and subsequently destroyed .
finally , the branching model is again self - avoiding . to investigate the infinite temperature spin - autocorrelation for the isotropic 2d compass model we have calculated a finite number of moments for two different branching models : one which includes the initial constraints , and one with out . in both cases ,
however , we have again omitted the self - avoidance requirement .
the @xmath24 parameters that we have calculated are shown in the inset of fig.[compassspectrum ]
. values . ]
the subsequent @xmath55 spectrum is shown in fig.[compassspectrum ] for the branching model including the initial branch - creation - order constraint ( c ) and without this initial constraint ( un ) .
these are compared to the spectrum obtained by a microcanonical lanczos calculation ( ed ) on a lattice of 4 by 6 spins .
it is interesting to see that even the non - monotonic behaviour of @xmath55 is reproduced when including the initial branch creation - order constraint .
in contrast , the unconstrained branching behaviour can not capture this , indicating that it arises due to the degeneracy of the first two @xmath56 values . by using the same approach of expanding the time dependance as a function of the liouville operator
, we can also consider further - range spin correlation in the high temperature limit .
@xmath57 consequently , we deduce , that for @xmath58 all correlations vanish in the infinite temperature limit .
this occurs because the initial pair of operators @xmath59 can not be reduced to the identity by application of @xmath31 .
next we can consider finite temperatures by expanding @xmath60 in powers of @xmath61 .
then we can plausibly argue that we obtain a finite value for the correlations @xmath62 , where @xmath63 is a nearest neighbor site in the @xmath44 lattice direction . however , further - neighbour correlations all vanish identically at all temperatures . again
, this is because it is impossible to reduce any of the resulting operator strings to the identity by application of either @xmath31 or @xmath61 .
this result has already been found for the kitaev model using the majorana fermion approach@xcite , but our analysis shows that it also holds for the 2d compass model too .
presumably this is also the case for other quantum compass models where the is no analytical solution is not known . in conclusion , the above analysis , to an extent heuristic , provides an interesting new perspective .
firstly , we have discovered that the infinite temperature spin - autocorrelation function of quantum compass models can be mapped onto a statistical - mechanical branching problem ( along with some additional , model - specific restrictions ) .
we have discussed the results from calculating a finite number of moments for two different isotropic compass models , investigating the effects of appropriately including the model - specific initial restricts on the branching - scheme . despite omitting the self - avoidance requirement ,
this provided a useful insight into the shape of these correlation functions .
these techniques also have the potential to be generalised to anisotropic couplings @xmath64 and the evaluation of finite temperature correlations .
it may be feasible to obtain analytic expressions for the moments of a constraint - free branching model , but it is clear that an exact evaluation of the autocorrelation function requires the implementation of self - avoidance , which is well - known to be a highly non - trivial problem .
this raises the interesting question of whether the inverse procedure : solving self - avoiding statistical mechanics problems by means of a mapping onto a corresponding quantum compass model could be used .
this appears to be another instance of corresponding between quantum mechanical and statistical mechanics problems .
finally , we note that other types of correlations functions , for example those related to transport properties , can also be studied using this technique .
this work was supported by the european union program no .
fp7-regpot-2012 - 2013 - 1 under grant no .
a.k.r.b . acknowledges the hospitality of mpi - pks extended to her during visits and x.z .
discussions with b. bchner and c. hess that motivated this study as well as generous support by the alexander von humboldt foundation .
a. banerjee , c.a .
bridges , j - q .
yan , a.a .
aczel , l. li , m.b .
stone , g.e .
granroth , m.d .
lumsden , y. yiu , j. knolle , d.l .
kovrizhin , s. bhattacharjee , r. moessner , d.a .
tennant , d.g .
mandrus , s.e .
nagler , nature materials * 15 * , 733 ( 2016 ) | we analyze the high temperature spin dynamics of compass models using a moment expansion .
we point out that the evaluation of moments maps to the enumeration of paths in a branching process on the lattice .
this mapping to a statistical mechanics combinatorics problem provides an elegant visualization of the analysis .
we present results for the time dependent spin correlation function ( which is of relevance to nmr experiments ) for two compass models : the kitaev honeycomb model and two - dimensional compass model . |
recent observational data imply -against any previous belief- that the current expansion of the universe is accelerating @xcite . since this discovery , the so - called dark energy issue has probably become the most ambitious and tantalizing field of research because of its implications in fundamental physics .
there exist several descriptions of the cosmic acceleration . among them ,
the simplest one is the introduction of small positive cosmological constant in the framework of general relativity ( gr ) , the so - called @xmath1cdm model , but it is well accepted the idea according to which this is not the ultimate theory of gravity , but an extremely good approximation valid in the present day range of detection .
a generalization of this simple modification of gr consists in considering modified gravitational theories @xcite . in the last years
the interest in modified gravity theories like @xmath2 and @xmath3-gravity as alternatives to the @xmath1cdm model grew up .
recently , a new modified gravity theory , namely the @xmath4-theory , has been proposed .
this is a generalized version of the teleparallel gravity originally proposed by einstein @xcite-@xcite .
it also may describe the current cosmic acceleration without invoking dark energy . unlike the framework of gr , where the levi - civita connection is used , in teleparallel gravity ( tg ) the used connection is the weitzenbckone . in principle , modification of gravity may contain a huge list of invariants and there is not any reason to restrict the gravitational theory to gr , tg , @xmath2 gravity and/or @xmath4 gravity . indeed , several generalizations of these theories have been proposed ( see e.g. the quite recent review @xcite ) . in this paper , we study some other generalizations of @xmath2 and @xmath4 gravity theories . at the beginning , we briefly review the formalism of @xmath2 gravity and @xmath4 gravity in friedmann - robertson - walker ( frw ) universe . the flat frw space - time is described by the metric @xmath5 where @xmath6 is the scale factor .
the orthonormal tetrad components @xmath7 are related to the metric through @xmath8 where the latin indices @xmath9 , @xmath10 run over 0 ... 3 for the tangent space of the manifold , while the greek letters @xmath11 , @xmath12 are the coordinate indices on the manifold , also running over 0 ... 3 .
@xmath2 and @xmath4 modified theories of gravity have been extensively explored and the possibility to construct viable models in their frameworks has been carefully analyzed in several papers ( see @xcite for a recent review ) . for such theories , the physical motivations are principally related to the possibility to reach a more realistic representation of the gravitational fields near curvature singularities and to create some first order approximation for the quantum theory of gravitational fields .
recently , it has been registred a renaissance of @xmath2 and @xmath4 gravity theories in the attempt to explain the late - time accelerated expansion of the universe @xcite . in the modern cosmology , in order to construct ( generalized ) gravity theories , three quantities the curvature scalar , the gauss bonnet scalar and the torsion scalar are usually used ( about our notations see below ) : @xmath13 in this paper , our aim is to replace these quantities with the other three variables in the form @xmath14 where @xmath15 , @xmath16 and @xmath17 are some functions to be defined . as a result
, we obtain some generalizations of the known modified gravity theories . with the frw metric ansatz
the three variables ( [ 1.3])-([1.5 ] ) become @xmath18 where @xmath19 . in the contrast , in this paper we will use the following three variables @xmath20 this paper is organized as follows . in sec .
2 , we briefly review the formalism of @xmath2 and @xmath4-gravity for frw metric .
in particular , the corresponding lagrangians are explicitly presented . in sec . 3
, we consider @xmath21 theory , where @xmath22 and @xmath23 will be generalized with respect to the usual notions of curvature scalar and torsion scalar .
some reductions of @xmath21 gravity are presented in sec .
4 . in sec . 5
, the specific model @xmath24 is * analized * and in sec . 6
the exact power - law solution is found ; some cosmological implications of the model will be here discussed .
the bianchi type i version of @xmath21 gravity is considered in sec .
8 is devoted to some generalizations of some modified gravity theories .
final conclusions and remarks are provided in sec .
at the beginning , we present the basic equations of @xmath2 , @xmath4 and @xmath3 modified gravity theories . for simplicity we mainly work in the frw spacetime .
the action of @xmath2 theory is given by @xmath25,\ ] ] where @xmath22 is the curvature scalar .
we work with the frw metric . in this case
@xmath22 assumes the form @xmath26 the action we rewrite as @xmath27 where the lagrangian is given by @xmath28 the corresponding field equations of @xmath2 gravity read @xmath29f_{rr}+[-2h^2 - 4a^{-1 } \ddot{a}+r]f_{r}-f & = & p,\label{2.6}\\ \dot{\rho}+3h(\rho+p)&=&0.\label{2.7 } \end{aligned}\ ] ] in the modified teleparallel gravity , the gravitational action is @xmath30,\ ] ] where @xmath31 , and for convenience we use the units @xmath32 throughout .
the torsion scalar @xmath23 is defined as @xmath33 where @xmath34 for a spatially flat frw metric , as a consequence of equations ( [ 2.9 ] ) and , we have that the torsion scalar assumes the form @xmath35 the action can be written as @xmath36 where the point - like lagrangian reads @xmath37 the equations of f(t ) gravity look like @xmath38 the action of @xmath3 theory is given by @xmath39,\ ] ] where the gauss bonnet scalar @xmath40 for the frw metric is @xmath41
our aim in this section is to present a naive version of @xmath21 gravity .
we assume that the relevant action of @xmath21 theory is given by @xcite @xmath42,\ ] ] where @xmath43 and @xmath44 are some dynamical geometrical variables to be defined , and @xmath45 and @xmath46 are the usual curvature scalar and the torsion scalar for the frw spacetime .
it is the so - called m@xmath47 - model @xcite . in this paper
we will restrict ourselves to the simple case where for frw spacetime @xmath22 and @xmath23 are given by @xmath48 as we can see these two variables ( @xmath49 ) are some analogies ( generalizations ) of the usual curvature scalar @xmath50 ) and torsion scalar ( @xmath46 ) and for obvious reasons we will still continue to call them as the `` curvature '' scalar `` and the ' ' torsion " scalar .
we note that , in general , @xmath51 and @xmath52 are some real functions , @xmath53 , while @xmath54 and @xmath55 are some unknown functions related with the geometry of the spacetime .
finally we can write the m@xmath47 - model for the frw spacetime as @xmath56,\label{3.4}\\ r&=&u+6(\dot{h}+2h^2),\label{3.5}\\ t&=&v-6h^2.\label{3.6}\end{aligned}\ ] ] in this paper we restrict ourselves to the case @xmath57 and @xmath58 .
the scale factor @xmath59 , the curvature scalar @xmath22 and the torsion scalar @xmath60 are taken as independent dynamical variables .
then , after some algebra the action becomes @xmath61 where the point - like lagrangian is given by @xmath62 the corresponding equations of the m@xmath47 - model assume the form @xcite @xmath63 here @xmath64 and @xmath65 we can rewrite the system in terms of @xmath66 as @xmath67 where @xmath68 and @xmath69
note that the system ( [ 3.9 ] ) or admits some important reductions .
let us now present these particular cases . now we consider the particular case @xmath70 .
thus , the system becomes @xmath71 or @xmath72 let us rewrite this system as @xmath73 where @xmath74 are the corrections to the energy denisty and pressure .
note that if @xmath75 we obtain the standard equations of gr , @xmath76 let us now to consider @xmath77 .
then the system leads to @xmath78 or @xmath79 the above system can be rewritten as @xmath80 where @xmath81 are the corrections to the energy density and pressure . obviously , if @xmath82 we obtain the standard equations of gr .
let us take @xmath83 .
then , the system becomes @xmath84 or @xmath85 this system can be rewritten as @xmath86 that is the same as - of @xmath4 gravity .
we get the second reduction if we consider the case where @xmath87 .
then the system ( [ 3.9 ] ) leads to @xmath88 where @xmath89 this system can be written as @xmath90f_r - a ^3f&=&-2a^3\rho,\label{4.30}\\ -6\dot{r}^2a^2f_{rrr}+[-12\dot{r}a \dot{a}-6\ddot{r}a^2]f_{rr}+[-6\dot{a}^2 - 12a \ddot{a}+3a^2r]f_{r}-3a^2f & = & 6a^2p,\label{4.31}\\ \dot{\rho}+3h(\rho+p)&=&0.\label{4.32 } \end{aligned}\ ] ] as a consequence , @xmath91f_{rr}+[-2h^2 - 4a^{-1 } \ddot{a}+r]f_{r}-f & = & 2p,\label{4.34}\\ \dot{\rho}+3h(\rho+p)&=&0.\label{4.35 } \end{aligned}\ ] ] this system corresponds to the one in equations ( [ 2.5])-([2.7 ] ) .
we have shown that our model contents @xmath2 and @xmath4 gravity models as particular cases . in this sense
it is the generalizations of these two known modified gravity theories .
for the m@xmath92 - model we have @xmath93 so that @xmath94,\label{4.36}\\ r&=&u+6(\dot{h}+2h^2),\label{4.37}\\ t&=&-6h^2.\label{4.38}\end{aligned}\ ] ]
if we consider the case @xmath96 , then we get the m@xmath95 - model with @xmath97,\label{4.39}\\ r&=&6(\dot{h}+2h^2),\label{4.40}\\ t&=&v-6h^2.\label{4.41}\end{aligned}\ ] ]
now we consider the case @xmath99 .
we get the m@xmath98 - model with @xmath97,\label{4.42}\\ r&=&u+6(\dot{h}+2h^2),\label{4.43}\\ t&=&\zeta(u)-6h^2,\label{4.44}\end{aligned}\ ] ] where in general @xmath100 is a function to be defined e.g. @xmath101 and @xmath102 is an unknown function .
now we consider the particular case of @xmath104 and we get the m@xmath103 - model with @xmath97,\label{4.45}\\ r&=&\xi(v)+6(\dot{h}+2h^2),\label{4.46}\\ t&=&v-6h^2,\label{4.47}\end{aligned}\ ] ] where in general @xmath105 is a function to be defined e.g. @xmath106 and @xmath102 is an unknown function .
finally we consider the case @xmath108 and we get the m@xmath107 - model with @xmath109,\label{4.48}\\ r&=&6(\dot{h}+2h^2),\label{4.49}\\ t&=&-6h^2.\label{4.50}\end{aligned}\ ] ] about this model we have some doubt related with the equation @xmath110 which follows from - by avoiding the variable @xmath66 .
this equation tell us that we have only one independent dynamical variable @xmath22 or @xmath23 .
it turns out that the model - is not of the type of @xmath21 gravity , but is equivalent to @xmath2 or @xmath4 gravity only .
this is why in this paper we introduced some new functions like @xmath111 and @xmath112 with the ( temporally ? ) unknown geometrical nature .
the m@xmath113 - model corresponds to the case @xmath114 that is @xmath115 as a consequence the m@xmath113 - model reads @xmath116,\label{4.54}\\ r&=&0,\label{4.55}\\ t&=&v-6h^2.\label{4.56}\end{aligned}\ ] ] we can see that the m@xmath113 - model is in fact a generalization of @xmath4 gravity .
we obtain the m@xmath117 - model by assuming @xmath118 that is @xmath119 in this way we write the m@xmath117 - model as @xmath120,\label{4.59}\\ r&=&u+6(\dot{h}+2h^2),\label{4.60}\\ t&=&0.\label{4.61}\end{aligned}\ ] ] this model is in fact a generalization of @xmath2 gravity .
the equations of @xmath21 gravity are much more complicated with respect to the ones of gr even for frw metric . for
this reason let us consider the following simplest particular model @xmath121 where @xmath11 and @xmath12 are some real constants .
the equations system of @xmath21 gravity becomes @xmath122 where @xmath123 we get@xmath124 may rewrite it as @xmath125 or @xmath126 this system contents 2 equations and 5 unknown functions ( @xmath127 ) . note that the eos parameter is given by @xmath128 now we assume @xmath129 where @xmath130 are some real constants so that we have @xmath131 then , the previous system - leads to @xmath132 or @xmath133
here we are interested in the cosmological implications of the model relating to the dark energy problem . in order to satisfy our interest ,
let us consider the power - law solution in the form @xmath134 where @xmath135 and @xmath136 are contants .
thus , @xmath137.\label{6.3 } \end{aligned}\ ] ] with respect of the cosmic time @xmath138 for eq .
( 125 ) ] the eos parameter reads @xmath139 these expressions still content some unknown constant parameters .
we assume that these parameters have the following values , namely @xmath140 , @xmath141 .
tin this case one has @xmath142 so that the eos takes the form @xmath143 furthermore , the eos parameter becomes @xmath144 hence , we see that @xmath145 , @xmath146 and @xmath147 , so that our particular case admits the phantom crossing for @xmath148 as @xmath149 .
in fig.1 we plot the evolution of the eos parameter with respect to the cosmic time @xmath138 .
it is interesting to compare this result with the torsionless case with @xmath150 , by taking the same values for all the other parameters , namely @xmath151 and @xmath141 , which is the case of gr . as a consequence @xmath152 and @xmath153 , which describe the dust matter .
the results of the section 3 can be extendent to the other metric . as an example , let us consider the m@xmath47 - model for the bianchi type spacetime .
the corresponding metric is given by @xmath154 in this case the m@xmath47 - model reads as @xmath155,\\ r&=&u+2\left(\frac{\ddot{a}}{a}+\frac{\ddot{b}}{b}+\frac{\ddot{c}}{c}+\frac{\dot{a}\dot{b}}{ab}+\frac{\dot{a}\dot{c}}{ac}+\frac{\dot{b}\dot{c}}{bc}\right),\label{7.3}\\ t&=&v-2\left(\frac{\dot{a}\dot{b}}{ab}+\frac{\dot{a}\dot{c}}{ac}+\frac{\dot{b}\dot{c}}{bc}\right).\label{7.4}\end{aligned}\ ] ] here @xmath156 and @xmath157
now we consider the m@xmath158 - model which looks like @xmath159,\label{8.1}\\ g&=&w+24h^2(\dot{h}+h^2),\label{8.2}\\ w&=&w(t , a,\dot{a } , \ddot{a},\dddot{a } , ... ; h_i),\label{8.3}\end{aligned}\ ] ]
where , again , @xmath160 is a real function and @xmath161 is an unknown function related to the geometry of the spacetime . if @xmath162 the m@xmath158 - model reduces to the usual @xmath3 gravity with @xmath163 .
now we consider the m@xmath164 - model which reads @xmath165,\ ] ] where @xmath166 here , @xmath51 and @xmath160 are some real functions and @xmath167 are some unknown functions relatedto the geometry of the spacetime .
note that if we put @xmath168 , the m@xmath164 - model reduces to the usual @xmath169 gravity .
let us consider the following action of the m@xmath170 - model @xmath171,\ ] ] where @xmath172 here in general @xmath160 and @xmath52 are some real functions and @xmath161 and @xmath55 are some unknown functions related with the geometry of the spacetime .
now we consider the m@xmath173 - model with the following action @xmath174,\ ] ] where @xmath175 here , again , @xmath51 , @xmath160 and @xmath52 are some real functions and @xmath167 are some unknown functions related to the geometry of the spacetime .
let us consider the m@xmath176 - model with the action @xmath177,\ ] ] where @xmath178 here @xmath179 , where @xmath180 and @xmath181 are some unknown functions .
this model admits ( at least ) two important particular cases . _
a ) the m@xmath182 model .
_ let us take @xmath183 .
then @xmath184 , @xmath185 and @xmath186 , so that we get purely @xmath4 gravity .
_ b ) the m@xmath187 model .
_ let us take now @xmath188 .
then , @xmath189 , @xmath190 and @xmath191 .
this case corresponds to the purely @xmath2 gravity .
as it is well known , modified gravity theories play an important role in modern cosmology . in particular , the well - known @xmath2 and @xmath4 theories are useful tools to study dark energy phenomena motivated at a fundamental level . in the present work ,
we have considered the more general theory , namely the @xmath21- models . at first , we have written the equations of the model and we have found their several reductions . in particular , the lagrangian has been explicitly constructed .
the corresponding exact solutions are found for the specific model @xmath24 theory , for which the universe expands as @xmath192 .
furthermore , we have considered the physical quantities corresponding to the exact solution , and we have found that it can describe the expansion of our universe in an accelerated way without introducing the dark energy .
some remarks are in order . of course many aspects of @xmath21 theory are actually unexplored .
for example , we do not have any realistic model which fits the cosmological data , unlike @xmath2 or @xmath4 theory .
we do not know viability conditions of the models , , what forms of @xmath21 can be derived from fundamental theories and so on ( it may be extremely important to reconstract a @xmath21-theory by starting from some basical principles ) . on the other hand , we have here shown that the @xmath21 models can be serious candidates as modified gravity models for the dark energy .
also we note that the behaviour and the results of @xmath21-gravity may be extremely different with respect to the ones of gr , @xmath2 and @xmath4 gravity theories , so that only the observation of our universe may discriminate between the various gravity theories .
we not want here discuss merits and demerits of the models above , since we think that it requires some more accurate investigations related to cosmological applications .
e. j. copeland , m. sami and s. tsujikawa , int .
j. mod .
d * 15 * , 1753 ( 2006 ) [ hep - th/0603057 ] ; + j. frieman , m. turner and d. huterer , ann .
astrophys . * 46 * , 385 ( 2008 ) [ arxiv:0803.0982 ] ; + s. tsujikawa , arxiv:1004.1493 [ astro-ph.co ] ; + m. li , x. d. li , s. wang and y. wang , commun . theor . phys . *
56 * , 525 ( 2011 ) [ arxiv:1103.5870 ] ; + y. f. cai , e. n. saridakis , m. r. setare and j. q. xia , phys . rept . *
493 * , 1 ( 2010 ) [ arxiv:0909.2776 ] .
a. de felice and s. tsujikawa , living rev .
* 13 * , 3 ( 2010 ) [ arxiv:1002.4928 ] ; + t. clifton , p. g. ferreira , a. padilla and c. skordis , arxiv:1106.2476 [ astro-ph.co ] ; + t. p. sotiriou and v. faraoni , rev .
* 82 * , 451 ( 2010 ) [ arxiv:0805.1726 ] ; + s. tsujikawa , lect .
notes phys .
* 800 * , 99 ( 2010 ) [ arxiv:1101.0191 ] ; + s. capozziello , m. de laurentis and v. faraoni , arxiv:0909.4672 [ gr - qc ] ; + r. durrer and r. maartens , arxiv:0811.4132 [ astro - ph ] ; + s. nojiri and s. d. odintsov , int .
j. geom .
phys . * 4 * , 115 ( 2007 ) [ hep - th/0601213 ] ; + s. nojiri and s. d. odintsov , arxiv:1011.0544 [ gr - qc ] .
p. x. wu and h. w. yu , phys .
b * 693 * , 415 ( 2010 ) [ arxiv:1006.0674 ] ; + p. x. wu and h. w. yu , phys .
b * 692 * , 176 ( 2010 ) [ arxiv:1007.2348 ] ; + p. x. wu and h. w. yu , eur .
j. c * 71 * , 1552 ( 2011 ) [ arxiv:1008.3669 ] ; + p. x. wu and h. w. yu , phys .
b * 703 * , 223 ( 2011 ) [ arxiv:1108.5908 ] ; + k. bamba , c. q. geng , c. c. lee and l. w. luo , jcap * 1101 * , 021 ( 2011 ) [ arxiv:1011.0508 ] ; + k. bamba , c. q. geng and c. c. lee , arxiv:1008.4036 [ astro-ph.co ] ; + y. zhang , h. li , y. g. gong and z. h. zhu , jcap * 1107 * , 015 ( 2011 ) [ arxiv:1103.0719 ] ; + y. f. cai , s. h. chen , j. b. dent , s. dutta and e. n. saridakis , arxiv:1104.4349 [ astro-ph.co ] ; + m. li , r. x. miao and y. g. miao , jhep * 1107 * , 108 ( 2011 ) [ arxiv:1105.5934 ] ; + r. x. miao , m. li and y. g. miao , jcap * 1111 * , 033 ( 2011 ) [ arxiv:1107.0515 ] ; + x. h. meng and y. b. wang , eur . phys .
j. c * 71 * , 1755 ( 2011 ) [ arxiv:1107.0629 ] ; + r. j. yang , eur .
j. c * 71 * , 1797 ( 2011 ) [ arxiv:1007.3571 ] ; + t. wang , phys .
d * 84 * , 024042 ( 2011 ) [ arxiv:1102.4410 ] ; + r. zheng and q. g. huang , jcap * 1103 * , 002 ( 2011 ) [ arxiv:1010.3512 ] .
b. li , t. p. sotiriou and j. d. barrow , phys .
d * 83 * , 104017 ( 2011 ) [ arxiv:1103.2786 ] ; + b. li , t. p. sotiriou and j. d. barrow , phys .
d * 83 * , 064035 ( 2011 ) [ arxiv:1010.1041 ] ; + t. p. sotiriou , b. li and j. d. barrow , phys .
d * 83 * , 104030 ( 2011 ) [ arxiv:1012.4039 ] .
g. r. bengochea , phys .
b * 695 * , 405 ( 2011 ) [ arxiv:1008.3188 ] ; + s. h. chen , j. b. dent , s. dutta and e. n. saridakis , phys .
d * 83 * , 023508 ( 2011 ) [ arxiv:1008.1250 ] ; + j. b. dent , s. dutta and e. n. saridakis , jcap * 1101 * , 009 ( 2011 ) [ arxiv:1010.2215 ] ; + r. ferraro and f. fiorini , phys
. lett .
b * 702 * , 75 ( 2011 ) [ arxiv:1103.0824 ] ; + r. ferraro and f. fiorini , arxiv:1106.6349 [ gr - qc ] ; + c. g. boehmer , a. mussa and n. tamanini , arxiv:1107.4455 [ gr - qc ] . c. deliduman and b. yapiskan , arxiv:1103.2225 [ gr - qc ] ; + r. myrzakulov , eur .
j. c * 71 * , 1752 ( 2011 ) [ arxiv:1006.1120 ] ; + k. k. yerzhanov , s. r. myrzakul , i. i. kulnazarov and r. myrzakulov , arxiv:1006.3879 [ gr - qc ] ; + r. myrzakulov , arxiv:1008.4486 [ astro-ph.co ] ; + k. karami and a. abdolmaleki , arxiv:1009.2459 [ gr - qc ] ; + k. karami and a. abdolmaleki , arxiv:1009.3587 [ physics.gen-ph ] ; + p. b. khatua , s. chakraborty and u. debnath , arxiv:1105.3393 [ physics.gen-ph ] ; + s. chattopadhyay and u. debnath , int . j. mod . phys .
d * 20 * , 1135 ( 2011 ) [ arxiv:1105.1091 ] ; + m. sharif and s. rani , mod .
a * 26 * , 1657 ( 2011 ) [ arxiv:1105.6228 ] .
m. h. daouda , m. e. rodrigues and m. j. s. houndjo , eur .
j. c * 71 * , 1817 ( 2011 ) [ arxiv:1108.2920 ] ; + m. h. daouda , m. e. rodrigues and m. j. s. houndjo , arxiv:1109.0528 [ physics.gen-ph ] ; + c. q. geng , c. c. lee , e. n. saridakis and y. p.
wu , phys .
b * 704 * , 384 ( 2011 ) [ arxiv:1109.1092 ] ; + c. q. geng , c. c. lee and e. n. saridakis , arxiv:1110.0913 [ astro-ph.co ] ; + y. p. wu and c. q. geng , arxiv:1110.3099 [ gr - qc ] ; + m. r. setare and f. darabi , arxiv:1110.3962 [ physics.gen-ph ] ; + r. ferraro and f. fiorini , phys . rev .
d * 84 * , 083518 ( 2011 ) [ arxiv:1109.4209 ] ; + p. a. gonzalez , e. n. saridakis and y. vasquez , arxiv:1110.4024 [ gr - qc ] ; + g. g. l. nashed , arxiv:1111.0003 [ physics.gen-ph ] ; + m. sharif and s. rani , phys .
scripta * 84 * , 055005 ( 2011 ) [ arxiv:1111.3599 ] ; + m. h. daouda , m. e. rodrigues and m. j. s. houndjo , arxiv:1111.6575 [ gr - qc ] ; + c.g .
boehmer , a. mussa , n. tamanini .
_ existence of relativistic stars in f(t ) gravity _ , arxiv:1107.4455 ; + n. tamanini , c. g. boehmer .
_ good and bad tetrads in f(t ) gravity _ , arxiv:1204.4593 .
r. myrzakulov , arxiv:1204.1093 ; arxiv:1201.4360 .
r. myrzakulov , arxiv:1008.4486 .
lopez - revelles , r. myrzakulov , d. saez - gomez , physical review d , * 85 * , n10 , 103521 ( 2012 ) .
k. bamba , r. myrzakulov , s. nojiri , s. d. odintsov , physical review d , * 85 * , n10 , 104036 ( 2012 ) .
k. bamba , s. capozziello , s. nojiri , s. d. odintsov , arxiv:1205.3421 . l. fabbri , s. vignolo . _ a modified theory of gravity with torsion and its applications to cosmology and particle physics _
, arxiv:1201.5498 .
. _ variational approach to gravitational theories with two independent connections _
, arxiv:1205.2511 . | since the discovery of cosmic acceleration , modified gravity theories play an important role in the modern cosmology . in particular , the well - known f(r)-theories reached great popularity motivated by the easier formalism and by the prospect to find a final theories of gravity for the dark scenarios . in the present work ,
we study some generalizations of f(r ) and f(t ) gravity theories . at the beginning
, we briefly review the formalism of such theories .
then , we will consider one of their generalizations , the so - called f(r , t)-theory .
the point - like lagrangian is explicitly presented .
based on this lagrangian , the field equations of f(r , t)-gravity are found .
for the specific model @xmath0 the corresponding exact solutions are derived .
furthermore , we will consider the physical quantities associated to such solutions and we will find how for some values of the parameters the expansion of our universe can be accelerated without introducing any dark component . |
the early pioneering sub - millimeter surveys in lensing clusters ( smail , ivison & blain 1997 ) and in blank fields ( hughes et al .
1998 , barger et al . 1998 ) demonstrated the feasibility of deep extragalactic surveys exploiting the favorable k - corrections in the sub - millimeter .
deep @xmath11 m imaging has now resolved around half of the @xmath11 m extragalactic background ( hughes et al .
1998 , blain et al .
1999 , cowie , barger & kneib 2002 ) .
these galaxies are often called scuba galaxies after the instrument with which they were first detected ( holland et al .
the @xmath20 @xmath11 m scuba beam makes identifications at other wavelengths difficult ; nevertheless , @xmath21 of @xmath11 m sources are identifiable in @xmath22jy @xmath23ghz imaging ( ivison et al . 2002 ) .
these radio identifications have led to optical identifications , morphologies and ultimately spectroscopic redshifts in multiwavelength follow - up campaigns ( e.g. chapman et al .
furthermore , the non - detection of scuba galaxies in hard x - ray imaging ( e.g. alexander et al . 2003 ) suggests that the bulk of the population has far - infrared luminosities dominated by star formation .
the morphologies , redshifts , clustering and molecular gas contents are so far consistent with at least some of the scuba population being the progenitors of giant elliptical galaxies ( e.g. dunlop 2002 ) , though other alternatives are still viable ( efstathiou & rowan - robinson 2003 ) and the scuba population is heterogeneous ( e.g. ivison et al .
1998 , 2000 ) .
finally , the k - correction effects in the sub - millimeter make the sub - millimeter extragalactic background sensitive to contributions from the far - infrared luminous energy densities at all redshifts @xmath24 .
the populations which contribute to the @xmath11 m extragalactic background are necessarily also significant contributors to the cosmic history of dust - shrouded star formation . following the _ iras _ mid - infrared surveys of the local universe ( e.g. rush , malkan & spinoglio 1993 ) ,
the mid - infrared was first made accessible to deep extragalactic surveys by the infrared space observatory ( _ iso _ , kessler et al .
1996 ) which conducted a suite of surveys with a variety of depths and areal coverages ( e.g. genzel & cesarsky 2000 and refs .
therein , rowan - robinson et al . 2004 ) .
the rapid upturn in the @xmath25 m extragalactic source counts clearly demonstrated the existence of a strongly evolving population of obscured starbursts and active galaxies ( e.g. serjeant et al .
2000 , franceschini et al .
2001 , gruppioni et al .
it has also been argued that the populations dominating the @xmath25 m extragalactic background light , which are resolved by _ iso _ , are also largely the same populations which dominate the unresolved @xmath26 m background ( elbaz et al .
if correct , this is a significant breakthrough in determining the populations which dominated the far - infrared luminous energy density throughout the relatively recent history of the universe ( e.g. @xmath27 ) . finding the population that supplies the luminous energy density at @xmath28 requires understanding the sub - millimeter background light .
however , it has been difficult to find sub - millimeter source counterparts in the mid - infrared .
very few sub - millimeter - selected sources have been detected by _
iso _ in the mid - infrared ( e.g. eales et al .
2000 , webb et al .
2003b , sato et al .
the reverse procedure of looking for mid - infrared sources in the sub - millimeter via stacking analyses have not fared much better .
serjeant et al . ( 2003a ) found no excess @xmath11 m flux at the locations of @xmath25 m sources in the hdf north .
lyman break galaxies , in contrast , are detectable statistically ( e.g. peacock et al . 2000 , webb et al .
if scuba galaxies are extreme star - forming galaxies in the most massive high - redshift halos , then their anomalously faint k - band identifications imply heavy obscuration in the observed - frame near - infrared ( serjeant et al .
2003b ) , suggesting that scuba galaxies may be detectable in @xmath22jy - level mid - infrared imaging . the _ spitzer space telescope _ ( hereafter _ spitzer _ , werner et al .
2004 ) is an enormous advance over _ iso _ in terms of mapping speed , sensitivity , and wavelength coverage .
it may now be possible to resolve the bulk of the extragalactic background light at @xmath3 m in exposures of the order @xmath29 minutes , an equivalent depth achieved at @xmath30 m in @xmath31 hours with _ iso _ by sato et al .
2003 . in this paper
we present statistical sub - millimeter detections of galaxies selected in a new _ spitzer _ survey .
the _ spitzer _
identifications of previously - published sub - millimeter sources are discussed by egami et al .
identifications of mambo sources are discussed by ivison et al .
the _ spitzer _ early release observations survey is one of the first extragalactic surveys conducted by _
spitzer_. besides the pragmatic goal of characterising the survey capabilities of the facility , the survey has the science goals of making the first constraints on the populations which dominate the extragalactic backgrounds in the shorter wavelength _ spitzer _
bands , and the links between these galaxies and other known populations . accordingly , the survey field was selected to lie in the lockman hole , an area with abundant multi - wavelength survey coverage , and in particular with @xmath32 galaxies from the @xmath11 m 8mjy survey @xmath33 catalog ( scott et al . 2002 , fox et al . 2002 , ivison et al .
2002 ; see below ) .
the _ spitzer _ imaging is described by egami et al .
( 2004 ) and huang et al .
in summary , irac ( fazio et al . 2004 ) imaged a @xmath34 field for @xmath35s per sky pixel at all four bands , resulting in median @xmath36 depths of @xmath37jy , @xmath38jy , @xmath39jy and @xmath40jy at @xmath14 m , @xmath15 m ,
@xmath6 m and @xmath7 m respectively , and sources were extracted to @xmath41 .
mips ( rieke et al . 2004 ) observed the field in photometry mode at @xmath5 m for @xmath42s per sky pixel , resulting in a typical @xmath36 depth of @xmath43jy . source confusion and blending
make the mips completeness and reliability more problematic than for scuba or irac .
mips source catalogs were extracted to @xmath44jy ( @xmath45 complete , @xmath46 reliable ) and to @xmath47jy ( @xmath48 complete , @xmath48 reliable ) .
we conservatively assume the catalogs are all extragalactic , as expected at these flux densities ; the deletion of any ( hypothetical ) contaminant population of galactic stars with negligible sub - millimeter fluxes would improve the confidence levels of our statistical detections . the scuba 8mjy survey data ( which covers all our _ spitzer _ field ) is described in detail by scott et al . ( 2002 ) .
we have reanalysed this data , modelling the intra - night scuba gain variations , improving the extinction corrections using improved fits to the @xmath49ghz skydips monitored at the caltech sub - millimeter observatory , and removing the cross - talking @xmath12 m bolometers a7 and a16 ( see mortier et al .
additional sources are detectable in the zero - sum chopped / nodded sub - millimeter maps . after point source detection using noise - weighted psf convolution
( serjeant et al .
2003a ) including the effect of negative chop / nod positions , the median @xmath36 depth ( total flux ) in the _ spitzer _ early release observations field is @xmath50mjy at @xmath11 m , and @xmath51mjy at @xmath12 m .
our reanalysis of the scuba 8mjy survey uncovered new candidate sub - millimeter sources ( mortier et al .
figure [ fig : ids ] shows the _ spitzer _ identifications of the most significant of these , at @xmath52 @xmath53 @xmath54 , @xmath55 @xmath56 @xmath57 ( j2000 ) , detected with a @xmath58 @xmath11 m flux of @xmath59mjy . at @xmath12
m the source is not detected , though the @xmath12 m flux at its position is @xmath60mjy ( @xmath61 ) ; there is also a hint of positive flux at @xmath62 significance at @xmath63 m ( greve et al .
2004 ) at the position of this source .
there are hints that the source is extended at @xmath11 m , suggesting a blend of more than one source , and indeed there are two candidate identifications in the _ spitzer _ imaging .
the probabilities of a random association ( @xmath64 where @xmath65 is the surface density of catalogued objects brighter than the identification at distance @xmath66 ) are @xmath67 and @xmath68 for the brighter and fainter @xmath15 m flux respectively .
interestingly , the low - significance contours of @xmath12 m emission are coincident with the brighter of the two candidates .
the _ spitzer _
fluxes of this source are @xmath69 , @xmath70 , @xmath71 , @xmath72jy and @xmath73jy at @xmath14 m , @xmath15 m , @xmath6 m , @xmath7 m and @xmath5 m respectively .
the fluxes of the second candidate identification are @xmath74jy and @xmath75jy at @xmath14 m , and @xmath15 m , but the source is not detected at longer wavelength _
spitzer _ bands . while deblending the fluxes from the two identifications is beyond the scope of this paper , it is interesting to note that either identification would have a sub - millimeter:_spitzer _ flux ratio which is redder than in the population as a whole , derived below .
the @xmath33 sub - millimeter sources are already known to have _
identifications ( egami et al . 2004 ) , so we construct a mask to remove the sources at these positions from the stacking analysis . at @xmath11 m we masked a @xmath76 radius ( @xmath77 the scuba fwhm , i.e. the fwhm of point sources in our psf - convolved maps ) around the @xmath33 @xmath11 m point sources from scott et al .
( 2002 ) , as well as the new point sources from our reanalysis ( mortier et al .
2004 ) . at @xmath12
m the masking radius was @xmath78 , which we also applied to new @xmath12 m sources from our reanalysis .
we also masked regions with high noise levels , arbitrarily selected as @xmath79mjy at @xmath11 m , or @xmath80mjy at @xmath12 m , and also masked all regions in the sub - millimeter maps without _ spitzer _ data .
the unmasked area at @xmath11 m ( @xmath12 m ) is @xmath81 ( @xmath82 ) arcmin@xmath83 for the irac catalogues . for mips ,
the corresponding area is @xmath84 ( @xmath85 ) arcmin@xmath83 .
the sub - millimeter maps give the best - fit point source flux at each location ( equation a4 of serjeant et al .
2003a ) , so we measure the values of the sub - millimeter images at the _ spitzer _
galaxy positions . figure [ fig : histograms ] shows the sub - millimeter signal - to - noise ratios ( equation a6 of serjeant et al . 2003a ) at the positions of the _ spitzer _ early release observations source catalogs at @xmath6 m and @xmath7 m , which lie in the unmasked regions .
these figures also show histograms for the whole of the unmasked maps .
note the clear positive skews in the sub - millimeter fluxes at the _
spitzer _ source positions , relative to the maps as a whole .
we used kolmogorov - smirnov tests to determine the confidence level of these relative positive skews , listed in table [ tab : coadd_results ] .
there are significant detections at @xmath6 m and @xmath7 m , and marginal detections at @xmath5 m .
no significant skew was found at @xmath14 m or @xmath15 m . we also calculated the mean fluxes in the sub - millimeter maps at the positions of the _ spitzer _ galaxies .
these are almost all positive and are listed in table [ tab : coadd_results ] .
this statistic is less efficient than the kolmogorov - smirnov test ( which does not just use the first moment ) , so the uncertainties on the mean fluxes are larger than suggested by the kolmogorov - smirnov confidence levels .
the _ spitzer _
galaxies are individually undetected in the sub - millimeter , so one must be careful in interpreting the stacking analysis ; for example , we do not use the _ noise - weighted _ mean fluxes because they can give false positives , as noted by serjeant et al .
also , many fluctuations in the sub - millimeter maps are blends ( e.g. scott et al .
2002 ) , so is there a risk of overestimating the flux of any given _
galaxy by also counting its neighbors ?
the answer is no , as peacock et al .
( 2000 ) showed : for a poissonian sampling of the sub - millimeter map , the expected total sub - millimeter flux from all neighbors equals the mean flux of the map , which is exactly zero for the chopped , nodded scuba maps .
( the chopped / nodded psf is also zero - sum , so the psf - convolved images are still zero - sum . ) blending slightly degrades the uncertainty on the mean , but it does not affect the mean fluxes .
another demonstration of this is to note that the total flux of any given source is exactly zero in these chopped , nodded maps .
even confusing sources within the same scuba beam will have no _ net _ effect , provided that there are also similar sources uniformly distributed over the rest of the map .
the histograms of the whole map in figure [ fig : histograms ] effectively act as the control sample .
nevertheless , as a test of the stacking analysis methodology , we performed the same tests on simulated _ spitzer _ catalogs of the same size as the observed catalogs , with a poissonian distribution in the unmasked regions .
no positive skews in the sub - millimeter flux distributions were detected relative to the maps as a whole .
these simulations also verified that the mean sub - millimeter flux around randomly chosen positions is consistent with zero .
although our _ spitzer _ @xmath6
m and @xmath7 m catalogues are small ( @xmath8 and @xmath9 respectively ) , we can use our statistical detections to obtain constraints on the contribution to the sub - millimeter extragalactic background light from the _ spitzer _ populations , by multiplying the mean sub - millimeter flux of a single _ spitzer _ galaxy ( table [ tab : coadd_results ] ) with the observed surface density of _ spitzer _ galaxies . at @xmath11 m
, the integrated contribution from the @xmath6 m population is @xmath16 nw m@xmath17 sr@xmath18 , and from the ( overlapping ) @xmath7 m population @xmath86 nw m@xmath17 sr@xmath18 .
this is already comparable to the total @xmath11 m extragalactic background reported by lagache et al .
( 2000 ) of @xmath87 nw m@xmath17 sr@xmath18 . at @xmath12
m the _ spitzer _ contributions can , within the uncertainties , account for all the observed @xmath88 nw m@xmath17 sr@xmath18 extragalactic background : @xmath19 nw m@xmath17 sr@xmath18 from the @xmath6m - selected sample , and @xmath89 nw m@xmath17 sr@xmath18 from the @xmath7 m galaxies .
our sub - millimeter faint source density is consistent with those in the deepest @xmath11 m maps ( cowie et al . 2002 ) . the sub - millimeter:_spitzer _ flux ratios in sub - millimeter point sources ( egami et al .
2004 , see above ) are much higher than average in the population ( table [ tab : coadd_results ] ) . despite the commonality between the scuba and _ spitzer _
galaxy populations , sub - millimeter galaxies are not at all representative of the _ spitzer _ population .
the sub - millimeter background independently suggests the same result : if the @xmath5 m population had sub - millimeter:_spitzer _ flux ratios as high as scuba galaxies ( e.g. median @xmath90 for the egami et al .
( 2004 ) sample ) , the @xmath44jy @xmath5 m sources alone would over - predict the @xmath11 m background by @xmath91 .
our weak sub - millimeter detection of the @xmath5 m population is less surprising in that context .
one possible interpretation is that the scuba population has fewer ( or more heavily obscured ) active nuclei than the _ spitzer _ population as a whole .
is our stacking analysis signal due to dusty galaxies , or to high-@xmath92 galaxies , or both ?
our samples are too small to distinguish these possibilities , but stacking analyses of subsets show that either option is credible .
almost all the @xmath6 m galaxies are detected at @xmath14 m , and as with the @xmath6 m population as whole ( table [ tab : coadd_results ] ) this subset ( @xmath93 galaxies ) is detected statistically .
the @xmath94 remaining @xmath14 m galaxies are not detected in the stacking analyses ( e.g. , only @xmath95 confidence in the @xmath12 m signal : noise maps ) , so the redder galaxies are more likely to be sub - millimeter emitters .
the lack of a stacking signal at @xmath14 m and @xmath15 m is at least in part due to dilution of the signal from the high density of bluer galaxies detected at these wavelengths .
the red colors could be due to heavy obscuration , but the colors also become redder quickly with increasing redshift .
simpson & eisenhardt ( 1999 ) show that @xmath96 or @xmath97 selects redshifts @xmath98 .
there are @xmath99 galaxies satisfying either criteria , with detections in the relevant bands , and these galaxies are detected in the stacking analysis ( e.g. @xmath100 confidence at @xmath12 m ) ; the corresponding @xmath101 foreground @xmath102 galaxies are not detected ( e.g. only @xmath103 confidence at @xmath12 m )
. further work on larger samples is needed to discriminate between high - redshift and high - obscuration sub - populations .
the sub - millimeter backgrounds are sensitive to contributions from the far - infrared luminous energy densities throughout the redshift range of favorable k - corrections .
it follows that the _ spitzer _ @xmath6 m and @xmath7m - selected galaxies are necessarily significant contributors to the comoving volume averaged star formation density .
many of the @xmath6 m and @xmath7 m galaxies are challenging targets for 8 - 10m - class optical / near - infrared spectroscopy , but redshift surveys of these populations , together with 2-d and 3-d clustering , would discriminate between competing semi - analytic descriptions of galaxy evolution , and determine whether ( or which ) scuba/_spitzer _ galaxies are the sites of the assembly of giant ellipticals .
finally , it is worth stressing that the conclusions in this paper could only have been reached with co - ordinated multi - wavelength surveys .
llllllllllll wave- & @xmath12
m & @xmath12 m & @xmath12 m & @xmath104 & @xmath105 & @xmath11 m & @xmath11 m & @xmath11 m & @xmath106 & @xmath107 + length & @xmath108 & k - s ( s : n ) & k - s ( flux ) & mjy & @xmath109 & @xmath108 & k - s ( s : n ) & k - s ( flux ) & mjy & @xmath109 + @xmath14 m & @xmath110 & @xmath111 & @xmath112 & @xmath113 & @xmath114 & @xmath115 & @xmath116 & @xmath117 & @xmath118 & @xmath119 + @xmath15 m & @xmath120 & @xmath121 & @xmath122 & @xmath123 & @xmath124 & @xmath125 & @xmath126 & @xmath127 & @xmath128 & @xmath129 + @xmath6 m & @xmath8 & @xmath130 & @xmath131 & @xmath132 & @xmath133 & @xmath134 & @xmath135 & @xmath136 & @xmath137 & @xmath138 + @xmath7 m & @xmath9 & @xmath131 & @xmath139 & @xmath140 & @xmath141 & @xmath141 & @xmath142 & @xmath143 & @xmath144 & @xmath145 + @xmath5m@xmath146 & @xmath147 & @xmath148 & @xmath149 & @xmath150 & @xmath151 & @xmath152 & @xmath153 & @xmath154 & @xmath155 & @xmath156 + @xmath5m@xmath157 & @xmath158 & @xmath159 & @xmath160 & @xmath161 & @xmath162 & @xmath114 & @xmath163 & @xmath164 & @xmath165 & @xmath166 + | we present sub - millimeter statistical detections of galaxies discovered in the @xmath0 spitzer early release observations ( to @xmath1jy @xmath2 at @xmath3 m , @xmath4jy at @xmath5 m ) through a stacking analysis of our reanalysed scuba 8mjy survey maps , and a spitzer identification of a new sub - millimeter point source in the 8mjy survey region . for sources detected at @xmath6 m or @xmath7 m ( @xmath8 and @xmath9 sources respectively ) ,
we detect positive skews in the sub - millimeter flux distributions at @xmath10 confidence using kolmogorov - smirnov tests , at both @xmath11 m and @xmath12 m . we also marginally detect the _ spitzer _ @xmath5 m galaxies at @xmath11 m at @xmath13 confidence , and place limits on the mean sub - millimeter fluxes of the @xmath14 m and @xmath15 m sources . integrating the sub - millimeter fluxes of the _ spitzer _ populations , we find the @xmath6 m galaxies contribute @xmath16 nw m@xmath17 sr@xmath18 to the @xmath11 m background , and @xmath19 nw m@xmath17 sr@xmath18 to the @xmath12 m background ; similar contributions are made by the @xmath7m - selected sample .
we infer that the populations dominating the @xmath6 m and @xmath7 m extragalactic background light also contribute around a quarter of the @xmath11 m background and the majority of the @xmath12 m background . |
in the standard cosmological framework , the universe is decribed by a friedmann - lemaitre solution , the spatial sections of which being usually assumed to be simply connected .
einstein equations being local , they allow us to determine the local geometry but they give no complete information about the global structure of the universe , i.e. about its topology , even if the geometry constrains to some extent the topology .
this was pointed out by friedmann himself just after he proposed its cosmological solution @xcite .
the indeterminacy about the global topology of the cosmological solutions was raised out as soon as einstein proposed the first cosmological solution of his equations ; the einstein static universe assumed spatial sections with the topology of the hypersphere @xmath1 @xcite , although de sitter stressed that the same geometry could admit as well the projective space @xmath2 as spatial sections @xcite .
these two solutions were locally identical ( same metric ) but differed by their topology , i.e. by the choice of boundary conditions .
many arguments were then advanced in favour of a simply connected universe , such as the simplicity or economy principle stating that one should introduce as few parameters as possible in physical modelling .
indeed , such an argument is very unclear .
there has been some tendency to favor spaces with finite volume .
the eternal and infinite space of newtonian physics was for instance leading to logical difficulties such as the olbers paradox @xcite , and the finite cosmological solution proposed by einstein in 1917 was received as a smart way of solving such paradoxes .
the mach s principle @xcite based on the idea that the local inertia is determined by the distribution of masses in the whole universe , also tends to favor universes with a finite volume .
if space is simply connected then it is finite if and only if it is locally elliptic , whereas if multi
connected it can also be locally hyperbolic or euclidean .
it has also been argued that infinite spaces were unaesthetic since `` all phenomenon with a non vanishing probability must happen somewhere else '' @xcite .
this argument was used by ellis to conclude that if space was infinite then one can avoid the former conclusion by dropping the assumption of spatial homogeneity and by arguing that we were living in a particular place of the universe @xcite .
to finish , many arguments coming from some ideas in quantum cosmology are in favor of a finite volume space @xcite . indeed
, none of these arguments can help us to determine the shape and size of our universe .
so , we are lead to the question , can we detect or constrain observationnally the topology of our universe ?
astonishingly enough , our century has on one hand seen the birth of the geometrical description of the universe and of a non static curved spacetime . on the other
, mathematicians have developped the classification of 3d manifolds .
the first motivation came from crystallography , which lead to the classification of euclidean 3d
manifolds @xcite and was achieved in 1934 @xcite .
the classification of locally elliptic 3-manifold was set by f. klein @xcite and w. killing @xcite and solved by j.a .
wolf in 1960 @xcite .
the classification of locally hyperbolic manifold was started in the 70 s by w. thurston @xcite and is not yet achieved @xcite . thus both elements necessary to answer the question of cosmic topology were developped at the same time without interacting so much . for more details concerning this interesting modern quest
, one can see @xcite .
in relativistic cosmology , our universe is described by a globally hyperbolic 4manifold @xmath3 @xcite , where the spatial sections @xmath4 are homogeneous and isotropic riemannian 3manifolds . from a topological point of view , it is convenient to decribe such a manifold by its fundamental polyhedron ( hereafter fp ) , which is convex .
its faces are associated by pairs through the elements of a holonomy group @xmath5 which is acting freely and discontinuously on @xmath4 ( see @xcite for mathematical definitions and @xcite for an introduction to topology in the cosmological context ) .
the holonomy group is isomorphic to the fundamental group @xmath6 . using the property ( see e.g. @xcite ) : @xmath7 we deduce that the study of the topology of the universe reduces to the study of the topology of its spatial sections .
what are the allowed homogeneous 3manifolds usable in cosmology ? according to the sign of the spatial curvature @xmath8 , the universal covering space of the spatial sections ( which will be referred to as @xmath9 ) can be described by the euclidean space @xmath10 , the hypersphere @xmath11 or the 3-hyperboloid @xmath12 if @xmath8 vanishes , is positive or negative respectively .
thus , the homogeneous and isotropic 3-manifolds will be of the form @xmath13 let us just summarize some of the properties of each family and also introduce @xmath14 , the full isometry group of @xmath9 keeping the metric invariant .
indeed @xmath15 will be a discrete sub - group of @xmath14 . 1 . : the covering space in the euclidean space @xmath16 and the isometry group is @xmath17 .
the metric can be written under the form @xmath18 where @xmath19 is the conformal time , @xmath20 the scale factor , @xmath21 the radial coordinate and @xmath22 the unit solid angle .
the generator of @xmath15 are the identity , the translations , the reflexions and the helicoidal motions .
+ these transformations generate 18 different spaces @xcite among which 17 are multi connected and correspond to the 17 crystallographic groups @xcite .
10 spaces are compact and among them 6 are orientable .
the description of these spaces can be found in e.g. @xcite .
2 . : the covering space in the elliptic space @xmath23 and the isometry group is @xmath24 .
the metric can be written as @xmath25 with the curvature radius as unit length , the volume of the hypersphere is @xmath26 wolf @xcite gives an exhaustive description of the allowed discrete groups @xmath15 .
they are the cyclic groups of order @xmath27 ( @xmath28 ) , the dihedral groups of order @xmath29 , and the symmetry groups of the tetrahedron , octaedron and icosahedron .
+ if we denote @xmath30 the order of the holonomy group , it is straigthforward to show that @xmath31 which tells us that the volume of an elliptic manifold is a topological invariant ( degenerate since two different groups with same order will have same volume ) .
3 . : the covering space in the hyperbolic space @xmath32 and the isometry group is @xmath33 .
the metric can be written as @xmath34 + the classification of compact hyperbolic manifolds is not achieved yet and we will just describe two of them and give some useful properties .
this classification relies on the rigidity theorem @xcite stating that the geometry is fixed by the topology , a consequence of which being that the volume and other characteristic lengths are topological invariants , so that @xmath35 it has also been shown that there is a minimal allowed volume @xcite @xmath36 we also define the outside radius @xmath37 , the radius of the smallest geodesic ball that contains the fp , the inside radius @xmath38 , the radius of the biggest geodesic ball contained in the fp , and the injectivity radius @xmath39 , half the length of the shortest closed geodesic .
@xmath37 , @xmath38 and @xmath39 are altogether topological invariants .
+ the classification and the description of the known compact hyperbolic manifolds can be obtained by using the software _ snappea _ @xcite which provides the fp , the generators of the holomy group and the characteristic lengths .
examples of such manifolds can be found in e.g. @xcite .
here we mention for a later use a manifold of special interest , the weeks space @xcite , which is the smallest known compact hypermabolic manifold .
its fp has 18 faces and its geometrical characteristics are @xmath40 at the time being , two classes of methods to detect and/or constrain the topology of the spatial sections of our universe have been proposed and investigated .
they use two classes of cosmological observations , namely the 2d cosmic microwave background data and the 3d catalogs of discrete sources . 1 . :
the cosmic microwave background is composed of all the photons emitted during the decoupling period between matter and radiation at a redshift of @xmath41 .
it is observed as a black body with a temperature of @xmath42 k with small fluctuations of order @xmath43 .
the temperature anisotropies reflect the small inhomogeneities that ultimately lead to the observed structures in the universe .
+ when studying these anisotropies in a multi connected universe , one has to take into account the fact that we are in a compact manifold and thus that the wavelengths must be discretized .
@xcite showed that in a cubic hypertorus one should see a cut off in the two point correlation function if @xmath44 .
this analysis was then generalised to non cubic hypertorus @xcite and to the six flat manifolds @xcite
. however , in compact hyperbolic manifolds there are super - curvature modes @xcite , which implies that no such cut off exists @xcite .
+ it has then been realised that the patterns on such an extended surface must be correlated if the universe is multi - connected .
@xcite studied these correlations in the flat manifolds and cornish _ et al . _
@xcite developped a topology independent method to look for a topological signature in the cosmic microwave background map as expected from the map and planck satellites .
+ all these studies have been performed assuming that the small initial inhomogeneities where generated during an inflationary phase .
there is however another process that could generate such inhomogeneities , namely topological defects produced during a symmetry breaking phase transition in the early universe .
uzan and peter @xcite showed that the topology implies a constraint on the topological defects network from which it was deduced @xcite that there will be a cut off in the angular power spectrum of the temperature anisotropies , the cut off value depending only on the characteristic size of the universe and on the cosmological parameters . note that such considerations also provide another method for constraining the topology , namely the observation of an extended defect at a redshift @xmath45
will give a lower bound on the size of the universe .
+ in conclusion , at the moment 2d methods constrain flat manifolds to @xmath44 , and there is no convincing constraint on compact hyperbolic manifolds @xcite .
some powerful methods can be hoped to be used when high resolution cosmic microwave background maps are achieved .
2 . : in the framework of pre relativistic cosmology , it was already pointed out by schwarzschild that if space is multi connected , one could see multiple images of astronomical objects @xcite , whereas friedmann @xcite did the same in the framework of general relativity .
all 3d methods for testing the space topology are based on this fact .
we shall develop their description in the following sections . in these proceedings , we describe the developments of the 3d methods , from the schwarzschild initial idea of the existence of multiple images of the same object , until its statistical implementation under the form of cosmic cristallography .
we then carefully study the method and explain why , contrary to what was previously thought , it wo nt work in locally hyperbolic universes .
we then present attempts to generalise it and finish by describing a new and promising ccp " method .
this will be examplified by numerical simulations with depiction of color pictures in order to make the subject as clear as possible .
as stressed above , if space is multi connected , one should see multiple topological " images of the same object .
this is illustrated on figure [ fig4 ] where we have used a 2d example . in the universal covering space
, the spacetime trajectories of the different images ( @xmath46 ) of a given object will intersect the observer past light cone at different times ( @xmath47 ) .
the same object will then been seen at different stages of its evolution . indeed in a 4d universe , these images will also be seen in different directions of the sky .
the question is indeed to find methods to implement such a property in order to detect or to put lower bounds on the size of the universe .
some specific objects have been tried to be recognised individually .
an extensive description of these attempts can be found in @xcite and we just stick to a brief description . * milky way : can we recognise our own galaxy ?
the maximum distance up to which we would be able to do that has been estimated as @xmath48 by sokolov and shvartsmann @xcite which gave the observational bound @xmath49 .
then , fagundes and wichoski @xcite proposed that our galaxy was a quasar long ago but indeed , for physical reasons due to realistic quasar models , our galaxy can not have been a quasar in its past .
fagundes @xcite also studied the occurrence of images of the milky way in the best hyperbolic space model @xcite .
presently , no source has been identified as an image of our own galaxy . *
galaxy clusters : gott used the coma cluster @xcite has a candidate for the search of ghosts images and obtained the bound @xmath50 .
he also performed simulations in @xmath51 to provide a pattern of clustering and a correlation function in agreement with observations of nearby galaxies .
roukema and edge @xcite proposed 3 candidates as ghost images for the coma cluster .
further investigations are needed .
+ sokoloff and shvartsmann @xcite considered the abell catalog of clusters and gave the bound @xmath52 in euclidean space , whereas demianski and lapucha @xcite searched unsuccessfully for opposite pairs in a catalog of 1889 clusters .
developped a statistical method that is presented below to obtain @xmath53 in euclidean space and fetisova _ et al .
_ reported a spike about @xmath54 in the correlation function of rich clusters , but such an evidence does not necessarily supports multi connectedness ( see e.g. @xcite ) . *
quasars : quasars occupy a large volume of the universe ( @xmath55 ) . however , they are probably short - lived compared to the expected time necessary for a photon to go around the universe . if @xmath56 , then one can hope to test topology on scale @xmath57 .
so , they are of interest even if @xmath58 .
pal @xcite remarked that although individual quasars may have short lifetime , they may be part of larger associations which survive longer .
+ roukema @xcite tried to identify quintuplets and quadruplets of quasars in a catalog of 4554 quasars .
he found 27 identifications whereas he found @xmath59 identifications in simulations of a simply connected universe .
fagundes @xcite tried to discuss the redshift contreversy in the best @xcite model .
until now , the best bound obtained by 3d methods was @xmath60 and hold only in euclidean universe .
there is still some observational room for the topologies of an hyperbolic universe , even on scales significantly shorter than the horizon radius .
all the methods trying to recognize a particular object suffer from the same major problems : 1 .
distance determination : catalogs give the angular position and the redshift .
one has then to go from redshift space to real space , which depends on the knowledge of the cosmological constant and of the density parameter .
this explained why most of the former studies were performed assuming @xmath61 .
this point will be develop later .
peculiar velocities : objects are not comoving , so that the position of their topological images are shifted . if their velocity is of order @xmath62 , then , assuming that the time for a photon to go around the universe can be estimated by @xmath63 , one obtains the maximal precision on the determination of position @xmath64 .
morphology : when one tries to associate two images to the same physical object , one has to be aware that , in general , the object will not be seen in the same angle and at the same stage of its evolution .
it will then be very difficult to recognise it and the identification will strongly depend on galactic evolution models .
conversely , the discovery of topology by any means could provide informations such as peculiar velocities with a better accuracy @xcite
. we will also be able to see the same object at different stages of its evolution and on different angles , which will constrain galactic evolution models .
since there is little chance to recognise different images of a given object , one can try to detect these images statistically , since for instance the number of topological images of a single object in a toroidal universe has been estimated in table [ tab ] of @xcite . *
4c + + + @xmath65 & @xmath66 ) & @xmath67 & @xmath68 + @xmath69 & 7000 & 1200 & 180 + 1500 & 279 & 45 & 7 + 2500 & 60 & 10 & 1.5 + the crystallographic method @xcite is based on a property of multi connected universes according to which each topological image of a given object is linked to each other one by the holonomies of space .
indeed , we do not know these holomies as far as we have not determined the topology , but we know that they are isometries . for instance in locally euclidean universes , to each holonomy is associated a distance @xmath70 , equal to the length of the translation by which the fundamental domain is moved to produce the tessellation in the covering space . assuming the fp contains @xmath71 objects ( e.g. galaxy clusters ) , if we calculate the mutual 3d distances between every pair of topological images ( inside the particle horizon )
, the distances @xmath70 will occur @xmath71 times for each copy of the fundamental domain , and all other distances will be spread in a smooth way between zero and two times the horizon distance . in a histogram
plotting the number of pairs versus their 3d separations , the distances @xmath70 will thus produce spikes .
simulations indeed showed that the pairs between two topological images of the same object drastically emerge from ordinary pairs @xcite in the histogram .
the applicability of the method in euclidean spaces has also been discussed by fagundes and gaussmann @xcite when the size of the physical space is comparable to the horizon size
. two kind of catalogs of astronomical objects can be thought of to apply this method : the galaxy cluster catalogs , which typically have a redshift depth @xmath72 , and the quasars catalogs , which typically extend to @xmath73 . concerning quasars , even if their lifetime is probably too short to be good candidates for producing topological images , they are usually part of systems that have a much larger lifetime @xcite .
let us stress that a recent survey @xcite in the hubble deep field south nicmos field found 17 galaxies with a redshift between 5 and 10 and 5 galaxies with a redshift above 10 , among a total of 323 galaxies .
this can let us hope to apply this method to deeper catalogs in the future .
the angular resolution needed is given by the fact that the objects have a peculiar velocity and that they will not be seen at exactly the same position @xcite .
note that the crystallographic method , contrary to the `` direct '' method which would try to recognize topological images of individual objects , is not plagued by the fact that topological images of the same object are seen at different stages of its evolution .
we present on figure [ fig5 ] an example from @xcite for the flat torus with a cubic fp .
now , we apply this method to a catalog @xcite containing more than 11,000 quasars up to a redshift of @xmath74 ( see figure [ catq ] where we have depicted its projection on the celestial sphere and the redshift distribution of its objects ) .
the pair separation histogram is plotted on figure [ psh ] and it exhibits no topological signature . since respectively 60 % , 80 % and 95 % of the sources are below the redshifts 2 , 2.3 and 3 , we only test scales smaller that @xmath75 . in a locally euclidean universe , the distance - redshift relation is given by @xcite @xmath76 we simulated a catalog with the same number of objects and the same redshift distribution to reproduce the real catalog depicted in figure [ catq ] . varying the size of cubic hypertorus
, a spike appears when @xmath77 for the simulated catalog and thus we obtain the new bound on the size of a locally euclidean manifold as @xmath78 which is the best constraint presently available with 3d
methods for locally euclidean manifolds .
note that it corresponds to @xmath79 which is comparable to the bound obtained with 2d methods .
note that this result has not been published elsewhere .
when one looks carefully at the crystallographic method , one must wonder about the origin of the spikes . as it has recently been explained @xcite , two kinds of pairs can create a spike , namely ( we keep the notation and vocabulary introduced in @xcite ) 1 .
_ type i pairs _ of the form @xmath80 , since @xmath81=\hbox{dist}[x , y]$ ] for all points and all elements @xmath82 of @xmath15 .
type ii pairs _ of the form @xmath83 if @xmath84=\hbox{dist}[y , g(y)]$ ] for at least some points and elements @xmath82 of @xmath15 . both families of pairs
are depicted on figure [ fig6 ] for the example of the 2torus .
a clifford translation is an isometry @xmath82 such that the displacement function @xmath85 is constant @xcite .
this is precisely what is required to get type ii pairs in the histogram . in compact hyperbolic manifolds , @xmath84
$ ] always depends on the position @xmath86 of the source ( see @xcite for details ) . in elliptic spaces ,
distances are position - independent whenever the holonomy is a clifford translation .
all finite groups of clifford translations of spheres are the cyclic group , the binary dihedral , tetrahedral , octahedral and icosahedral groups @xcite , from which we deduce @xcite that the covering transformations which move a source to its nearest neighbours are clifford translations ( although the transformations to more distant neighbours might not be ) , and type ii pairs can be produced . in locally euclidean universes ,
type i and type ii pairs are both present .
the reason is that the 3-torus has the very special property that the separation distance of gg - pairs ( i.e. any pair of images comprising an original and one of its ghosts , or two ghosts of the same object ) is independent of the location of the source . in other euclidean spaces
the spectrum of gg - pair distances varies with the location of the source .
however all closed euclidean 3-manifolds have the 3-torus as a covering space , so for each such manifold there will be some distances which are independent of the location of the source . as a consequence , the topological signal expected in the histogram from type i and type ii pairs clearly stands out , as was shown in the simulations of @xcite .
+ in conclusion , type ii pairs exist only in elliptic and euclidean universes .
the question we still have to answer to judge the efficiency of the crystallographic method in the general case is : what about the amplitude of the spikes related to type i pairs only
? if the injectivity radius of the space manifold is smaller than the calatog s length scale , type i pairs will always be present , whatever the curvature .
their number roughly equals the number of copies of the fundamental domain within the catalog s limits . following @xcite ,
this number can easily been estimated by computing the ratio between the volume of the geodesic sphere of radius @xmath87 and the volume of the manifold .
the calculation ( in the case of hyperbolic universes ) leads to @xmath88 where @xmath89 is the radial distance corresponding to the redshift @xmath90 , given @xmath91 where @xmath92 and @xmath93 are respectively the density parameter and the cosmological constant .
we have plotted on figure [ fig8 ] an estimation of the number of copies of an object taken respectively in a catalog of galaxy clusters and in a catalog of quasars , as a function of the cosmological parameters .
the small number of type i pairs in hyperbolic manifolds is due to the property that the volume of the manifold is fixed once the topology is determined ( the _ rigidity theorem _ ) , contrarily to euclidean spaces where the characteristic sizes and the volume of the fp can be chosen at will ( since @xmath94 , the geometry does notimpose any characteristic size ) . in conclusion ,
type i pairs will indeed contribute to spikes in the pair separation histogram but the amplitude of these spikes will be of the same order than the statistical noise .
the former conclusion can be tested numerically .
for that purpose , we concentrate on locally hyperbolic three dimensional manifolds . it can be first embedded in a four dimensional minkowski space by introducing the set of coordinates @xmath95 related to the intrinsic coordinates @xmath96 through ( see e.g. @xcite ) @xmath97 so that the three dimensional hyperboloid @xmath12 has the equation @xmath98 with these notations , the comoving spatial distance between two points of comoving coordinates @xmath86 and @xmath99 can be computed directly in the minkowski space by @xcite @xmath100=\arg\cosh{\left[x^\mu y_\mu\right]},\ ] ] where @xmath101 , @xmath102 being the minkowskian metric coefficients . to generate an idealised catalog , @xmath103 ,
we first distribute _ homogeneously _
@xmath71 objects in the fp , then we by unfold the catalog by applying the generators of the holonomy group . now , to generate a catalog with the required depth @xmath104 in redshift , the set of all points obtained as above
must be truncated in order to keep only the points @xmath86 such that @xmath105\leq \chi(\omega_{m0},\omega_{\lambda0};z_{\rm max})$ ] , given by equation ( [ chi1 ] ) .
this amounts to select objects located within the geodesic ball of radius @xmath106 centered onto an observer right at the centre .
the unfolding of weeks manifold is shown in figure [ fig9 ] , where we have depicted the central cell and its 18 nearest neighbours in klein coordinates . in figure
[ fig10 ] , we have plotted a simulated catalog showing all points , including those outside the horizon distance .
we then compute all the 3d separations between all the pairs of @xmath107 and we plot the histogram of the number of pairs with a given separation .
+ indeed the former procedure applies when the observer stands at the center of the polyhedron @xmath108 . in @xcite , we explain how to take into account an off - centered observer . + in lehoucq et al .
@xcite , we have generated some pair histograms in a universe whose spatial sections have the topology of the weeks manifold , using @xmath109 objects in the fp . in figure [ fig11 ]
we give an example wher @xmath110 , @xmath111 and where the observer stands at the center of the polyhedron ( @xmath112 ) .
the dependence on the cosmological parameters , on the position of the observer and on the catalog depth where studied in details in @xcite .
as expected , none of the plots exhibit spikes .
this is due to the combined effects of the geometry and of the cosmological parameters values . to summarize the failure of cosmic crystallography for hyperbolic manifolds : 1 .
hyperbolic manifolds are such that @xmath84 $ ] depends on @xmath86 , so that there is no amplification for the _ type ii pairs _
@xmath113 , whereas @xmath84=\hbox{dist}[y , g(y)]$ ] in the euclidean case .
this suppresses the corresponding spikes .
2 . the spikes associated to the isometries ( i.e. such that @xmath114=\hbox{dist}[x , y]$ ] ) must remain . but , given the cosmological parameters , we have shown that the number of topological images is too low to create such spikes associated to _ type i pairs_.
the next question is : can we generalise the crystallographic method in order to make it work in any case ?
two solutions were proposed almost immediately to improve the crystallographic method and to extract from 3d catalogs the signature of type i pairs : 1 .
fagundes and gaussman @xcite substracted the pair separation histogram for a simulated catalog ( with the same number of objects and the same cosmological parameters ) in a simply - connected universe to the observed pair separation histogram .
the result is `` a plot with much oscillation on small scale , modulated by a long wavelength quasi - sinusoidal pattern '' .
the statistical relevance of this signature still has to be investigated . to illustrate this point we show on figure [ fig12 ] the result of a simulation where we have substracted to realisations on a simply - connected friedmann - lemaitre universe
this has to be compared with figure 1 from @xcite and let us think that is method is not efficient for determining a relevant signature of the topology of the universe .
2 . gomero _ et al . _
@xcite proposed to split the catalog in `` smaller '' catalogs and average the pair separation histograms built from each sub catalog to reduce the statistical noise and extract a visible signal of the non - translational isometries .
the feasability of this method has not yet been demonstrated .
we now present a new method @xcite based on the property that whatever the topology , type i pairs will always exist . indeed ,
as shown ahead , this does not lead to any observable spike in the pair separation histogram .
we thus had to find a method to enhance the topological signal by `` collecting '' all the homologous pairs together .
this can be achieved by building a _ collecting correlated pairs _
method ( hereafter _ ccp _ ) .
such an approach ( described in details in @xcite and below ) is not able to determine the exact topology , but will provide a signature of the compactness of the spatial sections , which is indeed a first step toward the determination ( or the rejection ) of the cosmic topology . in this section , we recall the construction of the ccp index and the main resuts we have obtained .
as stressed in the previous section , type i pairs will always exist in multi
connected spaces as soon as one of the characteristic scales of the fundamental domain is smaller than the hubble scale . defining @xmath115 as the @xmath116 generators of @xmath15 and referring to @xmath86 as the position of the image in the universal covering space @xmath117
, we have : 1 .
@xmath118 , @xmath119 , @xmath120&=&\hbox{dist}[x , y].\label{zero1}\end{aligned}\ ] ] we will refer to these pairs as @xmath121pairs .
2 . @xmath122 , @xmath123 , @xmath124&=&\hbox{dist}[x , g_2(x)].\label{zero2}\end{aligned}\ ] ] we will refer to these pairs as @xmath125-pairs .
both the @xmath121pairs and the @xmath125-pairs are type i pairs . to collect all these pairs together and enhance the topological signal , we define the _ ccp index _ of a catalog containing @xmath71 objects as follows . 1 .
we compute all the 3d
distances @xmath126 $ ] for all points within the catalog s limit .
we order all these distances in a list @xmath127 , where @xmath128 is the number of pairs , such that @xmath129 .
we create a new list of increments defined by @xmath130,\quad \delta_i\equiv d_{i+1}-d_i\ ] ] ( keeping all the equal distances , if any , in the list ) .
we then define the ccp index @xmath131 as @xmath132
where @xmath133 , so that @xmath134 . with such a procedure , all type i pairs will contribute to @xmath135 .
for instance , if a given distance appears 4 times in the list @xmath127 , it will contribute to 3 counts in @xmath135 .
compared to the old crystallographic method , all the correlated pairs are gathered into a single spike , instead of being smoothed out into the noise of the histogram pair separation . indeed , in a more realistic situation
, one has to take into account bins of finite width @xmath136 and replace @xmath135 by @xmath137 in the computation of @xmath131 .
the effect of this `` binning '' is discussed below .
we now focus on the `` idealised '' version of the procedure by studying the amplitude of the ccp index in some multi - connected models .
+ for that purpose , let us assume that the catalog is obtained from an initial set of @xmath138 objects lying in the fundamental domain and that @xmath139 copies of the domain are within the catalog s limits ( @xmath140 if the whole observable universe up to the catalog s limit is included inside a fundamental domain ) .
the total number of images is @xmath141 .
indeed @xmath139 is usually not an integer but we assume it is , in order to estimate the amplitude of @xmath131 and compare it with the result in a simply - connected model . in @xcite , it has been shown that @xmath142 , \label{amplitude}\ ] ] where @xmath143 is a function characterising the manifold @xmath4 .
it is an increasing function of @xmath139 which vanishes for @xmath140 . indeed ,
if the holonomy group @xmath15 contains clifford translations allowing for type ii pairs , or if there are `` fake '' pairs ( i.e. such that @xmath126=\hbox{dist}[u , v]$ ] ) , @xmath144 computed from ( [ amplitude ] ) will give a lower bound for the true @xmath145 . the normalised ccp index ( [ ccp ] ) follows straightforwardly .
@xmath131 is a good index for extracting the topological signal since 1 . when @xmath146 ( i.e. when the fundamental domain is greater than the catalog s spatial scale ) , @xmath147 , 2 .
when the number of sources in the fundamental domain becomes large , it behaves as @xmath148 as shown in @xcite , one can compute analytically the ccp index in various examples and show that it is a fair indicator of the existence of at least one compact spatial dimension .
now that the ccp index can be computed for any multi - connected space , we can compare it with its value in a simply - connected friedmann - lemaitre model containing the same number objects ( i.e. @xmath149 ) .
@xmath131 is given as a function of @xmath150 , @xmath138 ( the number of sources in the fundamental domain ) and @xmath139 ( the number of copies of the domain within the observable universe ) by : @xmath151 } { ( b+1)^2a^2-(b+1)a-2}.\ ] ] indeed , when working with a real catalog , we do not know the radial distance of cosmic objects but only their redshift . as seen before , the determination of the radial distance requires the knowledge of the cosmological parameters @xmath152 and @xmath153 and can be obtained analytically only when @xmath154 .
thus , if the universe is multi - connected on sub horizon scale , the plot of @xmath131 in terms of @xmath152 and @xmath155 will exhibit a spike only when the cosmological parameters have the right value ( as shown on plot 7 of @xcite ) .
if the cosmological parameters are not exactly known , the distance determination ( [ chi1 ] ) will be wrong and the topological signature will be destroyed ( see figure 6 in @xcite ) .
two consequences follow : 1 . one should span the parameters space @xmath156 in order to detect the topological signal , plotting @xmath157 .
if there is any topological signal , the position of the spike gives the values of the cosmological parameters on the scale of the catalog s limit ( see figure [ testr ] ) .
we proceed as follows .
we first generate a catalog by choosing the number of objects in the fundamental domain ( @xmath158 ) , the topology ( weeks manifold ) and the cosmological parameters ( e.g. @xmath159 , @xmath160 ) , we then use a second code to apply the test and we draw @xmath131 in terms of the two cosmological parameters . as shown in @xcite , the method works pretty well in the sense that there is a strong spike which signals a non trivial topology and which determines the cosmological parameters .
but we also see that a slight deviation in the evaluation of the cosmological parameters would make the spike to disappear .
this effect is now discussed .
when one wants to apply the ccp method to real data , one has to face a number of problems .
first , we can not use a zero width bin , one of the reasons being that the sources are not comoving . in @xcite
, we estimate the precision needed on the cosmological parameters when working with a bin width @xmath136 such that @xmath161 indeed , using a catalog with a smaller depth @xmath162 will allow us to use a smaller resolution for the cosmological parameters .
one has thus to find a compromise between depth and resolution as discussed in @xcite . on figure [ testr ] , we give an example with a bin width of @xmath163 which produces a background noise .
now , we apply our test to the quasars catalog @xcite used ahead ( see figure [ catq ] ) .
no topological signature was found @xcite .
does it mean that there is no topological effect on scales smaller than @xmath164 ?
not necessarily , since we applied the test with presicions @xmath165 and were unable to span the full cosmological parameters space with the required accuracy .
the computational time is one of the main limitations of our technique ( this point is discussed extensively in @xcite ) now , another limitation comes from the peculiar velocities of the sources . in @xcite
, we discussed in details the effect of peculiar velocities both on @xmath121pairs and on @xmath125pairs , we and showed that even if the amplitude of the topological signal is reduced , it is not destroyed .
we have reviewed the main allowed topological structures for relativistic universe models and we discussed the various methods aimed to detect this topology . focusing on 3d - methods , we described the crystallographic method and we applied it to a quasars catalog to obtain a new constraint for euclidean manifolds , @xmath166 .
then we discussed the failure of this method for detecting the topology of compact hyperbolic manifolds .
we thus generalised the method by introducing a new technique based on the construction of a ccp index .
the main difference with the cosmic crystallography method is the collecting process of all correlated pairs of the catalog , which enhances the signal associated to the existence of a non trivial topology .
we gave examples of computation of such an index in order to show its statistical relevance .
we finally discussed the implementation difficulties of this new method when working with observational data .
99 a.a .
friedmann , zeitschr.fr phys .
* 21 * ( 1924 ) 326 a. einstein , preuss .
, sitzungsber .
( 1917 ) 142 .
w. de sitter , mon . not .
* 78 * ( 1917 ) 3 .
h. olbers , _ la transparence de lespace cosmique _ ( 1823 ) ; traduction j. merleau - ponty in _
lunivers lge du positivisme _ , 319 ( 1983 ) eds .
vrin , paris .
e. mach , _ conservation of energy _ ( 1872 ) ; traduction e. bertrand , ed .
hermann , paris ( 1904 ) .
e. mach , _ the science of mechanics _ ( 1883 ) .
( new edition open court publishing company , illinois , 1902 ) .
epicure , _ letter to herodotus _ g.f.r .
ellis , gen .
* 2 * ( 1971 ) 7 .
wheeler , ann .
* 2 * ( 1957 ) 604 .
d. aktaz et h. pagels , phys . rev
. * d25 * ( 1982 ) 2065 . s. hawking , nucl . phys . *
b239 * ( 1984 ) 257 .
gibbons , class .
. grav . * 15 * ( 1998 ) 2605 .
e. feodoroff , russian journ . for crystallography and mineralogy
* 21 * ( 1885 ) 1 .
l. bierberbach , mathematische annalen * 70 * , ( 1911 ) 297 ; _ ibid _ , * 72 * ( 1912 ) 400 .
w. novacki , commentarii mathematici helvetici * 7 * ( 1934 ) 81 .
f. klein , mathematisches annalen * 37 * ( 1890 ) 1 .
w. killing , mathematisches annalen * 39 * ( 1891 ) 1 .
wolf , comptes rendus de lacadmie des sciences de paris * 250 * ( 1960 ) 3443 . w.p .
thurston _ the geometry and topology of three manifolds _ , princeton lecture notes ( 1979 ) .
thurston and j.r .
weeks , sci . am . ,
july , 94 ( 1984 ) .
j. weeks , _ the shape of space : how to visualize surfaces and three dimensional manifolds _ , m. dekker , new - york ( 1985 ) .
thurston , _ three - dimensional geometry and topology _ vol . 1 ( 1997 ) princeton methematical series 35 , ed
. s. levy , princeton university press . w.p .
thurston , class .
. grav . * 15 * ( 1998 ) 2545 .
luminet , proceedings of `` concepts de lespace en physique '' , les houches , 29 sept - 3 oct 1997 ( 1998 ) , to appear in acta cosmologica , [ gr - qc/9804006 ] .
luminet , j - p .
luminet , b.f .
roukema , proceedings of cargese summer school `` theoretical and observational cosmology '' , cargese , corsica , august 1998 [ astro - ph/9901364 ] .
luminet , _
a. friedmann , g. lemaitre : essais de cosmologie _ , prcds de _ linvention du big - bang _
seuil , collection `` source du savoir '' ( 1997 ) .
luminet , g. starkman , j. weeks , scientific american , april 1999 , * 280 * 90 .
r. geroch , proc .
scool of physics ` enrico fermi ' , course xlvii , ed .
sachs , ny academic press ( 1971 ) 104 .
s. hawking and g.f.r .
ellis _ large scale structure of spacetime _ ( 1973 ) cambridge university press . j.a .
wolf , _ spaces of constant curvature _ , cinquime dition , publish or perish inc .
, wilmington usa ( 1967 ) .
beardon , _ the geometry of discrete groups _ , new york , springer ( 1983 ) .
m. nakahara , _ geometry , topology and physics _ , adam hilger , bristol , new - york ( 1990 ) .
m. lachize - rey and j .- p .
luminet , phys . rep . * 254 * ( 1995 ) 135 .
uzan , int .
journal of theor .
physics * 36 * , ( 1997 ) 2167 .
uzan , _ dfauts _ _ topologiques _ _ et _ _ conditions _ _ aux _ _ limites _ _ en _ _ cosmologie _ _ primordiale _ , thse de luniversit paris xi ( 1998 ) [ chapters 8 - 10 ] , [ http://milak.irfu.se/@xmath167uzan/pub.html ] .
thurston , bull .
* 6 * , 357 ( 1982 ) .
j. weeks , phd thesis , princeton university ( 1985 ) .
s.v . matveev and a.t .
fomenko , russian math .
surveys * 43 * ( 1988 ) 3 .
j. weeks : http://www.geom.umn.edu:80/software .
mostow , ann .
math . studies * 78 * ( 1973 ) princeton university press .
g. prasad , invent . math .
* 21 * ( 1973 ) 255 .
d. gabai , g.r .
meyerhoff and n. thurston , msri preprint 1996 - 058 ( 1996 ) .
k. schwarzschild , vierteljahrsschrift der astr . ges .
* 35 * ( 1900 ) 337 ; english translation in class .
grav * 15 * ( 1998 ) 2539 .
proceedings of the workshop _ cosmology and topology _ , cleveland 25 - 27 october 1997 , class .
. grav . * 15*(1998 ) .
roukema , month . not .
* 283 * ( 1996 ) 1147 .
b.f . roukema and a.c .
edge , month . not .
* 292 * ( 1997 ) 105 .
d. stevens , d. scott and j. silk , phys .
* 71 * ( 1993 ) 20 .
a. de oliveira costa , g.f .
smoot and a.a .
starobinsky , astrophys .
j. * 468 * ( 1996 ) 457 . n.j .
cornish , d.n .
spergel and g.d .
starkman , phys .
* d57 * ( 1998 ) 5982 .
j. levin , e. scannapieco , g. de gasperis , j. silk and j.d .
barrow , [ astro - ph/9807206 ] .
j. levin , e. scannapieco and j. silk , class .
. grav . * 15 * ( 1998 ) 2689 ; _ ibid .
_ phys.rev . *
d58 * ( 1998 ) 103516 .
inoue , [ astro - ph/9810034 ] .
bond , d. pogosyan and t. souradeep , class .
. grav . * 15 * ( 1998 ) 2671 .
uzan and p. peter , phys . lett . *
b406 * ( 1997 ) 20 .
uzan , phys . rev .
* d58 * ( 1998 ) 087301 ; _ ibid .
_ , class .
quant . grav . *
15 * ( 1998 ) 2711 .
r. aurich , [ astro - ph/9903032 ] .
fang and h. sato , comm .
( china ) * 2 * ( 1983 ) 1055 .
d.d . sokolov and v.f .
shvartsman , sov .
jetp * 39 * ( 1974 ) 196 .
t. fetisova _
et al . _ , astron
. lett . * 193 * ( 1993 ) , 198 .
m. demianski and m. lapucha , month . not .
* 224 * ( 1987 ) 527 .
fagundes and w.f .
wichoski , astrophys .
j. * 322 * ( 1987 ) l5 .
gott iii , month . not .
r. astron .
* 193 * ( 1980 ) 153 .
fagundes and e. gaussman , [ astro - ph/9704259 ] .
r. lehoucq , m. lachize - rey and j - p .
luminet , astron .
* 313 * ( 1996 ) 339 .
g. pal , acta .
hungaricae * 30 * ( 1971 ) 51 .
et al . _ , [ astro - ph/9812339 ] .
r. lehoucq , j - p .
luminet and j - p .
uzan , astron .
, to appear ( 1999 ) [ astro - ph/9810107 ] .
gomero , a.f.f .
teixeira , m.j .
rebouas and a. bernui , [ gr - qc/9811038 ] .
wolf , _ spaces of constant curvature _ , fifth edition , publish or perish inc .
, wilmington usa ( 1984 ) .
coxeter , _ non euclidean geometry _ , university of toronto press ( 1965 ) .
fagundes , astrophys .
j. * 338 * , 618 ( 1989 ) ; astrophys .
j. * 349 * ( 1960 ) 678 .
fagundes and e. gaussman , [ astroph/9811368 ] .
uzan , r. lehoucq and j - p .
luminet [ astroph/9903155 ] preprint n@xmath168 ugva - dpt 1999/03 - 1028 l.a .
best , canadian j. math .
* 23 * ( 1971 ) 451 .
b.f . roukema and s. bajtlik ,
to appear in month . not .
[ astro - ph/9903453 ] .
catalog n@xmath169 vii/207 _ quasars and active galactic nuclei _ ( 8th ed . )
( veron + 1998 ) .
available at http://cdsweb.u-strasbg.fr/viz-bin/vizier?-source=vii/207/ via the centre de donnes astronomiques de strasbourg [ http://cdsweb.u-strasbg.fr/cats/cats.html ] .
peebles , _ principles of physical cosmology _ , princeton university press ( 1993 ) . | multi connected universe models with space idenfication scales smaller than the size of the observable universe produce topological images in the catalogs of cosmic sources . in this review ,
we present the recent developments for the search of the topology of the universe focusing on three dimensional methods .
we present the crystallographic method , we give a new lower bound on the size of locally euclidean multi connected universe model of @xmath0 based on this method and a quasar catalog , we discuss its successes and failures , and the attemps to generalise it . we finally introduce a new statistical method based on a collecting correlated pair ( ccp ) technique .
email : [email protected] , [email protected] , [email protected] |
the theory of open quantum systems plays a central role in the description of realistic quantum systems due to unavoidable interaction with the environment .
as is well known , the system - environment interaction can lead to energy dissipation and decoherence @xcite , posing a major challenge to the development of modern technologies based on quantum coherence @xcite . due to its fundamental character and practical implications ,
the investigation of dissipative processes has been a subject of vigorous research , where the standard approach assumes a system - environment weak coupling and a memoryless quantum dynamics ( the born - markov approximation ) . under such assumptions ,
system dynamics are determined by a quantum markovian master equation , i.e. , a completely positive quantum dynamical map with a generator in the lindblad form @xcite .
although the markovian approach has been widely used , there is a growing interest in understanding and controlling non - markovianity . in quantum metrology ,
for example , entangled states can be used to overcome the shot noise limit @xcite in precision spectroscopy , even in the presence of decoherence @xcite . however , as suggested in refs .
@xcite , higher precision could be achieved in a non - markovian environment , since a small markovian noise would be enough to restore the shot noise limit .
non - markovian dynamics also play an important role in quantum biology @xcite , where interaction with a non - markovian environment can be used to optimize energy transport in photosynthetic complexes @xcite , and can be observed in condensed matter devices like quantum dots @xcite and superconducting qubits @xcite .
furthermore , as pointed out recently in studies involving quantum key distribution @xcite , quantum correlation generation @xcite , optimal control @xcite , and quantum communication @xcite , the use of non - markovian dynamics could offer an advantage over markovian dynamics . this scenario has motivated studies aimed at characterizing and quantifying non - markovian aspects of the time evolution of an open quantum system @xcite . however , unlike the classical case , the definition of non - markovianity in the scope of quantum dynamics is still a controversial issue .
for example , breuer , laine and piilo ( blp ) @xcite have proposed a measure for non - markovianity using the fact that all completely positive - trace preserving ( cptp ) maps increase the indistinguishability between quantum states . from a physical perspective
, a quantum dynamics would be non - markovian if there were a temporary back - flow of information from the environment to the system .
on the other hand , for rivas , huelga and plenio ( rhp ) @xcite , a quantum dynamics would be non - markovian if it could not be described by a _ divisible _ cptp map .
formally , for such cases , one could not find a cptp map @xmath0 , describing the evolution of the density operator @xmath1 from time @xmath2 to @xmath3 , such that @xmath4 , where @xmath5 and @xmath6 are two cptp maps .
therefore , the indivisibility of a map would be the signature of non - markovian dynamics .
these two different concepts of non - markovianity are not equivalent @xcite : although all divisible maps are markovian with respect to the blp criterion , the converse is not always valid @xcite . in this paper , we explore the idea of how one might manipulate the markovian nature of a dissipative subsystem , by exploiting features of its being a part of a composite system . for that
, we study the dynamics of interacting two - state systems ( tss ) coupled to a common thermal reservoir . by changing the composite initial state and/or the tss couplings ,
we show that it is possible to modify _ in situ _ the characteristics of the subsystem s dissipation , enabling one to induce a transition from markovian to non - markovian dynamics and _
vice versa_. moreover , we observe the possibility of having different behaviors for the composite and subsystem , even when they are coupled to a common thermal environment .
finally , we provide a qualitative and quantitative description of how the environmental tss acts as part of the subsystem environment .
we initiate our analysis by choosing an exactly soluble analytical model that is capable of presenting the physics we want to exploit from dissipative composite systems . therefore , our starting point is the dephasing model for two interacting two - state systems ( 2-tss ) @xmath7 with @xmath8 , where @xmath9 is the diagonal pauli matrix and @xmath10 . the choice of this model is also motivated by the possibility of implementation in different experimental settings . for example , it could be realized in superconducting qubits @xcite , trapped ions @xcite , ultracold atoms in an optical lattice @xcite , and nmr systems @xcite .
in addition , such a model , without tss - tss couplings , is also considered as a paradigm of quantum registers @xcite .
the bath of oscillators , introduced by the canonical bosonic creation and annihilation operators @xmath11 and @xmath12 , is characterized by its spectral density @xmath13@xcite , and is responsible for imposing a nonunitary evolution for the 2-tss . since @xmath14=0 $ ] ,
the populations of the eigenstates of @xmath15 are constants of motion and the coupling with the environment solely induces random dephasing between any superposition of those eigenstates .
the ( 2-tss)-bath time evolution operator can be determined as @xmath16 with @xmath17 .
consequently , if @xmath18 denotes the density matrix of the 2-tss plus bath , then @xmath19 . regarding the ( 2-tss)-bath initial correlations , the initial state @xmath20 is hereafter assumed to be separable , i.e. , @xmath21 , where the bath is considered to be in equilibrium at temperature @xmath22 and therefore @xmath23 .
the dynamics of the open 2-tss follows from the system s reduced density matrix , defined by @xmath24 $ ] , where @xmath25 $ ] means the trace of bath degrees of freedom . performing the calculation
, one can find that @xmath26 obeys the exact master equation @xmath27+\frac{\gamma(t)}{2}\left(s_z\rho(t ) s_z-\frac{1}{2 } \{s_z^2,\rho(t)\}\right ) , \label{meq}~~\end{aligned}\ ] ] where @xmath28$](@xmath29 ) denotes the standard ( anti)commutator , and @xmath30 is the time - dependent dephasing rate .
coupling is renormalized , such that @xmath31 , when working with the system s reduced dynamics . throughout the manuscript
, the @xmath32 values used for the system s reduced dynamics stand for their renormalized values . ]
since no approximation has been made whatsoever , it is worth noting that eq .
( [ meq ] ) constitutes a genuine cptp quantum dynamical map .
the master equation ( [ meq ] ) has a very suitable form for the analysis of quantum markovianity , since it can be directly compared with the well - known lindblad theory for open systems @xcite . indeed , if @xmath33 for all @xmath34 , the time - local master equation describes a divisible cptp map .
therefore , under this condition , the dynamics would fall into the class of problems considered as paradigms of quantum markovian processes , since there is only a single decoherence channel @xcite .
the simplest example one can find for the system hamiltonian eq .
( [ horiginal ] ) that has @xmath35 , is the case of an ohmic bath , i.e. , @xmath36 , one finds that , if @xmath37 , @xmath38 is a positive - valued function for any temperature and therefore the system dynamics is markovian . on the other hand ,
if @xmath39 , depending upon the temperature , @xmath38 can present negative values , given a non - divisible map . ] . indeed , as depicted in fig .
[ fig](a ) , for any bath temperature @xmath22 , the dephasing rate satisfies the condition of being non - negative for each fixed @xmath34 .
consequently , the 2-tss dissipative dynamics would be categorized as markovian .
another important case happens when the environment seen by the system of interest presents a pronounced peak ( often referred as a lorentzian peak @xcite ) at a characteristic frequency @xmath40 .
relevant examples are superconducting qubits coupled to readout dc - squids @xcite , electron transfer in biological and chemical systems @xcite and semicondutor quantum dots @xcite , to name just a few .
figure [ fig](b ) shows the dephasing rate assuming a lorentzian shape for the bath spectral density @xcite .
as one can observe , for the case in which the resonance width @xmath41 is of the same order as the frequency peak @xmath40 , the dephasing rate can present negative values ( solid curve ) or be a positive - valued function ( dashed and dot - dashed curves ) , depending upon the bath temperature . thus the composite dissipative dynamics would be markovian as long as the thermal energy scale is comparable to or larger than the resonance parameters ( dashed and dot - dashed curves ) , i.e. , @xmath42 .
thus far , we have only been concerned with characterizing the dissipative dynamics of the composite system , i.e. , the open 2-tss .
what we now want to address is whether or not the single tss dissipative dynamics can be tuned _ in situ _ such that its behavior would differ from that observed for the composite system .
in fact , one could envision that , due to its interaction with the other tss ( hereafter labeled the _ auxiliary _ tss ) , a single tss would be coupled to a structured bath ( auxiliary tss+environment ) , which would play the role of an effective bath , and could induce a different nature for the dissipative process .
if so , it might be possible to change dynamically the nature of such a structured bath by varying the tss - tss coupling @xmath32 and/or the initial state of the auxiliary tss . as matter of fact , the approach of considering the subsystem s environment to be composed a common environment plus the rest of the composite system was employed long ago .
indeed , regarding to the characterization of an effective dissipative mechanism , it was successfully applied to the class of problems mapped onto a tss coupled to a harmonic oscillator in the presence of a markov bath , where perturbative @xcite , non - perturbative @xcite , spectral density series representation @xcite and semi - infinite chain representation @xcite techniques were proposed to describe such an effective dynamics in several contexts .
in addition , a multi - spin environment coupled to local bosonic baths has also been studied @xcite in the context of markovianity .
it is also noteworthy that the presence of initial correlations in an environment composed of several subparts can lead to changes in the subsystem dynamics .
such an influence has been successfully investigated , with regard to the blp measure , theoretically @xcite and experimentally @xcite for the specific case of non interacting tss - tss systems , which interact locally with correlated multimode fields . in spite of the results mentioned above , no systematic investigation of their markovianity
has been undertaken using both rhp and blp measures . nor have interacting tss - tss composite systems in the presence of thermal baths been analyzed , where tunability seems more natural for manipulating the dissipative dynamics nature of several physical implementations .
we shall address this question by looking at the single tss reduced density matrix @xmath43={\rm tr}_2[\rho(t)]$ ] , where @xmath44 $ ] means the trace of the auxiliary tss degrees of freedom .
its matrix representation in the @xmath9 eigenbasis ( @xmath45 ) can be cast in the simple form@xmath46 where @xmath47 and the matrix elements are given by @xmath48 and @xmath49 .
hence , one finds that , in general , the dynamical map describing eq .
( [ redmatrix ] ) is non - linear , since it depends on the composite initial state .
indeed , such a feature becomes clear when one tries to write the reduced density matrix eq .
( [ redmatrix ] ) in a similar form to a kraus representation @xcite , i.e. , @xmath50 , with operation elements @xmath51 given by @xmath52 observe that those operators are dependent on the composite initial state , which would not be a genuine kraus representation . therefore , it is clear that , in general , the dynamical map describing the evolution of the single tss reduced matrix @xmath53 is itself dependent on the initial condition @xmath54 , which _ breaks _ the linearity of the map .
furthermore , the reduced dynamics are not necessarily described by a completely positive dynamical map .
it is worthy of mentioning that , under such circumstances , the characterization of the single tss dynamics as markovian or not is disputable with current measures , since both rhp and blp proposals rely on linear maps .
nevertheless , it is still possible to find conditions where the dynamical map of the single tss reduced matrix @xmath53 is a linear cptp map , thus being consistent with rhp and blp constructions .
those happen when i ) the composite initial state is a product state , i.e. , @xmath55 , and ii ) there is no tss - tss interaction ( @xmath56 ) , since for both cases one finds @xmath57 . in other words , in these cases the reduced dynamics is described by a genuine kraus representation .
it follows from eq .
( [ redmatrix ] ) that the master equation for the single tss reduced density matrix reads @xmath58+\frac{\tilde{\gamma}(t)}{2}\big(\sigma_1^z\tilde{\rho}_1(t ) \sigma_1^z-\tilde{\rho}_1(t)\big),~~ \label{redmeq}\end{aligned}\ ] ] where @xmath59 , with @xmath60 , and @xmath61 .
note the lindblad structure of the master equation ( [ redmeq ] ) , where @xmath62 and @xmath63 play the roles of effective single tss hamiltonian and dephasing rate , respectively , and are manifestly dependent on the composite system initial state condition .
moreover , since @xmath64 depends only on the initial state @xmath65 and the tss - tss coupling @xmath32 , neither the unitary part of eq .
( [ redmeq ] ) nor the extra term in its dephasing rate is influenced by the system - bath coupling .
consequently , the bath and auxiliary tss contributions for each term of eq .
( [ redmeq ] ) can be identified and quantified .
two results immediately become apparent : i ) as expected , if the tss - tss coupling is zero ( @xmath66 ) , the single tss dissipative dynamics is completely enforced by the environment ; and ii ) if the auxiliary tss initial state is set in an eigenstate of @xmath9 , one finds that @xmath67 and @xmath68 , where @xmath69 $ ] .
these results highlight the fact that the auxiliary tss dynamics will be locked into its initial state , since @xmath9 is a constant of motion .
therefore the interacting term @xmath70 will only play the role of a fixed external field for the single tss , in which dissipative dynamics would follow the process imposed by the environment .
thus , for such cases , both the composite and single tss dissipative dynamics would have the same behavior . in spite of the two cases discussed above , in general @xmath71
is not determined only by the environment dephasing rate @xmath72 and is not a positive - valued function .
therefore , with regard to the concordance between blp and rhp measures , it is not possible to infer from the master equations ( [ meq ] ) and ( [ redmeq ] ) whether the single tss dissipative dynamics will follow the same behavior as that observed for the composite system .
consequently , the two measures have to be determined in order to find the cases of agreement in our system .
as already mentioned , the figure of merit for blp is the distinguishability of quantum states @xcite .
an important tool that has been used as a measure for the distinguishability of quantum states is the trace distance @xmath73 between two quantum states @xmath1 and @xmath74 @xcite . according to blp ,
since the trace distance has the feature that under cptp maps @xmath6 its value can not increase beyond its initial value @xcite , i.e. , @xmath75 , where @xmath76 , the trace distance could also be used as a definition of non - markovianity @xcite .
that definition is based upon the idea that a markovian dynamics has to be a process in which any two quantum states become less and less distinguishable as the dynamics flows , leading necessarily to a monotonic decrease of its trace distance .
thus , a non - monotonic behavior is interpreted as a back - flow of information from the environment to the system . from an experimental perspective
, the trace distance could be calculated by the tomography of the density operator , as was recently done by bi - heng liu _
. _ @xcite using all - optical experimental setups , or inferred from current fluctuations @xcite . for states given by eq .
( [ redmatrix ] ) , the trace distance can be easily determined as @xmath77 which time behavior is manifestly dependent on the 2-tss initial state . for instance , panels ( c ) ( ohmic ) and ( d ) ( lorentzian bath ) of fig .
[ fig ] show a representative initial condition , in which the auxiliary tss is set in an equal superposition of @xmath9 eigenstates , i.e. , @xmath78 .
the bath temperature is chosen such that both ohmic and lorentzian baths produce a markovian process for the composite system .
however , even though it asymptotically vanishes for both cases , the trace distance for single tss states is non - monotonic , indicating a non - markovian process .
moreover , this non - monotonic behavior implies that the tss dynamics is indivisible @xcite .
thus the tss dynamics is non - markovian not only in terms of the back - flow of information , but also with respect to the rhp criterion . on the other hand
, one can find a situation in which the single tss dynamics is markovian even when the composite system presents a non - markovian behavior .
this unexpected scenario is shown in fig .
the negative values assumed by the dephasing rate @xmath38 , panel ( a ) , ensure that the 2-tss dynamics is indivisible .
furthermore , as shown in panel ( b ) , the trace distance can be found having a non - monotonic behavior .
therefore , the composite system conforms to a non - markovian quantum process with respect to the blp and rhp criteria .
however , since the effective rate @xmath63 is always positive for the fixed initial condition @xmath79 , inset of panel ( a ) , we have a divisible dynamics , implying that the trace distance is a monotonic function of time @xcite , as illustrated in panel ( b ) .
consequently , the single tss dynamics is a markovian process in the standard way .
a similar result was obtained for non - interacting tss - tss systems coupled locally with correlated multimode fields @xcite .
the results and examples offered here demonstrate i ) that the single tss hamiltonian has no influence regarding the markovianity of both single tss and composite system ; ii ) if one tries to push the characterization of the markovianity of the single tss dynamics for cases where the map is non - linear , it is found that a 2-tss entangled state is not a sufficient condition for non - markovianity of the single tss dynamics .
indeed , a simple example would be having one of the bell states , e.g. , @xmath80 , as the 2-tss initial state .
for those states , one finds @xmath81 in eq .
( [ redmatrix ] ) , which leads to a non - dissipative dynamics for the single tss dynamics .
this result explicitly shows that the single tss reduced dynamics has a convoluted dependence with the 2-tss initial state , which does not allow one drawing immediate conclusions about the role played by the presence of entanglement in the initial state ; and iii ) that the presence of the auxiliary tss not only has quantitative effects , but may also have , with regard to blp and rhp criteria , a decisive impact on the nature of the tss dynamics .
such an influence can be made more explicit if one focuses on cases where the composite initial state is a product state ( @xmath55 ) .
in fact , for those the extra term in @xmath71 , namely , @xmath82 , can be written in the simple form @xmath83 furthermore , if one considers problems having the same @xmath79 , the figure of merit for both criteria becomes @xmath71 , since the monotonicity of the trace distance eq .
( [ tracered ] ) is determined by @xmath84 . for this case , the agreement between the two criteria is guaranteed because eq .
( [ redmeq ] ) describes a single decoherence channel with the same @xmath63 for all @xmath85 @xcite .
thus , as @xmath86 , it is clear that can be established a competition when setting the nature of the reduced tss dynamics , which can be tailored through the knobs @xmath32 and @xmath87 due to the presence of the auxiliary tss . indeed , except for the trivial case @xmath88 , eq .
( [ gammaaux ] ) shows that the term due to coupling with the auxiliary tss will induce an _
ad infinitum _ pattern of momentary loss and recurrence of quantum coherence observed for the reduced tss dynamics , which is a characteristic of the entanglement created between a tiny number of degrees of freedom
. thus the coupling with the auxiliary tss constitutes a channel , here reversible , for exchanging information .
the reversibility of such a channel is only possible here because of the diagonal nature of the interactions , which maintains the amplitude of @xmath89 constant .
consequently , the irreversibility of the dephasing process observed for the reduced tss is only led by its direct coupling to the environment . following that reasoning
, one could expect that the same results would be found for the reduced tss dynamics if an independent bath model were assumed , instead of the common environment model eq .
( [ horiginal ] ) . as a matter of fact ,
as long as the diagonal nature of the interactions is preserved , the same results eqs .
( [ redmatrix])-([gammaaux ] ) are obtained , with @xmath38 reading the dephasing rates due to the individual baths coupled to the single tsss .
a final note on the single tss dissipative dynamics comes from the observation that , if @xmath79 is assumed to be a thermal state , our problem resembles very much with the one having a unique ( structured ) thermal bath initially decoupled from the single tss . despite several known examples of the tss - harmonic oscillator problem @xcite , because of the different character between the spin and bosonic degrees of freedom , it does not seem possible to assign an effective spectral density for the tss - tss case
nevertheless , for the case of a constant @xmath32 tss - tss interaction , one can single out a spectral density due to the spin character of the structured bath .
following from eq .
( [ gammaaux ] ) , one finds that @xmath90 , which leads to the spin component of the spectral density @xmath91 .
those finds reveal that the auxiliary tss behaves as a mode filter for the natural @xmath92 frequency and its harmonics , which are weighted by a boltzmann factor of effective temperature @xmath93 .
the analysis of the exact soluble model eq .
( [ horiginal ] ) gives evidence that the presence of an auxiliary system coupled to the system of interest could create a knob to manipulate _ in situ _ its markovianity .
however , since it was obtained for the particular situation where the system - environment interaction commutes with the system hamiltonian , one could wonder whether such a feature would be spoiled if the system - bath interaction would not commute with the system hamiltonian .
unfortunately , for such cases , exact solutions are rare , if not impossible .
nevertheless , we now present a case study , which is representative of some physical systems @xcite , where one finds that the idea of having a knob for markovianity , due to the presence of an auxiliary system , does also apply .
our case study has the feature that , _ by construction _ , the composite system dynamics is assumed to be markovian , but depending on its initial state , the single tss dynamics can be found to be both markovian or non - markovian .
the hamiltonian of interest is exactly the one given by eq .
( [ horiginal ] ) , except for the system - environment interaction , which now reads @xmath94 assuming that the system - environment interaction is weak and the composite system dynamics is markovian , such that the born - markov approximation is aplicable , the system s reduced master equation is found to have a lindblad form with positive rates @xcite .
as for the single tss dynamics , as expected , its markovian character is dependent on the 2-tss initial state .
for example , for the environment temperature @xmath95 , if the 2-tss initial state is the separable state @xmath96 , one can put the single tss master equation in a lindblad form given by @xmath97 + \left(\frac{\gamma_-e^{-\gamma_- t}}{1+e^{-\gamma_- t}}\right)\left(2\sigma_-\tilde{\rho}_1\sigma_+ -\tilde{\rho}_1\sigma_+\sigma_--\sigma_+\sigma_-\tilde{\rho}_1\right),\end{aligned}\ ] ] where @xmath98 . , @xmath99 and @xmath100 .
it is worthy of note that , for a general initial state , even for a separable state , it may not be possible writing the single tss master equation in a lindblad form . ] thus , for that initial condition , the single tss dynamics would be markovian . on other hand , if the 2-tss initial state is still separable , but given by another condition , e.g. , @xmath101 , with @xmath102 , one finds , using the trace distance , that the single tss dynamics would be characterized as non - markovian , according to both blp and rhp criteria ( see fig .
[ fig3 ] ) .
therefore , such examples make clear that the 2-tss initial state can be used as a knob to determine the markovianity of the single tss , even having the system - environment interaction not commuting with the system dynamics .
in summary , we have shown the possibility of manipulating the markovianity of a subsystem of interest without having to change the common thermal environment properties or to assume a correlated environment state , which can be daunting requirements for some physical implementations . by choosing an exactly soluble model and a case study for a non - commuting system - environment interaction , we illustrated how one can induce a transition from markovian to non - markovian dynamics , and _ vice versa _ , by changing the characteristics of the composite system .
such analises build evidences that a knob for markovianity can be introduced when one couples an auxiliary system to the system of interest . among the perspectives offered by this work
, one could envision points regarding whether it would be possible having the concept of markovianity and non - markovianity defined for those cases which dynamics is described by non - linear maps , the extension to other non - commutative tss - tss interactions and the role of the number of degree of freedom involved in the coupling auxiliary system . on this last point
, it would be interesting to investigate whether there would be a limitation on the maximum number of degrees of freedom allowing for the existence of the knob + f.b .
is supported by instituto nacional de cincia e tecnologia - informao quntica ( inct - iq ) and by fundao de amparo pesquisa do estado de so paulo ( fapesp ) under grant number 2012/51589 - 1 .
t.w . acknowledges financial support from cnpq ( brazil ) through grant no .
478682/2013 - 1 . s. f. huelga _
et al_. , phys .
. lett . * 108 * , 160402 ( 2012 ) ; f. f. fanchini _
et al . _ , phys . rev .
a * 81 * , 052107 ( 2010 ) ; jin - shi xu _
et al . _ ,
commun . * 4 * , 2851 ( 2013 ) ; a. darrigo _ et al .
_ , ann . phys . * 350 * , 211 ( 2014 ) ; a. orieux _
et al . _ , sci .
* 5 * , 8575 ( 2015 ) ; r. lo franco _ et al . _ ,
b * 27 * , 1345053 ( 2013 ) .
d. chruciski _ et al .
lett . * 112 * , 120404 ( 2014 ) ; b. bylicka _
et al . _ , sci
. rep . * 4 * , 5720 ( 2014 ) ; x .- m .
a * 82 * , 042103 ( 2010 ) ; s. luo _
et al . _ ,
a * 86 * , 044101 ( 2012 ) ; m. m. wolf et al .
lett . , 150402 ( 2008 ) .
a. friedenauer _
phys . * 4 * , 757 ( 2008 ) ; d. porras _
et al . _ ,
a * 78 * , 010101 ( 2008 ) ; k. kim _
_ new j. phys . * 13 * , 105003 ( 2011 ) ; p. schindler _ et al .
* 9 * , 361 ( 2013 ) ; j. simon _
et al . _ ,
nature * 472 * , 307 ( 2011 ) ; a. recati _
et al . _ ,
a * 94 * , 040404 ( 2005 ) ; p. p.
a * 77 * , 051601 ( 2008 ) ; p. haikka _
et al . _ ,
a * 84 * , 031602 ( 2011 ) . | we study the markovianity of a composite system and its subsystems .
we show how the dissipative nature of a subsystem s dynamics can be modified without having to change properties of the composite system environment . by preparing different system initial states or dynamically manipulating the subsystem coupling
, we find that it is possible to induce a transition from markov to non - markov behavior , and _ vice versa_. |
the _ ratchet _ effect , i. e. the possibility to obtain the directed transport by using zero - mean perturbations only , has induced notable scrutiny over the last decade [ 1 ] . initially , most studies have been focused on noisy overdamped models have been inspired by a molecular motors realm [ 1 ] .
then the ratchet s approach has been applied to a broad class of physical systems in which inertia effects are essential [ 2 ] .
the examples are josepshon junctions [ 3 ] , cold atoms systems [ 4 ] , and mechanical engines [ 5 ] .
recently it has been shown that the ratchet idea can be viewed as a part of a general symmetry - breaking approach [ 6 ] .
this approach is based on the analysis of all relevant symmetries which have to be broken in order to fulfill necessary conditions for a dc - output appearance .
the formalization of symmetry analysis for one particle s dynamics has been addressed in ref.[7 ] . in the present paper we aim at collective rectification effects which arise in a set of coupled single ratchet units .
although various examples of interacting ratchet systems have been proposed already in the context of molecular motors [ 8 ] , here the emphasis is put on the weak - noise underdamped limit .
we show that a coupling scheme determines a set of certain symmetries for the ratchet s collective and hence defines necessary conditions for a dc - current generation .
dynamical mechanisms of a current rectification are connected with a coherence between units , which depends not only on a coupling scheme , but also on a strength of interactions .
let us consider a set consisting of @xmath0 identical dynamical units , @xmath1 , that are linearly and symmetrically coupled .
the coupling scheme is described by some graph which can be encoded in the symmetrical @xmath2 binary matrix @xmath3 , @xmath4 .
the equations of motion are the following : @xmath5 where @xmath6 is the double periodic force function , @xmath7 is linear over all arguments coupling function , and @xmath8 is the strength of interactions .
the stochastic terms @xmath9 are mutually independent delta - correlated gaussian white noises , @xmath10 , where @xmath11 is the noise intensity .
we assume also that the force function , @xmath12 , is the same for all the units , but the phases , @xmath13 and @xmath14 , can be different for different units . finally , we are interested in the mean dc - current , @xmath15 following the symmetry analysis ideology [ 6 - 7 ] , in order to determine necessary conditions for a dc - output appearance , we have to check whether there exist symmetry transformations which allow to generate out for each trajectory of the system ( 1 ) another one with a reversal velocity .
the presence of the white noise , @xmath9 , does not change the symmetry properties of the system .
moreover , the coupling to a heat bath leads to an effective exploration of the whole phase space and produce an averaging in a case of several coexisting attractors [ 6,9 ] .
for the one - particle case , @xmath16 , a transformation of interest has to involve a change of the sign of @xmath17 ( and thus change of the current direction ) .
it allows also some shifts in time and space domains [ 6 - 7 ] : @xmath18 in the case of several coupled ratchets a symmetry operation should be performed in the global coordinate space @xmath19 and can also involve a permutation between different units .
the corresponding symmetry operation can be described as the linear transformation , @xmath20 where @xmath21 is a @xmath2 binary matrix with only one non - zero element at each row and line , and @xmath22 is the vector of shifts .
the matrix @xmath23 encodes permutation between units .
the properties of the permutation matrix @xmath21 ( and the existence of such a matrix at all ) are strongly depend on the structure of connections among units .
below , using simple examples , we will illustrate this statement .
_ breaking symmetries by connections .
_ let us first consider the case of two particles ( rotators ) in the standing - wave potential with the modulated amplitude [ 10 ] , coupled by the linear spring : @xmath24 where @xmath25 . in the uncoupled case , @xmath26 , both the systems posses symmetries of the type ( 3 ) : @xmath27 with @xmath28 for the first rotator , eq.(5 ) .
the symmetry transformations , eqs.(7 - 8 ) , are independent for each rotators .
the whole system , eqs.(5 - 6 ) , is symmetric with respect to the transformation @xmath29 for any choice of @xmath30 and @xmath31 .
so , the mean dc - output for the uncoupled case is zero ( line(1 ) in fig.1a )
. in the case of coupled particles , @xmath32 , abovementioned transformations , , eqs.(7 - 8 ) , should been conjugated and independent spatial and temporal shifts are forbidden now .
for @xmath33 and @xmath34 , the connection breaks both the symmetries and we can expect on nonzero current appearance ( see also ref . [ 11 ] for another example of an overdamped dimer with an additive driving force ) .
for the set of parameters @xmath35 , @xmath36 , @xmath37 , @xmath38 and @xmath39 , the symmetry violation is realized by the asymmetrical limit cycle with the negative winding number ( line(2 ) in fig.1a ) .
so , in this case the connection between the units destroy all the symmetries and leads to the dc - current generation .
_ restoring symmetries by connections .
_ let us now consider the dimer identical to the previous one , eq.(5 - 6 ) , but spatially shifted by @xmath40 ( half of the period ) : @xmath41 both the systems , eqs.(5 - 6 ) and eqs.(9 - 10 ) , can be transformed from one to another by the simple coordinate shift .
thus , the system in eqs.(9 - 10 ) produces the same mean current , as the system in eqs.(6 - 7 ) . as the next step we couple both the systems by additional connections @xmath42 ( see the inset in fig.1a ) .
the generalized symmetry transformation , eq.(4 ) , can be identified with the following permutation matrix : @xmath43 and the shift vector @xmath44 . here , despite to the previous case
, the introduction of additional connections leads to the appearance of the new symmetry and we may expect on the nullification of the dc - output ( line(3 ) in fig.1a ) . for the above set of parameters we found that the symmetry is realized in the phase space by two symmetry - coupled chaotic attractors with opposite mean velocities ( see fig.1b )
it is easy to check , that any other type of connection with two additional links does not restore the system s symmetry .
while the presence or the absence of the dc - output is clearly connected to the absence / presence of the symmetry transformation , the dc - current value is determined by dynamical mechanisms . from the previous studies [ 12 ]
it is known that global properties of a collective dynamics are defined by a coherence between units .
thus , we can expect that an efficiency of the current rectification is closely related to a degree of synchronization within a ratchet s collective . as a natural example we consider the model of @xmath0 globally coupled array of underdamped josepshon junctions ( jj ) , subjected to an ac - current @xmath45 [ 13 ] .
the equation for the superconducting phase difference @xmath46 across the single junction is @xmath47 where @xmath48 is the instantaneous mean array current . as the driving force we used the two - harmonics combination , @xmath49 , which ensures that all the relevant symmetries are broken [ 7 ] .
the numerically obtained dependence of the array dc - output on the strength of interaction @xmath8 is shown in fig.2(a ) .
for the set of parameters @xmath50 , the dependence demonstrates the presence of two successive current reversals , at @xmath51 and @xmath52 . in order to understand dynamical mechanisms of these events
, we introduced the mean decoherence , which we define as @xmath53 where @xmath54 means the averaging over the array and over the one period of ac - driving . the first result from the comparison of both the dependencies ( fig.2(b))is that the current reversal at @xmath55 is connected with the transition to the regime of the complete synchronization , @xmath56 .
lets now track briefly a relation between the current reversal and the synchronization .
initially , in the uncoupled limit , @xmath26 , all jj s work independently , each one as the single one - dimensional rectificator [ 1,6 ] .
the asymmetry of the ac - force @xmath45 is realized through the chaotic attractor with the positive mean current ( see fig.3a ) [ 14 ] .
the local phase space of each units , @xmath57 , has identical structure , but the relative phases of different units are randomly distributed ( depending on the initial conditions ) .
after the introduction of a nonzero coupling , @xmath32 , some coherence between the units occurs . from a point of view of the dynamics of the single junction , this leads to changes of an attractor structure in the phase space @xmath58 ( see fig.3b ) .
this causes changes in a projection of the attractor s invariant density on the velocity subspace @xmath59 and , as a result , leads to the current reversal at @xmath60 .
the further increase of the interaction strength up to @xmath55 results in the complete synchronization and to a shrinking of the global attractor in @xmath61 to the hyperplane @xmath58,@xmath62 .
the global attractor has now dimension equal to @xmath63 .
the attractor of the single unit now is the same as in the uncoupled limit but all the units have the same phase .
the mean dc - output returns to its value at the limit @xmath26 . here
some analogy with a current reversal in an one - dimensional deterministic ratchet [ 14 ] can be drawn . in the one - dimensional case
current reversals have been identified with tangent bifurcations from chaotic to regular ( limit cycles ) attractors [ 14 ] . in our case the current reversal at @xmath55 corresponds to the transition `` hyperchaos - chaos '' [ 15 ] , connected with the shrinking of a the system attractor in a global phase space @xmath61 .
it is interesting to note , that from a point of view of the local attractor of the single junction , this transition corresponds to a crisis , connected with _ expanding _ of the attractor in three - dimensional subspace @xmath64 ( compare fig.3a and fig.3b ) .
the presence of a weak noise ( which is equivalent to a weak coupling with a heat bath ) can strongly suppress correlations between units and may lead to a delayed synchronization transition [ 16 ] .
due to a strong conjugation between the synchronization and the current rectification process , we can manage a noise - induced current reversal for a fixed strength of interaction ( fig.2(b ) ) .
thus , the presence of a thermostat allows to control the dc - output by changing a temperature of the system .
finally , we have presented the symmetry approach to the problem of the collective current rectification by a set of coupled dynamical units .
this idea can be used for a more general problem such as an obtaining of a non - zero value of some relevant mean ensemble characteristic , @xmath65 .
this characteristic can correspond , for example , to a mean magnetization of a spin lattice with a complex geometry [ 17 ] .
the proposed collective ratchet s ideology may also be relevant in a context of a cooperative dynamics of neural networks , as an approach to a visual processing with directional selectivity [ 18 ] .
relations between symmetries of a collective ratchet and its coupling scheme on the one side , and relations between a topology of interactions and synchronization properties [ 19 ] , on the other one , may open an interesting perspective .
it has been found that in the thermodynamic limit , @xmath66 , the synchronization is impossible for nearest - neighbor coupled dynamical networks if the number of sites connected to a given site , @xmath67 , is a finite fraction of the total number of sites @xmath0 , @xmath68 [ 20 ] . on the other hand , in a case of small - world networks ,
i.e. sets with long ( in a sense of a topological distance ) connections , the synchronization can be achieved in the thermodynamic limit through a small fraction of distant connections [ 20 ] .
this nontrivial effect provides a tool for a control of the dc - output in a massive collective ratchet by changing a small number of relevant connections only .
p. hnggi and r. bartusek , lect . notes .
476 , ( 1996 ) 294 ; f. julicher , a. ajdari , and j. prost , rev .
69 ( 1997 ) 1269 ; r. d. astumian and p. hnggi , physics today 55 ( 2002 ) 33 ; p. reimann , phys .
361 ( 2002 ) 57 .
p. jung , j. g. kissner , and p. hnggi , phys .
76 ( 1996 ) 3436 ; m. barbi and m. salerno , phys .
e 62 ( 2000 ) 1988 ; l. machura et al , phys .
e 70 ( 2004 ) 061105 ; l. machura et al , j. phys .
c : condens .
matter 17 ( 2005 ) s3741 .
g. carapella , phys . rev .
b 63 ( 2001 ) 054515 ; g. carapella and g. costabile , phys
87 ( 2001 ) 077002 .
m. schiavoni , l. sanchez - palencia , f. renzoni , and g. grynberg , phys .
( 2003 ) 094101 ; p.
h. jones , m. goonasekera , and f. renzoni , phys .
93 ( 2004 ) 073904 ; r. gommers , s. denisov , and f. renzoni , phys .
96 ( 2006 ) 240604 .
b. norden , y. zolotaryuk , p. l. christiansen , and a. v. zolotaryuk , phys .
e 65 ( 2002 ) 011110 .
s. flach , o. yevtushenko , and y. zolotaryuk , phys .
84 ( 2000 ) 2358 .
s. denisov _
et all _ , phys .
e. 66 ( 2002 ) 041104 .
f. jlicher and j. prost , phys .
lett . 75 ( 1995 ) 2618 ; i. dernyi and t. vicsek , phys .
lett . 75 ( 1995 ) 374 ; p. reimann , r. kawai , c. van den broeck , and p. hnggi , europhys .
( 1999 ) 545 .
j. l. mateos , physica a 325 ( 2003 ) 92 .
r. z. sagdeev , d. a. usikov , and g. m. zaslavsky , nonlinear physics : from the pendulum to turbulence and chaos , harwood academic publ .
s. cilla , f. falo , and l. m. floria , phys .
e 63 ( 2001 ) 031110 . v. s. afraimovich , v. i. nekorkin , g. v. osipov , and v. d. shalfeev , stability , structures and chaos in nonlinear synchronization network , world scientific , singapore , 1994 .
k. wiesenfeld and p. hadley , phys .
let . 62 ( 1989 ) 1335 ;
k. y. tsang et al . , _ ibid . _ 66 ( 1991 ) 1094 ; k. y. tsang and i. b. schwartz , _ ibid . _ 68 ( 1992 ) 2265 .
j. l. mateos , phys .
84 ( 2000 ) 258 ; j. l. mateos , physica d , 168 - 169 ( 2002 ) 205 .
e. ott , chaos in dynamical systems , cambridge university press , cambridge , 1992 .
a. pikovsky , m. rosenblum , and j. kurths , synchronization : universal concept in nonlinear science , cambridge university press , cambridge , 2001 .
d. c. cabra and m. d. grynberg , phys .
82 ( 1999 ) 1768 , ph .
kurz , g. bihlmayer , k. hirai , and s. blugel , _ ibid . _ 86 ( 2001 ) 1106 ; h. sompolinsky and d. golomb , phys . rev .
a 43 ( 1991 ) 6990 .
u. e. vincent , a. n. njah , o. akinlade , and a. r. t. solarin , chaos 14 ( 2004 ) 1018 ; m. kostur , p. hnggi , p. talkner , and j. l. mateos , phys .
e 72 ( 2005 ) 036210 .
x. f. wang , int .
chaos 12 ( 2002 ) 885 ; b. ao and z. zheng , europhys .
74 ( 2006 ) 229 . | we consider a set of coupled underdamped ac - driven dynamical units exposed to a heat bath .
the coupling scheme defines the absence / presence of certain symmetries , which in turn cause a nonzero / zero value of a mean dc - output .
we discuss dynamical mechanisms of a dc - current appearance and identify current reversals with synchronization / desynchronization transitions in the collective ratchet s dynamics .
driven underdamped systems ; synchronization ; ratchet effect 05.45.ac , 05.60.cd |
one of the main motivations to study high energy heavy ion collisions is the creation of a new deconfined phase of strongly interacting matter , the so called quark - gluon plasma ( qgp ) @xcite . at the relativistic heavy ion collider ( rhic )
many experimental observations like , e.g. , jet quenching and high elliptic flow hint to the fact that a strongly coupled qgp ( sqgp ) might have been created @xcite . at cern - sps energies evidence for the creation of a new state of matter
has been published , e.g. , the enhanced k/@xmath0 ratio ( horn ) and the step in the mean transverse mass excitation function for pions , kaons and protons @xcite . especially the low energy ( high @xmath1 ) program at sps showed a culmination of exciting results
. therefore this energy regime will be the subject to further detailed studies at the cern - sps , bnl - rhic , jinr - nica and at the fair facility .
since the direct detection of free quarks and gluons is impossible due to the confining nature of qcd , it is important to model the dynamical evolution of heavy ion reactions to draw conclusions from the final state particle distributions about the interesting early stage of the reaction .
one approach which aims at the description of heavy ion reactions consistently from the initial state to the final state is relativistic transport theory @xcite .
this microscopic description has been applied quite successfully to the partonic as well as to the hadronic stage of the collision .
unfortunately , most transport approaches are restricted to @xmath2 scattering processes .
thus , if the particle density increases it becomes questionable if a restriction to two - particle interaction is still justified .
while first attempts to include multi - particle interactions have been proposed @xcite , this extension of transport theory is still in its infancy .
to explain hadronization and the phase transition between the hadronic and the partonic phase on a microscopic level is also one of the main open issues that still has to be resolved .
it is therefore difficult to find an appropriate prescription of the phase transition in such a microscopic approach .
first , however promising attempts to solve the microscopic hadronization problem can be found in the literature @xcite .
hydrodynamics , on the other hand , has been proposed many years ago as a tool for the description of the hot and dense stage of heavy ion reactions where the matter might behave like a locally thermalized ideal fluid @xcite . in this approach
it is possible to model phase transitions explicitly because one of the major inputs to a hydrodynamic calculation is the equation of state ( eos ) .
the hydrodynamic description has gained importance over the last few years because the high elliptic flow values that have been observed at rhic seem compatible with some ideal hydrodynamic predictions @xcite . the initial conditions and freeze - out prescription are the boundary conditions for a hydrodynamic calculation and therefore a further crucial input .
thus , the hydrodynamic results depend strongly on the initial and final state prescription that is applied in the specific calculation . to get a more consistent picture of the whole dynamics of heavy ion reactions various so called microscopic plus macroscopic ( micro+macro ) hybrid approaches
have been launched during the last decade .
most noteworthy in this respect are the pioneering studies related to a coupling between urqmd ( ultra - relativistic quantum molecular dynamics ) and hydrodynamics ( a detailed systematic investigation of this coupling procedure can be found in the following references @xcite ) .
other approaches in the same spirit are , e.g. , the nexspherio approach that uses initial conditions calculated in a non - equilibrium model ( nexus ) followed by an ideal hydrodynamic evolution @xcite or a hybrid approach by toneev et al . which uses qgsm initial conditions followed by a three - dimensional hydrodynamic evolution @xcite .
in this way event - by - event fluctuations are taken into account and the calculation mimics more realistically the experimental case . for
the freeze - out nexspherio employs a continuous emission scenario or a standard cooper - frye calculation .
other groups , e.g. , teaney et al .
@xcite , hirano et al .
@xcite , bass / nonaka @xcite , are using smooth glauber or color glass condensate initial conditions followed by a full two- or three - dimensional hydrodynamic evolution and calculate the freeze - out by a subsequent hadronic cascade .
the separation of chemical and kinetic freeze - out and final state interactions like resonance decays and rescatterings are taken into account .
there are two major conclusions from these previous studies : the treatment of the initial state fluctuations and the final decoupling is of major importance for a sound interpretation of the experimental data .
unfortunately , all presently existing micro+macro approaches rely on a complete separation of the three main ingredients ( initial conditions , hydrodynamic evolution , transport calculation ) .
thus , it is impossible to compare the evolution of the system between hydrodynamics and transport simulation directly and from the same initial conditions . this may provide essential new insights into the role of viscosity and local equilibration .
in addition , the usual separation of the program code does not allow for a dynamical coupling between hydrodynamics and transport calculation , which would be desirable to consistently solve the freeze - out puzzle @xcite .
to overcome these restrictions , we go forward and build a transport approach with an embedded three - dimensional ideal relativistic one fluid evolution for the hot and dense stage of the reaction .
this allows to reduce the parameters for the initial conditions and the freeze - out prescription .
the aim is to compare calculations with different eos within the same framework .
it will be possible to extract the effect of changes in the eos , e.g. , a phase transition from hadronic matter to the qgp , on observables . in this paper
we describe the specific micro+macro hybrid approach that embeds a hydrodynamic phase in the urqmd approach .
first we explain the initial conditions , then introduce the basics of the hydrodynamic evolution including the hadron gas eos and the transport calculation and illustrate how the freeze - out is treated . in the second part
, the sensitivity of the results on the parameters are tested and the time evolution of the baryon density and the particle numbers are compared .
results on particle multiplicities and rapidity as well as transverse mass spectra are presented in the third part . at present
we have calculated results imposing a hadron gas eos to provide a baseline calculation to disentangle the effects of the different assumptions for the underlying dynamics in a transport vs. hydrodynamic calculation .
the purely hadronic calculations can be compared in the broad energy regime from @xmath3gev where a vast amount of experimental data from bnl - ags and cern - sps exists and which will be explored in more detailed energy scans by the fair project near gsi and the critrhic program .
studies employing different eos are delayed to future work to concentrate on effects of the underlying dynamics first .
the modelling of the dynamical evolution of heavy ion reactions is essential to gain further insights about the properties of the newly produced hot and dense qcd matter .
transport theory aims at the description of all stages of the collision on the basis of an effective solution of the relativistic boltzmann equation @xcite @xmath4 this equation describes the time evolution of the distribution functions for particle species @xmath5 and includes the full collision term on the right hand side .
the interaction with external potentials leads to an additional term on the left hand side .
the influence of potentials gets small at higher energies compared to the energy that is transferred by collisions .
therefore , they are dropped in eqn .
[ boltzmann ] and are not further discussed here .
usually , the collision kernel is truncated on the level of binary collisions and @xmath6 processes to keep the calculation numerically tractable .
this microscopic approach has the advantage that it is applicable to non - equilibrium situations and the full phase space information is available at all stages of the heavy ion reaction .
the restriction to binary collisions assumes large mean free paths of the particles . between interactions the particle trajectories
are given by straight line trajectories and particles are assumed to be in asymptotic states between the collisions ( no `` memory effect '' ) .
this assumption might not be justified in the hot and very dense stage of heavy ion collisions anymore . in this regime
the continuum limit in form of relativistic hydrodynamics might fit better to the characteristics of the system .
the hydrodynamic evolution is governed by the energy and momentum conservation laws for given initial conditions , i.e. spatial distributions of energy and net baryon number densities .
the coordinate space is divided into small cells in which the distribution functions correspond to equilibrium distributions ( fermi or bose distribution ) . in this macroscopic approach
the propagated quantities are net baryon number and energy densities which can be translated into information about the temperature and chemical potential via the specific equation of state ( eos ) .
since the evolution is driven by pressure gradients and the pressure is determined via the eos , the eos is the essential ingredient for the hydrodynamical evolution .
thus , hydrodynamics is a good tool to describe collective behaviour .
ideal hydrodynamics applies to systems with small mean free path , otherwise viscous effects have to be taken into account @xcite .
a general advantage of hydrodynamics is the feature to explicitly incorporate phase transitions by changing the eos .
however , in the late stage of the heavy ion reaction the system gets too dilute to apply ideal fluid dynamics .
the hadronic rescatterings and decays of resonances have to be described , e.g. , by using a transport description .
overall , there are two crucial points one has to take care of when building up a transport+hydrodynamics hybrid approach .
the first is the initial switch from the microscopic to the macroscopic calculation where it has to be ensured that the local equilibrium assumption is fulfilled .
the second one is the so called freeze - out where the hydrodynamic fields are mapped to particles that are further propagated in a hadronic cascade .
the freeze - out transition must be placed in a region where both descriptions are valid at the same time , e.g . , the phase transition region . in the following the specific implementation developed here
will be discussed in more detail .
for our investigation , the urqmd model ( v2.3 ) @xcite is applied to heavy ion reactions from @xmath7 gev .
this non - equilibrium transport approach constitutes an effective solution of the relativistic boltzmann equation ( see eqn .
[ boltzmann ] ) .
the underlying degrees of freedom are hadrons and strings that are excited in high energetic binary collisions .
mean fields can in principle be taken into account in this framework , but the model is run in the so called cascade mode without inter - particle potentials . to omit the potentials is reasonable because the inclusion of mean fields would not change the results in the energy range that we are considering here .
note that this is consistent with the calculation of the equation of state for the hydrodynamic evolution where no mean field has been taken into account as well .
the projectile and target nuclei are initialised according to a woods - saxon profile in coordinate space and fermi momenta are assigned randomly for each nucleon in the rest frame of the corresponding nucleus .
the hadrons are propagated on straight lines until the collision criterium is fulfilled .
if the covariant relative distance @xmath8 between two particles gets smaller than a critical distance that is given by the corresponding total cross section a collision takes place , @xmath9 each collision process is calculated in the rest frame of the binary collision .
the reference frame that is used for the time ordering of the collisions and later on also for the switchings to and from the hydrodynamic phase is the equal speed - system of the nucleus - nucleus collision ( for symmetric systems the equal speed system is identical to the center of mass system ) . in urqmd 55 baryon and 32 meson species , ground state particles and all resonances with masses up to @xmath10 gev ,
are implemented with their specific properties and interaction cross sections .
in addition , full particle - antiparticle symmetry is applied .
isospin symmetry is assumed and only flavour - su(3 ) states are taken into account .
the elementary cross sections are calculated by detailed balance or the additive quark model or are fitted and parametrized according to the available experimental data . for resonance excitations and decays the breit - wigner formalism , utilizing their vacuum properties is employed . towards higher energies ,
the treatment of sub - hadronic degrees of freedom is of major importance . in the present model ,
these degrees of freedom enter via the introduction of a formation time for hadrons produced in the fragmentation of strings @xcite .
string excitation and fragmentation is treated according to the lund model . for hard collisions with large momentum transfer ( @xmath11 gev )
pythia is used for the calculation . a phase transition to a quark - gluon state
is not incorporated explicitly into the model dynamics . however , a detailed analysis of the model in equilibrium yields an effective equation of state of hagedorn type @xcite .
the urqmd transport model is successful in describing the yields and the @xmath12 spectra of various particles in proton - proton , proton - nucleus and nucleus - nucleus collisions @xcite .
a compilation of results of the actual version urqmd-2.3 compared to experimental data can be found in @xcite .
apart from the success of transport simulations to describe spectra and yields certain problems remain : * elliptic flow values above sps energies are too small @xcite , * hbt radii hint to a very small @xmath13 ratio @xcite , * strangeness , especially multi - strange baryons are not produced in sufficient amounts @xcite .
these observables that are sensitive to the early stage of the collision ( pressure ) or to the approach of thermal and chemical equilibrium during the collision history hint to the fact that a purely hadronic transport model may not be sufficient to describe the dynamics of the hot and dense stage of heavy ion reactions at higher energies @xcite .
therefore , these observations exemplify the need to embed a full three - dimensional relativistic fluid dynamics description for these stages of the reaction . for the results that are shown in this paper , the reference calculations are always performed employing the state - of - the - art urqmd-2.3 model .
the ultra - relativistic quantum molecular dynamics model is used to calculate the initial state of a heavy ion collision for the hydrodynamical evolution @xcite .
this is necessary to account for the non - equilibrium nature of the very early stage of the collision .
event - by - event fluctuations of the initial state are naturally included by this set - up .
the coupling between the urqmd initial state and the hydrodynamical evolution takes place when the two lorentz - contracted nuclei have passed through each other . the initial time to begin with the hydrodynamical evolution
is calculated via eqn .
[ eqn_tstart ] ( and is assumed to be at least 1 fm / c ) :
@xmath14 where r is the radius of the nucleus , @xmath15 is the nucleon mass and @xmath16 is the kinetic beam energy .
this assures that ( essentially ) all initial baryon - baryon scatterings have proceeded and that the energy deposition has taken place .
this is the earliest possible transition time where thermalization might be achieved @xcite .
it is also convenient from the hydrodynamical point of view since at that time the two baryon currents that fly into oppposite directions have separated again . in general ,
it is not well - established how and when chemical and kinetic equilibrium might have been reached in the early stage of the collision .
one of the problems is , e.g. , that the local equilibrium assumption might not apply equally well to all parts of the system at the same time in the computational frame which corresponds to the center of mass system of the two colliding nuclei . as a consequence ,
the faster particles have had less time in their local rest frame to equilibrate .
for the bulk part and the high density region at midrapidity the difference between the two frames is small .
these problems are present in all hydrodynamic / macroscopic approaches that rely on an equilibrium assumption and it is not our attempt to resolve these difficulties in this paper .
one perspective might be the dynamical coupling between the initial transport calculation and the hydrodynamic evolution including source terms on both sides of the transition surface . to allow for a consistent and numerically stable mapping of the point like particles from urqmd to the 3-dimensional spatial - grid with a cell size of @xmath17 , each hadron
is represented by a gaussian with a finite width .
`` pre - formed '' hadrons in the process of string fragmentation are also included in the transformation to the hydrodynamic quantities .
i.e. each particle is described by a three - dimensional gaussian distribution of its total energy , momentum ( in x- , y- , and z - direction ) and baryon number density .
the width of these gaussians is chosen to be @xmath18fm .
a smaller gaussian widths leads to numerical instabilities ( e.g. , entropy production ) in the further hydrodynamical evolution , while a broader width would smear out the initial fluctuations to a large extent . to account for the lorentz - contraction of the nuclei in the longitudinal direction ,
a lorentz - gamma - factor is included .
the resulting distribution function in the computational frame ( cf ) , e.g. , for the energy density , reads : @xmath19 where @xmath20 provides the proper normalisation , @xmath21 and @xmath22 are the energy density and total energy of the particle in the computational frame , while @xmath23 is the position vector of the particle .
summing over all single particle distribution functions leads to distributions of energy , momentum and baryon number densities in each cell . to allow for calculations at finite impact parameter the spectators - nucleons that have not interacted at all before the start time of the hydrodynamic evolution @xmath24 - are propagated separately from the hydrodynamic evolution
the spectators are propagated on straight line trajectories in the usual cascade mode until the end of the hydrodynamic phase has been reached . instead of smearing out the initial distributions by describing the point like hadrons as gaussian distributions
, one could also obtain a smooth distribution by averaging over a large sample of urqmd events .
our procedure of creating a new initial state for each event is motivated by the fact , that the experimental results all relate to observed ( averaged ) final , and not initial , states .
thus , event - by - event fluctuations of the initial state can be observed , ( e.g. , in @xmath25 fluctuations ) and have therefore been taken into account properly ( for discussion of the importance of these fluctuations see , e.g. , @xcite ) .
( color online ) initial energy density distribution in the reaction plane ( @xmath26 plane ) of one central ( @xmath27 fm ) pb+pb collision at @xmath28gev .
@xmath29 corresponds to the beam direction and @xmath30 to the in - plane axis ( direction of the impact parameter ) of the collision.,scaledwidth=50.0% ] ( color online ) initial net baryon number density distribution in the reaction plane ( @xmath26 plane ) of one central ( @xmath27 fm ) pb+pb collision at @xmath28gev .
@xmath29 corresponds to the beam direction and @xmath30 to the in - plane axis ( direction of the impact parameter ) of the collision.,scaledwidth=50.0% ] as an example , figs .
[ fig_epsilon_plane ] and [ fig_rhob_plane ] show the energy and baryon number densities obtained in one single central ( @xmath27 fm ) pb+pb collision at @xmath28 gev after the initialisation of the hydrodynamic fields .
the starting time is in this case @xmath31 fm and the densities in the figures correspond to the same time .
all quantities are given in the local rest frame .
the maximum values reach 6 gev/@xmath32 for the energy density and around 8 times the nuclear ground state density for the baryon number density .
the distributions are quite smooth which is necessary to provide possible initial conditions for the hydrodynamic evolution .
one can see some peaks that correspond to local maxima of the distributions ( `` hot spots '' ) .
it is further clear that the single event distributions are not symmetric , neither in the transverse nor the longitudinal direction .
the remaining question is if the system is thermalized enough to assure that the local equilibrium assumption of ideal hydrodynamics is fulfilled . in our case
, the hydrodynamic code transforms all the given quantities from the computational frame to the local rest frame of the energy momentum tensor which is also known as the landau frame .
this frame coincides in ideal hydrodynamics with the eckart frame which is defined as the local rest frame of the baryon number current .
the iterative calculation of the cell velocity succeeds if those two frames are close enough to each other . by this transformation
the system is forced to local equilibrium .
ideal relativistic one fluid dynamics is based on the conservation of energy , momentum and the net baryon number current . for the hydrodynamical evolution local equilibrium
is assumed and zero viscosity which corresponds to zero mean free path .
the two conservation equations that govern the evolution are @xcite @xmath33 where @xmath34 is the energy - momentum tensor and @xmath35 is the baryon current . for an ideal fluid the energy - momentum tensor and the net baryon number current take the simple form @xmath36 where @xmath37 and @xmath38 are the local rest frame energy density , pressure and net baryon density , respectively .
@xmath39 is the four velocity of the cell and @xmath40 is the metric tensor . the local rest frame is defined as the frame where @xmath34 has diagonal form , ( i.e. all off - diagonal elements vanish ) .
the four - velocity of the cells is calculated via the transformation into the local rest frame .
the equations of motion are solved in the following form by employing computational frame quantities @xmath41 and @xmath42 for the energy , momentum and net baryon number densities .
@xmath43 in our case , the full ( 3 + 1 ) dimensional hydrodynamic evolution is performed using the shasta algorithm @xcite .
the partial differential equations are solved on a three - dimensional spatial eulerian grid with fixed position and size in the computational frame .
the standard size of the grid is 200 cells in each direction while the cell size has been chosen to be @xmath44 which leads to timesteps of @xmath45 fm .
depending on the beam energy , the cell sizes may require adjustment to assure a stable solution of the differential equation .
the equation of state is needed as an additional input to calculate the pressure , temperature and chemical potential corresponding to the energy and the baryon number densities .
since the evolution of the system is driven by pressure gradients the eos has the most important influence on the evolution . to solve the hydrodynamical equations , the eos , the pressure as a function of energy and net - baryon number density , is needed as an input . since the actual eos of hot and dense qcd matter is still
not precisely known , it may seem disadvantageous to have this additional uncertainty in the model . on the contrary it may prove to be an important trait of the model to be able to study changes on the dynamics of the bulk matter when changing the eos thus finding observables for a phase transition in hot qcd matter . for recent discussions of different eos and how to obtain eos from lattice calculations ,
the reader is referred to @xcite .
+ the eos used in the present calculations is a grand canonical description of a free , non interacting gas of hadrons
. we will refer to it as the hadron gas ( hg ) .
it follows from the hadronic chiral model presented in @xcite .
the chiral hadronic @xmath46 lagrangian incorporates the complete set of baryons from the lowest flavour-@xmath46 octet , as well as the entire multiplets of scalar , pseudo - scalar , vector and axial - vector mesons @xcite . in mean - field approximation ,
the expectation values of the scalar fields relevant for symmetric nuclear matter correspond to the non - strange and strange chiral quark condensates , namely the @xmath47 and its @xmath48 counterpart @xmath49 , respectively , and further the @xmath50 and @xmath51 vector meson fields .
another scalar iso - scalar field , the dilaton @xmath52 , is introduced to model the qcd scale anomaly . however
, if @xmath52 does not couple strongly to baryonic degrees of freedom then it remains essentially `` frozen '' below the chiral transition @xcite .
consequently , we focus here on the role of the quark condensates . interactions between baryons and scalar ( bm ) or vector ( bv ) mesons , respectively , are introduced as @xmath53 here , @xmath5 sums over the baryon octet ( @xmath54 , @xmath55 , @xmath56 , @xmath57 ) .
a term @xmath58 with mass terms and quartic self - interaction of the vector mesons is also added : @xmath59 the scalar self - interactions are @xmath60 interactions between the scalar mesons induce the spontaneous breaking of chiral symmetry ( first line ) and the scale breaking via the dilaton field @xmath52 ( last two terms ) .
non - zero current quark masses break chiral symmetry explicitly in qcd . in the effective lagrangian
this corresponds to terms such as @xmath61~.\end{aligned}\ ] ] according to @xmath62 ( [ l_bm+v ] ) , the effective masses of the baryons , @xmath63 , are generated through their coupling to the chiral condensates , which attain non - zero vacuum expectation values due to their self - interactions @xcite in @xmath64 ( [ l_0 ] ) .
the effective masses of the mesons are obtained as the second derivatives of the mesonic potential @xmath65 about its minimum .
+ + all parameters of the chiral model discussed so far are fixed by either symmetry relations , hadronic vacuum observables or nuclear matter saturation properties ( for details see @xcite ) .
in addition , the model also provides a satisfactory description of realistic ( finite - size and isospin asymmetric ) nuclei and of neutron stars @xcite . if the baryonic degrees of freedom are restricted to the members of the lowest lying octet , the model exhibits a smooth decrease of the chiral condensates ( crossover ) for both high @xmath66 and high @xmath1 @xcite . however , additional baryonic degrees of freedom changes this into a first - order phase transition in certain regimes of the @xmath66-@xmath1 plane , depending on the couplings @xcite .
+ + in what follows , the meson fields are replaced by their ( classical ) expectation values , which corresponds to neglecting quantum and thermal fluctuations .
fermions have to be integrated out to one - loop .
the grand canonical potential can then be written as @xmath67}\right ) } \right ] \nonumber \\ & & { } + t \sum_{l\in m } \frac{\gamma_l}{(2 \pi)^3 } \int d^3k \left[\ln{\left(1 - e^{-\frac 1t[e^{\ast}_l(k)-\mu^{\ast}_l]}\right ) } \right ] , \nonumber\end{aligned}\ ] ] where @xmath68 denote the baryonic and mesonic spin - isospin degeneracy factors and @xmath69 are the corresponding single particle energies .
the effective baryon - chemical potentials are @xmath70 , with @xmath71 . here
@xmath72 is the quark , and @xmath73 the strange quark , chemical potential .
the potentials of the mesons are given by the sum of the corresponding quark and anti - quark chemical potentials .
the vacuum energy @xmath74 ( the potential at @xmath75 ) has been subtracted . by extremizing @xmath76 one obtains self - consistent gap equations for the meson fields . here ,
globally non - strange matter is considered and @xmath77 for any given @xmath66 and @xmath1 is adjusted to obtain a vanishing net strangeness .
+ + once the grand canonic potential is known as a function of @xmath66 and @xmath1 , all other thermodynamic quantities are derived straightforward . in its minima
the grand canonic potential corresponds to @xmath78 , the pressure .
the entropy density @xmath79 , number density @xmath80 and energy density @xmath81 then follow from the euler relation , @xmath82 where the sum runs over all included hadron species . + + setting all hadron masses and chemical potentials to their vacuum values , and adding all reliably known heavy resonance states - with masses up to 2 gev @xcite - as free particles into ( [ thermpot ] ) , yields the above mentioned hadron gas eos @xcite . hence , the hadronic degrees of freedom included in this eos are consistent with the active degrees of freedom in the urqmd model .
this enables us to directly compare the dynamics of the hydrodynamic model with the transport simulation .
+ + using the chiral model and adding additional baryonic degrees of freedom as well as adjusting their scalar and vector coupling , an eos with a phase structure including a first order phase transition and even a critical endpoint at finite @xmath1 can be obtained @xcite .
this chiral eos has already been successfully applied to a hydrodynamic calculation @xcite . here
the essentially different equation of state leads to distinguishable different results on the properties of bulk matter .
+ to emphasize these differences even more , a mit - bag model eos matched to an interacting hadron gas @xcite , generating a phase structure with a broad first order phase transition at all @xmath1 , can also be applied in our model . + consequently these results will be compared to our present calculations , constituting observables for a phase transition in hot and dense qcd matter , even suggesting the order of this phase transition .
comparisons between the different eos ( i.e. free hadron gas , chiral eos , and bag model eos ) will be presented in a follow - up paper .
presently , the hydrodynamic evolution is stopped , if the energy density drops below five times the nuclear ground state energy density ( i.e. @xmath83 mev/@xmath84 ) in all cells .
this criterium corresponds to a t-@xmath1-configuration where the phase transition is expected ( see dotted line in fig .
[ fig_tmu_width ] ) , i.e. a region where the hydrodynamic and the transport description are valid at the same time .
the hydrodynamic fields are mapped to hadrons according to the cooper - frye equation @xcite @xmath85 where @xmath86 are the boosted fermi or bose distributions corresponding to the respective particle species .
since we are dealing with an isochronous freeze - out , the normal vector on the hypersurface is @xmath87 .
let us note that it is of utmost importance to consider the same degrees of freedom on both sides of the hypersurface because otherwise energy and momentum conservation is violated . in our case ,
this is assured by the inclusion of the same particle species in the equation of state for the hydrodynamic calculation as in the transport calculation . in principle
, it might also happen that particles are moving back into the hydrodynamic phase , however , the explosive character of heavy ion reactions , i.e. the rapid expansion flow suppresses the back - streaming effect .
therefore , this effect is negligible in our situation @xcite . the assumption of an isochronous freeze - out leads to fluctuations of the temperature and baryo - chemical potential distributions and not to single values for some of the thermodynamic quantities on the hypersurface . the quality of this approach and that the two ( hydrodynamics and transport ) prescriptions are valid through the applied range of switching temperatures is shown in the next section , where parameter tests and the time evolutions of the particle yields are explored in detail .
( color online ) distribution of the energy in the cells at freeze - out at @xmath28gev.,scaledwidth=50.0% ] fig .
[ fig_dedtempdmu ] shows the distribution of energy in the cells at freeze - out with respect to temperature and baryo - chemical potential at @xmath28gev .
we present here the energy distribution and not the number of cells because it is more interesting to know where the energy of the system sits than considering the many almost empty cells that do essentially not contribute to particle production . from fig .
[ fig_dedtempdmu ] one obtains the mean values of the distributions that are in line with results from statistical model fits .
the mean values of , e.g. , the temperature can be calculated as @xmath88 where @xmath89 are the cell indices and the sum runs over all cells of one event .
the net baryon number density @xmath90 has been used as a weighting factor .
( color online ) mean values of temperatures and baryo - chemical potentials at freeze - out for different beam energies are depicted as red circles ( starting in the lower right corner at @xmath91gev and going through 4,6,8,11,20,30,40 and 80 to @xmath92gev in the upper left ) .
the error bars indicate the width of the distribution .
the dotted line depicts the line of constant energy density ( @xmath93 ) that corresponds to our freeze - out criterium . for comparison the freeze - out line calculated by cleymans et al .
@xcite ( full line ) and results from dumitru et al .
@xcite(green open squares ) are shown.,scaledwidth=50.0% ] to illuminate this finding in more detail , fig .
[ fig_tmu_width ] shows the mean values of the temperature and baryo - chemical potential distributions at different energies in the @xmath94-plane for central ( @xmath27 fm ) au+au
/ pb+pb collisions .
also , the widths of the distributions are depicted as `` error '' bars .
[ fig_tmu_width ] shows that the present freeze - out distributions are similar to the parametrized curve for chemical freeze - out as calculated by cleymans et al . from statistical model
fits to final particle multiplicities .
the calculation by dumitru et al . shows mean values as well as widths of temperature and baryo - chemical potential distributions that have been obtained by statistical model fits to final particle yields employing the assumptions of an inhomogeneous freeze - out hypersurface .
this calculation also leads to similar values as our calculation .
the effect that the mean temperature at the transition to the transport prescription saturates or even drops down a little at higher beam energies is related to the rapidity distribution of the temperature in the hydrodynamic cells at freeze - out which is shown in fig .
[ fig_temprap ] for three different beam energies . at low beam energies
the midrapidity region coincides with the hottest region at freeze - out . at higher sps energies
the situation changes .
the hottest cells are at high rapidities while the midrapidity region has already cooled down well below the temperature of 170 mev .
this problem might be resolved by a different freeze - out prescription on another hypersurface ( e.g.,isotherm , iso-@xmath81 ) and is subject to future investigations .
the best solution will be the dynamical coupling between hydrodynamics and transport which allows also for back - streaming contributions .
( color online ) rapidity profile of the freeze - out temperatures in the spatial plane with @xmath95 fm for central au+au / pb+pb collisions at three different beam energies ( @xmath96 and @xmath97gev).,scaledwidth=50.0% ] in the following the practical implementation will be explained in more detail .
the implementation is based on a monte carlo sampling of eqn .
[ cooper_frye ] and follows the general steps : 1 .
the particle numbers @xmath98 are calculated according to the following formula , @xmath99 where the index @xmath5 runs over the different particle species like , e.g. , @xmath0 , p , @xmath100 or @xmath101 .
@xmath102 is the boost factor between the computational frame and the cell .
@xmath103 is the volume of the cell in the computational frame and @xmath80 is the particle number density .
all cells with temperatures that are lower than 3 mev are discarded from the following procedure because of numerical reasons .
the local rest frame equilibrium distribution function is denoted by @xmath104 . to simplify the calculation ,
a relativistic boltzmann distribution is used for all particles , except pions .
it has been checked , that the boltzmann approximation is sufficient to describe all particle species it is applied to . for the boltzmann distribution
the momentum integration leads to the following result for the particle number density + @xmath105 + where @xmath106 is the degeneracy factor for the respective particle species , @xmath107 is the mass of the particle to be produced , @xmath66 the temperature of the cell and @xmath108 is the modified bessel function .
the chemical potential @xmath109 includes the baryo - chemical potential and the strangeness chemical potential in the following way + @xmath110 + where @xmath111 is the quantum number for strangeness and @xmath112 is the baryon number .
+ for pions the bose distribution has to be taken into account because the pion mass is on the order of the temperature of the system . in this case
, the momentum integration involves an infinite sum over modified bessel functions + @xmath113 + to calculate the number of particles in the computational frame the particle number density has to be multiplied with the lorentz - stretched volume of the cell ( @xmath114 ) .
the average total number of particles in the cell , @xmath115 , is the sum over all particle numbers @xmath116 + @xmath117 3 .
the total number of particles emitted from a cell , @xmath98 , is obtained from a poisson distribution according to @xmath118 .
+ in the limit of small mean values , the poisson distribution becomes @xmath119 .
thus it can be decided by one random number between 0 and 1 if a particle is produced in the respective cell .
if the random number is smaller than @xmath115 one particle is produced and there is no particle production otherwise .
the full poisson distribution is used , if the particle number @xmath115 is larger than ( @xmath120 ) .
this assures an accuracy better than 1 % .
the particle type is chosen according to the probabilities @xmath121 .
the @xmath122 component of the isospin is distributed randomly because urqmd assumes full isospin symmetry . to conserve the overall charge of the system and the initial isospin - asymmetry the probability to generate the isospin component that leads to the right value of the charge that should be obtained in the end
is favoured .
the other isospin components are exponentially suppressed .
the power of the exponential is proportional to the difference of the total charge generated by this produced particle and the required value .
+ ( color online ) pion and kaon multiplicities ( upper panel ) and mean transverse mass of pions and kaons at midrapidity ( @xmath123 ) ( lower panel ) for four different starting times for central ( @xmath124 fm )
au+au / pb+pb collisions .
the open symbols depict the result at @xmath125gev while the filled symbols are the results at @xmath28gev.,scaledwidth=50.0% ] 6 .
the 4-momenta of the particles are generated according to the cooper - frye equation ( see eqn .
[ cooper_frye ] ) . for baryons and strange mesons
the chemical potentials for baryon number and strangeness are taken into account .
the particle vector information is transferred back into the urqmd model .
the subsequent hadronic cascade calculation incorporates important final state effects as , e.g. , rescatterings of the particles and resonance decays .
( color online ) pion and kaon multiplicities ( upper panel ) and mean transverse mass of pions and kaons at midrapidity ( @xmath123 ) ( lower panel ) for four different freeze - out criteria for central ( @xmath124 fm )
au+au / pb+pb collisions .
the open symbols depict the result at @xmath125gev while the filled symbols are the results at @xmath28gev.,scaledwidth=50.0% ] the above mentioned steps are pursued on random cells until the initial net baryon number is reached .
strangeness and charge are also conserved in each event separately , energy conservation is fulfilled for the mean values averaged over several events .
aiming at a realistic description of heavy ion reactions we perform the freeze - out for each event separately and do not average over many freeze - outs for one hydrodynamical evolution .
in this section we investigate the dependences of observables on parameters of the implementation .
two important parameters have to be determined .
the first one is the starting time @xmath24 which defines the initial switch from urqmd to the hydrodynamic evolution .
the second parameter is the freeze - out criterium which is parametrized as an energy density criterium .
while varying one parameter we have fixed the other one to the default value ( @xmath126 or @xmath127 ) .
[ fig_intestmul ] ( top ) shows calculations of the total , i.e. pion and kaon ( @xmath128 ) multiplicities , for four different starting times at two beam energies .
the open symbols depict always the result at the highest ags energy ( @xmath125gev ) and the filled symbols are the results at the sps energy ( @xmath28gev ) . the starting time is varied in factors of the default value that has been calculated via eqn .
[ eqn_tstart ] . displayed
are results from halved to doubled initial time .
one observes a higher pion production for earlier starting times compared to the pion production in the standard setup ( 1 @xmath129 ) .
this may be explained by the fact that the system is forced more strongly to equilibrium and the cascade evolution lasts longer .
if the hydrodynamic evolution is started at later times ( 1.5 or 2 @xmath24 ) the resulting pion multiplicities are not affected anymore .
the kaon yield is essentially not sensitive to the switching time . to summarize , varying the starting time by a factor of 4 results in a change in the pion and kaon production of less than @xmath130 compared to the pion and kaon production in the default configuration ( 1 @xmath129 ) .
[ fig_intestmul ] ( bottom ) the mean transverse mass of pions and kaons at midrapidity is shown .
the mean transverse mass values are calculated for the same four different starting times at the two exemplary beam energies as before . here
the results do also not change more than @xmath131 for a large spectrum of starting times .
therefore , our choice of the starting time as the geometrical criterium when the nuclei have passed through each other is sensible and stable .
it is the earliest possible time where thermalization may have been achieved and the baryon currents have disconnected .
[ fig_frtestmul ] ( top ) is the equivalent picture to fig .
[ fig_intestmul ] ( top ) , however displaying the dependence of the total pion and kaon multiplicities on the freeze - out criterium .
the default value for the transition energy density is @xmath132 while we have varied it from @xmath133 .
the higher the freeze - out energy density the earlier the hydrodynamic evolution is stopped because the cells reach this critical energy density value earlier . as a consequence
the kaon yields raises with an increase of the energy density criterium , while the pions remain virtually unchanged for all investigated transition criteria .
[ fig_frtestmul ] ( bottom ) shows the results for the pions and kaons mean transverse mass for two beam energies and different freeze - out criteria .
again one observes only a very weak dependence on the freeze - out criterion .
furthermore , the mean transverse mass values for the two different meson species are very similar since they acquire the same transverse flow .
these findings confirm that our choice for the freeze - out criterium as @xmath132 is robust . in this section
the time scales that are important will be explored .
let us start with a table that summarizes the mean durations of the hydrodynamic and the hadronic phase of the collision for different starting times and freeze - out parameters at @xmath28gev .
.[tab_params40 ] this table contains the mean durations of the hydrodynamic evolution and the hadronic calculation afterwards for different starting times and freeze - out criteria for central ( @xmath124 fm ) pb+pb collisions at @xmath28gev . [ cols="^,^,^,^",options="header " , ] ( color online ) beam energy dependence of the starting time @xmath24 ( blue full line ) , the averaged time for the hydrodynamic evolution @xmath134 ( red dotted line ) and the mean duration of the hadronic stage @xmath135 ( green dashed line ) of central ( @xmath124 fm ) au+au
/ pb+pb collisions.,scaledwidth=50.0% ] the duration of the hydrodynamic evolution is a well - defined period for each event because of the isochronous freeze - out .
the average is therefore an average over 100 events .
the hadronic stage starts when the hydrodynamic evolution is stopped and it ends when the particles have undergone their last interaction .
an interaction can be an inelastic or an elastic collision or a decay .
the averaged time duration of the stages of the reaction ( given in table [ tab_params40 ] ) reflect the expectations . the lower the freeze - out energy density the later the hydrodynamic freeze - out proceeds and therefore the hydrodynamic evolution lasts longer while the hadronic stage is shortened .
the later the hydrodynamic evolution starts , the bigger @xmath24 is , the shorter the hydrodynamic evolution lasts .
the hadronic phase does not show a clear trend .
to first approximation the final urqmd stage lasts for @xmath136 fm independent of the parameters .
[ fig_timescales ] shows the beam energy dependences of the timescales for the chosen values of @xmath24 and the freeze - out energy density , from @xmath137 gev .
the starting time decreases as a function of beam energy from more than 10 fm at low energies to less than 2 fm at higher sps energies .
the mean duration of the hydrodynamic as well as the hadronic phase of the reaction grow with raising energy .
( color online ) time evolution of the net baryon number density in the local rest frame for central ( @xmath27 fm ) pb+pb collisions at @xmath138gev.,scaledwidth=50.0% ] in the following section we investigate the time evolution of different quantities and compare the results of the hybrid model calculation to the urqmd calculation without an hydrodynamic stage .
since the net baryon density is directly propagated on the hydrodynamic grid , it serves as a good example to compare to the default urqmd calculation . in the microscopic approach
the local rest frame density is calculated as the zero component of the net baryon number current in the frame where the corresponding local velocity vanishes @xcite .
( color online ) time evolution of the total particle number and the midrapidity ( @xmath123 ) yield ( upper panel ) , of the total number of pions and nucleons ( middle panel ) and of the conserved quantities ( lower panel ) for central ( @xmath27 fm ) pb+pb collisions at @xmath138gev .
results of the hybrid model calculation urqmd+hydro ( hg ) are depicted with symbols while urqmd-2.3 results are represented by lines .
the total energy of the system ( red circles and line ) has been divided by eight for visibility reasons .
the other conserved quantum numbers are net baryon number ( orange triangles and line ) , the overall charge ( blue squares and line ) and the strangeness ( green diamonds and line ) .
the total strangeness ( black dots and line ) is given by the sum of s- and @xmath139-quarks.,scaledwidth=50.0% ] fig .
[ fig_rhobtime ] shows the time evolution of the net baryon number density at the center of a central pb+pb collision at @xmath138gev .
the blue full line depicts the default urqmd calculation while the red full line depicts the result of the hybrid model calculation .
the result has some spikes because here we compare single events .
there are two important observations : the absolute values of the net baryon number densities are very similar in both cases and there are no obvious discontinuities at the switching points to and from the hydrodynamic model calculation .
the smoothness of the curve confirms our choice of parameters . in fig .
[ fig_noptime ] ( top and middle ) the time evolution of the particle yields in the two different models for the most central pb+pb collisions at @xmath28gev is compared .
the multiplicities at different timesteps are extracted from the hydrodynamic evolution by converting the number densities to particle numbers via the freeze - out procedure ( see section [ freeze - out ] ) .
[ fig_noptime ] ( top ) depicts the total particle multiplicity ( red circles and full line ) and the midrapidity multiplicity ( blue squares and full line ) .
the full lines indicate the default urqmd calculation , while the symbols show the results of the hybrid model .
the multiplicities increase rapidly in the initial 3 fm / c and then decrease a little , followed by a slower constant raise until the final multiplicity is reached .
this qualitative behaviour is very similar in both approaches .
the decrease of the multiplicity can be associated with the thermalization because absorption and production processes are on the same order .
note again that there are no discontinuities at the switching times in the hybrid calculation .
next , we explore the time evolution for two particle species in more detail . in fig [ fig_noptime ] ( middle ) , the pions ( red circles and full line ) represent the newly produced particles while the nucleons ( blue squares and full line ) are already there in the beginning as the constituents of the two incoming nuclei .
the qualitative behaviour of the temporal evolution of the pion yield is similar to that discussed above for the total multiplicity .
the decrease at the starting time of the hydrodynamic evolution @xmath140 fm / c is much stronger than in the model without hydrodynamic phase , because of the instant thermalization at the transition time .
the default urqmd transport calculation results in a similar , but much smoother , shape of the curve .
this similarity hints to the fact that the microscopic calculation also equilibrates the hot and dense matter to a rather large degree .
the number of nucleons decreases in the beginning due to the production of resonances and string excitations . at the thermalization
the minimum is reached and the number of nucleons increases slowly until the final value is reached .
in this case , not only the qualitative behaviour is independent of the underlying dynamics but also the absolute values are very close to each other . to check quantum number and energy conservation the temporal evolution of all the important quantities in both approaches for central pb+pb collisions at @xmath141gev
is depicted in fig .
[ fig_noptime ] ( bottom ) .
the default urqmd calculations are indicated by full lines while the hybrid model calculations with integrated hydrodynamic evolution are depicted by symbols . the net baryon number ( orange triangles and full line ) , the charge ( blue squares and full line ) and the net strangeness ( green diamonds and full line ) are exactly conserved in both approaches .
the total energy ( red circles and full line ) is only on average over several events conserved in the hybrid model calculation due to the freeze - out prescription .
but the fluctuations are on a @xmath142 level .
note however , that the total strangeness in the system ( @xmath143 quarks ) is very different in both approaches . in the default transport calculation ( black line )
the total strangeness increases in the early stage of the collision and remains constant .
this is contrasted by the hybrid calculation ( dots ) . due to the local thermal equilibration and the thermal production of strange particles in the hybrid calculation
the yield of strange quarks jumps to a higher value at the switching time ( @xmath24 ) .
the total strangeness then decreases as the system cools down , but the final value remains @xmath144 higher than in the default transport calculation .
gev , the middle plot at @xmath28gev and the lowest plot at the highest sps energy ( @xmath92gev ) for central ( @xmath124 fm ) au+au
/ pb+pb collisions .
the full line refers to baryon - baryon collisions ( b+b ) , the dotted line to baryon - meson collisions ( b+m ) and the dashed line to meson - meson collisions ( m+m ) .
the grey shaded area depicts the averaged duration of the hydrodynamic evolution.,scaledwidth=50.0% ] values for the binary collisions in the hadronic cascade calculation after the hydrodynamic evolution .
the upper plot depicts the result at @xmath125gev , the middle plot at @xmath28gev and the lowest plot at the highest sps energy ( @xmath92gev ) for central ( @xmath124 fm ) au+au
/ pb+pb collisions .
the full line refers to baryon - baryon collisions ( b+b ) , the dotted line to baryon - meson collisions ( b+m ) and the dashed line to meson - meson collisions ( m+m).,scaledwidth=50.0% ] the last step is the analysis of the freeze - out process , i.e. how much hadronic interactions happen after the hydrodynamic evolution . for this purpose , we have calculated the number of collisions during the hadronic cascade calculation in dependence of time and @xmath145 of the elementary collisions .
the corresponding distributions are shown in figs .
[ fig_dncolldt ] and [ fig_dncolldsqrts ] , respectively , for three different beam energies ( 11 , 40 and 160@xmath146 gev , from top to bottom ) .
[ fig_dncolldt ] shows the collision rates for meson - meson , meson - baryon and baryon - baryon interactions .
the grey area indicates the average time span of the hydrodynamic phase .
one observes that substantial collision rates are present directly after the transition to urqmd .
the collision rates stay high for 5 fm / c , and after 30 - 40 fm / c the system is completely frozen out . only some resonance decays proceed for longer . according to the composition of the system the baryon - baryon and baryon - meson interactions dominate the lower beam energy result , while at the highest sps energy the meson - meson together with the meson - baryon interactions are the most abundant type of collision indicating the transition from baryon dominated to meson dominated systems .
note that the overlap of the hadronic interaction phase with the hydrodynamic evolution results from the fact that the duration of the different stages fluctuates in the present approach .
shown here is the average duration of the hydrodynamic evolution .
[ fig_dncolldsqrts ] shows the @xmath145 distribution of the elementary collision in the freeze - out process .
one nicely observes all the resonance peaks in the corresponding channels .
this figure suggests that the most abundant meson - baryon collisions are excitations of the @xmath101 resonance ( i.e. @xmath147 interactions ) since there is a sharp peak at the @xmath101 mass ( @xmath148 gev ) . for meson - meson reactions the @xmath100 and the @xmath50 peaks are clearly present
this result indicates that there is still resonance regeneration even at this late stage of the system s evolution .
we start with a comparison of the multiplicities and particle spectra in the two frameworks .
calculations with the embedded hydrodynamic evolution employing a hadron gas equation of state for the high density stage of the collisions are compared to the reference results of the default transport calculation ( urqmd-2.3 ) .
since both calculations use the same initial conditions and freeze - out prescription it allows to extract , which observables are sensitive to the change in the underlying dynamics , thus allowing to explore the effect of local equilibration , of viscosities and heat conductivity .
the hybrid model calculation is in the following always depicted as full lines while the default urqmd-2.3 calculations are depicted as dotted lines .
note that we do not tune any parameters for different energies or centralities for the hybrid model calculation .
the starting time for the hydrodynamic expansion is always calculated using eqn .
[ eqn_tstart ] and the fixed energy density criterium ( @xmath132 ) for the freeze - out ( as explained in section [ freeze - out ] ) is always employed . ) in au+au / pb+pb collisions from @xmath91gev to @xmath149 gev .
urqmd+hydro ( hg ) calculations are depicted with full lines , while urqmd-2.3 calculations are depicted with dotted lines . the corresponding data from different experiments @xcite are depicted with symbols.,scaledwidth=50.0% ] ) in au+au / pb+pb collisions from @xmath91gev to @xmath149 gev .
urqmd+hydro ( hg ) calculations are depicted with full lines , while urqmd-2.3 calculations are depicted with dotted lines . the corresponding data from different experiments @xcite are depicted with symbols.,scaledwidth=50.0% ] in fig .
[ fig_mulally ] the excitation functions of the total multiplicities are shown for central au+au / pb+pb collisions for @xmath3gev .
the present hybrid approach simulations have been restricted to this energy range because for calculations at higher energies some numerical subtleties have to be resolved , e.g. a dynamical grid size for the hydrodynamical evolution .
compared to the default simulation , the pion and proton multiplicities are decreased over the whole energy range in the hybrid model calculation due to the conservation of entropy in the ideal hydrodynamic evolution .
the non - equilibrium transport calculation produces entropy and therefore the yields of nonstrange particles are higher . the production of strange particles however , is enhanced due to the establishment of full local thermal equilibrium in the hybrid model calculation . since the abundance of strange particles is relatively small
they survive the interactions in the urqmd evolution that follows the hydrodynamic freeze - out almost without re - thermalization .
[ fig_mulmidy ] shows the midrapidity yields of protons , pions , @xmath55 s and kaons as a function of the beam energy for central au+au / pb+pb collisions at @xmath3gev . for pions , kaons and @xmath55 s the same trend as for the @xmath150 multiplicities
is observed .
there are less pions produced in the hybrid model calculation due to entropy conservation , but more strange particles because of the production according to the local thermal equilibrium distributions .
the proton yield at midrapidity is very similar in both calculations while there are less antiprotons produced in the hybrid calculation .
also for the strange antiparticles a reduction of the midrapidity yield in the hybrid calculation at higher sps energies can be seen . for central ( @xmath124 fm ) au+au
/ pb+pb collisions for @xmath151gev .
urqmd+hydro ( hg ) calculations are depicted with full lines , while urqmd-2.3 calculations are depicted with dotted lines . the corresponding data from different experiments @xcite are depicted with symbols.,scaledwidth=50.0% ] to explore the kinetics of the system in more detail , fig .
[ fig_dndypiminus ] shows the rapidity distribution for @xmath152 at three different energies ( @xmath153 and @xmath97gev ) .
the general shape of the distribution is very similar in both approaches and in line with the experimental data . at higher energies
even the absolute yields become very close to each other in both approaches . for central ( @xmath124 fm ) au+au
/ pb+pb collisions for @xmath151gev .
urqmd+hydro ( hg ) calculations are depicted with full lines , while urqmd-2.3 calculations are depicted with dotted lines . the corresponding data from different experiments @xcite are depicted with symbols.,scaledwidth=50.0% ] fig .
[ fig_dndykaplus ] shows the @xmath154 rapidity distributions . in this case , the yield is higher in the hybrid calculation as already discussed above and also the shape of the distribution fits very nicely to the experimental data at sps energies .
overall the rapidity distributions seem not to be too sensitive to the details of the dynamics for the hot and dense stage , but strangeness yields are influenced by the local equilibrium assumption .
it seems that the local equilibrium assumption provides similar strangeness enhancement as previous calculations including additional strong color fields @xcite .
it is remarkable how well the hybrid calculation matches the rapidity spectra at lower energies ( @xmath125gev ) , even though the transport calculation provides a slightly better description to the experimental data at this energy .
one might still conclude from the rapidity spectra that the local equilibrium is not a good assumption for ags energies . after the longitudinal dynamics which reflects more the stopping power in the initial state we turn now to the transverse dynamics of the system .
transverse spectra are a promising candidate to be sensitive to the change in the underlying dynamics because they emerge from the transverse expansion which is mostly dominated by the evolution in the hot and dense stage of the reaction . , @xmath154 and protons at midrapidity ( @xmath123 ) for central ( @xmath124 fm ) au+au collisions at @xmath125gev .
urqmd+hydro ( hg ) calculations are depicted with full lines , while urqmd-2.3 calculations are depicted with dotted lines .
the corresponding data from the e866 experiment @xcite are depicted with symbols.,scaledwidth=50.0% ] , @xmath154 and protons at midrapidity ( @xmath123 ) for central ( @xmath124 fm ) pb+pb collisions at @xmath28gev .
urqmd+hydro ( hg ) calculations are depicted with full lines , while urqmd-2.3 calculations are depicted with dotted lines .
the corresponding data from the na49 experiment @xcite are depicted with symbols.,scaledwidth=50.0% ] , @xmath154 and protons at midrapidity ( @xmath123 ) for central ( @xmath124 fm ) pb+pb collisions for @xmath92gev .
urqmd+hydro ( hg ) calculations are depicted with full lines , while urqmd-2.3 calculations are depicted with dotted lines .
the corresponding data from the na49 experiment @xcite are depicted with symbols.,scaledwidth=50.0% ] figs .
[ fig_dndmt11 ] , [ fig_dndmt40 ] and [ fig_dndmt160 ] display the transverse mass spectra for pions , protons and kaons at midrapidity for central au+au / pb+pb reactions at three different beam energies . at @xmath125gev ( fig .
[ fig_dndmt11 ] ) the differential transverse mass spectra are very similar for both calculations and are in line with the experimental data . at @xmath28gev ( fig .
[ fig_dndmt40 ] ) first differences become visible . most notably is the strong flow of protons in the hybrid approach , that results in an overestimate of protons at high transverse momenta . ) for central ( @xmath124 fm ) au+au
/ pb+pb collisions from @xmath3gev .
urqmd+hydro ( hg ) calculations are depicted with full lines , while urqmd-2.3 calculations are depicted with dotted lines . the corresponding data from different experiments @xcite are depicted with symbols.,scaledwidth=50.0% ] at the highest sps energy ( @xmath92gev , fig .
[ fig_dndmt160 ] ) all the transverse mass spectra are flatter in the hybrid approach .
the initial pressure gradients are higher in the hydrodynamic calculation due to the hadronic equation of state without phase transition .
therefore , the matter expands faster in the transverse plane and higher transverse masses are reached . at this energy either the introduction of a mixed phase ( first order phase transition ) or non - equilibrium
effects are necessary to explain the experimental data . ) as a function of the particle mass for pions , kaons , protons , @xmath55 , @xmath57 and @xmath155 in central ( @xmath124 fm ) pb+pb collisions at @xmath28gev .
urqmd+hydro ( hg ) calculations are depicted with blue squares , while urqmd-2.3 calculations are depicted with red circles.,scaledwidth=50.0% ] fig .
[ fig_mmt ] shows the mean transverse mass excitation function for pions .
it confirms the observations from the differential spectra . up
to @xmath156 gev beam energy the hybrid model calculation leads to similar results as the default urqmd calculation and is in line with the experimental data .
the mean value of the transverse mass of pions is proportional to the temperature of the system and very different in the two calculations at higher energies .
the urqmd approach shows a softening of the equation of state in the region where the phase transition is expected because of non - equilibrium effects , while the hadron gas hydrodynamic calculation continuously rises as a function of the energy .
this behaviour is well known , see , e.g. , @xcite . finally fig .
[ fig_mptmass ] shows the mean transverse momenta as a function of particle mass for @xmath157 and @xmath155 particles .
here we observe the behaviour known from previous hybrid studies , that with increased strangeness , baryons accumulate less flow than in a complete hydrodynamic approach .
this effect can be traced back to the small cross sections of multi strange baryons in the hadronic cascade , thus showing , that the freeze - out / decoupling process proceeds gradually .
we have presented the first fully integrated boltzmann+hydrodynamics approach to relativistic heavy ion reactions .
this hybrid approach is based on the ultra - relativistic quantum molecular dynamics ( urqmd ) transport approach with an intermediate hydrodynamical evolution for the hot and dense stage of the collision . the specific coupling procedure including the initial conditions and the freeze - out prescription have been explained .
the event - by - event character of the hybrid approach has been emphasized .
the present implementation allows to compare pure microscopic transport calculations with hydrodynamic calculations using exactly the same initial conditions and freeze - out procedure .
the parameter dependences of the model have been investigated and the time evolution of different quantities were explored .
these tests led to the conclusion that the choice of the starting time and the freeze - out criterium does generally alter the multiplicities and transverse mass spectra only on a @xmath158 level .
the time evolution has shown that there are no discontinuities at the switching times in the hybrid model calculation .
the importance of the final state interactions has been emphasized by demonstrating that there is still resonance regeneration after the hydrodynamic evolution .
the effects of the change in the underlying dynamics - ideal fluid dynamics vs. non - equilibrium transport theory - have been explored .
the final pion and proton multiplicities are lower in the hybrid model calculation due to the isentropic hydrodynamic expansion while the yields for strange particles are enhanced due to the local equilibrium in the hydrodynamic evolution .
the results of the different calculations for the mean transverse mass excitation function , rapidity and transverse mass spectra for different particle species at three different beam energies have been discussed in the context of the available data .
the transverse expansion of the system is much faster in the hybrid model calculation , especially at higher energies which leads to differences in the observables that are sensitive to the transverse dynamics .
this finding indicates qualitatively that `` new '' physical effects like , e.g. , non - equilibrium effects or a phase transition have to be taken into account .
forthcoming work will be devoted to the study of different equations of state and the effect of changes in the equation of state on observables .
also in progress are calculations within the hybrid model at higher beam energies ( rhic and lhc ) , but therefore specific numerical subtleties have to be resolved like , e.g. , a dynamical grid size for the hydrodynamical evolution
. we are grateful to the center for scientific computing ( csc ) at frankfurt for providing the computing resources .
the authors thank dirk rischke for providing the 1-fluid hydrodynamics code .
h. petersen gratefully acknowledges financial support by the deutsche telekom - stiftung and support from the helmholtz research school on quark matter studies .
this work was supported by gsi and bmbf .
j. w. harris and b. muller , ann .
nucl . part .
sci . * 46 * , 71 ( 1996 ) s. a. bass , m. gyulassy , h. stoecker and w. greiner , j. phys .
g * 25 * , r1 ( 1999 ) j. adams _ et al . _ [ star collaboration ] , nucl .
phys . a * 757 * , 102 ( 2005 ) b. b. back _ et al . _ , nucl . phys . a * 757 * , 28 ( 2005 ) i. arsene _ et al . _ [ brahms collaboration ] , nucl . phys . a * 757 * , 1 ( 2005 ) k. adcox _ et al . _ [ phenix collaboration ] , nucl . phys . a * 757 * , 184 ( 2005 ) c. alt _ et al . _ [ na49 collaboration ] , phys . rev .
c * 77 * , 024903 ( 2008 ) s. a. bass _ et al .
_ , prog . part
. nucl .
* 41 * , 255 ( 1998 ) [ prog .
phys . * 41 * , 225 ( 1998 ) ] m. bleicher _ et al .
_ , j. phys .
g * 25 * , 1859 ( 1999 ) d. molnar and p. huovinen , phys .
lett . * 94 * , 012302 ( 2005 ) z. xu and c. greiner , phys .
c * 71 * , 064901 ( 2005 ) z. w. lin , c. m. ko , b. a. li , b. zhang and s. pal , phys .
c * 72 * , 064901 ( 2005 ) g. burau , j. bleibel , c. fuchs , a. faessler , l. v. bravina and e. e. zabrodin , phys .
c * 71 * , 054905 ( 2005 ) h. w. barz and b. kampfer , nucl .
a * 683 * , 594 ( 2001 ) a. b. larionov , o. buss , k. gallmeister and u. mosel , phys .
c * 76 * , 044909 ( 2007 ) j. bleibel , g. burau , a. faessler and c. fuchs , phys .
c * 76 * , 024912 ( 2007 ) j. bleibel , g. burau and c. fuchs , phys .
b * 659 * , 520 ( 2008 ) b. andersson , g. gustafson and c. peterson , nucl . phys .
b * 135 * , 273 ( 1978 ) .
j. r. ellis and k. geiger , phys .
d * 52 * , 1500 ( 1995 ) t. s. biro , p. levai and j. zimanyi , phys .
c * 59 * , 1574 ( 1999 ) c. t. traxler , u. mosel and t. s. biro , phys . rev .
c * 59 * , 1620 ( 1999 ) m. hofmann , m. bleicher , s. scherer , l. neise , h. stoecker and w. greiner , phys .
b * 478 * , 161 ( 2000 ) t. hirano , phys .
c * 65 * , 011901 ( 2002 ) p. f. kolb and u. w. heinz , arxiv : nucl - th/0305084 . c. nonaka and s. a. bass , phys .
c * 75 * , 014902 ( 2007 ) p. f. kolb , p. huovinen , u. w. heinz and h. heiselberg , phys .
b * 500 * , 232 ( 2001 ) p. f. kolb , u. w. heinz , p. huovinen , k. j. eskola and k. tuominen , nucl .
a * 696 * , 197 ( 2001 ) p. huovinen , p. f. kolb , u. w. heinz , p. v. ruuskanen and s. a. voloshin , phys .
b * 503 * , 58 ( 2001 ) a. dumitru , s. a. bass , m. bleicher , h. stoecker and w. greiner , phys .
b * 460 * , 411 ( 1999 ) s. a. bass , a. dumitru , m. bleicher , l. bravina , e. zabrodin , h. stoecker and w. greiner , phys . rev .
c * 60 * , 021902 ( 1999 ) s. a. bass and a. dumitru , phys .
c * 61 * , 064909 ( 2000 ) s. soff , s. a. bass and a. dumitru , phys .
lett . * 86 * , 3981 ( 2001 ) s. soff , s. a. bass , d. h. hardtke and s. y. panitkin , phys .
lett . * 88 * , 072301 ( 2002 ) c. nonaka and s. a. bass , nucl .
a * 774 * , 873 ( 2006 ) j. steinheimer , m. bleicher , h. petersen , s. schramm , h. stocker and d. zschiesche , phys .
c * 77 * , 034901 ( 2008 ) s. paiva , y. hama and t. kodama , phys . rev .
c * 55 * , 1455 ( 1997 ) . c. e. aguiar
, y. hama , t. kodama and t. osada , nucl .
a * 698 * , 639 ( 2002 ) o. .
j. socolowski , f. grassi , y. hama and t. kodama , phys .
lett . * 93 * , 182301 ( 2004 ) f. grassi , y. hama , o. socolowski and t. kodama , j. phys .
g * 31 * , s1041 ( 2005 ) .
r. andrade , f. grassi , y. hama , t. kodama , o. .
j. socolowski and b. tavares , eur .
j. a * 29 * , 23 ( 2006 ) r. andrade , f. grassi , y. hama , t. kodama and o. .
j. socolowski , phys .
lett . * 97 * , 202302 ( 2006 ) c. e. aguiar , t. kodama , t. koide and y. hama , braz .
j. phys .
* 37 * , 95 ( 2007 ) . v. v. skokov and v. d. toneev , phys .
c * 73 * , 021902 ( 2006 ) d. teaney , j. lauret and e. v. shuryak , arxiv : nucl - th/0110037
. t. hirano , u. w. heinz , d. kharzeev , r. lacey and y. nara , phys .
b * 636 * , 299 ( 2006 ) t. hirano , u. w. heinz , d. kharzeev , r. lacey and y. nara , phys .
c * 77 * , 044909 ( 2008 ) c. anderlik _ et al .
_ , phys .
c * 59 * , 3309 ( 1999 ) v. k. magas _ et al .
_ , heavy ion phys . * 9 * , 193 ( 1999 ) k. a. bugaev and m. i. gorenstein , arxiv : nucl - th/9903072 . k. a. bugaev , phys .
lett . * 90 * , 252301 ( 2003 ) s. r. de groot , w. a. van leeuwen and c. g. van weert , _ amsterdam , netherlands : north - holland ( 1980 ) 417p _ l. d. landau , izv . akad .
nauk ser .
* 17 * , 51 ( 1953 ) .
b. andersson , g. gustafson and b. nilsson - almqvist , nucl .
b * 281 * , 289 ( 1987 ) .
b. nilsson - almqvist and e. stenlund , comput .
commun .
* 43 * , 387 ( 1987 ) .
t. sjostrand , comput .
commun .
* 82 * , 74 ( 1994 ) .
s. a. bass _ et al .
_ , phys .
lett . *
81 * , 4092 ( 1998 ) l. v. bravina _ et al .
_ , j. phys .
g * 25 * , 351 ( 1999 ) m. belkacem _ et al .
_ , phys . rev
. c * 58 * , 1727 ( 1998 ) l. v. bravina _ et al .
_ , phys .
c * 60 * , 024904 ( 1999 ) l. v. bravina _ et al .
_ , arxiv:0804.1484 [ hep - ph ] .
e. l. bratkovskaya _ et al .
_ , phys .
c * 69 * , 054907 ( 2004 ) h. petersen , m. bleicher , s. a. bass and h. stocker , arxiv:0805.0567 [ hep - ph ] . m. bleicher and h. stoecker , phys .
b * 526 * , 309 ( 2002 ) h. petersen , q. li , x. zhu and m. bleicher , phys .
c * 74 * , 064908 ( 2006 ) s. kniege _ et al . _ [ na49 collaboration ] , aip conf .
* 828 * , 473 ( 2006 ) q. li , m. bleicher and h. stocker , phys .
b * 659 * , 525 ( 2008 ) s. soff _ et al .
. lett .
b * 471 * , 89 ( 1999 ) d. molnar and m. gyulassy , phys .
lett . * 92 * , 052301 ( 2004 ) u. w. heinz and p. f. kolb , arxiv : hep - ph/0204061 . h. j. drescher and y. nara , phys .
c * 75 * , 034905 ( 2007 ) r. b. clare and d. strottman , phys .
* 141 * , 177 ( 1986 ) .
d. h. rischke , s. bernard and j. a. maruhn , nucl .
a * 595 * , 346 ( 1995 ) d. h. rischke , y. pursun and j. a. maruhn , nucl .
a * 595 * , 383 ( 1995 ) [ erratum - ibid .
a * 596 * , 717 ( 1996 ) ] m. bluhm , b. kampfer , r. schulze , d. seipt and u. heinz , phys . rev .
c * 76 * , 034901 ( 2007 ) k. redlich , b. friman and c. sasaki , j. phys .
g * 35 * , 044013 ( 2008 ) t. s. biro , p. levai , p. van and j. zimanyi , j. phys .
g * 32 * , s205 ( 2006 ) p. papazoglou , d. zschiesche , s. schramm , j. schaffner - bielich , h. stoecker and w. greiner , phys .
c * 59 * , 411 ( 1999 ) d. zschiesche , g. zeeb and s. schramm , j. phys .
g * 34 * , 1665 ( 2007 ) s. schramm , phys . lett .
b * 560 * , 164 ( 2003 ) s. schramm , phys .
c * 66 * , 064310 ( 2002 ) d. zschiesche , s. schramm , h. stoecker and w. greiner , phys .
c * 65 * , 064902 ( 2002 ) j. theis , g. graebner , g. buchwald , j. a. maruhn , w. greiner , h. stoecker and j. polonyi , phys .
d * 28 * , 2286 ( 1983 ) .
d. zschiesche , g. zeeb , s. schramm and h. stoecker , j. phys .
g * 31 * , 935 ( 2005 ) s. eidelman _ et al . _
[ particle data group ] , phys .
b * 592 * , 1 ( 2004 ) .
d. zschiesche , s. schramm , j. schaffner - bielich , h. stoecker and w. greiner , phys .
b * 547 * , 7 ( 2002 ) f. cooper and g. frye , phys .
d * 10 * , 186 ( 1974 ) . k. a. bugaev , phys
. rev .
c * 70 * , 034903 ( 2004 ) j. cleymans and k. redlich , phys .
lett . * 81 * , 5284 ( 1998 ) j. cleymans , h. oeschler , k. redlich and s. wheaton , j. phys .
g * 32 * , s165 ( 2006 ) a. dumitru , l. portugal and d. zschiesche , phys .
c * 73 * , 024902 ( 2006 ) s. vogel , h. petersen , j. aichelin and m. bleicher , arxiv:0710.4463 [ hep - ph ] .
j. l. klay _ et al . _
[ e-0895 collaboration ] , phys .
c * 68 * , 054905 ( 2003 ) c. pinkenburg _ et al .
_ [ e895 collaboration ] , nucl .
a * 698 * , 495 ( 2002 ) p. chung _ et al . _
[ e895 collaboration ] , phys .
lett . * 91 * , 202301 ( 2003 ) s. v. afanasiev _ et al . _
[ the na49 collaboration ] , phys .
c * 66 * , 054902 ( 2002 ) m. bleicher , w. greiner , h. stoecker and n. xu , phys .
c * 62 * , 061901 ( 2000 ) t. anticic _ et al .
_ [ na49 collaboration ] , phys . rev .
lett . * 93 * , 022302 ( 2004 ) a. richard [ na49 collaboration ] , j. phys .
g * 31 * , s155 ( 2005 ) .
m. k. mitrovski _ et al . _
[ na49 collaboration ] , j. phys .
g * 32 * , s43 ( 2006 ) c. alt _ et al .
_ [ na49 collaboration ] , arxiv:0804.3770 [ nucl - ex ] . c. blume [ na49 collaboration ] , j. phys .
g * 31 * , s685 ( 2005 ) s. v. afanasiev _ et al .
_ [ na49 collaboration ] , phys .
b * 538 * , 275 ( 2002 ) l. ahle _ et al .
_ [ e866 collaboration ] , phys .
b * 476 * , 1 ( 2000 ) k. adcox _ et al .
_ [ phenix collaboration ] , phys .
c * 69 * , 024904 ( 2004 ) j. h. lee _ et al . _ [ brahms collaboration ] , j. phys .
g * 30 * , s85 ( 2004 ) .
l. ahle _ et al . _
[ e866 collaboration ] , phys .
b * 490 * , 53 ( 2000 ) d. ouerdane [ brahms collaboration ] , nucl .
a * 715 * , 478 ( 2003 ) s. ahmad _ et al . _ ,
b * 382 * , 35 ( 1996 ) .
a. mischke _ et al .
_ [ na49 collaboration ] , j. phys .
g * 28 * , 1761 ( 2002 ) j. adams _ et al . _ [ star collaboration ] , phys .
lett . * 92 * , 112301 ( 2004 ) y. akiba _ et al . _
[ e802 collaboration ] , nucl .
a * 610 * , 139c ( 1996 ) . c. alt _ et al . _
[ na49 collaboration ] , phys .
c * 73 * , 044910 ( 2006 ) .
h. sorge , phys .
b * 402 * , 251 ( 1997 ) | we present a coupled boltzmann and hydrodynamics approach to relativistic heavy ion reactions .
this hybrid approach is based on the ultra - relativistic quantum molecular dynamics ( urqmd ) transport approach with an intermediate hydrodynamical evolution for the hot and dense stage of the collision .
event - by - event fluctuations are directly taken into account via the non - equilibrium initial conditions generated by the initial collisions and string fragmentations in the microscopic urqmd model .
after a ( 3 + 1)-dimensional ideal hydrodynamic evolution , the hydrodynamical fields are mapped to hadrons via the cooper - frye equation and the subsequent hadronic cascade calculation within urqmd proceeds to incorporate the important final state effects for a realistic freeze - out .
this implementation allows to compare pure microscopic transport calculations with hydrodynamic calculations using exactly the same initial conditions and freeze - out procedure .
the effects of the change in the underlying dynamics - ideal fluid dynamics vs. non - equilibrium transport theory - will be explored .
the freeze - out and initial state parameter dependences are investigated for different observables .
furthermore , the time evolution of the baryon density and particle yields are discussed .
we find that the final pion and proton multiplicities are lower in the hybrid model calculation due to the isentropic hydrodynamic expansion while the yields for strange particles are enhanced due to the local equilibrium in the hydrodynamic evolution .
the results of the different calculations for the mean transverse mass excitation function , rapidity and transverse mass spectra for different particle species at three different beam energies are discussed in the context of the available data . |
ambipolar diffusion is the joint drift of charged particles and the associated magnetic flux with respect to the neutral particles in a partially ionized plasma . @xcite
first proposed it in order to explain the loss of magnetic flux from the dense cores of molecular clouds , required for the formation of stars , starting an active field of research in this area .
later , it was suggested to also play a role in the decay of the magnetic fields of neutron stars ( @xcite , hereafter gr-92 ) which became particularly relevant with the identification of `` magnetars '' , neutron stars whose main power source appears to be the dissipation of their magnetic field @xcite . in a previous paper ( @xcite ; hereafter paper i ) , we established a multifluid formalism in which it is possible to study the long - term evolution of magnetic fields in neutron stars ( see @xcite for a discussion of the main properties of the magnetic field equilibria and their subsequent long - term evolution ) . in that work , and following the ideas developed by gr-92 , we included the effects of several physical processes that are also relevant for star formation and protoplanetary disks , including _
ambipolar diffusion _ , _ hall drift _
( non - dissipative advection of the magnetic field by the associated electrical current ) , and _ ohmic diffusion _ ( dissipation of currents through the electrical resistivity ) .
here we continue this study and concentrate our analysis on the long - term evolution of the magnetic field caused by ambipolar diffusion aided by beta decays .
following the same philosophy of paper i , and as a first approach to the understanding of our general formalism , we focus on a simplified , one - dimensional configuration in which the magnetic field points in one cartesian direction @xmath0 but varies only along an orthogonal direction @xmath1 , i. e. , @xmath2 .
such models have also been considered in several studies of ambipolar diffusion in the interstellar medium @xcite , although some of the assumptions differ from case to case . in our analysis
, we consider separately two relevant limits , similar in spirit , though not exactly equivalent to those of @xcite : * in the _ weak coupling _
limit , there are few collisions between the particles , and the beta decays proceed slowly .
therefore , the particles can reach the diffusive equilibrium easily , but it takes much longer to reach the chemical equilibrium . * in the _ strong coupling _
limit , there are many collisions between the particles , and the beta decays proceed fast , so the ( local ) chemical equilibrium is reached much more quickly than the diffusive equilibrium . for each of these cases , we find that the long - term evolution of the magnetic field can be modelled by a single equation that gives the time - derivative @xmath3 at a given instant @xmath4 only in terms of the configuration of the magnetic field at the same instant , @xmath5 .
this makes it easy to carry out numerical simulations of the evolution of some selected non - linear magnetic field profiles , and even find some exact , analytical solutions . in sect .
[ 1dmodel ] of this paper , we briefly review the one - dimensional model of neutron star magnetic field evolution introduced in paper i , paying particular attention to its characteristic evolutionary time - scales . in sects .
[ asy1 ] and [ asy2 ] , we obtain the equations for the long - term , asymptotic magnetic field evolution promoted by ambipolar diffusion in each of the two opposite regimes mentioned above , and we make numerical simulations of the evolution of different initial magnetic field configurations .
we show that , in both cases , the magnetic flux of a given sign tends to spread out , but singularities develop at the null points where regions with different signs meet , as previously found by @xcite . in the weak coupling case , these singularities correspond to current sheets that are dissipated by resistive effects , in this way leading to reconnection . in the strong coupling case ,
the singularities have a somewhat different character ( a smoothly diverging current density ) and might lead directly to reconnection even in the case of no ohmic resistivity ( but see @xcite ) . finally , in sect . [ conc ] ,
we give the main conclusions of our study .
we model the neutron star interior as an electrically neutral and slightly ionized plasma composed of three mobile , strongly degenerate , particle species : neutrons @xmath6 , protons @xmath7 , and electrons @xmath8 .
we account for binary collisions and weak interactions ( causing nuclear beta decays ) between the particles , and allow for strong interactions between neutrons and protons by writing each of their chemical potentials as a function of both of their number densities : @xmath9 , while considering the electrons as an ideal , relativistic fermi gas , whose chemical potential is a function only of their number density , @xmath10 .
we study a one - dimensional geometry in which the magnetic field points in one cartesian direction @xmath0 , but varies only along an orthogonal direction @xmath1 as @xmath11 , and assume that all physical variables vary only along @xmath1 . since , in neutron star conditions , the ratio of the magnetic pressure @xmath12 to the pressure of the charged particles is very small , the magnetic force causes only small perturbations to the hydrostatic equilibrium state of the non - magnetized star
. these assumptions are generally not true in molecular cloud cores , where the ionization fraction tends to be extremely low , whereas the magnetic field can be near equipartition with the neutral gas pressure . in this sense
, our derivation will be valid only for the case of neutron stars , although we will see that some of the results agree with those of other authors , obtained under somewhat different assumptions .
a more general treatment appears to be difficult and not to yield simple results . for the reasons stated
, we consider a non - magnetized , fixed background system in hydrostatic and chemical equilibrium and introduce small perturbations to the number density of each species @xmath13 as @xmath14 , with the subscript zero labelling the background number densities and @xmath15 .
the associated chemical potential perturbations are given by @xmath16 .
the long - term magnetic field evolution implies small particle velocities that change over long time - scales , much longer than the very short dynamical times that are only relevant shortly after the formation of the star ( i.e , sound or alfvn time - scales ) .
therefore , at all times we use a slow - motion approximation in which we neglect the acceleration terms in the equations of motion for the particles .
taking account of all these considerations , the system of non - linear partial differential equations governing the evolution is ( see paper i for the derivation ) @xmath17
@xmath18 @xmath19 where @xmath20,\end{aligned}\ ] ] and @xmath21.\ ] ] the magnetic field evolution is given by eq .
( [ bz ] ) , where @xmath22 denotes the velocity of the charged particles ( the same for electrons and protons ) , @xmath23 is the velocity of the neutrons , and @xmath24 is the ambipolar diffusion velocity .- components of the respective velocities , which cause changes in the distributions of particles and magnetic flux .
the charged - particle velocities must also have a @xmath25-component , responsible for the current that acts as the source of the magnetic field . ] the first term on the right - hand side of this equation causes an advection of the magnetic flux with a velocity @xmath22 , while the second term describes the ohmic diffusion of the field , where @xmath26 is the electrical conductivity . in neutron star core conditions , the electrical conductivity is @xmath27 , thus , the evolution of the large - scale magnetic field through ohmic diffusion proceeds very slowly , with a time - scale @xmath28 , longer than the age of the universe @xcite .
qualitatively similar conditions hold in essentially all astrophysical settings .
therefore , in the rest of this paper , we neglect the ohmic term in the magnetic field evolution equation and focus only on the magnetic evolution due to the advective term .
however , the time - scale for ohmic diffusion scales with the square of the characteristic length of the magnetic field variations ( see paper i ) , thus , the influence of ohmic dissipation can be important in regions with strong spatial magnetic field variations .
the evolution of the particle number density perturbations , for the different species , is given by eqs .
( [ nb ] ) and ( [ nc ] ) , where @xmath29 is the perturbation of the charged particle number density keeping charge neutrality and @xmath30 is the perturbation of the baryon number density .
weak interactions cause beta decays ( conversion of charged particles into neutrons and the opposite ) that tend to reduce deviations from the _ chemical equilibrium _ state between charged particles and neutrons , with a net rate coefficient @xmath31 and @xmath32 .
the chemical equilibrium is achieved when @xmath33 .
continuing our description , eq .
( [ va ] ) shows that the ambipolar diffusion velocity @xmath34 is controlled by the collision rate between the charged particles and neutrons , which is proportional to the parametre @xmath35 .
this velocity is driven by the lorentz force but choked by the charged particle pressure gradient it produces @xcite . regarding the neutrons , we see from the right - hand side of eq .
( [ vn ] ) that they move with a velocity @xmath23 as long as the system is not in a state of _ magneto - hydrostatic equilibrium_. in this state , the lorentz force is balanced by the pressure gradients of all the particles ( charged particles and neutrons ) .
we also see that the neutron velocity is controlled by the parametre @xmath36 , whose meaning we explain in the next paragraph .
since we are interested in a numerical solution of our equations , we have to take into account that modelling the true dynamical time - scales ( sound or alfvn time - scales of milliseconds to seconds ) would require a time step many orders of magnitude shorter than that required to simulate the long - term evolution , making the simulation computationally unfeasible . for this reason , in paper i we introduced a _
slow - motion approximation _ , neglecting the acceleration terms and instead introducing a small , artificial , friction - like force acting on the neutrons ( the most abundant species ) of the form @xmath37 , where the parametre @xmath36 is chosen in such a way that the time to reach magneto - hydrostatic equilibrium is long enough for the numerical code to be able to deal with it ( and therefore , much longer than the real dynamical times ) , but shorter than the long time - scales over which the magnetic field evolves .
we showed in paper i that the latter time - scales , which are the astrophysically interesting ones , are unaffected by the choice of @xmath36 .
we require the conservation of both the magnetic flux @xmath38 and the baryon number perturbation @xmath39 during the evolution of our system , which spans the segment @xmath40 .
we ensure this through the boundary conditions @xmath41 @xmath42 @xmath43 the system of eqs .
( [ bz]-[nc ] ) describes the evolution of three coupled variables : the magnetic field , the charged particle density perturbation , and the baryon density perturbation . in paper
i , we estimated the three associated characteristic evolutionary time - scales of this set of equations corresponding to exponentially decaying eigenmodes in the linear approximation and showed that they characterize the approach to three successive equilibrium states . here
we briefly describe the evolutionary stages and summarize the relevant time - scales with the main goal of establishing the basic ideas that will be used in our subsequent analysis ( see paper i for more details about their derivation ) .
the shortest time corresponds to the approach of the magneto - hydrostatic equilibrium , controlled by @xmath36 in our model , as already explained . in a longer time , two alternative processes compete : 1 .
the particle species move relative to each other , controlled by the inter - particle collisions , in a tendency to reach a diffusive equilibrium state in which the fluid forces acting on each species separately are balanced .
the weak interactions convert particles from one species into another , tending towards a chemical equilibrium state . if the relative motion of charged and neutral particles proceeds much faster than the conversion from one into another , a diffusive equilibrium is achieved in the system , characterized by the balance equations @xmath44 and @xmath45 the inter - particle collision frequency controls the time - scale to reach the diffusive equilibrium , which is given by paper i , @xmath46 where @xmath47 and @xmath48 .
if the beta decays are much faster , the chemical equilibrium state @xmath49 is reached in a time - scale controlled by the beta decay rate , @xmath50 thus , we have two relevant limits : 1 . in the _ weak coupling limit _ , @xmath51 and the weak interactions operate slowly , therefore , the system relaxes to a diffusive equilibrium in a time - scale @xmath52 , but remains out of chemical equilibrium .
2 . in the opposite , _ strong coupling limit
_ , @xmath53 , the inter - particle collisions are very frequent which impedes a fast achievement of the diffusive equilibrium , while the system can relax to the chemical equilibrium in the time - scale @xmath54 .
the transition between these limits is achieved when @xmath55 .
from eqs .
( [ tnccgen ] ) and ( [ dmu1gen ] ) we infer that the condition for this transition is @xmath56 , which gives @xmath57 .
since the density perturbations are assumed to be small , these will not involve large motions of the particles , and therefore not cause a substantial change in the magnetic flux distribution .
note , however , that these two equilibria are incompatible with each other as long as a spatially non - uniform magnetic field is present , therefore a full equilibrium will only be reached in a much longer time - scale , on which the magnetic field is made uniform ( in our model ) or expelled from the system ( likely more realistic in a true astrophysical setting ) . for the determination of this much longer time - scale , on which the magnetic field does evolve substantially , we again consider the two opposite regimes discussed above . in the _ weak coupling limit _ , the system reached the diffusive equilibrium but not the chemical equilibrium , during the previous stage . during this much longer stage , the weak interactions slowly convert charged particles into neutrons in a tendency to reduce the charged - particle pressure gradient that counterbalances the magnetic pressure gradient .
this causes a slight deviation from the diffusive equilibrium , producing a joint transport of the charged particles and the magnetic flux at a small ambipolar diffusion velocity @xmath34 , always keeping the system very close to diffusive equilibrium .
this interplay continues until both the pressure and magnetic field gradients dissapear , which occurs in a long time - scale that depends on the magnetic field strength and the weak interaction rate : @xmath58 with @xmath59 , which is roughly the ratio of the charged particle pressure to the magnetic pressure and we defined @xmath60 . in the _ strong coupling limit _
, the relative motion of the charged particles and neutrons is strongly suppressed by the inter - particle collisions , which delay the diffusive equilibrium state , while in comparison the chemical equilibrium is reached quickly .
the deviation from diffusive equilibrium promotes , as before , a joint motion of the charged particles and the magnetic flux , with a very small ambipolar diffusion velocity @xmath34 .
this movement yields the diffusive equilibrium in a very lon time - scale controlled by the collision rate between particles and by the magnetic - field amplitude , @xmath61 note that at the transition between the strong and weak coupling limits ( @xmath56 ) , the ambipolar diffusion timescales are of the same order : @xmath62 this value corresponds to the shortest possible ambipolar diffusion time , since the relevant timescales increase both towards higher and lower temperatures . although the hall effect is not present in our one - dimensional calculations , it is important to asses its potential importance relative to ambipolar diffusion . consider its timescale , : @xmath63 compared to the minimum ambipolar diffusion timescale estimated above .
we find that , if @xmath64 , the hall drift is likely to play a dominant role .
it might not be important in magnetars , where @xmath65 , except possibly in reconnection layers , where @xmath66 .
in the present paper , we are interested in the details of the evolution of the magnetic field in the strong and weak coupling limits .
therefore , in the next sections we will obtain the differential equations modelling the evolution of the magnetic field in the time - scales given by eqs .
( [ tbp2 ] ) and ( [ tambi22p2 ] ) .
in this section , we derive a single integro - differential equation that models the magnetic field evolution in the _ weak coupling limit _ ( @xmath51 ) , in which neutral and charged particles drift easily with respect to each other , and the main bottleneck is the ( slow ) rate at which they can be converted into each other , in order to decrease the charged - particle pressure gradients that balance the lorentz force , impeding the magnetic flux to spread .
this limit is not likely to be relevant in the interstellar medium @xcite , although it roughly corresponds to one of the limits considered by @xcite . in neutron stars ,
it becomes important at later stages , once their temperature is low enough .
hereafter , we take the background properties to be homogeneous , i.e. the variables with a subscript @xmath67 do not depend on the position .
we assume that the system has already reached the diffusive equilibrium , which also implies the magneto - hydrostatic equilibrium . however , since weak interactions are slow , the system may be out of chemical equilibrium , namely , @xmath68 . from eq .
( [ neut - diff ] ) we infer a spatially uniform distribution of the neutrons , @xmath69 with @xmath70 an arbitrary function that depends only on the time variable . from eq .
( [ charg - diff ] ) , the diffusive equilibrium of the charged particles implies @xmath71 since @xmath72 and the time - scale for the evolution is @xmath73 , we can compare the terms in eq .
( [ nc ] ) ,
@xmath74 so we can neglect the time derivative and write this equation as @xmath75 integrating this equation between @xmath76 and an arbitrary internal point with coordinate @xmath1 , we obtain @xmath77dx^\prime,\ ] ] where we used the boundary condition @xmath78 . replacing eqs .
( [ diffusive - eq - neu ] ) and ( [ diffusive - eq - cha ] ) in eq .
( [ vcasyint1 ] ) and using the boundary condition @xmath79 to eliminate @xmath80 and @xmath70 , we obtain @xmath81 note that the velocity at any given point is proportional to the parametre @xmath31 controlling the weak interaction rate and to the difference of the _ average _ magnetic pressure to the left and to the right of this point . this is because , in diffusive equilibrium , a stronger magnetic pressure corresponds to a lower density of charged particles , and the magnetic flux can only spread as the reactions modify the charged - particle density .
replacing @xmath22 in eq .
( [ bz ] ) , we obtain the integro - differential equation governing the magnetic field evolution @xmath82.\end{aligned}\ ] ] the characteristic time - scale from this equation is the same as estimated in paper i. if we scale the position variable @xmath1 in eq .
( [ magn - equati1 ] ) to the total length of the system @xmath83 , the time variable to @xmath84 , and the magnetic field to some characteristic magnetic field ( for our subsequent numerical analysis it is chosen as the maximum of the initial magnetic field profile ) , we can write this equation with dimensionless variables as @xmath85.\ ] ] below , we solve eq .
( [ magn - equati1-nodime ] ) by using a numerical finite - difference scheme .
however , and with the goal of testing the adequate performance of the numerical method , it is desirable to know some analytical solutions to compare with the numerical results . following this philosophy , we try an analytical solution of eq .
( [ magn - equati1-nodime ] ) in the form of a `` step '' profile , motivated by the fact that we observed it as a generic asymptotic state in the evolution of some of the initial profiles that we study in the following paragraphs .
thus , we make the _ ansatz _ @xmath86 using the condition of flux conservation , @xmath87 , and replacing eq .
( [ explicit - step1 ] ) in the dimensionless version of eq .
( [ ambvelr1 ] ) evaluated at @xmath88 , we get @xmath89 integrating eq .
( [ diffasr1 ] ) yields @xmath90,\ ] ] where @xmath91 is some reference time . inverting this equation
gives @xmath92 , while @xmath93 can be obtained through the condition of flux conservation .
the step profile eq .
( [ explicit - step1 ] ) presents a sharp gradient around @xmath92 . since we are neglecting the ohmic dissipation , any attempt that we made to use this step profile as an initial condition in the finite - difference scheme results in a numerical instability . in order to overcome this difficulty , we compare in fig .
[ g1 ] the analytical evolution of eq .
( [ explicit - step1 ] ) with the finite - difference solution of eq .
( [ magn - equati1-nodime ] ) , using as initial condition a smooth profile , @xmath94,\ ] ] which is a good approximation to eq .
( [ explicit - step1 ] ) . for this comparison , the initial parametres @xmath95 and @xmath96
are chosen so that the step profile given by eq .
( [ explicit - step1 ] ) and the smooth profile given by eq .
( [ perta ] ) share the same initial magnetic flux .
we see from fig .
[ g1 ] that , except for the differences around @xmath92 , the evolution of both profiles is similar .
now , we intend to verify whether the asymptotic evolutionary equation eq . ( [ magn - equati1-nodime ] ) correctly describes the long - term behavior of the magnetic field .
this requires a comparison of the asymptotic evolution of the magnetic field as given by the full set of eqs .
( [ bz ] ) , ( [ nb ] ) , and ( [ nc ] ) with the corresponding evolution that arises when using only eq .
( [ magn - equati1-nodime ] ) . in fig .
[ g2 ] , we carry out this comparison for the evolution of a gaussian initial magnetic field profile given by @xmath97.\ ] ] as initial condition for the particle densities we set @xmath98 .
note that this comparison is valid at late times , when this profile has evolved to be consistent with the diffusive equilibrium .
so , we start this comparison from the instant labelled with number @xmath99 in this figure , which corresponds to @xmath100 .
we see that the asymptotic behavior of the magnetic field ( labels ( 3 ) and ( 4 ) in the figure ) calculated from these two methods is the same . with the purpose of characterizing this asymptotic behavior analytically
, we construct an analytic `` box''-type solution of the form @xmath101 with @xmath102 where the magnetic flux is @xmath103 , and @xmath104 is given by @xmath105.\ ] ] for the instants ( 3 ) and ( 4 ) in fig .
[ g2 ] , we show a full line that represents the solution given by eq .
( [ explicit - step ] ) .
we see that this analytical characterization works well . in fig .
[ g3 ] , we show the evolution of a harmonic initial magnetic field profile @xmath106 in this figure , we compare the evolution of this profile given by eq .
( [ magn - equati1-nodime ] ) with that given by the full system of coupled eqs .
( [ bz ] ) , ( [ nb ] ) , and ( [ nc ] ) . for the latter ,
the initial conditions for the particle density perturbations are @xmath107 . at the instant
labelled with number ( 2 ) , @xmath108 , this initial profile has evolved to be consistent with the diffusive equilibrium , therefore , from this instant we also calculate the evolution of the magnetic field through eq .
( [ magn - equati1-nodime ] ) .
we see in this figure that the asymptotic behavior of the magnetic field [ instant ( 4 ) ] calculated from these two methods is the same , which again verifies the validity of eq .
( [ magn - equati1-nodime ] ) to model the asymptotic evolution of the magnetic field . in summary ,
the last results confirm the adequacy of eq .
( [ magn - equati1-nodime ] ) to describe the asymptotic evolution of the magnetic field promoted by ambipolar diffusion but controlled by beta decays .
so , in what follows we explore the evolution of different magnetic profiles using only this equation .
this also has the advantage that we can increase the time - step with respect to that needed when solving the full set of eqs .
( [ bz ] ) , ( [ nb ] ) , and ( [ nc ] ) without generating numerical instabilities . in figs .
[ p1arbs1]-[p1arbs2 ] , we show the evolution of different initial profiles . from the numerical results , we can infer some generic properties : * the ambipolar diffusion process tends to spread out the magnetic flux and the total magnetic flux is conserved as it is expected from our boundary conditions . * contrary to the normal ohmic diffusion , in fig .
[ p1arbs2 ] it is observed that the magnetic field is not smoothed over all the space .
it becomes uniform only over regions whose boundaries are points where the magnetic field is zero , and across which it jumps between values of the same magnitude , but opposite signs , making the magnetic pressure @xmath109 equal on both sides .
* the magnetic field nulls move according to the local values of the charged particle velocity , as given by eq .
( [ ambvelr1 ] ) .
this velocity is a continuous function of @xmath1 , so the flux can be spread out or compressed , but no reconnection ( mutual elimination of opposite field lines ) occurs in the absence of ohmic dissipation , contrary to the strong coupling limit to be studied in the next section .
of course , the formation of steep gradients makes it possible for ohmic dissipation and therefore reconnection to occur in a realistic astrophysical setting .
in this section , we study the magnetic field evolution in the opposite limit of the last section , namely , the _ strong coupling limit _ , @xmath53 , in which the conversion from charged to neutral particles and vice - versa is essentially instantaneous , but their relative motion is impeded by a strong mutual collisional drag force .
this limit is relevant for young , hot neutron stars , as well as in the interstellar medium @xcite . in order to study the long - term evolution , we consider that the system has reached both the magneto - hydrostatic and chemical equilibria . from eq .
( [ vn ] ) , the magneto - hydrostatic equilibrium condition implies @xmath110 where @xmath111 is an arbitrary function depending only on the time variable . on the other hand
, the condition for chemical equilibrium implies @xmath112 combining eqs .
( [ equiltmhd ] ) and ( [ chemicalt ] ) , we obtain @xmath113.\ ] ] as done in eq .
( [ orderan ] ) , we can compare the time derivative to one of the spatial derivatives in eq .
( [ nb ] ) , @xmath114 thus , we can neglect the time derivative of @xmath115 in comparison with the spatial derivative term in eq .
( [ nb ] ) and write @xmath116 integrating eq .
( [ vnvx2 ] ) between @xmath76 and an inner point @xmath1 with the boundary conditions @xmath117 , and using eq .
( [ dncr2 ] ) , we obtain @xmath118 we see from eq .
( [ vasegr ] ) that the lorentz force drives the motion of magnetic flux and charged particles at the velocity @xmath22 , which is controlled by the inter - particle collisions through the factor @xmath119 . after including eq .
( [ vasegr ] ) in eq .
( [ bz ] ) , we obtain the equation for the magnetic field evolution as @xmath120\ ] ] scaling the time variable to @xmath121 and the position variable to the total length of the system @xmath83 , this equation can be written in dimensionless form as @xmath122.\ ] ] eq .
( [ magn - equatiadimensi2 ] ) belongs to a group of _ non - linear diffusion equations _ or _ porous medium equations _ whose mathematical properties have been studied by different authors ( see e.g. @xcite ) .
@xcite obtained it for ambipolar diffusion in interstellar gas , in a similar , strong - coupling regime as considered here , but ignoring the pressure of the charged particles , while allowing for arbitrarily large density perturbations , and using a lagrangian coordinate system moving with the neutral fluid .
it was re - derived under slightly different assumptions by @xcite .
as an analytical solution of eq .
( [ magn - equatiadimensi2 ] ) that will allow us to validate our numerical results , we follow @xcite and consider @xmath123 this solution is a semi - ellipse in the @xmath124 plane , centered at @xmath125 , with semi - axes @xmath126 and @xmath93 . replacing eq .
( [ expicit - nonlin - diff ] ) in eq .
( [ magn - equatiadimensi2 ] ) and using the flux conservation condition @xmath127 , we obtain the differential equation for @xmath126 , namely , @xmath128 with the solution @xmath129^{1/4},\ ] ] where @xmath130 is a reference time . in fig .
[ g5 ] , we compare this analytical solution to the corresponding evolution obtained from the finite - difference numerical integration of eq .
( [ magn - equatiadimensi2 ] ) .
we see that there is a good agreement between the two methods except at the last instant , labelled with @xmath131 in the figure .
this is expected , since at this instant the elliptic part of the profile has crossed the boundaries of the system , so the analytical solution is no longer valid .
note also that the latter does not conserve magnetic flux once the ends of the ellipse have reached the boundaries .
we also see at the instant labelled with @xmath131 how the magnetic field given by eq .
( [ magn - equatiadimensi2 ] ) is almost homogeneously distributed across the system , as expected from the magnetic - flux conserving boundary conditions that we are using for this equation .
thus , the finite - difference method that we are using to solve eq .
( [ magn - equatiadimensi2 ] ) appears to be working well .
hereafter , and with the main goal of establishing a comparison , we follow the evolution of the same initial profiles that we used in the previous section , where we dealt with the opposite regime . in fig .
[ g6 ] , we show the evolution of an initial gaussian magnetic field profile given by eq .
( [ nonlinerabp2 ] ) , as given by the full coupled system of eqs .
( [ bz ] ) , ( [ nb ] ) , and ( [ nc ] ) with @xmath107 . at the instant
( 2 ) , @xmath132 , the system should be very close to chemical equilibrium .
thus , from this instant onwards , eq .
( [ magn - equatiadimensi2 ] ) should be valid , so we also follow the evolution of the magnetic field obtained from the finite - difference solution of this equation .
we see that the asymptotic behavior of the magnetic field ( labels ( 2 ) and ( 3 ) in the figure ) calculated from these two models is the same .
in addition , for the instants ( 2 ) and ( 3 ) , we show for comparison the analytical solution [ eq .
( [ expicit - nonlin - diff ] ) ] .
we see that this explicit solution is an adequate description of the asymptotic evolution .
the previous results confirm the adequacy of eq .
( [ magn - equatiadimensi2 ] ) in describing the asymptotic evolution of the magnetic field promoted by ambipolar diffusion and controlled by inter - particle collisions .
we now examine the evolution of different magnetic profiles using only this equation . in figs .
[ g66]-[g70 ] we see the evolution of different magnetic field profiles . from these numerical results
, we can infer the following properties : * as in the weak coupling limit , the ambipolar diffusion process acts in a tendency to spread out the magnetic flux .
again , as expected , the total magnetic flux is conserved .
* at the null points , the magnetic field vanishes continuously , but with a high ( possibly infinite ) derivative , as expected from the example of the solution given by eq .
( [ expicit - nonlin - diff ] ) .
* contrary to the weak coupling limit , in fig .
[ g666 ] , it is observed that the magnetic flux is not preserved in each one of the regions separated by the magnetic null points .
therefore , there is a transfer of magnetic flux through these null points , leading to reconnection of magnetic field lines even in the absence of ohmic diffusion , as will be discussed further below . in order to understand the formation and behaviour of the singularities at the null points , we first consider a magnetic field with a finite derivative at a null point @xmath133 , and follow the evolution of its derivative @xmath134 using the _ ansatz _ @xmath135 the transport velocity , @xmath136 vanishes at the null point as long as @xmath137 is finite , so there is no reconnection at the null point in this regime . from eq .
( [ magn - equatiadimensi2 ] ) , the differential equation for @xmath134 is @xmath138 with the solution@xmath139 which diverges in a finite time @xmath140 .
@xcite found that this divergence ( which they identified numerically ) can lead to a stationary solution for the magnetic field .
we re - derive it by noting that , to have @xmath141 in eq .
( [ magn - equatiadimensi2 ] ) , the term in parenthesis on the right - hand side , which corresponds to ( minus ) the `` flux of flux '' ( amount of magnetic flux crossing any point @xmath1 per unit time ) , must be uniform in space , @xmath142 leading to @xmath143^{1/3}.\ ] ] the infinite derivative at @xmath133 allows @xmath144 to remain finite at this point , so there is magnetic flux crossing the singularity and causing reconnection , without having included ohmic diffusion in the model . of course
, this simple solution is not compatible with our boundary conditions , which were constructed so as to enforce @xmath145 at the boundaries .
therefore , for our numerical solutions , the reconnection must produce a decrease in the absolute value of the magnetic flux in each of the regions separated by the singularity , and therefore in the amplitude of the magnetic field . in order to explore this behaviour
, we will study the evolution of an initial magnetic field profile of the form @xmath146 this magnetic field profile has a null at @xmath147 and is an odd function respect to this point , a property that is preserved by the evolution according to eq .
( [ magn - equatiadimensi2 ] ) .
thus , in particular , the position of the null point will remain fixed . based on this fact
, we try a solution of eq .
( [ magn - equatiadimensi2 ] ) by separation of variables in the form @xmath148 with @xmath80 and @xmath149 satisfying @xmath150 and @xmath151=-\frac{c}{2}g,\ ] ] where @xmath152 is a separation constant . the magnetic flux between the left boundary of the system and the null point at @xmath147 is @xmath153 by symmetry
, the flux in the other half of the interval will be @xmath154 if we normalize , @xmath155 we will have @xmath156 . in order to solve eq .
( [ sep3 ] ) subject to the boundary conditions of eq .
( [ boundaries ] ) , we use the new variable @xmath157 , which must satisfy the conditions @xmath158 and @xmath159 . thus , eq . ( [ sep3 ] )
is rewritten as @xmath160 combining eq .
( [ sepu1 ] ) with eq .
( [ sep67 ] ) , we obtain the auxiliary condition @xmath161 , which indicates that the magnetic field close to the null point has the shape given by eq .
( [ asyu ] ) .
the numerical solution of eq .
( [ sepu1 ] ) must satisfy conditions at different points , which can be intricate from the numerical point of view .
thus , we rescale of our variables as @xmath162 and @xmath163 , where @xmath164 and @xmath165 are constants to be determined .
this allows us to write eq .
( [ sepu1 ] ) as @xmath166 which requires @xmath167 .
the boundary conditions can be set as @xmath168 , @xmath169 . from the numerical solution of eq .
( [ sepu2 ] ) we obtain @xmath170 at @xmath171 . since the null point of @xmath172 is at @xmath173 , we get @xmath174 . using @xmath162 and the conditions on the first derivatives , @xmath161 and @xmath175 ( obtained numerically )
, we get @xmath176 . finally , comparing eqs .
( [ sepu1 ] ) and ( [ sepu2 ] ) we obtain @xmath177 , which determines the full evolution of the magnetic flux in each half of the interval , which is given according to eq .
( [ sep2 ] ) as @xmath178 ^ 2 t}}~\underset{t \to \infty}{\to } \frac{1}{\sqrt{2ct}}.\ ] ] the flux calculated from the numerical solution of eq .
( [ magn - equatiadimensi2 ] ) agrees with eq .
( [ sep4 ] ) with a percentage error of order @xmath179 .
( [ sep4 ] ) implies that positive magnetic flux coming from the left of the singularity annhilates with the negative flux coming from the right .
this is reconnection in the absence of ohmic resistivity . in fig .
[ g70 ] we observe that the magnetic field as given by the numerical solution of eq .
( [ magn - equatiadimensi2 ] ) converges after a initial transient to the field of eq .
( [ sep1 ] ) ( obtained from the solution of eq .
( [ sepu1 ] ) .
this magnetic field profile has a strong divergence of the velocity field at the null point , and we are assuming that the excess charged particles instantaneously recombine when they reach this point , which is not realistic even in the `` strong coupling limit '' , in which @xmath180 .
in other words , even if the beta decays can be considered as instantaneous everywhere else in the system , this approximation will break down close enough to the singularity .
the behaviour close to the singularity , including both a fast , but finite recombination rate and a small , but finite ohmic diffusivity , was discussed in detail by @xcite .
we have studied the asymptotic magnetic field evolution promoted by ambipolar diffusion in a one - dimensional geometry for two opposite , limiting regimes . in the _ weak coupling limit _ , in which neutral and charged particles drift easily with respect to each other , the bottleneck for the evolution is the conversion of one species into another , which is required in order to eliminate the charged - particle pressure gradients caused by the magnetic field , which impede its evolution . in the _ strong coupling limit _ , conversions are easy , but the inter - particle collisions are the corresponding bottleneck . in molecular clouds , the second regime appears to be generally relevant @xcite .
neutron stars in their hot , early phase will also be in the strong - coupling regime , and evolve to the weak coupling regime as they cool . in the _ weak coupling limit _
, the magnetic field evolution is described by a non - linear , partial integro - differential equation , while in the _ strong coupling limit _ this evolution is described by a non - linear diffusion equation .
we made numerical simulations of the evolution of different initial magnetic field profiles in each of these limits and found agreement between our numerical results and some analytic solutions that can be found for these differential equations . from our results , we infer that , in both limits , the ambipolar diffusion process operates in a tendency to spread out the magnetic flux , but contrary to the normal ohmic diffusion this process asymptotically produces singular points with sharp magnetic field gradients .
these sharp gradients develop around those points where the magnetic field is null , and separate regions of magnetic fields with opposite signs .
we observe some generic properties of this process , as follows : in the _ weak coupling limit _ , the resulting discontinuities can be modelled as step solutions [ eqs .
( [ explicit - step ] ) , ( [ explicit - step1 ] ) ] .
the asymptotic magnetic field is spatially uniform in each of the regions separated by these singularities , and its absolute value ( and thus the magnetic pressure ) is the same in each region .
ambipolar diffusion by itself does not cause magnetic flux transfer ( and thus reconnection ) across the singularities , but the associated current sheets will easily be dissipated by ohmic diffusion , so reconnection will occur in a realistic system . in the _ strong coupling limit _
, at the singular points the magnetic field vanishes continuously but with infinite spatial derivative .
ambipolar diffusion acts in a tendency to spread out the magnetic flux , but , contrary to the weak coupling limit , the magnetic flux is not preserved in each one of the regions separated by the magnetic null points .
therefore , there is a transfer of magnetic flux through these null points , i. e. reconnection without ohmic resistivity ( but see @xcite ) .
the main limitation in applying the present formalism to realistic systems ( either neutron stars or molecular cloud cores ) is the very restrictive , one - dimensional geometry .
an extension to more realistic geometries ( i. e. , axial symmetry ) will be attempted in further work .
we are grateful to mara cristina depassier and cristbal petrovich for valuable information about the non - linear diffusion equation .
we also thank the referee for a very helpful report .
this work was financed by the gemini - conicyt fund , project @xmath181 32070014 ; fondecyt regular projects 1060644 and 1070854 , the fondap center for astrophysics ( 15010003 ) , proyecto basal pfb-06/2007 and the joint project `` estudio computacional del decaimiento de campos magnticos en estrellas de neutrones '' between universidad de medelln ( summa group ) , pontificia universidad catlica de chile and universidad de chile .
the postdoctoral stay of j.h.h . at astrophysikalisches
institut potsdam was possible due to the financial support of deutscher akademischer austauschdienst ( daad ) - germany and comisin nacional de investigacin cientifica y tecnolgica ( conicyt ) - chile through the postdoctoral fellowship @xmath181 a0772255 - 2007 - 07-docdaad-25 .
we also thank the fondecyt international cooperation project 7090020 .
99 aguilera d.n . , pons j.a . &
miralles j.a .
2008 , a&a , 486 , 255 akmal a. , pandharipande v.r . & ravenhall d.g .
1998 , phys . rev
. c , 58 , 1804 arras p. , cumming a. , & thompson c. , 2004 , apj , 608 , l49 baym g. , pethick c. , & pines d. 1969 , nat , 224 , 872 brandenburg a. , zweibel e.g , 1994 , a&a , 427 , 91 brandenburg a. , zweibel e.g , 1995 , apj , 427 , 91 duncan r.c . ,
thompson c. , 1992 , apj , 392 , l9 galli d. , & shu f.h .
, 1993 , apj , 417,220 goldreich p. , & reisenegger a. 1992 , apj , 395 , 250 haensel p. , urpin v.a . , & yakovlev d.g .
1990 , a&a , 229 , 133 harrison e. 1991 , mnras , 248 , 419 heitsch f. & zweibel e.g. 2003a , apj , 583 , 229 heitsch f. & zweibel e.g. 2003b , apj , 590 , 291 hoyos j. , reisenegger a. , & valdivia j.a .
2008 , a&a , 487 , 789 jones p.b .
1987 , mnras , 228 , 513 mestel l. 1966 , mnras , 133 , 265 mestel l. & spitzer l. 1956 , mnras , 116 , 503 mouschovias t. , 1991 , apj , 371 , 296 mouschovias t. ch . & paleologou e. v. 1981 , apj , 246 , 48 niebergal b. , ouyed r. , & leahy d. 2006 , apj , 646 , l17 pethick c.j . , 1992 , in structure and evolution of neutron stars , ed .
d.pines , r. tamagaki , & s.tsuruta ( redwood city : addison - wesley ) , 115 reisenegger a. , 2009 , a&a , 449 , 557 shu f. h. 1983 , apj , 273 , 202 spitzer l. 1978 , physical processes in the interstellar medium , willey , new york thompson c. , duncan r.c . , 1996 , apj , 473 , 322 tuck b. 1976 , j. phys .
phys . , 9 , 1559 vzquez j.l . , 2007 , the porous medium equation : mathematical theory , oxford mathematical monographs , oxford science publications , oxford university press inc . , new york zweibel e.g. 1989 , apj , 340 , 550 | we study the effect of the non - linear process of ambipolar diffusion ( joint transport of magnetic flux and charged particles relative to neutral particles ) on the long - term behavior of a non - uniform magnetic field in a one - dimensional geometry .
our main focus is the dissipation of magnetic energy inside neutron stars ( particularly magnetars ) , but our results have a wider application , particularly to the interstellar medium and the loss of magnetic flux from collapsing molecular cloud cores .
our system is a weakly ionized plasma in which neutral and charged particles can be converted into each other through nuclear beta decays ( or ionization - recombination processes ) . in the `` weak - coupling '' limit of infrequent inter - particle interactions ,
the evolution of the magnetic field is controlled by the beta decay rate and can be described by a non - linear partial integro - differential equation . in the opposite , `` strong - coupling '' regime , the evolution is controlled by the inter - particle collisions and can be modelled through a non - linear diffusion equation .
we show numerically that , in both regimes , ambipolar diffusion tends to spread out the magnetic flux , but , contrary to the normal ohmic diffusion , it produces sharp magnetic field gradients with associated current sheets around those regions where the magnetic field is weak .
diffusion plasmas magnetohydrodynamics ( mhd ) ism : magnetic fields stars : magnetic fields stars : neutron |
understanding the processes involved in the supernova explosions , in the creation of stellar compact objects and in their temporal evolution requires a huge multidisciplinary effort with investigations in areas as distinct as nuclear and particle physics , thermodynamics , quantum field theory and astrophysics . in the present work we concentrate on the description of neutron stars . from very low densities up to the high densities present in their core
, the constitution of these compact objects is a great source of speculation . at low densities
there can be neutrons , protons , electrons and possibly neutrinos ( at finite temperatures ) . at high densities ,
stellar matter can be much more complex , including hyperons , kaons and even deconfined quarks .
many works considering the construction of equations of state ( eos ) used to describe compact objects have already been done @xcite .
once a certain eos is obtained , it serves as input to the tolman - oppenheimer - volkoff equations ( tov ) @xcite and the output gives the structure of the compact stars , characterized by their mass and radius .
an appropriate eos or an inadequate one can only be chosen or ruled out once astronomical observations are used as constraints .
although some observational results are known , many uncertainties exist .
it is still unknown whether the neutron stars are composed only of hadrons and leptons , necessary to ensure chemical equilibrium and charge neutrality @xcite , if they are quark stars @xcite or even hybrid stars , containing both hadron and quark matter in their interior @xcite .
each one of these possibilities is represented by a great variety of relativistic and even non - relativistic models used to built the eos .
we next investigate hybrid stars only , whose existence is a source of intense discussions in the literature @xcite .
the discussion presented in @xcite is particularly interesting because the existence of quark stars is shown to be questionable within the calculations performed ( which depend strongly on a specific parametrization ) .
moreover , it is also pointed out that the possibility of a mixed population ( or hybrid stars ) is compatible with the calculations of model dependent quark matter nucleation , what reinforces the interest in the calculations of hybrid stars as compact objects .
recent calculations show the importance of the nucleation mechanism in the process of phase transition from hadronic to quark matter @xcite .
the main reason for the present work is the fact that many astrophysicists claim that the mixed phase is only a hypothetical choice and can not be checked .
moreover , some authors calculated macroscopic quantities as radii and masses for hybrid stars with and without the mixed phase and concluded that the differences were not significant @xcite or that the region corresponding to the hadron - quark mixed phase is too narrow @xcite .
although hybrid stars have been obtained with different combinations of models for the hadron and the quark phases , most of the discussions on the use of gibbs and maxwell constructions have been based on the mit bag model @xcite for the description of the quark phase .
the mit bag model @xcite is a very simple model that does not reproduce some of the necessary features of qcd at high densities , as chiral symmetry , for instance .
as it is easily checked on the literature , all results for compact stars are model dependent .
hence , before completely ruling out the need for the gibbs construction and the consequent existence of the mixed phase in hybrid stars , it is desirable that another calculation with a different model for the quark phase is considered .
that is the basis of the calculations and discussion that follows . in the present paper ,
the hadron phase is described by the non - linear walecka model ( nlw ) @xcite and the quark phase by the nambu - jona - lasinio model ( njl ) @xcite .
two different constructions are made : one with a mixed phase ( mp ) and another without the mixed phase , where the hadron and quark phases are in direct contact . in the first case ,
neutron and electron chemical potentials are continuous throughout the stellar matter , based on the standard thermodynamical rules for phase coexistence known as gibbs conditions . in the second case ,
the electron chemical potential suffers a discontinuity because only the neutron chemical potential is imposed to be continuous .
the condition underlying the fact that only a single chemical potential is common to both phases is known as maxwell construction . in our approach
we ignore surface and coulomb effects for the structure in the mixed phase so the leptons are taken into account as free fermi gases . however , it is worthy pointing out that the energy density in mixed phases should depend on the eletromagnetic and surface contributions and this is commonly known as finite size effects . in @xcite it was shown that for a weak surface tension the eos resembles the one obtained with a gibbs construction while for a strong surface tension , the maxwell construction was reproduced .
unfortunatelly , the surface energy coefficient is not well described yet @xcite .
the differences between stellar structures obtained with both constructions are discussed through out the present paper .
a similar calculation was done in @xcite , where the effects of different lepton fractions on protoneutron stars with trapped neutrinos were investigated .
although the result for zero temperature was also presented , its validity when trapped neutrinos are enforced is only academic because the neutrino mean free path at t=0 is larger than the neutron star radius .
while in @xcite no hyperons were included in the hadronic phase , they are also taken into account in the present paper for two parametrizations of the nlw model .
notice , however , that @xmath0 quarks were also considered in the quark phase described in @xcite . in works involving quark @xcite or hybrid stars @xcite
, it is seen that the njl model gives results that are quite different from the ones obtained with the mit model .
the fact that the njl model incorporates chiral symmetry and that the strange quark appears only in densities much higher than the @xmath1 and @xmath2 quarks are the main reasons for the differences .
hence , the calculations for the hybrid stars are here done with the njl model so the previous conclusions on the mixed phase are confirmed or refuted .
the consequences of the inclusion of the @xmath0-quark in the njl model at quite high densities is also seen once a comparison between the two versions of the njl model , i.e. , su(2 ) and su(3 ) is performed .
whenever the su(2 ) version of the njl is used to describe quark matter , the corresponding hadron phase is strangeness free , i.e. , no hyperons are considered .
two parameter sets are used for each case considered so that the model dependence can be established .
the paper is organized as follows : in sec .
2 we show the lagrangian densities of the models considered and describe the formalism used ; in sec .
3 we present and discuss the results ; in sec .
4 we draw our final conclusions .
we next give some of the main equations related to the two models used in our investigation .
detailed calculations are extensively available in the literature and hence are omitted in the present paper .
two possible systems are studied : one comprehends 8 baryons in the hadron phase and 3 quarks in the quark phase and the other includes only protons and neutrons in the hadron phase and the corresponding @xmath1 an @xmath2 quarks in the quark phase . in most cases ,
our studies refer to a hadron matter with all 8 baryons and a quark matter with the 3 possible quarks .
the njl model is defined by the lagrangian density @xmath3\ ] ] @xmath4 + det[\bar{q}_i(1-\gamma_5)q_j ] \ } , \label{lagnjl}\ ] ] where @xmath5 are the quark fields and @xmath6 @xmath7 , are the u(3 ) flavour matrices .
the model parameters are the current quark mass matrix @xmath8 , the coupling constants @xmath9 and @xmath10 , and the cutoff in 3-momentum space @xmath11 .
the thermodynamical potential density is given by @xmath12 where the energy density is @xmath13 @xmath14 in the above expressions , @xmath15 , @xmath16 , @xmath17 is the chemical potential ( number density ) of particles of type @xmath18 , and @xmath19 and @xmath20 are included in order to ensure @xmath21 in the vacuum .
the quark condensates and the quark densities are defined , for each of the flavors @xmath22 , respectively , as : @xmath23 @xmath24 minimizing the thermodynamical potential @xmath25 with respect to the constituent quark masses @xmath26 leads to three gap equations for the masses @xmath26
@xmath27 with cyclic permutations of @xmath28 . the pressure can be found from @xmath29 the relations between the chemical potentials of the different particles required by @xmath30- equilibrium are given by @xmath31 and for charge neutrality we must impose @xmath32 in order to obtain the njl su(2 ) model we just need to neglect the terms related to the strange quark in equations 2 , 4 and 5 .
it means that @xmath33 .
the parameter sets of the njl model used in the present work are given in table 1 .
the lagrangian density for the nlw model reads @xmath34
@xmath35\psi_b + \frac{1}{2}(\partial_{\mu}\phi\partial^{\mu}\phi -m_s^2 \phi^2 ) -\frac{1}{3!}\kappa
\phi ^{3 } -\frac{1}{4!}\lambda \phi ^{4}\ ] ] @xmath36 with @xmath37 and @xmath38 the hyperon coupling constants are defined as @xmath39 . in a mean field approximation
the energy density reads @xmath40 @xmath41 and the pressure becomes @xmath42 @xmath43 where @xmath44 is the spin degeneracy factor .
the conditions of chemical equilibrium are also imposed through the two independent chemical potentials @xmath45 and @xmath46 and it implies that @xmath47 @xmath48 @xmath49 for the charge neutrality , we must have @xmath50 where @xmath51 and @xmath52 stand , respectively , for the electric charges of baryons and leptons . when the system is constituted only of protons and neutrons , the sum on the baryons appearing in the equations above
are restricted to the nucleons and the only condition for chemical equilibrium is the last one in ( [ betahadron ] ) .
two sets of parameters were chosen and they are given in table 2 .
the following parameters are equal for both sets : @xmath53 , @xmath54 , @xmath55 mev , @xmath56 mev and @xmath57 mev . as we are dealing with neutral stellar matter in @xmath30- equilibrium in both quark and hadron phases according to eqs .
( [ betaquark])-([chargequark ] ) and ( [ betahadron])-([chargehadron ] ) respectively , the electrons and muons have to be introduced .
they are normally included as free fermi gases obeying the following lagrangian density : @xmath58 expressions for energy density and pressure in a mft become : @xmath59 and @xmath60 we next build a mixed phase ( mp ) constituted of hadrons and quarks , which interpolates between the hadron ( hp ) and the quark phase ( qp ) . in the mixed phase charge neutrality
is not imposed locally but only globally .
this means that quark and hadron phases are not neutral separately , but rather , the system prefers to rearrange itself so that @xmath61 where @xmath62 is the charge density of the phase @xmath18 , @xmath63 is the volume fraction occupied by the quark phase , and @xmath64 is the electric charge density of leptons .
according to the gibbs conditions for phase coexistence , the neutron chemical potentials , the electron chemical potentials and pressures have to be identical in both phases , i.e. @xcite , @xmath65 as a consequence , the energy density and total baryon density ( no leptons included ) in the mixed phase read @xmath66 and @xmath67 much simpler than the case above , we just need to find the point where @xmath68 and then construct the eos . in this case the electron chemical potential suffers a discontinuity when passing from the hadron to the quark phase as expected from the simple use of the maxwell conditions .
in the graphs shown next su(3 ) stands for the quark phase taking into account the strange quark and su(2 ) represents the njl model without the strange quark .
gm1 and gm3 represent the hadron phase with their respective set of parameters .
systems without strangeness are described by protons and neutrons in the hadron phase and quarks @xmath1 and @xmath2 in the quark phase .
systems with strangeness also accommodate the hyperons in the hadron phase and quark @xmath0 in the quark phase . in fig .
1 the eos for the pure hadron and pure quark matter are shown .
they are the base to built the eos of the hybrid stars .
the kick , or variation , in curvature in the quark system is due to the appearance of the @xmath0-quark . in figs . 2 and 3 the eos for the two types of hybrid stars
are shown with parametrizations gm1 and gm3 respectively for the hadron phase .
two parametrizations ( su(3 ) hk and su(3 ) rkh ) are used for the quark phase in both figures . in both figures the hyperons and strange quarks are included .
the eos of the hybrid stars with mixed phase are built by the superposition of the eos for the quark and hadron matter , plus the eos for the mixed phase .
the plateau in the eos of the hybrid stars without mixed phase shows a vivid phase transition from hadron to quark matter .
one can see that gm1 produces a much larger quark phase in both constructions while gm3 gives rise to a very large mixed phase when it is present and a quark phase much smaller than the hadron phase if a maxwell construction is used .
this consideration is true independently of the parametrization used in the quark phase , what means that the size of each phase is basically dependent on the hadron phase parametrization , at least for the choices we have considered .
this fact has obvious consequences in the constituents of the stellar matter . in figs . 4 and 5 we compare hybrid stars without the mixed phase built with the su(2 ) and su(3 ) njl models , to understand the role played by the strange quark .
whenever su(2)set1 is used , the hadron phase is minimal and the eos is practically given only by the quark phase .
in all the other cases the size of each phase is mainly determined by the hadron phase parametrization , with small variations in the size of each phase .
notice that the @xmath1 and @xmath2 quark vacuum masses are not identical , being the smallest for the su(2)set1 parametrization , as seen in table 1 .
we use all the eos studied and commented before as input to the tov equations to obtain the neutron star profiles that are shown in fig .
6 , 7 and table 3 .
the tails of the hadron and hybrid stars were obtained with the insertion of the bps eos @xcite .
as expected , the maximum masses for the hadron stars are larger than for the hybrid stars .
when the maxwell construction is used , the resulting mass - radius curve shows a kink , produced by the sharp transition in the eos .
a similar result is shown in @xcite and it is easier to see in fig . 7 , where we once more compare the hybrid stars without mixed phase built with the njl su(2 ) and su(3 ) models .
as a consequence of the eos , gm1 and gm3 produce identical stellar profiles if only protons , neutrons and consequently quarks @xmath1 and @xmath2 are considered with su(2)set1 .
su(2)set2 produces different results .
if only stars with strangeness are considered , gm1 always result in stars with larger maximum masses and radii than gm3 . in table 3
we also show the results for the central energy density @xmath19 .
@xmath69 corresponds to the point where the hadron phase disappears , either because of the onset of a mixed phase ( whenever @xmath70 is also shown ) or giving rise to the quark phase ( otherwise ) .
if strangeness are considered , in hybrid stars without mixed phase , the central energy density may lie within the hadron phase as in gm3@xmath71su(3)hk and gm3@xmath71su(3)rkh .
this means that if a phase transition to the quark phase occurs , the star becomes unstable . on the other hand ,
if the parametrization gm1@xmath71su(3)rkh or gm1@xmath71su(3)hk is used , the central energy density lies in the quark phase .
if a mixed phase is considered , the central energy density always lies in its interior . as a consequence ,
the pure quark phase is never present .
this is the only effect ( not possible to infer from astronomical observations ) that depends strongly on the choice between gibbs and maxwell constructions .
nevertheless , it is well known that the strange quark condensate is very large within the su(3 ) njl model and our results are a consequence of this behaviour . for this specific reason
, we have also checked the results for hybrid stars built without strangeness .
if strangeness is not included , su(2)set1 gives rise to hybrid stars with central energy densities in the quark phase and su(2)set2 shows unstable solutions after the onset of the quark phase with the gm1 parametrization .
analyzing the results shown in table 3 and based on the accuracy of our calculations and the experimental difficulties in the measurements of neutron stars radii , it is fair to say that the method used to built the eos , i.e. , the more rigorous gibbs conditions or the simple use of the maxwell construction give almost indistinguishable results for gravitational masses and radii . finally , in figs .
6 and 7 we have added three lines corresponding to observational constraints .
some properties of the neutron stars are determinated by measuring the gravitational redshift of spectral lines produced in neutron star photosphere which provides a direct constraint on the mass - to - radius ratio ( m / r ) .
a redshift of @xmath72 from three different transitions of the spectra of the x - ray binary exo0748 - 676 was obtained in @xcite .
this redshift corresponds to @xmath73 .
the top line corresponds to this constraint , whose validity remains controversial @xcite . on the other hand , the 1e 1207.4 - 5209 neutron star , which is in the center of the supernova remnant pks 1209 - 51/52 was also observed and two absorption features in the source spectrum were detected @xcite .
these features were associated with atomic transitions of once - ionized helium in the neutron star atmosphere with a strong magnetic field .
this interpretation leads to a redshift of the order of @xmath74 .
this redshift imposes another constraint to the mass to radius ratio given by @xmath75 to @xmath76 .
this constraint is represented by the two lowest lines .
one can see in figs .
[ fig:6 ] and [ fig:7 ] that all the curves obtained are consistent with the measurements of @xcite and @xcite by crossing the 3 lines .
assuming that hybrid stars are possible remnants of supernova explosions , their constitution becomes important only if their macroscopic quantities can be constrained to astronomical observations .
while some calculations practically exclude the existence of hybrid stars @xcite favoring quark stars , others tend to rule out quark stars and favor hybrid stars @xcite .
all those conclusions are obviously model dependent and were reached based on the use of the mit bag model to describe quark matter .
hybrid stars have a hadron phase , in this paper described by the non - linear walecka model ( nlw ) @xcite and a quark phase . instead of using the usual mit bag model @xcite to build the quark phase ,
we have opted to use the njl model @xcite to check some of the previous results on the existence of the mixed phase inside hybrid stars .
if the hadron phase is constituted of protons and neutrons , the corresponding quark phase has quarks @xmath1 and @xmath2 only and the su(2 ) version of the njl model is used .
if the baryonic octet is possible in the hadron phase , quarks @xmath1 , @xmath2 and @xmath0 are present in the quark phase described by the su(3 ) njl model .
we have concluded that the results are very model dependent , as expected .
the onset of a stable quark phase is practically ruled out .
our calculations suggest that stable neutron stars are either of hadronic nature only or bear a mixed phase in their core . concerning the existence of the mixed phase ( mp )
, one can see that the stellar measurable results calculated in the present paper ( mass , radii , central energy density ) depend very little on the choice of the maxwell or the gibbs construction .
hence , it is reasonable to claim that the maxwell construction gives satisfactory results .
the effects of colour superconductivity are out of the scope of the present work , but it is important to mention that they may play an important role in the description of neutron star matter @xcite . the colour - flavour - locked phase ( cfl ) could turn into a superconducting phase ( 2sc ) before matter is hadronized when we read the qcd phase diagram from high to low densities ( see a figures @xcite , for example ) .
this non - continuos transition from the cfl to the 2sc phase in the presence of realistic strange quark masses would certainly affect the description of hybrid stars .
this work was partially supported by cnpq ( brazil ) .
the authors would like to thank fruitful discussions with dr .
constana providncia and two anonymous referees for suggestions that improved the presentation of our results .
k. schertler et al . , phys .
c * 60 * ( 1999 ) 025801 ; m. baldo et al .
, phys . lett .
b * 562 * ( 2003 ) 153 ; i.a .
shovkovy et al . , phys .
d * 67 * ( 2003 ) 103004 ; m. buballa et al .
b * 595 * ( 2004 ) 36 ; t. kahn et al .
b * 654 * ( 2007 ) 170 ; a. steiner et al .
b * 486 * ( 2000 ) 239 ; f. yang and h. shen , phys .
c * 77 * , 025801 ( 2008 ) . | we investigate the structure of hybrid stars based on two different constructions : one is based on the gibbs condition for phase coexistence and considers the existence of a mixed phase ( mp ) , and the other is based on the maxwell construction and no mixed phase is obtained .
the hadron phase is described by the non - linear walecka model ( nlw ) and the quark phase by the nambu - jona - lasinio model ( njl ) .
we conclude that the masses and radii obtained are model dependent but not significantly different for both constructions .
pacs number(s ) : 24.10.jv , 26.60+c |
the current success in the experimental study of decays with the cp - violation in the gold - plated mode of neutral @xmath2-meson by the babar and belle collaborations @xcite allows one to extract the ckm - matrix angle @xmath3 in the unitarity triangle by the model - independent method .
the intensive efforts are intended in the physical programs on the @xmath2 and @xmath4 mesons at the hadron colliders both the active tevatron @xcite and prospective lhc . due to the relatively high cross - sections
the doubly heavy hadrons such as the @xmath5 meson and baryons @xmath6 , @xmath7 and @xmath8 , @xmath7 would be copiously produced at such the machines @xcite .
in addition to the indirect or model - dependent measurements of unitarity triangle @xcite , there is an intriguing opportunity to extract the angle @xmath0 in the model - independent way using the strategy of reference triangles @xcite in the decays of doubly heavy hadrons .
this ideology for the study of cp - violation in @xmath5 decays was originally offered by m.masetti @xcite and investigated by r.fleischer and d.wyler @xcite . in this letter
we extend the method to study the decays of doubly heavy baryons containing the charmed and beauty quarks . to begin , we mention the necessary conditions for extracting the cp - violation effects in the model - independent way . 1 . interference .
the measured quantities have to involve the amplitudes including both the cp - odd and cp - even phases .
exclusive channels .
the hadronic final state has to be fixed in order to isolate the definite matrix elements of ckm matrix , which can exclude the interference of two cp - odd phases with indefinite cp - even phases due to strong interactions at both levels of the quark structure and the interactions in the final state .
3 . oscillations .
the definite involvement of the cp - even phase is ensured by the oscillations taking place in the systems of neutral @xmath2 or @xmath9 mesons , wherein the cp - breaking effects can be systematically implemented .
4 . tagging .
once the oscillations are involved , the tagging of both the flavor and @xmath10 eigenstates is necessary for the complete procedure .
the gold - plated modes in the decays of neutral @xmath2 mesons involve the oscillations of mesons themselves and , hence , they require the time - dependent measurements .
in contrast , the decays of doubly heavy hadrons such as the @xmath5 meson and @xmath6 baryons with the neutral @xmath11 or @xmath12 meson in the final state do not require the time - dependent measurements .
the triangle ideology is based on the direct determination of absolute values for the set of six decays : the decays of baryon in the tagged @xmath11 meson , the tagged @xmath12 meson , the tagged cp - even state , and those of the anti - baryon . to illustrate ,
let us consider the decays of @xmath13 the corresponding diagrams with the decay of @xmath14-quark are shown in fig .
[ fig:1 ] .
we stress that two diagrams of the baryon decay to @xmath11 has the additional negative sign caused by the pauli interference of two charmed quarks , while the color factors is analyzed in the next section .
( 120,25 ) ( -15,0)=55 ( 33,0)=55 ( 85,0)=55 ( -5,7.5)@xmath15 ( 29.5,7.5)@xmath16 ( 33,22)@xmath12 ( 43,7.5)@xmath15 ( 77.5,7.5)@xmath16 ( 81,22)@xmath11 ( 95,7.5)@xmath15 ( 129.5,7.5)@xmath16 ( 133,22)@xmath11 ( 1.5,16.5)@xmath17 ( 12.5,10.5)@xmath18 ( 12.5,3.7)@xmath19 ( 22.5,23)@xmath20 ( 23,18.8)@xmath18 ( 26,15.7)@xmath21 ( 49.5,16.5)@xmath17 ( 60.5,10.5)@xmath18 ( 60.5,3.7)@xmath19 ( 70.5,23)@xmath18 ( 71,18.8)@xmath20 ( 74,15.7)@xmath21 ( 101.5,16.5)@xmath18 ( 112.5,10.5)@xmath17 ( 112.5,3.7)@xmath19 ( 122.5,23)@xmath18 ( 123,18.8)@xmath20 ( 126,15.7)@xmath21 ( 123.5,10.5)@xmath18 the exclusive modes make the penguin terms to be excluded , since the penguins add an even number of charmed quarks , i.e. two or zero , while the final state contains two charmed quarks including one from the @xmath14 decay and one from the initial state . by the same reason the diagrams with the weak scattering of two constituents , i.e. the charmed and beauty quarks in the @xmath15 baryon , are also excluded for the given final state ( see fig .
[ fig:2 ] ) .
( 120,25 ) ( -15,0)=55 ( 33,0)=55 ( 85,0)=55 ( -5,7.5)@xmath15 ( 43,7.5)@xmath15 ( 95,7.5)@xmath15 ( 1.5,16.5)@xmath17 ( 12.5,10.5)@xmath18 ( 12.5,3.7)@xmath19 ( 21.5,23)@xmath22 ( 23,18.8)@xmath18 ( 26,15.7)@xmath18 ( 60.5,16.5)@xmath17 ( 60.5,10.5)@xmath18 ( 60.5,3.7)@xmath19 ( 71,10.5)@xmath18 ( 70,16.5)@xmath22 ( 112.5,16.5)@xmath18 ( 112.5,10.5)@xmath17 ( 112.5,3.7)@xmath19 ( 122.5,16.5)@xmath22 ( 123.5,10.5)@xmath18 the weak scattering of @xmath14 quark off the charmed quark in the initial state can contribute in the next order in @xmath23 as shown in fig . [ fig:2a ] .
nevertheless , we see that such the diagrams have the same weak - interaction structure as at the tree level .
therefore , they do not break the symmetries under consideration .
the magnitude of @xmath23-correction to the absolute values of corresponding decay widths is discussed in section [ sec:2 ] .
( 90,25 ) ( -15,0)=55 ( 35,0)=55 ( -9,7.5)@xmath24 ( 29.5,7.5)@xmath25 ( 33,22)@xmath12 ( 41,7.5)@xmath24 ( 79.5,7.5)@xmath25 ( 83,22)@xmath12 ( -.5,16.5)@xmath17 ( 6.5,16.5)@xmath20 ( 12.5,10.5)@xmath21 ( -.5,10.5)@xmath18 ( 12.5,3.7)@xmath19 ( 22.5,23)@xmath20 ( 23,18.8)@xmath18 ( 26,15.7)@xmath18 ( 54.5,16.5)@xmath17 ( 54.5,10.5)@xmath18 ( 62.5,17.7)@xmath17 ( 62.5,3.7)@xmath19 ( 72.5,23)@xmath20 ( 73,18.8)@xmath18 ( 76,15.7)@xmath18 ( 72.,10.5)@xmath21 thus , the cp - odd phases of decays under consideration are determined by the tree - level diagrams shown in fig .
[ fig:1 ] .
therefore , we can write down the amplitudes in the following form : @xmath26 where @xmath27 denote the cp - even factors depending on the dynamics of strong interactions . using the definition of angle @xmath0 @xmath28,\ ] ] for the cp - conjugated channels , which corresponds to fixing the representation of the ckm matrix , e.g. by the wolfenstein form @xcite . ]
we find @xmath29 we see that the corresponding widths for the decays to the flavor tagged modes coincide with the cp - conjugated ones .
however , the story can be continued by using the definition of cp - eigenstates for the oscillating @xmath1 system mesons are irrelevant here , and we can safely neglect them . ] , @xmath30 so that we straightforwardly get @xmath31 \sqrt{2}{\cal a}(\bar \xi^0_{bc}\to \bar \xi^0_c d_1 ) \stackrel{\mbox{\tiny def}}{= } \sqrt{2 } { \cal a}_{d_1}^{\mbox{\sc cp } } & = & e^{2{\rm i}\gamma } { \cal a}_{\bar d}+{\cal a}_{d}. \label{eq:6}\end{aligned}\ ] ] the complex numbers entering ( [ eq:5 ] ) and ( [ eq:6 ] ) establish two triangles with the definite angle @xmath32 between the vertex positions as shown in fig .
[ fig:3 ] .
( 80,50 ) ( 10,7)=75 ( 32,4)@xmath33 ( 68,22)@xmath34 ( 52,14)@xmath32 ( 44.5,33)@xmath35 ( 7,23)@xmath36 thus , due to the unitarity , the measurement of four absolute values @xmath37 |{\cal a}_{d_1}| = |{\cal a}(\xi^0_{bc}\to \xi^0_c d_1)| , & \quad & |{\cal a}_{d_1}^{\mbox{\sc cp}}| = |{\cal a}(\bar\xi^0_{bc}\to \bar \xi^0_c d_1)|,\end{aligned}\ ] ] can constructively reproduce the angle @xmath0 in the model - independent way .
the above triangle - ideology can be implemented for the analogous decays to the excited states of charmed hyperons in the final state .
the residual theoretical challenge is to evaluate the characteristic widths or branching fractions .
we address this problem and analyze the color structure of amplitudes .
so , we find that the matrix elements under consideration have the same magnitude of color suppression @xmath38 , while the ratio of relevant ckm - matrix elements , @xmath39 with respect to the small parameter of cabibbo angle , @xmath40 , which one can easily find in the wolfenstein parametrization @xmath41 v_{cd } & v_{cs } & v_{cb } \\[2 mm ] v_{td } & v_{ts } & v_{tb } \end{array}\right ) = \left(\begin{array}{ccc } 1-\frac{1}{2}\lambda^2 & \lambda & a \lambda^3 ( \rho-{\rm i}\eta ) \\[2 mm ] -\lambda & 1-\frac{1}{2}\lambda^2 & a \lambda^2 \\[2 mm ] a \lambda^3 ( 1-\rho-{\rm i}\eta ) & -a \lambda^2 & 1 \end{array}\right).\ ] ] thus , we expect that the sides of the reference - triangles are of the same order of magnitude , which makes the method to be a realistic way for extracting the angle @xmath0 . in section [ sec:2 ]
we classify the diagrams for the decays of doubly heavy baryons @xmath42 and @xmath43 by the color and weak - interaction structures .
section [ sec:3 ] is devoted to the numerical estimates in the framework of a potential model .
the results are summarized in conclusion .
let us remind a general framework of the @xmath44-expansion .
its physical meaning at @xmath45 quite reasonably implies that the quarks are bound by the gluon string , which is not broken by the quark - antiquark pair creation .
so , the excitations of the ground states are quasi - stable under the strong interactions producing the decays described by the @xmath44-suppressed terms . in this method
we have got the following scaling rules of color structures : 1 .
the meson wavefunction @xmath46 2 .
the baryon wavefunction @xmath47\ldots i[n_c]}.\ ] ] 3 . the coupling constant @xmath48 4 .
the casimir operators @xmath49 5 .
the fierz relation for the generators of su(@xmath50 ) group in the fundamental representation @xmath51 6 .
the baryon structure constant @xmath52 it gives the color factor emerging in the connection of two quark lines entering the baryon by the gluon line ( see fig .
[ fig:4 ] ) . + ( 80,25 )
( 0,0)=75 next , the non - leptonic weak lagrangian has a form typically given by the following term @xcite : @xmath53 where @xmath54 , and the wilson coefficients @xmath55 in the @xmath44-expansion .
then , we can proceed with the analysis of decays under consideration .
all three diagrams shown in fig .
[ fig:1 ] have the same order in @xmath44 , i.e. @xmath57 more definitely we get the color factors @xmath58 where @xmath59,\\[1 mm ] a_2 & = & \frac{1}{2n_c}\,\left[c_+(n_c+1)-c_-(n_c-1)\right].\end{aligned}\ ] ] thus , we have to calculate the three diagrams given above in the leading order in @xmath44 . the decay modes @xmath61 are described by the diagrams shown in fig .
[ fig:5 ] similar to those of fig .
[ fig:1 ] .
the only difference is the replacement of @xmath62 quark by the strange one , that should be taken into account by the anti - symmetrization of wavefunction in the final state , i.e. the baryon structure of @xmath63 , which results in the vector - spin state of the doubly strange diquark .
( 120,25 ) ( -15,0)=55 ( 33,0)=55 ( 85,0)=55 ( -5,7.5)@xmath43 ( 29.5,7.5)@xmath63 ( 33,22)@xmath12 ( 43,7.5)@xmath43 ( 77.5,7.5)@xmath63 ( 81,22)@xmath11 ( 95,7.5)@xmath43 ( 129.5,7.5)@xmath63 ( 133,22)@xmath11 ( 1.5,16.5)@xmath17 ( 12.5,10.5)@xmath18 ( 12.5,3.7)@xmath21 ( 22.5,23)@xmath20 ( 23,18.8)@xmath18 ( 26,15.7)@xmath21 ( 49.5,16.5)@xmath17 ( 60.5,10.5)@xmath18 ( 60.5,3.7)@xmath21 ( 70.5,23)@xmath18 ( 71,18.8)@xmath20 ( 74,15.7)@xmath21 ( 101.5,16.5)@xmath18 ( 112.5,10.5)@xmath17 ( 112.5,3.7)@xmath21 ( 122.5,23)@xmath18 ( 123,18.8)@xmath20 ( 126,15.7)@xmath21 ( 123.5,10.5)@xmath18 then , the appropriate color factors are given in ( [ eq:8 ] ) .
the diagrams for the decay modes @xmath64 can be obtained from those of fig .
[ fig:5 ] by the replacement of weak currents @xmath65 by @xmath66 and @xmath67 by @xmath68 .
the triangle ideology is effective in the case under consideration too .
the color factors are given by ( [ eq:8 ] ) again .
however , the amplitude of @xmath69 is suppressed by the ckm - matrix factor of @xmath70 which implies that the corresponding side of reference - triangle will be much less than another . in practice , the relatively large branching of decay to the @xmath11 meson should be measured with extremely high accuracy in order to make a sense in the reconstruction of the triangle with the relatively small side determined by the branching of decay to @xmath12 meson .
however , with an expected event rate discussed below , in the nearest future there is no opportunity to observe the @xmath71 triangles in practice .
the diagrams for the decay @xmath73 are shown in fig .
[ fig:6 ] , where the negative relative sign caused by the pauli interference should be taken into account .
( 90,25 ) ( -15,0)=55 ( 35,0)=55 ( -5,7.5)@xmath24 ( 29.5,7.5)@xmath25 ( 33,22)@xmath12 ( 45,7.5)@xmath24 ( 79.5,7.5)@xmath25 ( 83,22)@xmath12 ( 1.5,16.5)@xmath17 ( 12.5,10.5)@xmath18 ( 12.5,3.7)@xmath20 ( 22.5,23)@xmath20 ( 23,18.8)@xmath18 ( 26,15.7)@xmath21 ( 51.5,16.5)@xmath20 ( 62.5,10.5)@xmath17 ( 62.5,3.7)@xmath18 ( 72.5,23)@xmath20 ( 73,18.8)@xmath18 ( 76,15.7)@xmath21 ( 73.5,10.5)@xmath20 the corresponding color factors are given by @xmath74 and @xmath75 in ( [ eq:8 ] ) .
the consideration of decays into the @xmath11 meson is more complicated because of the weak scattering of constituent @xmath14 and @xmath76 quarks as shown in fig .
[ fig:7 ] .
( 90,25 ) ( -15,0)=55 ( 35,0)=55 ( -9,7.5)@xmath24 ( 29.5,7.5)@xmath25 ( 33,22)@xmath11 ( 41,7.5)@xmath24 ( 79.5,7.5)@xmath25 ( 83,22)@xmath11 ( -.5,16.5)@xmath17 ( 6.5,16.5)@xmath18 ( 12.5,10.5)@xmath21 ( -.5,10.5)@xmath20 ( 12.5,3.7)@xmath18 ( 22.5,23)@xmath18 ( 23,18.8)@xmath20 ( 26,15.7)@xmath20 ( 54.5,16.5)@xmath17 ( 54.5,10.5)@xmath20 ( 62.5,17.7)@xmath17 ( 62.5,3.7)@xmath18 ( 72.5,23)@xmath18 ( 73,18.8)@xmath20 ( 76,15.7)@xmath20 ( 72.,10.5)@xmath21 we can easily find that the gluon emission from the other quark lines is suppressed by the color factor since such the exchange by the gluon leads to the baryon color - structure factor @xmath77 in contrast to the @xmath78 in the diagrams shown in fig .
[ fig:7 ] .
the spectator charmed quark can not emit the virtual gluon , since the quark line should be on mass shell up to the small virtualities about the relative momentum of the quark inside the baryon .
further , the color factors for the diagrams in fig . [ fig:7 ]
have the form @xmath79 thus , these factors are of the same order of magnitude as for the decay amplitudes shown in fig .
[ fig:8 ] .
( 90,25 ) ( -15,0)=55 ( 35,0)=55 ( -5,7.5)@xmath24 ( 29.5,7.5)@xmath25 ( 33,22)@xmath11 ( 45,7.5)@xmath24 ( 79.5,7.5)@xmath25 ( 83,22)@xmath11 ( 1.5,16.5)@xmath17 ( 12.5,10.5)@xmath18 ( 12.5,3.7)@xmath20 ( 22.5,23)@xmath18 ( 23,18.8)@xmath20 ( 26,15.7)@xmath21 ( 51.5,16.5)@xmath18 ( 62.5,10.5)@xmath17 ( 62.5,3.7)@xmath20 ( 72.5,23)@xmath18 ( 73,18.8)@xmath20 ( 76,15.7)@xmath21 ( 73.5,10.5)@xmath18 finally , in this section we have analyzed the color and weak - interaction structures of decay amplitudes and isolate those of the largest magnitude , while an illustrative numerical estimate is presented in the next section .
in this section we formulate the framework of a potential model , which allows us to evaluate the characteristic widths and branching ratios for the modes under study .
let us consider the decay of @xmath80 .
so , we define the doubly heavy baryon state in its rest frame by the following form : @xmath81 & & \hspace*{1.7 cm } a^{\dagger}_{i}[b](\boldsymbol k)\,a^{\dagger}_{j}[c](-\boldsymbol k)\ , a^{\dagger}_{k}[d](\boldsymbol q)|0\rangle , \label{eq:10}\end{aligned}\ ] ] where @xmath82 is the charge - conjugation matrix , and we have introduced the following notations : * the relativistic normalization - factor @xmath83 with @xmath84 being the baryon mass in the initial state , * the wavefunction @xmath85 given by two factors of the doubly heavy diquark and the @xmath62 quark , * the color wavefunction determined by @xmath86 , * the spin wavefunction of scalar doubly heavy diquark as determined by the spinor factor @xmath87 , * the quark spinors depending on the four - velocity of the baryon , @xmath88 , as normalized by the condition of sum over the polarization states , say , @xmath89 * @xmath90 denoting the creation operator marked by the flavor , color and momentum in an appropriate way . in ( [ eq:10 ] )
, we use the momentum - space wavefunction in the form of @xmath91,\ ] ] where the four - vector satisfies the condition @xmath92 the model parameter @xmath93 is related with the wavefunction @xmath94 at the origin in the configuration space .
such the quantities were calculated by solving the schrdinger equation with the static potential as described in @xcite , so that @xmath95 while for the @xmath16 baryon we take @xmath96 we will see that for the mode under consideration the absolute values of above parameters are not so critical , while their ratios are important .
then the matrix element is equal to @xmath97 where the mass and wavefunction factors as well as the spinors and four - velocities are transparently denoted , while @xmath98 is given by the spin structure of the matrix element @xmath99,\ ] ] and @xmath100 represents the overlapping of wavefunctions , which can be calculated in the model specified .
so , it is presented by the product of @xmath101 factors @xmath102 where the diquark transition is determined by @xmath103 , \label{eq:13}\ ] ] with @xmath104 and @xmath105 in the given kinematics the product of four - velocities is fixed by @xmath106 the factor of @xmath107 has the form similar to ( [ eq:13 ] ) under the appropriate substitutions for the values of @xmath93 as well as for @xmath108 @xcite .
having written down eq.([eq:11 ] ) , we have neglected the @xmath23-corrections following from the diagrams shown in fig .
[ fig:2a ] . this approximation is theoretically sound .
indeed , the gluon virtuality is determined by the expression @xmath109 the suppression factor of appropriate diagrams is given by @xmath110 where @xmath111 is the characteristic value of wavefunction , and the virtuality of heavy quark line connected to the virtual gluon is of the order of @xmath112 therefore , @xmath113 similar relations can be found , if we consider the value of @xmath23-correction due to the weak scattering shown in fig .
[ fig:7 ] .
one can easily find that the corresponding suppression is given by @xmath114 where we have suggested the constituent mass of light quark @xmath115 therefore , the above @xmath23-corrections can be neglected to the leading order in the @xmath116-expansion .
further , @xmath117 and we can get the matrix element squared @xmath118 where we have expressed the wavefunction of @xmath9 meson in terms of its effective leptonic constant @xmath119 , @xmath120 then , the width is given by @xmath121 = \frac{|\boldsymbol k_d|}{16\pi m_1 ^ 2}\ , \left|{\cal a}_{\bar d}\right|^2,\ ] ] where @xmath122 is the momentum of d meson in the @xmath15 rest - frame .
numerically , at @xmath123 gev @xcite we find @xmath124 \approx 1.3\cdot 10^{-6}\;{\rm ps}^{-1 } \times \left|\frac{v^2_{ub}}{0.003 ^ 2}\,\frac{v^2_{cs}}{0.975 ^ 2}\right|\ , \frac{f^2_d}{(0.222\;{\rm gev})^2}\cdot { \cal o}^2\cdot \frac{(n_c a_2)^2}{1}.\ ] ] putting @xcite @xmath125 = 0.27\;{\rm ps},\ ] ] we get @xmath126 \approx 3.6\cdot 10^{-6 } \times \left|\frac{v^2_{ub}}{0.003 ^ 2}\,\frac{v^2_{cs}}{0.975 ^ 2}\right|\ , \frac{f^2_d}{(0.22\;{\rm gev})^2}\cdot { \cal o}^2\cdot \frac{(n_c a_2)^2}{1}.\ ] ] the calculation of overlap between the wavefunctions is model - dependent , though one can expect that @xmath127 . in the framework of potential model
described we get @xmath128 which gives our final estimate @xmath129 \approx 0.7\cdot 10^{-6 } \times \left|\frac{v^2_{ub}}{0.003 ^ 2}\,\frac{v^2_{cs}}{0.975 ^ 2}\right|\ , \frac{f^2_d}{(0.22\;{\rm gev})^2}\cdot \frac{{\cal o}^2}{0.2}\cdot \frac{(n_c a_2)^2}{1}.\ ] ] therefore , we could expect that the other branching ratios are of the same order of magnitude .
in this work we have extended the reference - triangle ideology to the model - independent extraction of ckm - matrix angle @xmath0 from the set of branching ratios of doubly heavy baryons exclusively decaying to the neutral @xmath9 mesons .
tagging the flavor and cp - eigenstates of such the @xmath9 mesons allows one to avoid the uncertainties caused by the qcd dynamics of quarks .
we have estimated the characteristic branching ratios in the framework of a potential model , which yields , for example , @xmath130 \approx 0.7\cdot 10^{-6}.\ ] ] accepting the above value , we can estimate the rate of events at the lhc collider in the experiment lhcb or in the btev facility at fnal .
so , the production cross section yields the characteristic value of about @xmath131 doubly charmed baryons per year @xcite .
next , the estimated branching rations for the charmed strange baryons are measured for @xmath132 , so that the detection of charged particles in the final state would cover about 50% of decay events with @xmath132 , which we accept for the optimistic estimates . the efficiency of observing the neutral charmed meson crucially depends on the possibility for detecting the neutral kaons and pions
so , removing the neutral kaons and pions could kill the opportunity offered in the present paper .
we consider the optimistic case with the required detection at work , which gives the efficiency for observing the @xmath11 decay at the level of 25% .
therefore , putting the vertex reconstruction efficiency equal to 10% , we can expect the observation of about @xmath133 events per year .
however , the situation is less optimistic for the detection of cp - eigenstates of @xmath11 .
indeed , the cp - even mode with the @xmath134 or @xmath135 final states covers only a @xmath136 fraction of @xmath11 decays , which makes the observation unreachable .
the cp - odd events with @xmath137 can be detected with the branching fraction about @xmath138 , which downs the rate to 1 event per year .
thus , the reconstruction of reference triangle after 10 years of data taking would give , at least , the 30% accuracy for the characteristic triangle side , which makes the extraction of angle @xmath0 rather academic exercise in the method offered .
the authors thank the organizing committee of the heavy quarkonium workshop held in cern , nov . 8 - 11 , 2002 , and personally to antonio vairo and nora brambilla for the invitation and a kind hospitality . v.v.k .
also thanks prof .
r.dzhelyadin for the possibility to visit cern in collaboration with the lhcb group , to which members he expresses a gratitude for a hospitality .
we thank prof .
a.k.likhoded , drs .
yu.gouz and v.romanovsky for useful discussions and valuable remarks .
* * k. abe _ et al . _ [ belle collaboration ] , phys . rev .
d * 66 * , 071102 ( 2002 ) [ arxiv : hep - ex/0208025 ] ; + b. aubert _ et al .
_ [ babar collaboration ] , phys .
lett . * 89 * , 201802 ( 2002 ) [ arxiv : hep - ex/0207042 ] .
k. anikeev _ et al .
_ , arxiv : hep - ph/0201071 . s. s. gershtein , v. v. kiselev , a. k. likhoded and a. v. tkabladze , phys .
usp . * 38 * , 1 ( 1995 ) [ usp . fiz .
nauk * 165 * , 3 ( 1995 ) ] [ arxiv : hep - ph/9504319 ] ; + e.eichten , c.quigg , phys . rev . * d49 * , 5845 ( 1994 ) ; + s.s.gershtein et al . , phys
. rev . * d51 * , 3613 ( 1995 ) ; + a. v. berezhnoi , v. v. kiselev , a. k. likhoded and a. i. onishchenko , phys .
nucl . * 60 * , 1729 ( 1997 ) [ yad .
* 60n10 * , ( 19 ? ? ) ] [ arxiv : hep - ph/9703341 ] .
v. v. kiselev and a. k. likhoded , usp .
fiz . nauk * 172 * , 497 ( 2002 ) [ arxiv : hep - ph/0103169 ] ; + a. v. berezhnoi , v. v. kiselev and a. k. likhoded , phys . atom .
nucl . * 59 * , 870 ( 1996 ) [ yad .
fiz . * 59 * , 909 ( 1996 ) ] [ arxiv : hep - ph/9507242 ] ; + s. p. baranov , phys .
d * 54 * , 3228 ( 1996 ) ; + m. a. doncheski , j. steegborn and m. l. stong , phys .
d * 53 * , 1247 ( 1996 ) [ arxiv : hep - ph/9507220 ] ; + a. v. berezhnoi , v. v. kiselev , a. k. likhoded and a. i. onishchenko , phys .
d * 57 * , 4385 ( 1998 ) [ arxiv : hep - ph/9710339 ] ; + v. v. kiselev and a. e. kovalsky , phys .
nucl . * 63 * , 1640 ( 2000 ) [ yad .
fiz . * 63 * , 1728 ( 2000 ) ] [ arxiv : hep - ph/9908321 ] ; + v. v. braguta , v. v. kiselev and a. e. chalov , phys .
nucl . * 65 * , 1537 ( 2002 ) [ yad .
fiz . * 65 * , 1575 ( 2002 ) ] .
a. j. buras , arxiv : hep - ph/0210291 ; + a. j. buras , f. parodi and a. stocchi , arxiv : hep - ph/0207101 . m. gronau and d. wyler , phys
. lett .
b * 265 * , 172 ( 1991 ) .
m. masetti , phys .
b * 286 * , 160 ( 1992 ) .
r. fleischer and d. wyler , phys .
d * 62 * , 057503 ( 2000 ) [ arxiv : hep - ph/0004010 ] .
l. wolfenstein , phys .
lett . * 51 * , 1945 ( 1983 ) .
g. buchalla , a. j. buras and m. e. lautenbacher , rev .
phys . * 68 * , 1125 ( 1996 ) [ arxiv : hep - ph/9512380 ] .
v. v. kiselev , mod .
a * 10 * , 1049 ( 1995 ) [ arxiv : hep - ph/9409348 ] .
v. v. kiselev , a. e. kovalsky and a. i. onishchenko , phys .
d * 64 * , 054009 ( 2001 ) [ arxiv : hep - ph/0005020 ] ; + v. v. kiselev , a. k. likhoded , o. n. pakhomova and v. a. saleev , phys .
d * 65 * , 034013 ( 2002 ) [ arxiv : hep - ph/0105190 ] .
v. v. kiselev , a. k. likhoded , o. n. pakhomova and v. a. saleev , phys .
d * 66 * , 034030 ( 2002 ) [ arxiv : hep - ph/0206140 ] ; + s. s. gershtein , v. v. kiselev , a. k. likhoded and a. i. onishchenko , phys . rev .
d * 62 * , 054021 ( 2000 ) ; + s. s. gershtein , v. v. kiselev , a. k. likhoded and a. i. onishchenko , heavy ion phys . *
9 * , 133 ( 1999 ) [ phys . atom .
nucl . * 63 * , 274 ( 2000 yafia,63,334 - 347.2000 ) ] [ arxiv : hep - ph/9811212 ] ; + v. v. kiselev and a. i. onishchenko , nucl .
b * 581 * , 432 ( 2000 ) [ arxiv : hep - ph/9909337 ] ; + v. v. kiselev and a. e. kovalsky , phys . rev .
d * 64 * , 014002 ( 2001 ) [ arxiv : hep - ph/0005019 ] .
m. j. savage and m. b. wise , phys .
b * 248 * , 177 ( 1990 ) ; + m. j. savage and r. p. springer , int .
j. mod .
phys . a * 6 * , 1701 ( 1991 ) ; + s. fleck and j. m. richard , prog .
phys . * 82 * , 760 ( 1989 ) ; + d. b. lichtenberg , r. roncaglia and e. predazzi , phys .
d * 53 * , 6678 ( 1996 ) [ arxiv : hep - ph/9511461 ] ; + m. l. stong , arxiv : hep - ph/9505217 ; + j. m. richard , phys .
* 212 * , 1 ( 1992 ) ; + d. ebert , r. n. faustov , v. o. galkin , a. p. martynenko and v. a. saleev , z. phys .
c * 76 * , 111 ( 1997 ) [ arxiv : hep - ph/9607314 ] ; + s. p. tong , y. b. ding , x. h. guo , h. y. jin , x. q. li , p. n. shen and r. zhang , phys .
d * 62 * , 054024 ( 2000 ) [ arxiv : hep - ph/9910259 ] ; + c. itoh , t. minamikawa , k. miura and t. watanabe , phys .
d * 61 * , 057502 ( 2000 ) ; + h. kaur and m. p. khanna , arxiv : hep - ph/0005077 ; + c. p. singh and m. p. khanna , lett .
nuovo cim . * 30 * , 276 ( 1981 ) ; + d. ebert , r. n. faustov , v. o. galkin and a. p. martynenko , arxiv : hep - ph/0201217 ; + r. lewis , n. mathur and r. m. woloshyn , phys .
d * 64 * , 094509 ( 2001 ) [ arxiv : hep - ph/0107037 ] , arxiv : hep - ph/0203253 . v. v. kiselev , a. k. likhoded and a. i. onishchenko , eur . phys .
j. c * 16 * , 461 ( 2000 ) [ arxiv : hep - ph/9901224 ] ; + b. guberina , b. melic and h. stefancic , eur .
j. c * 9 * , 213 ( 1999 ) [ eur .
j. c * 13 * , 551 ( 1999 ) ] [ arxiv : hep - ph/9901323 ] .
a. i. onishchenko , arxiv : hep - ph/0006295 , arxiv : hep - ph/0006271 ; + a. faessler , t. gutsche , m. a. ivanov , j. g. korner and v. e. lyubovitskij , phys .
b * 518 * , 55 ( 2001 ) [ arxiv : hep - ph/0107205 ] ; + a. k. likhoded and a. i. onishchenko , arxiv : hep - ph/9912425 ; + a. i. onishchenko , arxiv : hep - ph/9912424 ; + m. a. sanchis - lozano , nucl .
b * 440 * , 251 ( 1995 ) [ arxiv : hep - ph/9502359 ] ; + d. a. egolf , r. p. springer , j. urban , [ arxiv : hep - ph/0211360 ] . | the model - independent method based on the triangle ideology is implemented to extract the ckm - matrix angle @xmath0 in the decays of doubly heavy baryons containing the charmed and beauty quarks .
we analyze a color structure of diagrams and conditions for reconstructing two reference - triangles by tagging the flavor and cp eigenstates of @xmath1 mesons in the fixed exclusive channels .
the characteristic branching ratios are evaluated in the framework of a potential model setting a parametric dependence on the hadronic matrix elements for the decay rates . |
the theory of quantum transport in nano - scale devices has evolved rapidly over the past decade , as advances in experimental techniques have made it possible to probe transport properties ( at different temperatures ) down to the single - molecule scale .
furthermore simultaneous measurement of charge and heat transport through single molecules is now also possible @xcite .
the development of accurate theoretical methods for the description of quantum transport at the single - molecule level is essential for continued progress in a number of areas including molecular electronics , spintronics , and thermoelectrics .
one of the longstanding problems of quantum charge transport is the establishment of a theoretical framework which allows for quantitatively accurate predictions of conductance from first principles .
the need for methods going beyond the standard approach based on density functional theory combined with landauer - like elastic scattering @xcite has been clear for a number of years .
it is only recently that more advanced methods to treat electronic interaction have appeared , for example those based on the many - body @xmath0 approximation @xcite .
alternative frameworks to deal with the steady - state or time - dependent transport are given by many - body perturbation theory based on the non - equilibrium ( ne ) green s function ( gf ) formalism : in these approaches , the interactions and ( initial ) correlations are taken into account by using conserving approximations for the many - body self - energy @xcite .
other kinds of interactions , e.g. electron - vibron coupling , also play an important role in single - molecule quantum transport .
inelastic tunneling spectroscopy constitutes an important basis for spectroscopy of molecular junctions , yielding insight into the vibrational modes and ultimately the atomic structure of the junction @xcite .
there have been many theoretical investigations focusing on the effect of electron - vibron coupling in molecular and atomic scale wires @xcite .
in all these studies , the interactions have always been considered to be present in the central region ( i.e. the molecule ) only , and the latter is connected to two non - interacting terminals .
interactions are also assumed not to cross at the contracts between the central region and the leads .
when electronic interactions are present throughout the system , as within density - functional theory calculations , they are treated at the mean - field level and do not allow for any inelastic scattering events .
however , there are good reasons to believe that such approximations are only valid in a very limited number of practical cases .
the interactions , in principle , exist throughout the entire system . in a recent paper we derived a general expression for the current in nano - scale junctions with interaction present everywhere in the system @xcite .
with such a formalism , we can calculate the transport properties in those systems where the interaction is present everywhere .
the importance of extended interaction in nano - scale devices has also been addressed , for electron - electron interaction , in recently developed approaches such as refs . [ ] . in the present paper , we also consider interactions existing beyond the central region .
we apply our recently developed formalism @xcite for fully interacting systems to a specific model of a single - molecule nanojunction .
we focuss on a model system in the presence of electron - vibron interaction within the molecule and between the molecule and one of the leads .
we show how the interaction crossing at one interface of the molecular nanojunctions affects the transport properties by renormalising the coupling at the interface in a bias - dependent manner .
we also study the relationship between the non - equilibrium charge susceptibility @xcite and the dynamical conductance for the present model of interaction crossing at the contacts .
the paper is organised as follows : in sec .
[ sec : transport ] , we briefly recall the main result of our current expression for fully interacting systems . in sec .
[ sec : interac ] , we present the model hamiltonian for the system which include two kinds of electron - vibron interaction , an holstein - like hamiltonian combined with a su - schrieffer - heeger - like hamiltonian . in this section , we also describe how the corresponding self - energies are calculated and the implications of such approximations on the current expression at the left and right interfaces . in sec .
[ sec : res ] , we show that our approximations are fully consistent with the constraint of current conservation .
then the effects of the static non - equilibrium ( i.e. energy - independent but bias - dependent ) renormalisation of the coupling at the contact on both the current and the dynamical conductance are studied for a wide range of parameters .
we also show that the ne charge susceptibility is still related to the dynamical conductance even in the presence of crossing interaction at the contact .
we finally conclude and discuss extensions of the present work in sec .
[ sec : ccl ] .
we consider a two - terminal device , made of three regions left - central - right , in the steady - state regime .
in such a device , labelled @xmath1 , the interaction which we specifically leave undefined ( e.g. electron - electron or electron - phonon)is assumed to be well described in terms of the single - particle self - energy @xmath2 and spreads over the entire system . we use a compact notation for the green s function @xmath3 and the self - energy @xmath4 matrix elements @xmath5 .
they are annotated @xmath6 ( @xmath7 or @xmath8 ) for the elements in the central region @xmath9 ( left @xmath10 , right @xmath11 region respectively ) , and @xmath12 ( or @xmath13 ) and @xmath14 ( or @xmath15 ) for the elements between region @xmath9 and region @xmath10 or @xmath11 .
there are no direct interactions between the two electrodes , i.e. @xmath16 . in refs .
[ ] , we showed that for a finite applied bias @xmath17 the steady - state current @xmath18 flowing through the left @xmath19 interface is given by : @xmath20 \\ + & { { \rm tr}}_{\{l\}}\left[\sigma^{{{\rm mb}},>}_l g^<_l - \sigma^{{{\rm mb}},<}_l g^>_l \right ] \end{split}\ ] ] where the @xmath21 quantities are @xmath22 by definition @xmath23 ( similarly for the @xmath24 components ) where @xmath25 are the nominal coupling matrix elements between the @xmath10 and @xmath9 regions .
@xmath26 are the gf of the region @xmath10 renormalised by the interaction _ inside _ that region , where @xmath27 stands for the retarded , advanced and lesser gf components .
for example , for the advanded and retarded components , we have @xmath28 where all quantities are defined only in the subspace @xmath10 . the first line in the current equation eq .
( [ eq : ilfinal ] ) corresponds to a generalisation of the meir and wingreen @xcite result to the cases for which the interactions are present in the three @xmath29 regions as well as in between the @xmath30 and @xmath31 regions .
the second trace in eq .
( [ eq : ilfinal ] ) corresponds to inelastic events induced by the interaction in the @xmath10 lead . when a local detailed balance equation helds , this terms vanishes since locally one has @xmath32 .
( [ eq : ilfinal ] ) bears some resemblance to the expression derived by meir and wingreen @xcite when written as : @xmath33 .
\end{split}\ ] ] where we use the definitions @xmath34 and @xmath35 .
hence @xmath36 becomes hence one can see by comparing eq .
( [ eq : ilfinal ] ) and eq .
( [ eq : il_meirwingreen_bis ] ) that the quantities @xmath37 ( @xmath38 ) and @xmath39 are playing the role of the @xmath10 lead self - energy @xmath40 ( @xmath41 ) and @xmath42 respectively when the interactions cross at the @xmath19 interface . in meir and wingreen model
, the leads are non - interacting , hence the second trace @xmath43 $ ] in eq .
( [ eq : ilfinal ] ) does not exist .
we consider a single - molecule junction in the presence of electron - vibron interaction inside the central region and crossing at the contacts .
using a model system to reduce these calculations to a tractable size , we concentrate on a single molecular level coupled to a single vibrational mode . a full description of our methodology , for the interaction inside the region @xmath9 ,
is provided in refs .
[ ] . furthermore , we consider that the electron - vibron interaction exist also at one contact ( the left @xmath10 electrode for instance ) .
this model typically corresponds to an experiment for a molecule chemisorbed onto a surface ( the left electrode ) with a tunneling barrier to the right @xmath11 lead . in the following model , we consider two kinds of electron - vibron coupling : a local coupling in the sense of an holstein - like coupling of the electron charge density with a internal degree of freedom of vibration inside the central region , and an off - diagonal coupling in the sense of a su - schrieffer - heeger - like coupling @xcite to another vibration mode involving the hopping of an electron between the central @xmath9 region and the @xmath10 electrode .
the hamiltonian for the region @xmath9 is @xmath44 where @xmath45 ( @xmath46 ) creates ( annihilates ) an electron in the molecular level @xmath47 .
the electron charge density in the molecular level is coupled to the vibration mode of energy @xmath48 via the coupling constant @xmath49 , and @xmath50 ( @xmath51 ) creates ( annihilates ) a vibration quantum in the vibron mode @xmath48 .
the central region @xmath9 is nominally connected to two ( left and right ) one - dimensional tight - binding chains via the hopping integral @xmath52 and @xmath53 .
the corresponding electrode @xmath54 self - energy is @xmath55 with the dispersion relation @xmath56 where @xmath57 and @xmath58 are the tight - binding on - site and off - diagonal elements of the electrode chains . to describe the electron - vibron interaction existing at the left contact
, we consider that the hopping integral @xmath52 is actually dependent on some generalised coordinate @xmath59 .
the latter represents either the displacement of the centre - of - mass of the molecule or of some chemical group at the end of the molecule link to the @xmath10 electrode . at the lowest order
, the matrix element can be linearised as @xmath60 .
hence the hopping of an electron from the @xmath9 region to the @xmath10 region ( and _ vice versa _ ) is coupled to a vibration mode ( of energy @xmath61 ) via the coupling constant @xmath62 ( itself related to @xmath63 ) .
the corresponding hamiltonian is given by @xmath64 where @xmath65 ( @xmath66 ) creates ( annihilates ) a vibration quantum in the vibron mode @xmath61 , the generalised coordinate is @xmath67 , and @xmath68 ( @xmath69 ) creates ( annihilates ) an electron in the level @xmath70 of the @xmath10 electrode . the hamiltonians eq . ( [ eq : h_central ] ) and eq . ( [ eq : vcrossing ] )
are used to calculate the corresponding electron self - energies at different orders of the interaction @xmath49 and @xmath62 using conventional non - equilibrium diagrammatics techniques @xcite .
furthermore , at equilibrium , the whole system has a single and well - defined fermi level @xmath71 .
a finite bias @xmath17 , applied across the junction , lifts the fermi levels as @xmath72 .
the fraction of potential drop @xcite at the left contact is @xmath73 and @xmath74 at the right contact , with @xmath75 and @xmath76 $ ] .
the electron - vibron self - energies in the central region @xmath9 are calculated within the born approximation .
the details of the calculations are reported elsewhere @xcite so we briefly recall the different expressions for the self - energies @xmath77 with @xmath78 with @xmath79 and @xmath80 and @xmath81 with the usual definitions for the bare vibron gf @xmath82 : @xmath83 \\
d_0^r(\omega ) & = \frac{1}{\omega - \omega_0 + i 0^+ } - \frac{1}{\omega + \omega_0 + i 0^+ } \ , \end{split}\ ] ] where @xmath84 is the averaged number of excitations in the vibration mode of frequency @xmath48 given by the bose - einstein distribution at temperature @xmath85 . in the following ,
we work in the limit of low temperature for which @xmath86 . as a first application of our transport formalism for crossing interactions
, we consider a mean - field approximation for the electron - vibron coupling at the @xmath19 interface .
this leads to the hartree - like expressions for the many - body self - energies at the @xmath19 interface : @xmath87 where @xmath88 similarly the self - energy @xmath89 is obtained from @xmath90 .
one can see that the interaction crossing at the @xmath19 interface induces a static ( however bias - dependent ) renormalisation of the nominal coupling @xmath91 between the @xmath10 and @xmath9 regions .
this non - equilibrium renormalisation will induce , amongst other effects , a bias - dependent modification of the broadening of the spectral features of the @xmath9 region .
since the renormalisation is static at the hartree level , we use a `` potential '' notation to represent the normalised coupling : @xmath92 with @xmath93 and similarly for @xmath94 .
the static renormalisation of the nominal coupling @xmath52 is driven by the ratio @xmath95 . in the following numerical applications ,
we consider small to larger renormalisation effects , for which the ratio @xmath96 is ranging from @xmath97 to @xmath98 .
one should note , however , that in all the calculations we have performed , the density matrix element @xmath99 is always of the order of @xmath100 . therefore the renormalisation effects are always smaller than the nominal coupling @xmath52 itself . in order to get the renormalised couplings @xmath101 , we need the off - diagonal elements @xmath102 and @xmath103
. the closed expression for the @xmath102 gf matrix element is obtained from the corresponding dyson equation @xmath104^<_{lc}$ ] .
after formal manipulation , we find that @xmath105 with @xmath106 is the renormalised advanced gf of the central region @xmath107^{-1 } -\sigma^a_c - v_{cr } g^a_r v_{rc } - \tilde v_{cl } g^a_l \tilde v_{lc } \right]^{-1 } , \ ] ] and the off - diagonal element @xmath108 is given by @xmath109^{-1 } - v_{cr } g^r_r v_{rc } \right]^{-1 } , \ ] ] where @xmath110 is the renormalised gf of the disconnected c region @xmath111^{-1}= [ g^r_c]^{-1 } - \sigma^r_c$ ] .
after further manipulation , we get the following expression for @xmath108 by using the notation @xmath112 for the non - interacting @xmath11 lead self - energy : @xmath113^{-1 } - \tilde v_{lc } \left [ [ \tilde g^r_c]^{-1 } - y_c^{r , r } \right]^{-1 } \tilde v_{cl } \right]^{-1 } \tilde v_{lc } \left [ [ \tilde g^r_c]^{-1 } - y_c^{r , r } \right]^{-1 } .\ ] ] similarily we can calculate the off - diagonal element @xmath103 from the corresponding dyson equations @xmath114^<_{cl}$ ] .
we find @xmath115 where @xmath116 is the retarded version of eq .
( [ eq : gac ] ) and @xmath117^{-1 } - y_c^{r , a } \right]^{-1 } \tilde v_{cl } g^a_l .\ ] ] as expected from the definition of the different gfs , we can see that indeed @xmath118 and @xmath119 .
calculations are performed in a self - consistent manner .
there are different ways to calculate the gf and the self - energies in a self - consistent way : in the present work , we first perform self - consistent calculations for the central region @xmath9 .
the new hopping matrix elements between the @xmath9 and @xmath10 regions given by eq .
( [ eq : vlc_renorm ] ) are then updated via the use of eqs .
( [ eq : glessercl ] ) and ( [ eq : glesserlc ] ) .
the calculations are then re - iterated until full self - consistency for all self - energies and gf in the region @xmath9 and at the left contact @xmath120 is obtained . for the model of crossing interaction we are considering here ,
further simplications can be introduced in the calculation of the current . since there is no other interaction inside the @xmath10 and @xmath11 regions , the current expression
is given only by the first line of eq.([eq : ilfinal ] ) .
furthermore , there are no lesser and greater components for the self - energy @xmath121 at the mean - field level . the current expression eq .
( [ eq : ilfinal ] ) reduces to : @xmath122 \ , \end{split}\ ] ] with the following simplified expressions for the @xmath123 quantities @xmath124 since @xmath125 and @xmath126 .
furthermore , the current at the @xmath127 interface has a conventional meir and wingreen expression , since there are no crossing interactions : @xmath128 \ , \end{split}\ ] ] where we use the conventional definition for the @xmath11 lead self - energies : @xmath129 with @xmath130 .
the main differences between the expression for the left and right currents @xmath131 and @xmath132 arise from the non - equilibrium static renormalisation of the coupling between the central region @xmath9 and the @xmath10 lead given in eq .
( [ eq : vlc_renorm ] ) .
( black line with dots ) and @xmath133 ( red line ) given respectively by eq .
( [ eq : il_hartree ] ) and eq .
( [ eq : ir_meirwingreen ] ) .
the current is fully conserved as can be seen by the green dashed line which shows @xmath134 magnified by a factor of 2000 .
the parameters are @xmath135 , @xmath136 , @xmath137 , @xmath138 , @xmath139 , @xmath140 , @xmath141.,width=272 ] .
current @xmath142 ( left panel ) , dynamical conductance @xmath143 in logarithmic scale ( top - right panel ) and the real part of non - equilibrium renormalized static coupling @xmath144 ( bottom - right panel ) for different values of the coupling @xmath62 strength at the left interface between @xmath145 and @xmath146 , corresponding to the follow ratio : @xmath147 .
the ne renormalization of @xmath148 corresponds to a non - monotonic reduction of the nominal hopping integral @xmath52 with applied bias , which induces a diminution of the current , and modifications of the widths of the conductance peaks accompanied with a slight shift of the conductance peaks at high bias .
see text for a complete discussion .
the other parameters are @xmath136 , @xmath137 , @xmath138 , @xmath149 , @xmath141.,width=272 ]
we have perfomed calculations for many different values of the parameters in the hamiltonians .
we present below the most characteristic results for different transport regimes and for different coupling strengths @xmath62 , while the interaction in the region @xmath9 is taken to be in the intermediate coupling regime @xmath150 .
the nominal couplings between the central region and the electrodes @xmath151 , before ne renormalisation , are not too large , so that we can discrimate clearly between the different vibron side - band peaks in the spectral functions .
the values chosen for the parameters are typical values for realistic molecular junctions @xcite .
the different transport regimes considered bellow are called the far off - resonant , the off - resonant , the intermediate regime and the quasi - resonant regime .
they correspond to different position of the molecular level @xmath47 with respect to the fermi level @xmath71 at equilibrium . with the static renormalised molecular level @xmath152 , the far off - resonant regime corresponding to @xmath153 and @xmath154 several @xmath48
, the off - resonant regime corresponds to @xmath153 and @xmath155 one or two @xmath48 , the intermediate regime corresponds to @xmath156 and the quasi - resonant regime corresponds to @xmath157 . in the following the current
is given in units of charge per time , the conductance in unit of quantum of conductance @xmath158 and the bias @xmath17 and the normalised coupling @xmath94 in natural units of energy where @xmath159 and @xmath160 .
one of the most important physical conditions that our formalism needs to fulfil is the constraint of current conservation .
we use conserving approximations to calculate the interaction self - energies in the central region @xmath9 and for the crossing interaction at the left interface .
however , since there is no interaction crossing at the @xmath127 interface while the interaction _ is _ crossing at the @xmath19 interface , we have to check that the current given by eq .
( [ eq : il_hartree ] ) for @xmath18 is equal to the meir and wingreen current given by eq .
( [ eq : ir_meirwingreen ] ) for @xmath133 , i.e. @xmath161 . figure [ fig : ex_iconserv ] shows that the condition of current conservation is indeed fulfilled , as expected .
we have carefully checked that the current is conserved for all the calculations presented in the present paper , i.e. that @xmath162 . in figures [ fig : eps090 ] to [ fig : eps020 ] , we show the dependence of the current @xmath142 , of the dynamical conductance @xmath143 and of the renormalised coupling @xmath148 on the applied bias @xmath17 , for different values of the interaction strength @xmath62 at the left contact .
we consider different transport regimes ranging from the far off - resonant regime ( fig .
[ fig : eps090 ] ) , the off - resonant regime ( fig . [ fig : eps050 ] ) , the intermediate regime ( fig . [ fig : eps030 ] ) and finally the quasi - resonant regime ( fig . [ fig : eps020 ] ) . the ne renormalisation of the coupling at the left contact given by eq .
( [ eq : vlc_renorm ] ) corresponds to an effective decrease of the hopping integral leading to a decrease of the current for increasing values of @xmath62 , as can be seen in the left panels of figs .
[ fig : eps090 ] to [ fig : eps020 ] .
the real part of @xmath148 shows a non - monotonic behaviour with the applied bias and presents features ( local dips ) at applied biases corresponding to peaks in the dynamical conductance ( bottom right panels of figs .
[ fig : eps090 ] to [ fig : eps020 ] ) .
therefore , the interaction crossing at the left contact not only decreases the value of the current but also affects the width of the peaks in the conductance , as can be seen in the top right panels of figs .
[ fig : eps090 ] to [ fig : eps020 ] . in all our calculations
, it appears that each conductance peak now has an asymmetric shape , i.e. a different broadening on each side of the peak , which is due to the non - monotonic and asymmetric behaviour of @xmath94 versus applied bias .
the detailed understanding of such a behaviour is rather complex .
however , one can obtain a qualitative understanding of our results by considering the following analysis . .
the other parameters are the same as in figure [ fig : eps090 ] .
see text for a complete discussion.,width=272 ] .
the other parameters are the same as in figure [ fig : eps090 ] .
see text for a complete discussion.,width=272 ] .
same parameters as in figure [ fig : eps090 ] .
see text for a complete discussion.,width=272 ] we can use the current expressions for @xmath131 and @xmath163 given by eq . ( [ eq : il_hartree ] ) and eq . ( [ eq : ir_meirwingreen ] ) respectively to obtain a symmetrized current @xmath164 expression within a good approximation @xcite : @xmath165 where @xmath166 and @xmath167^{-1 } \ , \ ] ] with the self - energy in the central region @xmath168 given by the sum of eq .
( [ eq : sigmac_hartree ] ) and eq .
( [ eq : sigmacr_fock ] ) ; and we recall that @xmath169 in the off - resonant regime and for small bias @xmath170 , one can consider that @xmath171 ( see ref . [ ] ) , hence @xmath172 we can now use eq .
( [ eq : isym_bis ] ) to understand the behaviour of the current . in figure
[ fig : parametrici_vcl ] we show the parametric curves @xmath173 obtained for the far off - resonant transport regime ( shown in fig .
[ fig : eps090 ] ) .
figure [ fig : parametrici_vcl ] shows a complex dependence of the current versus the ne renormalised coupling @xmath94 . however at low bias , one can consider that @xmath174 and @xmath175 are independent of @xmath176 and hence take such quantities out of the integral in eq .
( [ eq : isym_bis ] ) .
therefore we have @xmath177 and the current is quadratic in @xmath94 since @xmath178 .
such a dependence can be clearly seen in figure [ fig : parametrici_vcl ] for the low bias regime where @xmath179 . at larger bias
, one can no longer neglect the @xmath176 dependence of @xmath174 ( and @xmath175 ) and more importantly the effects of the interaction in the central region , i.e. @xmath180 . hence the quadratic dependence of @xmath181 on @xmath182 is lost . for the far off - resonant transport regime shown in fig .
[ fig : eps090 ] ( using the same legend for the different @xmath62 values ) .
the starting points of the parametric curves , i.e. @xmath183 , are at the bottom of the graph ( zero current values ) , and increasing @xmath17 corresponds to increasing values of the current . at low bias ( @xmath179 ) , @xmath131 is quadratic in @xmath94.,width=272 ] we now turn onto the effect of the strength of the nominal hopping integral @xmath52 on the renormalised coupling @xmath148 at fixed values of the crossing interaction strength @xmath62 .
figure [ fig : vcl_varyt0l ] shows the relative dependence of @xmath148 on the nominal coupling @xmath52 at the left interface versus applied bias .
the figure shows that there is no simple relationship between @xmath148 and @xmath52 for the whole range of parameters explored .
there is a small linear regime at low applied bias , otherwise the dependence of @xmath94 on both @xmath184 and @xmath17 is highly non - linear .
there is a progressive washing - out of the features in @xmath148 for increasing values of @xmath184 , since a general increase of the coupling to the leads generates to a global broadening of the features in the spectral functions . in a recent paper @xcite ,
we have developed the concept of the generalized susceptibilities for nonlinear systems @xcite and applied it to the charge transport properties in two - terminal nano - devices .
we have introduced the non - equilibrium charge susceptibility @xmath185 concept , defined by @xmath186 .
we have shown that @xmath185 is related to the dynamical conductance @xmath187 .
the relationship is formally different than the one obtained at equilibrium . in spectroscopic terms , both @xmath185 and @xmath187 contain features versus applied bias when charge fluctuation occurs in the corresponding electronic resonance .
this relationship has been demonstrated for model calculations of interacting nanoscale devices but only when the interaction is present in the central region @xcite .
we now check the validity of such a relationship between @xmath185 and @xmath187 when interaction crossing at the left contact is also taken into account .
figure [ fig : nechigv ] shows the results we obtain for @xmath185 and @xmath187 for different sets of parameters corresponding to the off - resonant and quasi - resonant transport regimes .
once more , we find that @xmath185 and @xmath187 present features at the same applied bias .
such a behaviour validates the relationship between the non - equilibrium charge susceptibility and the dynamical conductance even when interactions cross at the contacts ( at least at the static mean - field level ) . for the off - resonant transport regime ( @xmath188 ) and @xmath189 versus applied bias and for different values of the nominal hopping integrals @xmath184 ( @xmath54 ) .
the relative dependence of @xmath94 , shown here for @xmath190 , is linear only for small bias .
otherwise the dependence of @xmath94 on both @xmath184 and @xmath17 is highly non - linear .
the other parameters are the same as in fig .
[ fig : eps050].,width=272 ] and dynamical conductance @xmath187 versus applied bias for different sets of parameters
. @xmath185 is rescaled by a factor of 7 .
( a ) off - resonant regime @xmath191 and @xmath192 .
( b ) off - resonant regime @xmath191 and @xmath189 , ( c ) quasi - resonant regime @xmath193 and @xmath194 . @xmath185 and @xmath187 present features at the same applied bias .
note the negative contribution to @xmath195 around @xmath196 .
see main text for a complete discussion .
the other parameters are the same as in fig .
[ fig : eps050].,width=272 ] finally , one should note the negative contribution to @xmath185 in panel ( c ) of figure [ fig : nechigv ] .
such a behaviour originates from the properties of @xmath195 and not from the approximation used to calculate the crossing interaction @xcite .
in fact , there are always two contributions to @xmath195 , one is positive and corresponds to electron fluctuation and the other is negative and corresponds to hole fluctuation . by electron ( hole )
fluctuations , we mean the variation of the occupancy ( versus applied bias ) of electronic resonances located nominally above ( below ) the fermi level at equilibrium ( or at small bias ) .
therefore , for the off - resonant transport regime with @xmath197 , the features in @xmath185 corresponds to positive peaks ( as shown in panels ( a ) and ( b ) in figure [ fig : nechigv ] and in the different figures of ref .
[ ] ) . for the off - resonant regime with @xmath198 , one would get negative peaks in @xmath185 .
for the intermediate and quasi - resonant transport regime , one obtains both positive and negative contributions in @xmath185 as shown in panel ( c ) of figure [ fig : nechigv ] .
the most peculiar case corresponds to a fully electron - hole symmetric system , i.e. the spectral function @xmath199 is fully symmetric around the equilibrium fermi level and for any applied bias @xmath17 with @xmath200 . in that case
, one can easily show that @xmath201 is actually given by @xmath202 for symmetry reasons .
hence @xmath203 is independent of the applied bias for conserving approximations for the self - energies , and consequently @xmath204 .
such a behaviour can also be interpreted with the previous picture : for a fully electron - hole symmetric system , any contributions from electron fluctuation is exactly cancelled out by the opposite contribution from hole fluctuation , and @xmath185 is flat and equal to zero for each bias @xcite .
we have studied the transport properties through a two - terminal nanoscale device with interactions present not only in the central region but also with interaction crossing at the interface between the left lead and the central region . to calculate the current for such an fully interacting system
, we have used our recently developed quantum transport formula @xcite based on the negf formalism . as a first practical application ,
we have considered a prototypical single - molecule nanojunction with electron - vibron interaction . in terms of the electron density matrix ,
the interaction is diagonal in the central region for the first vibron mode and off - diagonal between the central region and the left electrode for the second vibron mode .
the interaction self - energies are calculated in a self - consistent manner using the lowest order hartree - fock - like diagram in the central region and the hartree - like diagram for the crossing interaction .
our calculations were performed for different transport regimes ranging from the far off - resonance to the quasi - resonant regime , and for a wide range of parameter values .
they shown that , for this model , we obtain a non - equilibrium ( i.e. bias dependent ) static ( i.e. energy independent ) renormalisation of the nominal hopping matrix element between the left electrode and the central region .
the renormalisation is such that the amplitude of the current is reduced in comparison with the current values obtained when the interaction is only present in the central region .
such a result could provide an partial explanation for the fact in conventional density - functional based calculations , the values of the current are always much larger than in the corresponding experiments , since no non - equilibrium renormalisation of the contacts is taken into account in those calculations .
however , even though it provides the right trends , the decrease in the current obtained by ne renormalisation of the coupling to the leads is not as important as the effects obtained from a proper calculation of the band - gap and band - alignment in realistic molecular system @xcite .
the ne static renormalisation of the contact is highly non - linear and non - monotonic in function of the applied bias , and the larger effects occur at applied bias corresponding to resonance peaks in the dynamical conductance .
the conductance is also affected by the ne renormalisation of the contact , showing asymmetric broadening around the resonance peaks and some slight displacement of the peaks at large bias in function of the coupling strengh @xmath62 .
furthermore , we have also shown that , even in the presence of crossing interactions , the relationship between the ne charge susceptibility and dynamical conductance @xcite still holds for the different transport regimes considered here .
extensions of the present study are now considered .
one route is to develop more accurate ne renormalisation by considering , for example , a quasi - particle approach within a dynamical mean - field - like treatement of the crossing interaction self - energy @xcite .
another route is to go beyond the hartree approximation for the crossing interaction self - energy using other many - body diagrams @xcite .
this would lead to dynamical ne renormalisation of the contact involving inelastic scattering processes .
finally , it should be noted that for the model system we considered here , one could also solve the problem by using more standard approaches ( typically the original meir and wingreen approach ) by extending the size of the central region to include all the interaction .
the results then obtained should be strictly equivalent to our calculations .
we have already commented in detail on this point ( in a formal and theoretical point of view ) in refs .
[ ] . however our approach offers a more intuitive physical interpretation of the results , i.e. renormalisation of the contact ( in the present case in a static ne mean - field scheme ) .
using an extended central region will not provide an easy physical interpretation of the results ; and as a matter of principle , it will not always be possible to increase at will the size of the central region , more especially when one considers future application of the method to much large molecular system with much more complex coupling to the leads . | we present an application of a new formalism to treat the quantum transport properties of fully interacting nanoscale junctions [ phys .
rev .
b * 84 * , 235428 ( 2011 ) ] .
we consider a model single - molecule nanojunction in the presence of two kinds of electron - vibron interactions . in terms of electron density matrix ,
one interaction is diagonal in the central region and the second is off - diagonal in between the central region and the left electrode .
we use a non - equilibrium green s function technique to calculate the system s properties in a self - consistent manner .
the interaction self - energies are calculated at the hartree - fock level in the central region and at the hartree level for the crossing interaction .
our calculations are performed for different transport regimes ranging from the far off - resonance to the quasi - resonant regime , and for a wide range of parameters .
they show that a non - equilibrium ( i.e. bias dependent ) static ( i.e. energy independent ) renormalisation is obtained for the nominal hopping matrix element between the left electrode and the central region .
such a renormalisation is highly non - linear and non - monotonic with the applied bias , however it always lead to a reduction of the current , and also affects the resonances in the conductance .
furthermore , we show that the relationship between the non - equilibrium charge susceptibility and dynamical conductance still holds even in the presence of crossing interaction . |
phase transition in a given physico - chemical system is characterized by parameters like the range of the microscopic interactions , the space dimensionality @xmath1 and the dimensionality of the order parameter , often referred to the spin dimensionality @xmath2 .
there are features whose qualitative nature is determined by the universality class to which the system belongs .
short - range interactions , double and super - exchange nearest - neighbor type , classical and quantum spins @xmath2 in @xmath1-dimensional systems have been studied @xcite .
double - exchange ( de ) interaction or indirect exchange , is the source of a variety of magnetic behavior in transition metal and rare - earth compounds@xcite .
the origin of de lies in the intra - atomic coupling of the spin of itinerant electrons with localized spins @xmath3 .
this coupling favors a ferromagnetic ( f ) background of local spins and may lead to interesting transport properties such as colossal magnetoresistance .
this mechanism has been widely used in the context of manganites @xcite .
this f tendency is expected to be frustrated by anti - ferromagnetic ( af ) inter - atomic super - exchange ( se ) interactions between localized spins @xmath3 as first discussed by de gennes@xcite who conjectured the existence of canted states.in spite of recent interesting advances , our knowledge of magnetic ordering resulting from this competition is still incomplete .
although it may look academic , the one - dimensional ( 1d ) version of this model is very illustrative and helpful in building an unifying picture . on the other hand , the number of pertinent real 1d systems as the nickelate one - dimensional metal oxide carrier - doped compound @xmath0@xcite is increasing .
haldane gap @xmath4 has been observed for the parental compound @xmath5 @xmath6 ( s=1 ) from susceptibility and neutron scattering measurements . in these compounds
, carriers are essentially constrained to move parallel to @xmath7 chains and a spin - glass - like behavior was found at very low temperature @xmath8 for typical dopings @xmath9 , @xmath10 and @xmath11 . at high temperature curie - like behavior of the magnetic susceptibility was found .
the question is how physical properties change by introducing @xmath12 holes in the system . in the doped case
the itineracy of doped electrons or holes plays an important role taken into account by the double - exchange mechanism .
recently , it has been shown that three - leg ladders in the oxyborate system fe@xmath13bo@xmath14 may provide evidence for the existence of spin and charge ordering resulting from such a competition@xcite .
naturally , the strength of the magnetic interactions depends significantly on the conduction electron band filling , @xmath15 . at low conduction electron density , f polarons
have been found for localized @xmath16 quantum spins batista19982000 .
island phases , periodic arrangement of f polarons coupled anti - ferromagnetically , have been clearly identified at commensurate fillings both for quantum spins in one dimension @xcite and for classical spins in onekoshibae1999 and two dimensions @xcite .
phase separation between hole - undoped antiferromagnetic and hole - rich ferromagnetic domains has been obtained in the ferromagnetic kondo model @xcite .
phase separation and small ferromagnetic polarons have been also identified for localized @xmath17 quantum spins @xcite .
in addition to the expected f - af phase separation appearing for small super - exchange coupling , a new phase separation between small polarons ordered ( one electron within two or three sites ) and af regions for larger se coupling was found @xcite .
these phase separations are degenerate with phases where the polarons can be ordered or not giving a natural response to the instability at the fermi energy and to an infinite compressibility as well .
wigner crystallization and spin - glass - like behavior were also obtained and could explain the spin - glass - like behavior observed in the nickelate 1d doped compound @xmath0 @xcite . in this paper , we present a study of the parallel static magnetic susceptibility in an ising - like exchange model .
short - range spin - spin correlations are also presented .
our results are compared with the curie - like behavior observed at high temperature in the nickelate one - dimensional compound @xmath0 @xcite .
the paper is organized as follows . in section ii a brief description of the model is given . in section iii ,
results and a discussion are presented .
finally , our results are summarized in section iv .
the de hamiltonian is originally of the form , @xmath18where @xmath19 are the fermions creation ( annihilation ) operators of the conduction electrons at site @xmath20 , @xmath21 is the hopping parameter and @xmath22 is the electronic conduction band spin operator . in the second term
, @xmath23 is the hund s exchange coupling . here
, hund s exchange coupling is an intra - atomic exchange coupling between the spins of conduction electrons @xmath24 and the spin of localized electrons @xmath3 .
this hamiltonian simplifies in the strong coupling limit @xmath25 , a limit commonly called itself the de model . in this strong coupling limit
itinerant electrons are now either parallel or anti - parallel to local spins and are thus spinless .
the complete one dimensional de+se hamiltonian becomes , @xmath26 @xmath27 is the relative angle between localized spins at sites @xmath20 , @xmath28 defined with respect to a z - axis taken as the spin quantization axis of the itinerant electrons .
the super - exchange coupling is an anti - ferromagnetic inter - atomic exchange coupling between localized spins @xmath3 .
this coupling is given in the second term of the former equation . here
@xmath29 is the super - exchange interaction energy .
an ising - like model with itinerant electrons will be considered in this paper , i. e. @xmath30 and @xmath31 or @xmath32 . for itinerant electrons ( holes )
an electron ( hole)-single approximation will be used .
the nickelate one - dimensional parental compound @xmath33 , is basically formed of quasi one - dimensional chains of @xmath6 . @xmath34 and @xmath35 are two relevant @xmath6 orbitals in this system .
@xmath35 is basically localized while @xmath34 has finite overlap with @xmath36 orbital of the o @xcite .
so , to make contact with the nickelate one - dimensional compound @xmath0 , @xmath37 localized s=1/2 spins in the @xmath35 orbital will be considered . on the other hand
itinerant electrons @xmath38 or holes @xmath12 will be placed in the @xmath34 orbital .
the role of these electrons ( holes ) within the parental compound @xmath5 , will be considered by the de mechanism . within our ising - like model
there is an electron - hole symmetry .
exact parallel static magnetic susceptibility @xmath39 and short - range spin - spin correlations are presented using a standard canonical ensemble . to obtain @xmath39 within the electron ( hole)-single approximation is necessary to calculate eigenvalues of the following matrix @xmath40 where @xmath41 in the former equation first term is super - exchange interaction and the second one is the zeeman coupling of the localized background of s=1/2 spins .
third term is the coupling between the magnetic moment @xmath42 of the itinerant electron and the magnetic field @xmath43 .
a magnetic field was introduced to calculate @xmath39 .
@xmath44 with eigenvalues of equation ( [ hamil ] ) is easy to obtain partition function @xmath45 in the canonical ensemble within the electron - single approximation @xmath46 for one ( @xmath20 ) , two ( @xmath20 and @xmath47 ) , three ( @xmath20 , @xmath47 and @xmath48 ) and ( @xmath49 ) itinerant electrons respectively .
@xmath50 being @xmath51 boltzmann constant and @xmath52 temperature@xmath53 magnetic susceptibility is related with partition function as @xmath54 mean value of all operators can be related to partition function i. e. @xmath55 @xmath56 on the other hand , the phenomenological ising - like model was proposed because of our previous results using classical localized spins lead basically to an ising - like model @xcite .
high temperature @xmath39 will be compared with experimental results of the nickelate one - dimensional compound @xmath0 @xcite .
in this section , phase diagram , parallel static magnetic susceptibility ms and short - range spin - spin correlations are presented for a particular open linear chain of @xmath57 sites . in the thermodynamic limit
, phase diagram is shown in figure [ fig1cl ] .
this phase diagram is similar to our previous one using classical localized spins ( s=3 ) @xcite .
phase separation between ferromagnetic ( f ) @xmath58 and anti - ferromagnetic ( af ) @xmath59 phases is found for low super - exchange interaction energy . on the other hand phase separation between p2 @xmath60 and p3 @xmath61 phases and
the af phase was obtained for high @xmath62 .
because of the scalar @xmath63 spin character used in this paper canted cp3 , cp2 and t phases are not obtained in this paper @xcite .
the af phase observed at @xmath64 was previously studied for an ising ( s=1 ) and classical ( s=3 ) model respectively in references @xcite .
vs super - exchange interaction energy @xmath62 phase diagram . ] vs temperature @xmath65 for @xmath66 i.e. one itinerant electron and a typical value of the super - exchange interaction energy @xmath67 .
curie - weiss like behavior at high temperature limit can be observed .
solid line represents @xmath68 limit . ]
but for two itinerant electrons @xmath69 . ]
but for three itinerant electrons @xmath70 . ]
figures [ fig2cl ] , [ fig3cl ] and [ fig4cl ] show the inverse of the magnetic susceptibility vs temperature for one , two and three itinerant electrons respectively . solid lines in those figures represent high temperature @xmath68 limit .
curie - weiss behavior can be easily observed in those figures as @xmath71 being @xmath72 curie constant and @xmath73 curie - weiss like temperature .
( @xmath74 @xmath75 ) , ( @xmath76 @xmath77 ) and ( @xmath78 @xmath79 ) for one , two and three itinerant electrons respectively .
curie constant can be rigorously extracted for the former limit @xmath25 and @xmath80 . for this goal
it is considered @xmath37 localized spins and @xmath81 itinerant electrons . because of @xmath25 limit hilbert space
is reduced .
so there are @xmath81 and @xmath82 free particles with @xmath83 and @xmath84 energies respectively . being @xmath43 the magnetic field .
the former gives @xmath85 .
curie constant is identified like @xmath86 .
it gives @xmath87 , @xmath88 and @xmath89 for one , two and three itinerant electrons respectively i. e. @xmath90 and @xmath91 .
these values are very close to those obtained in figures [ fig2cl]-[fig4cl ] .
now , we can use our curie constant 1 + 3x to make contact with results of the nickelate one - dimensional compound @xmath0 .
@xmath92 ( s=1/2 ) for curie constant was proposed by kojima _
@xcite kojima _ et al .
_ proposed that each ca - atom introduces three @xmath16 spins .
they studied hole dopings @xmath93 and @xmath94 . in our case
these itinerant holes correspond to @xmath95 and @xmath96 itinerant electrons studied here .
it means curie constant ( 1 + 3x ) as @xmath97 and @xmath98 or simply @xmath99 if we introduce holes as kojima .
@xmath5 corresponds to @xmath100 or ne = n electrons coupled with n localized spins s=1/2 by an infinite hund s coupling . on the other hand
, @xmath101 is exactly n localized spins s=1/2 with @xmath102 .
so , the effect to introduce holes in our itinerant electron system is to reduce curie constant .
for low temperature the model proposed by kojima _
et al . _ is very close to our p3+af phase separation .
on the other hand , curie - weiss like temperature @xmath73 decreases as itinerant electron density increases .
itinerant electrons are responsible for the former f behavior because of our de interaction . short range spin - spin correlations @xmath103 at zero magnetic field can be observed in figures [ fig5cl]-[fig7cl ] for a typical value of @xmath67 and four different temperatures @xmath104 solid circles , cross , large open circles and plus symbols respectively were used . to obtain these short range correlations negative in - site
@xmath105 energies were used to pin one , two and three polarons in the linear chain as can be observed in figures [ fig5cl]-[fig7cl ] respectively . these negative in - site
energies can be related with impurities in our linear chain . for low temperature
can be clearly seen polarons of three sites in an af background .
similar polarons were found in reference @xcite by using quantum s=3/2 core spins .
this phase with disordered polarons is degenerated to our p3+af phase separation .
it means ordered polarons of three sites in an af background . at high temperature
@xmath106 polarons disperse and a very low correlation is observed . and @xmath67 .
solid circles , cross , large open circles and plus symbols respresent four different temperatures @xmath107 and @xmath108 respectively . ] but for two itinerant electrons @xmath69 . ]
but for three itinerant electrons @xmath70 .
] in the same way , figures [ fig8cl]-[fig10cl ] show short range spin - spin correlations @xmath103 for another typical value of @xmath109 and three different temperatures @xmath110 and @xmath111 . in this case
only one in - site @xmath105 energy was utilized to pin de f phase as can be seen in figures [ fig8cl]-[fig10cl ] .
for low temperature f - af phase separation can be observed .
the f phase increases as the itinerant electron density @xmath38 increases , see figures [ fig8cl]-[fig10cl ] .
the former is because of de interaction . at high temperature
the f phase disperses and a very low correlation is observed . and @xmath109 .
solid circles , cross and large open circles respresent three different temperatures @xmath110 and @xmath111 respectively . ] but for two itinerant electrons @xmath69 . ]
but for three itinerant electrons @xmath70 . ] and @xmath67 .
solid circles , cross , plus and open circles respresent four different temperatures @xmath112 and @xmath113 respectively .
fitting solid lines are also shown in the same figure . ]
but for an itinerant electron density of @xmath114 . ]
but for a chain of @xmath115 sites , one itinerant electron @xmath69 is presented . ]
it is tempting to apply our results to the magnetic properties of the hole doped @xmath0 .
doing so raises the question of the relation between quantum spins and classical spins cases .
it is clear that some properties are specific to the quantum character of the spins , in particular the haldane gap occurring in heisenberg @xmath116 chains , as in the case of un - doped @xmath33 .
however , in the doped case the itineracy of doped electrons or holes plays an important role taken into account by the double - exchange mechanism .
the essential behavior of the spin correlations in the quantum level is similar in the classical case . for the commensurate filling @xmath117 the polaronic phase @xmath118
reference @xcite is qualitatively similar to the quantum @xmath16 case .
we have calculated magnetic susceptibility for typical values of the conduction electron density to make contact with experiments@xcite . the inverse of magnetic susceptibility ( @xmath119 ) vs @xmath52 presents a complicated behavior as described in the former lines . at high temperature curie - weiss behavior was obtained . as shown , curie constant is basically t - j independent .
our ising - like results give @xmath120 or @xmath99 .
et al . _ from experimental results proposed @xmath121 ( s=1/2 ) . in our case we remove electrons from an s=1 system @xmath5 . in the case of kojima
, holes are added . in this case
our ising - like model may be can be related with experimental results .
curie - weiss temperature @xmath73 is
t - j dependent and can be related with curie - like behavior observed in this compound @xcite .
it is important to mention that the contribution related to the haldane gap in @xmath116 spin chains decreases exponentially with decreasing temperature and becomes negligible at low temperature @xmath122 @xcite .
it is difficult to identify the different contributions to the magnetic susceptibility in such a complex magnetic ground state .
of course , our comparison with the experimental results becomes irrelevant below the spin - glass transition identified to be @xmath123 .
finite size effects are taken into account to show that our @xmath57 sites are of relevance .
inverse of magnetic susceptibility vs inverse of n sites for different temperatures are shown in figures [ fig11cl ] and [ fig12cl ] for an itinerant electron density of @xmath124 and @xmath114 respectively .
fitting solid lines @xmath125 with an error of @xmath126 , 95 per cent of confidence levels are shown in the same figures . as can be seen in the same figures an error of @xmath127
is obtained if @xmath57 sites are taken into account .
the @xmath80 limit , that is n - site independent , is also compared with these thermodynamic limits , giving an error of @xmath128 @xcite .
finite size effects for a heisenberg and an ising model ( without itinerant electrons @xmath64 ) were studied in reference @xcite .
as can be seen in that reference , magnetic susceptibility is almost n - site independent at high temperature limit . in our model , because of itinerant electrons , both high and low temperature limits lead to the same qualitative behavior .
it is also presented , in figure [ fig13cl ] , short - range spin - spin correlations for one itinerant electron and @xmath115 sites @xmath129 and @xmath67 .
these results can be compared with results shown in figure [ fig6cl ] for @xmath57 sites and two itinerant electrons . the same spin - spin correlations behavior can be observed .
magnetic phase diagram for classical localized spins and an exchange model , as used in this paper , is compared with the thermodynamic limit in reference @xcite . as can be observed in that reference ,
the same magnetic phases were obtained .
of course that because of our exact results very long systems can not be studied easily because of a huge cpu time used .
in this work , we presented exact parallel static magnetic susceptibility calculations and short - range spin - spin correlations of an equivalent ising - like de+se model using large hund s coupling .
magnetic susceptibility was calculated in a region where p3-af and f - af phase separation can be found . at high temperature curie - weiss behavior and a very low correlated system
were obtained .
curie constant is basically t - j independent and could be related with the curie - like behavior observed in the nickelate one - dimensional compound @xmath0 .
finite size effects were considered to show the relevance of our finite @xmath57 system .
it is important to mention that another itinerant electronic densities and super - exchange couplings were considered .
almost the same errors were obtained and the same qualitative behavior in magnetic susceptibility and spin - spin correlations . | spin - polarons are obtained using an ising - like exchange model consisting of double and super - exchange interactions in low dimensional systems . at zero temperature , a new phase separation between small magnetic polarons , one conduction electron self - trapped in a magnetic domain of two or three sites , and the anti - ferromagnetic phase was previously reported . on the other hand
the important effect of temperature was missed .
temperature diminishes boltzmann probability allowing excited states in the system .
static magnetic susceptibility and short - range spin - spin correlations at zero magnetic field were calculated to explore the spin - polaron formation . at high temperature curie - weiss behavior
is obtained and compared with the curie - like behavior observed in the nickelate one - dimensional compound @xmath0 .
exchange and super - exchange interactions , classical spin models , phase separation 75.30.et,75.10.hk,64.75.+g |
in the 1950ies , fermi and coworkers started to measure pion - nucleon cross sections and to analyze the data in terms of partial waves , leading the way to the discovery of a large number of baryon resonances . in 1963 , in a partial - wave analysis performed at the lawrence livermore national laboratory , l. d. roper found a @xmath2 resonance at @xmath3 gev ( @xmath4 mev pion laboratory kinetic energy ) @xcite .
the result was surprizing as there were no hints for such a state and the @xmath2 scattering length is rather large and negative . in words of roper : _ i spent a much time trying to eliminate the @xmath2 resonance _ @xcite .
the particle data group estimates for the main @xmath5 properties are listed in table [ tab1 ] .
l r + + + [ 0.05 cm ] + + [ 0.2 cm ] decay modes & fraction ( @xmath6 ) + @xmath7 & 0.55 to 0.75 + @xmath8 & @xmath9 % + @xmath10 & @xmath11 % + @xmath12 & @xmath13 8 % + @xmath14 & @xmath15 % + @xmath16 & @xmath17 % + @xmath18 & @xmath19 % + considerable uncertainties are apparent , specially in the full breit - wigner width and the branching ratios to the strong - decay channels .
indeed , different values are obtained with different models , most of them built in terms of breit - wigner resonances plus background , meson - exchange or @xmath20-matrix formalisms .
for example , the recent @xmath20-matrix multichannel analysis of ref .
@xcite , which combines single and double - pion production data induced by pions and photons finds a @xmath21 % , in agreement with the pdg , but a smaller @xmath22 % and a considerably larger @xmath23 % ( to be compared to the @xmath24 5 - 10 % pdg estimate ) .
pole positions and residues allow for a parameterization of resonances in a well - defined way , free of assumptions for the background and energy dependence of the resonance part @xcite .
actually , many different studies find for the roper resonance two almost degenerate poles close to the @xmath25 threshold on two different riemann sheets of the @xmath25 channel @xcite .
the pole positions are stable against larger variations of parameters in meson - exchange mechanisms , with averaged values of @xmath26 mev and @xmath27 mev @xcite .
the second pole is a replica or shadow of the first one without strong physical implications rather than a new structure @xcite . in spite of this agreement ,
the dynamical origin of the roper poles is not clear : while in the jlms model of ref .
@xcite , they evolve from a single bare state that also gives rise to the @xmath28 , no genuine pole term is required in the jlich model @xcite .
in a simple quark model with a harmonic oscillator potential it is easy to understand why it is unexpected to have a radial excitation of the nucleon as the first @xmath0 .
the energy spectrum is given by @xmath29 with @xmath30 .
if the lowest state with @xmath31 , @xmath32 is associated with the nucleon ( @xmath33 ) , then the first excited state with @xmath34 , @xmath35 is @xmath36 and only the next one with @xmath37 , @xmath32 is an @xmath38 like the roper .
however , the first negative parity state @xmath39 @xmath40 turns out to be heavier than the @xmath1 @xmath2 .
this parity reversal pattern can not be described by successful quark models based on @xmath41 symmetry with residual color - spin interactions between quarks ( see for instance fig .
9 of ref .
@xcite ) .
some authors argue that reverse parity is an indication that at low energies the interactions among constituent quarks could be dominated by flavor - dependent goldstone boson exchange ( gbe ) ( see ref .
@xcite for a review ) . with this assumption
it is possible to obtain a good description of the low - lying baryon spectrum and , in particular , the correct level ordering between the @xmath5 and the @xmath42 , as can be seen in fig . 4 of ref .
the model has been extended to include the exchange of vector and scalar mesons to account for correlated multiple gbe , although the special nature of pseudoscalar goldstone bosons does not extend to other mesons .
besides , the special status of mesons in this model makes it difficult to achieve a unified description of both mesons and baryons @xcite .
further understanding of the nature of the roper resonance and the level ordering may be provided by lattice qcd . in a recent study , the first positive and negative parity excited states of the nucleon
have been obtained with variational analysis in quenched qcd @xcite .
the @xmath43 state is below the @xmath44 one for heavy quark masses , but the physical ordering is recovered for pion masses below 380 mev ( see fig.10 of ref .
caution should be exercised in the interpretation of this result obtained in quenched qcd and for which the identification of the @xmath43 at low quark masses , where finite lattice volume effects become significant , still remains .
if confirmed , this level crossing could support the hypothesis that there is a transition from heavy quarks , where @xmath41 symmetry with color - spin interactions works well , to light quarks where flavor - spin interactions due to gbe prevail @xcite . to circumvent the parity reversal problem ,
alternative descriptions in which the roper resonance is not ( only ) a @xmath45 state have also been proposed .
for instance , it could have a large gluonic component @xmath46 , although the masses of such hybrid states calculated with the flux - tube model are quite large ( @xmath47 mev ) @xcite . in one of its oldest representations , the roper appears as a collective vibration of the bag surface , a breathing mode .
indeed , with the skyrme model , where baryons are topological solitons of the meson nonlinear fields , a resonance was found in the breathing mode spectrum with a mass of @xmath48 mev @xcite . in line with the collective picture ,
juli - daz and riska explored the presence of @xmath49 components in the roper resonance @xcite .
they found that the confining interaction mixes the @xmath45 and @xmath50 components .
the @xmath50 admixture in the roper ranges from 3 to 25 % depending on the constituent quark mass while the @xmath51 components are negligible
. the @xmath45 component could even be totally absent in the @xmath5 as suggested by the fact that the resonance shape is dynamically generated in the jlich model from meson - baryon interactions in coupled channels , mostly from the @xmath52 s - wave interaction @xcite . finally , if the baryons are regarded as many - body systems of quarks and gluons , it is natural to expect that they could be deformed .
such a possibility was investigated in ref .
@xcite , with a deformed oscillator potential .
it was shown that low lying masses fit well to rotational spectra with the roper as an @xmath37 rotational state .
although the vast majority of the information about the @xmath5 has been extracted from the @xmath53 reaction , there are many other processes where the resonance properties can be studied and/or where the reaction mechanism can not be understood without taking it into account . some of these processes are reviewed in this section .
valuable information about nucleon resonances is encoded in the electromagnetic @xmath54 transitions , often presented in terms of helicity amplitudes connecting states with well defined helicities . in the case of the @xmath55 transition ,
two such amplitudes should be introduced , @xmath56 and @xmath57 , defined as @xmath58 here , @xmath59 is fine - structure constant , @xmath60 , @xmath61 is the four - momentum transfered to the nucleon and @xmath62 stand for the transverse and longitudinal polarizations of the virtual photon . the @xmath55 transition electromagnetic current can be parametrized with two form factors @xmath63 u(p ) \ , .
\label{current}\ ] ] this current is very similar to the nucleon one , except for the @xmath64 part . in the nucleon case ,
the form factor associated with this operator has to vanish to ensure current conservation , but not for the @xmath65 transition because the roper mass differs from the nucleon one .
introducing electric and magnetic form factors , in analogy to the sachs form factors of the nucleon and substituting eq .
( [ current ] ) in the expressions for the helicity amplitudes , one obtains that up to well known factors @xmath66 and @xmath67 @xcite .
the @xmath55 helicity amplitudes have been studied using various models with a wide diversity of results .
some of these are shown in fig .
[ hel ] , namely , the prediction from the non - relativistic quark model ( nrqm ) @xcite , the hybrid model @xcite , the light - front relativistic quark model ( lf ) calculation of ref .
@xcite , the chiral chromodielectric ( chd ) model @xcite and the extended vector - meson dominance ( evmd ) model of ref .
@xcite . ) and longitudinal ( @xmath57 ) helicity amplitudes for the @xmath55 transition calculated with various models : nrqm ( solid line ) @xcite , hybrid model ( dash - dotted line ) @xcite , lf ( dotted line ) @xcite , chd ( dashed line ) @xcite and evmd ( dash - double - dotted line ) @xcite .
the result of the global maid07 analysis @xcite is given by the thick solid line.,scaledwidth=65.0% ] the extensive @xmath0 program at jlab has provided a large amount of precision data on pion electroproduction which , together with the data from previous experiments at mit / bates and mami / mainz , has made possible the extraction of the transition helicity amplitudes at @xmath68 @xmath69 the for several resonances and , in particular , for the roper @xcite .
the result from the global maid07 analysis is also shown in fig . [ hel ] .
the comparison with the models reveals that none of them is really satisfactory .
this is an indication of the difficulties that quark models encounter in the description of the low @xmath70 @xmath69 region . at @xmath71 @xmath69 , where @xmath72 and @xmath73 are positive and decreasing ,
good agreement is obtained with relativistic quark model calculations assuming that the roper is the first radial excitation of the nucleon @xcite .
the discrepancies at low @xmath74 are interpreted as due to the missing meson cloud effects .
the importance of the pion cloud , particularly at low @xmath74 , has also been demonstrated in a recent study of electroproduction amplitudes with the simple cloudy bag model @xcite .
the pion cloud is found to be responsible for the large and negative value of @xmath72 at the photon point , while the quark dynamics becomes progressively relevant as @xmath74 increases , causing @xmath72 to change sign .
it is important to bare in mind that extraction of helicities amplitudes in both the maid @xcite and clas @xcite analyses imply certain model dependent assumptions about the resonant and non - resonant parts of the pion electroproduction amplitudes . for this reason ,
alternative methods are being pursued , like the extraction of transition form factors at the resonance poles using analytic continuation @xcite .
the excitation of the roper resonance in @xmath75 and @xmath76 reactions can only be assessed with partial wave analyses ; in the reaction cross section , the @xmath1 @xmath2 overlaps with the @xmath77 @xmath78 and the @xmath39 @xmath40 forming the so called second resonance region .
moreover , all these @xmath0 states might be masked by the prominent @xmath79 excitation since @xmath75 and @xmath76 interactions mix isospin 1/2 and 3/2 .
however , certain reactions act as filters , making the direct observation of the roper excitation possible .
an example is the @xmath80 reaction of proton target studied at saturne with a beam energy of 4.2 gev @xcite .
as the projectile has @xmath81 , the @xmath82 excitation can occur on the projectile but not on the target .
for this reason the roper excitation appears as small peak on the tail of the dominant @xmath82 excitation ( see fig . 2 of ref .
the theoretical study of hirenzaki et al .
@xcite showed that the isoscalar excitation on the proton is the dominant @xmath5 production mechanism and extracted its strength from data .
the fact that the interference with the @xmath82 excitation on the @xmath59 is important allowed to establish also the relative sign of the amplitudes .
an even clearer case of direct @xmath5 observation has been made by the bes collaboration with the decay @xmath83
@xcite . here , because of isospin conservation , the @xmath75 system is in pure isospin @xmath84 .
several @xmath0 were observed in the @xmath75 invariant mass distribution , the first of them corresponding to the roper resonance .
its mass and width , estimated with a simple breit - wigner function were found to be @xmath85 mev and @xmath86 mev respectively . as a constant width
was used in the breit - wigner , the extracted mass is close to the pole value .
the roper resonance is a vital ingredient in double - pion production reaction mechanisms . in spite of its small branching ratio , the s - wave character of the @xmath87 decay ( or @xmath88 as often denoted in the literature )
makes it a very important nonvanishing contribution at threshold .
this is the case for the @xmath89 reaction , as was shown long ago in ref .
@xcite and supported by other models .
for instance , in fig .
10 of ref .
@xcite the dotted lines denoting the results without @xmath5 are well below the full model ( and the data ) in the channels where the @xmath87 decay is allowed .
the relevance of the roper is even more dramatic in @xmath90 , where according to the model of ref .
@xcite , the isoscalar excitation of the resonance , followed by its decay into @xmath91 appears to be dominant at laboratory kinetic energies of the incident proton @xmath92 gev .
the other two important reaction mechanisms : @xmath93 excitation and @xmath94 are negligible at threshold but rise fast to become important above @xmath95 gev . in recent years , this reaction has been accurately measured at celsius and cosy . at low energies , the main features predicted by the model of ref . @xcite have been confirmed ( see for instance ref .
the situation is more involved at higher energies : an isospin analysis of the data @xcite indicates that the contribution from heavier @xmath96 states might be important .
resonances with masses up to 1.72 gev have been incorporated in the relativistic model of cao et al .
@xcite , finding large contributions from the @xmath97 and @xmath98 states .
the agreement to data is improved by reducing the @xmath99 branching ratio , in line with the findings of ref .
@xcite . the @xmath90 model of ref .
@xcite does not include interferences but , in particular , the interference between the @xmath91 and @xmath100 decay modes of the roper has been found to explain some details of the invariant mass and angular distributions for @xmath89 ( fig . 12 of ref .
@xcite ) , @xmath90 ( fig . 4 of ref .
@xcite ) and specially @xmath101 . for this later reaction
, it has been shown that the shape of the double differential cross sections measured at lampf with a neutron beam of @xmath102 gev@xmath103 @xcite can be explained by the above mentioned interference between the two - pion decay modes of the roper resonance @xcite . as shown in fig [ deut ] , by taking into account the roper
one obtains a good description of the size and energy dependence of the total @xmath101 cross section even with a rather simple model as the one of ref .
the @xmath101 reaction close to threshold has been recently investigated in the framework of chiral perturbation theory @xcite .
the reported results for the total cross section are considerably smaller than those of fig .
[ deut ] even at lower energies . as a function of the neutron laboratory momentum ( solid line ) .
the dotted line corresponds to the @xmath87 mechanism , the short - dashed line stands for the @xmath99 and the long - dashed one for the double-@xmath96 excitation ( see ref .
@xcite for details ) .
the data are from refs .
@xcite ( circle ) , @xcite ( square ) and @xcite ( triangles).,scaledwidth=64.0% ] c. amsler _ et al .
_ [ particle data group ] , phys .
b * 667 * ( 2008 ) 1 .
a. v. sarantsev _ et al .
_ , phys .
b * 659 * ( 2008 ) 94 [ arxiv:0707.3591 [ hep - ph ] ] .
m. dring , c. hanhart , f. huang , s. krewald and u. g. meissner , nucl .
a * 829 * ( 2009 ) 170 [ arxiv:0903.4337 [ nucl - th ] ] .
r. a. arndt , j. m. ford and l. d. roper , phys .
d * 32 * ( 1985 ) 1085 .
n. suzuki , b. julia - diaz , h. kamano , t. s. lee , a. matsuyama and t. sato , phys .
* 104 * ( 2010 ) 042302 [ arxiv:0909.1356 [ nucl - th ] ] . h. kamano , s. x. nakamura , t. s. lee and t. sato , phys .
c * 81 * ( 2010 ) 065207 [ arxiv:1001.5083 [ nucl - th ] ] .
s. capstick and w. roberts , prog .
* 45 * ( 2000 ) s241 [ arxiv : nucl - th/0008028 ] .
l. y. glozman and d. o. riska , phys .
* 268 * ( 1996 ) 263 [ arxiv : hep - ph/9505422 ] .
l. y. glozman , nucl .
a * 663 * ( 2000 ) 103 [ arxiv : hep - ph/9908423 ] . m. s. mahbub , a. o. cais , w. kamleh , b. g. lasscock , d. b. leinweber and a. g. williams , phys .
b * 679 * ( 2009 ) 418 [ arxiv:0906.5433 [ hep - lat ] ] .
m. s. mahbub , w. kamleh , d. b. leinweber , a. o. cais and a. g. williams , phys .
b * 693 * ( 2010 ) 351 [ arxiv:1007.4871 [ hep - lat ] ] .
n. mathur _ et al .
_ , phys .
b * 605 * ( 2005 ) 137 [ arxiv : hep - ph/0306199 ] .
s. capstick and p. r. page , phys .
d * 60 * ( 1999 ) 111501 [ arxiv : nucl - th/9904041 ] .
u. b. kaulfuss and u. g. meissner , phys .
b * 154 * ( 1985 ) 193 .
b. julia - diaz and d. o. riska , nucl .
a * 780 * ( 2006 ) 175 [ arxiv : nucl - th/0609064 ] .
o. krehl , c. hanhart , s. krewald and j. speth , phys .
c * 62 * ( 2000 ) 025207 [ arxiv : nucl - th/9911080 ] .
a. hosaka , h. toki and h. ejiri , nucl .
a * 629 * ( 1998 ) 160c . f. cardarelli , e. pace , g. salme and s. simula , phys . lett .
b * 397 * ( 1997 ) 13 [ arxiv : nucl - th/9609047 ] .
l. alvarez - ruso , m. b. barbaro , t. w. donnelly and a. molinari , nucl .
a * 724 * ( 2003 ) 157 [ arxiv : nucl - th/0303027 ] .
z. p. li , v. burkert and z. j. li , phys .
d * 46 * ( 1992 ) 70 .
p. alberto , m. fiolhais , b. golli and j. marques , phys .
b * 523 * ( 2001 ) 273 [ arxiv : hep - ph/0103171 ] .
f. cano and p. gonzalez , phys .
b * 431 * ( 1998 ) 270 [ arxiv : nucl - th/9804071 ] .
d. drechsel , s. s. kamalov and l. tiator , eur .
j. a * 34 * ( 2007 ) 69 [ arxiv:0710.0306 [ nucl - th ] ] . i. g. aznauryan _ et al . _ [ clas collaboration ] , phys .
c * 80 * ( 2009 ) 055203 [ arxiv:0909.2349 [ nucl - ex ] ] .
g. ramalho and k. tsushima , phys .
d * 81 * ( 2010 ) 074020 [ arxiv:1002.3386 [ hep - ph ] ] .
b. golli , s. sirca and m. fiolhais , eur .
j. a * 42 * ( 2009 ) 185 [ arxiv:0906.2066 [ nucl - th ] ] .
n. suzuki , t. sato and t. s. lee , arxiv:1006.2196 [ nucl - th ] .
h. p. morsch _ et al .
_ , phys .
* 69 * ( 1992 ) 1336 .
s. hirenzaki , p. fernandez de cordoba and e. oset , phys .
c * 53 * ( 1996 ) 277 [ arxiv : nucl - th/9511036 ] .
m. ablikim _ et al . _
[ bes collaboration ] , phys .
* 97 * ( 2006 ) 062001 [ arxiv : hep - ex/0405030 ] .
e. oset and m. j. vicente - vacas , nucl .
a * 446 * ( 1985 ) 584 .
h. kamano and m. arima , phys .
c * 73 * ( 2006 ) 055203 [ arxiv : nucl - th/0601057 ] .
l. alvarez - ruso , e. oset and e. hernandez , nucl .
phys . a * 633 * ( 1998 ) 519 [ arxiv : nucl - th/9706046 ] . j. patzold _ et al .
_ , phys .
c * 67 * ( 2003 ) 052202 [ arxiv : nucl - ex/0301019 ]
. t. skorodko _ et al .
_ , phys .
b * 679 * ( 2009 ) 30 [ arxiv:0906.3087 [ nucl - ex ] ] .
x. cao , b. s. zou and h. s. xu , phys .
c * 81 * ( 2010 ) 065201 [ arxiv:1004.0140 [ nucl - th ] ] .
c. l. hollas , c. r. newsom , p. j. riley , b. e. bonner and g. glass , phys . rev .
c * 25 * ( 1982 ) 2614 . l. alvarez - ruso , phys .
b * 452 * ( 1999 ) 207 [ arxiv : nucl - th/9811058 ] .
b. liu , v. baru , j. haidenbauer and c. hanhart , arxiv:1007.1382 [ nucl - th ] . f. plouin _ et al .
_ , nucl .
a * 302 * ( 1978 ) 413 .
i. bar - nir _ et al .
_ , nucl .
b * 54 * ( 1973 ) 17 . | the lightest @xmath0 state , @xmath1 @xmath2 , also known as roper resonance , has puzzled physicists for decades . a large variety of theoretical models aimed to understand its properties
have been proposed .
some of them are briefly reviewed here , together with the hadronic processes where the roper resonance is revealed or plays an important role . |
blends of polymers and nanoparticles , commonly called `` polymer nanocomposites '' ( pnc ) , have garnered much attention due to the possibility of dramatic improvement of polymeric properties with the addition of a relatively small fraction of nanoparticles @xcite . successfully making use of these materials
depends upon a firm understanding of both their mechanical and flow properties .
numerous computational and theoretical studies have examined the clustering and network formation of nanoparticles and their effect on both the structural and rheological properties of pncs @xcite . the vast majority of these efforts have focused on nanoparticles that are either spherical , polyhedral or otherwise relatively symmetric , although there are some notable exceptions @xcite . in contrast
, experiments have tended to emphasize highly asymmetric nanoparticles @xcite , such as layered silicates or carbon nanotubes .
it is generally appreciated that these highly asymmetric nanoparticles have the potential to be even more effective than spherical ( or nearly spherical ) nanoparticles in changing the properties of the polymer matrix to which they are added .
in addition to the large enhancements in viscosity and shear modulus expected from continuum hydrodynamic and elasticity theories , extended nanoparticles can more easily form network structures both through direct interaction between the nanoparticles , or through chain bridging between the nanoparticles @xcite , where a `` bridging '' chain is a chain in contact with at least two different nanoparticles .
these non - continuum mechanisms are believed to play a significant role in property enhancement , though the dominant mechanism depends on the properties considered , particle - polymer and particle - particle interactions , sample preparation , etc .
given that the majority of previous computational efforts have focused on symmetric nanoparticles , we wish to elucidate the role of _ nanoparticle shape _ in determining basic material properties , such as the viscosity @xmath0 , and material `` strength '' , ( i.e. , breaking stress ) .
computer simulations are well suited to examine the role of nanoparticle shape , since it is possible to probe the effects of changing the shape without the alteration of any of the intermolecular interactions . in this way
, the changes due to nanoparticle shape can be isolated from other effects .
such a task is complicated experimentally , since it is difficult to modify the shape of a nanoparticle without dramatically altering its intermolecular interactions . in this paper
we evaluate the viscosity @xmath0 and ultimate isotropic tensile strength @xmath1 of model pnc systems with either ( i ) highly symmetric icosahedral nanoparticles ( compact particles ) , ( ii ) elongated rod - like nanoparticles , and ( iii ) sheet - like nanoparticles .
these nanoparticles can be thought of as idealizations of common nanoparticles , such as gold nanoparticles and fullerenes ( polyhedral ) , nanotubes and fibers , and nanoclay and graphene sheet materials , respectively .
our results are based on molecular dynamics ( md ) computer simulations , using non - equilibrium methods to evaluate @xmath0 @xcite , and exploiting the `` inherent structure '' formalism to determine @xmath1 @xcite .
we find that the rod - like nanoparticles give the largest enhancement to @xmath0 , which we correlate with the presence of chains that bridge between the nanoparticles .
the sheet nanoparticles offer the weakest increase in @xmath0 , and correspondingly have the smallest fraction of bridging chains . for the ultimate isotropic strength @xmath1
, we find opposite results : the sheets provide the greatest reinforcement , while the rods the least . for both of these properties , the property changes induced by the icosahedral nanoparticles fall between those of the extended nanoparticles .
the present simulations are idealized mixtures of polymers and nanoparticles in which the polymer - nanoparticle interactions are highly favorable so as to promote nanoparticle dispersion .
moreover , we have chosen to work at relatively high temperature in order to avoid contributions to @xmath0 from the complex physics of slowing dynamics approaching the glass transition .
previous work @xcite has shown that polymer - surface interaction effects in this low temperature range can alter , and potentially dominate the nanocomposite properties .
we also limit the range of chain length @xmath3 studied to avoid effects of significant polymer entanglement .
these limitations on interaction , temperature , and chain length are advantageous in order to develop a clear understanding of the origin of the observed changes in properties .
such a reference calculation provides a reference starting point to understand behavior when these constraints are relaxed . with this in mind , caution is needed when comparing these results with experimental data where these complicating additional factors may be present along with other possible effects , such as crystallization or phase separation .
we organize this paper as follows : in section [ sec : simulation ] , we describe the details of the model and method , focusing on the differences between the nanoparticle types used in each system .
section [ sec : composite_rheology ] describes our investigation of the rheological properties of the nanocomposites , while section [ sec : isotropic_tensile_strength ] considers the effects of shape on @xmath1 .
we conclude in section [ sec : conclusion ] .
to directly compare to experiments , it is desirable to use as realistic a molecular model as possible . while a chemically accurate md simulation is possible in principle , it is often more difficult to identify basic physical trends with such models .
such attempts at chemical realism are also demanding in terms of the computational times required , which restricts the class of problems which can be investigated .
coarse - grained models of polymeric materials provide a good compromise between the the opposing needs of realism and computational feasibility .
such models can reproduce qualitative experimental trends of nanocomposites , but precise quantitative predictions can not be expected @xcite . building on nanocomposite models introduced before @xcite , we study a coarse - grained model for polymers @xcite and nanoparticles that allow us to consider pnc systems over a wide range of physically interesting systems . here
we consider several nanoparticle shapes built by connecting spherical force sites .
we perform md simulations of systems consisting of a small fraction of model nanoparticles in a dense polymer melt . for reference
we also simulate the corresponding pure polymer melt .
the polymers are modeled via the common `` bead - spring '' approach , where polymers are represented by chains of monomers ( beads ) connected by bond potentials ( springs ) @xcite .
all monomers interact via a modified lennard jones ( lj ) potential @xmath4 - v_{\rm lj}(r_c ) & r_{\rm ij } \le r_c \\ 0 & r_{\rm ij } > r_c \end{array } \right .
, \label{equation : vlj}\ ] ] where @xmath5 is the distance between two monomers , where @xmath6 is the depth of the well of the lj potential and @xmath7 is the monomer size
. the potential is truncated and shifted at @xmath8 , so that the potential and force are continuous at the cutoff . bonded monomers in a chain
interact via a finitely extensible , non - linear elastic ( fene ) spring potential @xcite @xmath9 , \label{equation : vfene}\ ] ] where @xmath10 and @xmath11 are adjustable parameters that have been chosen as in ref .
@xcite . since we do not aim to study a specific polymer , we use reduced units where @xmath12 ( @xmath13 is the monomer mass ) .
length is defined in dimensionless units relative to @xmath7 , time in units of @xmath14 and temperature @xmath15 is expressed in units of @xmath16 , where @xmath17 is boltzmann s constant .
we use three different types of nanoparticles for our calculations .
[ fig : nanoparticles ] shows representative images of the nanoparticles .
the first type of nanoparticle , an icosahedron , was previously studied in ref .
@xcite which focused on factors controlling nanoparticle dispersion and low temperature effects on transport for a similar polymer matrix , respectively .
practical realizations include fullerene particles , primary carbon black particles , quantum dots , and metal nanoparticle additives .
the nanoparticle force sites interact with each other via an identical @xmath18 given in eq .
( [ equation : vlj ] ) , with @xmath12 . to maintain the icosahedral shape , the force site at each vertex
is bonded to its 5 nearest neighbors via a harmonic spring potential @xmath19 where @xmath20 and @xmath21 equals the minimum of the force - shifted lj potential , approximately @xmath22 . to further reinforce the icosahedral geometry ,
a central particle is bonded to the vertices with the same potential eq .
( [ equation : vharm ] ) with the same value of @xmath23 , but with a slightly smaller preferred bond length , @xmath24 , the radius of the sphere circumscribed around an icosahedron .
the resulting nanoparticles have some flexibility , but are largely rigid , thereby preserving their icosahedral shape .
the second type of nanoparticle is a semiflexible rod represented by 10 lj force sites with neighboring monomers bonded by the same @xmath25 as used for the polymers .
we choose the shape to represent nanoparticles such as carbon nanotubes or nanofibers .
carbon nanotubes typically have lengths up to several @xmath26 m and a diameter of 1 nm to 2 nm ( single - walled tubes ) or 2 nm to 25 nm ( multiwalled tubes can have even larger diameters ) @xcite .
unfortunately , such large rods are not feasible to simulate , as the system size needed to avoid finite size effects exceeds current computational resources . as a compromise , we simulate rods with a length - to - diameter ratio of 10 .
there is an additional bond potential between nanoparticle force sites @xmath27 where @xmath28 is the angle between three consecutive force sites .
this imparts stiffness to the rod .
we choose @xmath29 , so that like the icosahedra , the rods have some flexibility , but are largely rigid .
the third type of nanoparticle is square `` sheet '' comprised of 100 force sites . while the sheets are obviously different from the rods , the sheets have the same aspect ratio as the rods , where aspect ratio is defined by the ratio of the largest and smallest length scales .
aspect ratio is often considered to be one of the most important properties for anisotropic particles , apart from the interactions with the surrounding matrix .
the sheets represent a coarse - grained model of clay - silicate nanoparticles @xcite , and in the literature these objects are also termed `` tethered membranes '' @xcite .
each monomer in the @xmath30 array has non - bonded interactions described by the same lj forces as the polymers .
the monomers are also bonded to their 4 nearest neighbors via the same fene bond potential of eq .
( [ equation : vlin ] ) .
the particles at edges and corners of the sheet are only bonded to 3 and 2 neighboring particles , respectively . like the rods
, the sheets are stiffened with a @xmath31 potential to prevent the nanoparticle from folding in on itself .
the sheets also include a perpendicular bonding potential @xmath32 which limits distortions from a square geometry . without the potential @xmath33 , the sheet can deform into a rhombus . as in ref .
@xcite , we choose @xmath34 and @xmath35 for the sheets .
thus far , we have not defined a nanoparticle - monomer potential . in previous work
@xcite , the same lj potential of eq .
( [ equation : vlj ] ) was used for nanoparticle - monomer interactions , with the well depth @xmath36 replaced by @xmath37 .
the soft @xmath38 form of the attraction makes it relatively easy for monomers which are first and second nearest neighbors of a nanoparticle to exchange . as a result ,
the chains can `` slide ''
relatively easily along the nanoparticle surface , thereby reducing any benefits of networks built by chain bridging @xcite . to make monomer exchange at the surface less favorable
, we use a `` 12 - 24 '' potential @xmath39 rather than the standard `` 6 - 12 '' powers of @xmath40 .
thus , the the nanoparticle - monomer interactions are of the form @xmath41 - v_{12 - 24}(r_c ) & r_{\rm ij } \le r_c \\ 0 & r_{\rm ij } > r_c \end{array } \right . , \label{equation : deepwell}\ ] ] where @xmath42 and @xmath43 .
this potential has shorter range and stronger forces binding the first neighbor monomers to the nanoparticle . so that the total energy of the potential well , i.e. @xmath44 , is the same as that of the lj potential used in ref .
@xcite , we choose @xmath45 . in this way , the total potential energy is comparable to that used in ref .
@xcite with @xmath46 , which led to well - dispersed nanoparticles in that study .
thus , we expect our systems will be well - dispersed .
however , we shall see this is _ not _ entirely the case for the systems with sheet nanoparticles .
figure [ fig : systems ] shows snapshots of a typical configuration for each system studied .
we generate initial configurations using the same approach as in ref .
@xcite , namely growing vacancies so that the nanoparticles can be accommodated . however , this process generates artificial initial configurations , and so we generate subsequent `` seed '' configurations by simulating at @xmath47 where reorganization occurs on relatively short time scales . depending on the nanoparticle type , we extract independent starting configurations every @xmath48 to @xmath49 time steps .
subsequently , we make the necessary changes in density and chain length before cooling and relaxing at @xmath50 .
a possible concern is that at @xmath50 , monomers will stick to the nanoparticle surface on a time scale that is long compared to the simulation , since @xmath45 ; however , we confirmed that monomers of chains exchange with the nanoparticle surface many times over the simulation .
the equilibration time depends on the system type and is directly related to particle size .
pure systems relax relatively quickly , needing only @xmath48 time steps .
icosahedral systems need roughly @xmath49 time steps before reaching thermodynamic equilibrium .
the rods and sheet nanoparticles are much larger , and thus diffuse more slowly . as a result , these systems require in excess of @xmath51 time steps to reach equilibrium .
given that equilibration requires more than @xmath49 time steps at @xmath50 , equilibration under conditions of entanglement or supercooling would be computationally prohibitive .
we integrate the equations of motion via the reversible reference system propagator algorithm ( rrespa ) , a multiple time step algorithm used to improve simulation speed @xcite .
we use a basic time step of 0.002 for a 3-cycle velocity verlet version of rrespa with forces divided into `` fast '' bonded ( @xmath25 , @xmath52 , @xmath31 , @xmath33 ) and `` slow '' non - bonded ( @xmath18,@xmath53 ) components .
the temperature is controlled using the nose - hoover method where the adjustable `` mass '' of the thermostat is selected to match the intrinsic frequency obtained from theoretical calculations of a face - centered cubic lj system @xcite .
to study rheological properties , we shear equilibrated configurations using the sslod equations of motion @xcite , integrated using the same rrespa algorithm used for equilibrium simulation . the sllod method generates the velocity profile for couette flow . if we choose the flow along the @xmath54-axis and the gradient of the flow along the @xmath55-axis , the shear rate dependent viscosity is given by @xmath56 where @xmath57 is the average of the @xmath54-@xmath55 components of the pressure tensor ( sometimes also called the stress tensor ) , and @xmath58 is the shear rate @xcite .
we limit our simulations to small enough shear rates to avoid potential instabilities associated with breaking fene bonds in the simulation @xcite .
we choose a loading fraction @xmath59 , defined by the ratio of the number of nanoparticle force sites to the total number of system force sites ; values of this order of magnitude are common in experimental studies .
@xcite since all force sites have the same diameter , @xmath60 should also be roughly equal to the volume fraction . due to the fact both polymers and nanoparticles consist of a discrete number of force sites
, @xmath60 varies slightly between systems . for the icosahedral systems
@xmath61 , and for the rods and sheets @xmath62 . the system size for all three nanoparticle systems is much larger than either the radius of gyration of the polymers , or the largest nanoparticle dimension , to avoid finite size effects . the number of chains @xmath63 ranges from @xmath64 in the smallest system up to @xmath65 for the largest systems . for each system
we examine chain lengths @xmath66 , @xmath67 , and @xmath68 . for this polymer model , the entanglement length @xmath69 30 to 40 @xcite .
hence , only the longest chains studied may exhibit any effects of entanglement and the effects should be small in the present work .
we reiterate that effects of dynamics due to proximity to the glass transition should not play a significant role at the relative high @xmath15 of our systems .
a summary of the system parameters can be found in table [ table1 ] .
in this section , we evaluate the role of nanoparticle geometry on the magnitude of @xmath0 and clarify the influence of chain bridging on @xmath0 for a range of shear rates ( @xmath70 , 0.007 , 0.01 , and 0.02 ) and chain lengths . in the nanocomposite simulations of @xcite ,
@xmath0 became nearly constant at @xmath71 in other words , we are near the newtonian limit where @xmath0 approaches a stationary value for @xmath72 .
thus , we expect our lowest shear rates approach the newtonian regime ; in this sense , the shear rates we use are fairly low . however
, if we consider @xmath73 in physical units , these rates would be extremely high by experimental standards .
ideally , we would estimate @xmath0 in the @xmath74 limit directly ( for example , by using an einstein or green - kubo relation ) , but the accurate evaluation of @xmath0 using these methods is difficult @xcite .
figure [ fig : shear_figure_1 ] shows @xmath0 for the three nanocomposites and pure polymer as a function of @xmath58 .
these data show a clear change in @xmath0 between the systems for all @xmath58 simulated , with the rods having the largest @xmath0 , followed by the icosahedra , sheets , and finally the bulk polymer .
the decrease of @xmath0 with increasing @xmath58 is indicative of shear thinning .
we note that @xmath0 is not constant at the lowest @xmath58 , indicating we have not yet reached the newtonian regime .
nonetheless , it is clear that the pure polymer has a viscosity approximately an order of magnitude smaller than any of the composites .
comparable differences in @xmath0 have been observed in experiments @xcite .
this demonstrates an enhancement of @xmath0 through the addition of nanoparticles , which is often a desired goal of adding nanoparticles to a polymer melt .
while the figure only shows @xmath0 for the @xmath75 and @xmath76 , other chain lengths and densities indicate the same trend .
figure [ fig : shear_figure_2 ] ( a ) shows the effect of varying @xmath3 on @xmath0 at @xmath76 and @xmath77 .
an increase in @xmath0 with increasing @xmath3 is expected from basic polymer physics since the chain friction coefficient increases with @xmath3 @xcite .
interchain interactions and `` entanglement '' interactions enhance this rate of increase since the friction coefficient of each chain increases linearly with @xmath3 . while it is clear that addition of nanoparticles increases @xmath0 of the resulting composite , the reasons for this are not obvious . since the pure polymer has a relatively low @xmath0 it is reasonable to assume that the rigid nanoparticles themselves inherently raise @xmath0 as in any suspension of particles in fluid matrix . to separate out this continuum hydrodynamic effect of the nanoparticles from the effects specifically due to polymer - nanoparticle interactions ,
we calculate intrinsic viscosity @xmath78 $ ] for the individual nanoparticles .
specifically , continuum hydrodynamics provides an estimate of the incremental change of @xmath79 through the addition of particles to a fluid , @xmath80 \phi + \mathcal{o}(\phi^2 ) \label{equation : intrinsic_viscosity}\ ] ] where the velocity of the fluid is taken to be zero on the surface of the particle ( `` stick '' boundary conditions ) .
such expansions are only quantitatively reliable for @xmath81 for nearly spherical particles ; for larger @xmath60 , we expect eq .
( [ equation : intrinsic_viscosity ] ) to recover only qualitative variations in @xmath82 . for reference ,
@xmath78 = 5/2 $ ] for spheres in three dimensions @xcite .
it is also known that @xmath78 $ ] is generally larger for isotropically oriented spherical shapes @xcite .
@xcite have reviewed the modeling of polymer composites containing extended particles based on continuum mechanics .
this work also considers the effect of particle clustering on @xmath0 which is important under poor dispersion conditions . to compare the predictions of hydrodynamic theory to our simulation
, we use the zeno package to calculate @xmath78 $ ] for our particles @xcite .
the computational method involves enclosing an arbitrary - shaped object within a sphere and launching random walks from the border or launch surface .
the probing trajectories either hit the object or return to the launch surface . from these path integration calculations ,
@xmath78 $ ] for the rigid particles can be estimated for objects of essentially arbitrary shape @xcite .
we determine @xmath78 $ ] of the nanoparticles as an average over @xmath83 different confirmations found in the equilibrium system as they adopt a range of shapes due to bond flexibility . using zeno ,
we find that the icosahedral nanoparticles have @xmath78 = 3.93 $ ] , the rods have @xmath78 = 16.3 $ ] , and the sheets have @xmath78 = 10.8 $ ] .
the intrinsic viscosity of a regular icosahedron has been estimated to be 2.47 @xcite .
the discrepancy between this value and our icosahedral nanoparticles arises from the fact that our nanoparticles are not regular solids , but rather consist of a group of spherical force sites with icosahedral symmetry .
we can use eq .
( [ equation : intrinsic_viscosity ] ) to predict @xmath79 for the md simulations , and find ( for the @xmath75 systems ) @xmath84 for the icosahedra , @xmath85 for the rods , and @xmath86 for the sheets .
such a prediction underestimates the @xmath75 md results by @xmath87 for the icosahedra , @xmath88 for the rods , and @xmath89 for the sheets .
clearly this hydrodynamic estimate of @xmath79 for our nanocomposites is not adequate , even accounting for the order - of - magnitude nature of the estimate based on eq .
( [ equation : intrinsic_viscosity ] ) .
consistent with the full md simulations , the rods have the largest value , but in contrast to the md simulation , the order of the sheets and icosahedra are reversed . while the larger @xmath78 $ ] value of the rods are generally consistent with a larger value of @xmath0 , the reversal of the sheet and icosahedra @xmath78 $ ] values indicates that nanoparticle and/or polymer interactions play a role in the overall viscosity .
moreover , the continuum theory indicates that @xmath79 should be _ independent _ of @xmath3 , while we see from fig .
[ fig : shear_figure_2 ] ( b ) that @xmath79 strongly depends on @xmath3 .
hence we conclude that the continuum model of @xmath0 does not provide an adequate explanation of the changes we see in our simulations .
evidently , we must also examine other contributions to the nanoparticle viscosity .
one possibility that has recently been discussed @xcite is that slip rather than stick boundary conditions might provide a better description of nanoparticle boundary conditions .
however , this would result in a _ decrease _ of the predicted value of @xmath79 @xcite relative to the stick case , which is in the wrong direction for explaining our results . at a molecular level
, the boundary conditions are clearly neither stick nor slip , as particles have finite residence times at the surface , although it is difficult to determine hydrodynamic boundary conditions directly from molecular considerations .
nanoparticle - polymer interactions and nanoparticle clustering are evidently factors that might cause the property enhancements we observe , and in the next section we focus on this possibility .
we next turn to the structural results obtained from md simulation to better quantify the relationship of polymer and nanoparticle structure to @xmath0 .
we first calculate the fraction of nanoparticles in contact with other nanoparticles , @xmath90 .
nanoparticles are said to be `` in contact '' if one or more comprising force sites of a nanoparticle are within the nearest neighbor distance of a force site belonging to another nanoparticle .
the nearest - neighbor distance is defined by the first minimum in the radial particle distribution function @xmath91 ; the first minima is at @xmath92 for all systems .
figure [ fig : np - np ] shows @xmath90 as a function of @xmath3 for @xmath93 and for both @xmath94 and @xmath95 . while it is not readily apparent from fig .
[ fig : systems ] ( c ) , the polymers `` intercalate '' between the sheets to some degree , and hence , the nanoparticles are not actually in direct contract with each other , resulting in a very low value of @xmath90 for the sheets . with the exception of the @xmath96 equilibrium system , the sheets have the fewest nanoparticle - nanoparticle contacts . for the systems under steady shear
, @xmath90 evidently stays fairly constant as @xmath3 varies .
while the order of the systems for @xmath90 is consistent with the trend for @xmath0 , the similarity between the rods and icosahedra suggests that the fraction of nanoparticles alone is not responsible for determining composite viscosity . of note , is the large fraction of nanoparticle contacts for the @xmath96 sheet system at equilibrium .
more in - depth analysis shows that the sheets favor a clustered state at lower density , i.e. chains no longer fully intercalate between the sheets .
both the @xmath97 and @xmath98 sheet systems have intercalating polymers , while the @xmath96 case does not .
the tendency of the sheets to stack is due in part to entropic interactions that result from depletion attractions between the sheet ; the depletion interactions arise due to physically adsorbed polymers on the sheet surface , much like the depletion interactions of polymer coated colloids .
@xcite this interaction is most pronounced for the sheets , presumably due to the large and relatively flat surface of a sheet , which more effectively reduces the entropy of chains at the surface .
given that nanoparticle - nanoparticle contacts do not appear to be the origin of the relative ordering , we next consider the role of polymer - nanoparticle interactions . to gauge the possible role of such interactions
we calculate the fraction of polymer chains in direct contact with a nanoparticle , @xmath99 . using the same nearest - neighbor criteria as used to determine nanoparticle - nanoparticle contacts , we compute @xmath99 as a function of @xmath3 for @xmath76 at @xmath94 and @xmath95 in fig .
[ fig : fraction ] .
we find that @xmath99 is approximately the same for both the quiescent and sheared systems .
there are also clear trends : @xmath99 increases with increasing @xmath3 and for every value of @xmath3 the system containing the rods have the largest value of @xmath99 , followed by the system containing the icosahedra , and finally the system containing the sheets .
these same trends in @xmath0 appear in fig .
[ fig : shear_figure_2 ] suggesting that the nanoparticle - polymer contacts indeed have a significant effect on relative magnitude of @xmath0 .
the correlation of @xmath0 with @xmath99 is consistent with the idea that bridging between nanoparticles play a major role in determining the rheological properties of polymer nanocomposites @xcite .
thus , we extend our analysis to test for a correlation between @xmath0 and chain bridging .
we define a `` bridging chain '' as a chain that is simultaneously in contact with two or more nanoparticles . figure [ fig : bridging ] shows @xmath100 as a function of @xmath3 for @xmath76 .
the trend in @xmath100 is consistent with both @xmath101 and @xmath0 , showing an increase in @xmath102 with increasing @xmath3 as well as a clear ordering between systems for every chain length .
although not shown , the results seen in fig . [ fig : bridging ] occur for every value of @xmath103 that we have simulated .
hence , the fraction of bridging chains seems to be a useful `` order parameter '' for characterizing the polymer - nanoparticle interactions .
if the polymer - nanoparticle interactions were sufficiently small , the bridging would not be expected to play significant role , and it is likely that the continuum hydrodynamic approach would be applicable .
similar to the notion of bridging is the idea that the nanoparticles can act as transient cross - linkers that can lead to effects equivalent with the formation of higher molecular weight chains @xcite . to test this idea
, we define an `` effective chain '' as the collection of chains that are connected by nanoparticles .
we then define an effective chain length @xmath104 as the mean mass of chains connected by the nanoparticles .
[ fig : effective_chains ] shows that @xmath104 is almost an order of magnitude larger than @xmath3 .
hence , we expect that the largest contribution to the @xmath0 increase comes from this effect . even in the simplest rouse theory ,
an order of magnitude increase in @xmath3 would lead to an order of magnitude increase in @xmath0 .
the formation of longer effective chains is expected to lead to entanglement interactions that would further amplify the @xmath0 increase , as emphasize by ref .
@xcite . however , this entanglement contribution is hard to quantitatively interpret in the present context .
the ultimate isotropic tensile strength @xmath1 of a material is defined as the maximum tension a homogeneously stretched material can sustain before fracture . while this definition can be directly probed experimentally , the situation is less straightforward and computationally accessible in an md simulation . ref .
@xcite has developed an approach to estimate @xmath1 based on potential energy - landscape ( pel ) formalism that is accessible by md simulation .
although the pel is complex and multidimensional , one can envision it as a series of energy minima connected by higher energy transition pathways . by definition ,
the minima , or inherent structures ( is ) are mechanically stable , i.e. there is no net force at the minimum .
the method of ref .
@xcite relates @xmath1 to the maximum tension the is can sustain , determined by an appropriate mapping from equilibrium configurations on the pel . in other words , @xmath1 is the upper bound on the tension at the breaking point of an ideal glass state in the @xmath105 limit . here
we evaluate @xmath1 for the various possible nanoparticle geometries . as a cautionary note ,
we point out that this method has not been directly validated by comparison with experiments ; thus conclusions drawn from these results should be considered tentative . ref .
@xcite discusses an alternate approach using the pel to evaluate the elastic constants . to determine @xmath1
, we must first generate the energy minimized configurations from the equilibrium configurations . for a system of @xmath3 atoms in the canonical ensemble , the most commonly used mapping from equilibrium configurations to the is takes each force site along the steepest decent path of the energy of the system .
this configurational mapping procedure corresponds to the physical process of instantaneous cooling to the @xmath106 limit to obtain an ideal glass with no kinetic energy .
therefore , by sampling thermally equilibrated configurations and minimizing their energies we can estimate the average inherent structure pressure , @xmath107 , which will be negative when the system is under tension .
the maximum tension then defines @xmath1 .
the value of @xmath107 has been found to be weakly dependent on the @xmath15 at which the sampling is performed .
for example , while cyclopentane has a strongly temperature dependent structure , ref .
@xcite has shown that equilibrium @xmath15 only weakly effects @xmath107 .
since the structures in our systems do not display such dependence , we expect our results also to be independent of the starting equilibrium @xmath15 .
thus , we can use the same systems at @xmath50 that we used to determine @xmath0 and still obtain reliable results for @xmath1 even though the starting configuration is a highly fluid state .
this calculation of @xmath1 should be an _ upper bound _ for the tensile strength that would be obtained for any finite @xmath15 system , and hence is referred to the as the ultimate tensile strength .
the energy minimization process eliminates the high frequency effects that would normally be associated with such high @xmath15 states and thus the method refers to the strength of an ideal glass material having essentially no configurational entropy .
figure [ fig : congrad_n20 ] shows the @xmath107 curves generated for the @xmath108 variant of each system .
we find that @xmath107 decreases at large @xmath103 , until a minimum value is reached .
the density of the minimum , referred to as the `` sastry density '' @xmath109 @xcite , characterizes the density where the system fractures , and voids first start to appear in the minimized configurations . as @xmath103 decreases below this point , @xmath107 begins to rise .
since this is the greatest tension achievable , we have @xmath110 . for pure , icosahedral , and rod systems
we find @xmath111 , while for the sheets it is much lower @xmath112 .
similarly , @xmath1 is considerably larger for the sheets than for the other systems .
we find that the increase in @xmath1 for the sheets is not as sizable as observed experimentally @xcite .
this may be related to the fact that ( experimentally ) the chains often form stronger associations with layered silicates than in our simulations ; additionally , the chain lengths we examine are small compared to experiments , and we will see that @xmath1 increases with @xmath3 .
we note that at this chain length , the icosahedra and rods actually _ reduce _ the strength of the nanocomposite as compared to the pure melt . to quantify the chain length dependence of @xmath1 , we plot the minimum of each @xmath107 curve as a function of @xmath3 in fig . [
fig : total ] . for the pure polymers ,
we find that @xmath1 _ decreases _ with increasing chain length .
this is non - trivial , since one might navely expect longer chains to exhibit more interchain coupling .
this is the same trend with chain length as observed in ref .
@xcite for @xmath113-alkanes .
it is reassuring that simple bead - spring model leads to similar results but this does not help us build intuition about the physical meaning of these results .
reference @xcite finds a maximum @xmath1 for @xmath114 , and a tendency for @xmath115 to saturate near @xmath116 . since the smallest chain we simulate is @xmath66 , we are above the regime where this feature occurs .
consistent with these facts , @xmath1 decreases with @xmath3 and @xmath109 is roughly independent of @xmath3 .
most importantly , fig .
[ fig : total ] shows that the addition of the nanoparticles _ reverses _ the @xmath3 dependence of @xmath1 when compared to the pure system .
thus , while the presence of icosahedral or rod - like nanoparticles decreases the material strength for most chain lengths studied , the trend of @xmath1 is increasing , and if this continues , @xmath1 will surpass the pure melt for all nanoparticle shapes at large enough @xmath3 .
indeed , @xmath1 for the icosahedra nanocomposite already exceeds that of the melt at @xmath117 . to better understand the predicted @xmath3 dependence of @xmath1
, we perform a parallel analysis to the structural analysis of shear runs discussed above . the analysis of figs .
[ fig : np - np]-[fig : bridging ] focused on @xmath76 .
we relate the structure to @xmath1 by examining the structure at @xmath109 .
evidently , @xmath109 differs for the sheets in relation to the other systems studied , and it is possible that @xmath103 dependent changes in connectivity properties are responsible for the difference of @xmath1 .
thus , we calculate the quantities @xmath100 , @xmath90 , and @xmath99 at @xmath109 for each system using the equilibrium and is configurations at the sastry density @xmath109 of the system .
the results for the equilibrium and is configurations show the same qualitative trends , so we will present only data for the is .
we first focus on @xmath100 because our investigations into the rheological properties suggested that bridging chains played a large role in determining @xmath0 . however , when we plot @xmath100 in fig .
[ fig : bridging_minimized ] for each system at @xmath109 , we see that the @xmath3 dependence is the _ reverse _ of that seen for @xmath1 , even though this quantity evidently follows @xmath0 . even for the sheets which have a @xmath109 significantly lower value of than the other systems ,
the ordering of @xmath100 among the systems is the same as for @xmath100 in the case of the @xmath0 calculations .
the sheet composites have the smallest @xmath100 , followed by those with icosahedra , and those with rods .
while bridging chains do increase with increasing @xmath3 , @xmath100 does _ not _ seem to be a major factor in the relative value of @xmath1 between the composites .
for example , the @xmath66 nanocomposites with icosahedra and sheets have almost the same value of @xmath100 , yet they have significantly different values for @xmath1 .
the fewer bridging chains in the systems with square sheets suggests that the sheets may be clustered , and we find that this is indeed the case for @xmath117 ( fig . [
fig : np - np ] ( a ) ) .
thus , we plot @xmath90 in fig .
[ fig : np - np_minimized ] and find indeed that @xmath118 for the sheets .
the sheets prefer a stacked state at low @xmath103 due to entropic interactions with polymers which leads to a reduced explicit energetic interaction with the surrounding polymers .
in particular , the relative ordering of @xmath90 matches that of @xmath1 .
navely , this would seem to suggest a potential correlation between @xmath90 and @xmath1 .
however , this apparent correlation is problematic .
firstly , the @xmath3 dependence of @xmath1 and @xmath90 are different , namely @xmath1 monotonically increases , while @xmath90 is roughly constant .
secondly , if the nanoparticles interactions are the origin of the increased strength , then one might expect to find the largest effect when the nanoparticles are well dispersed @xcite .
this is not the case , since the clustered sheets give the _
largest _ effect . ref .
@xcite demonstrated that the rupture of the system occurs in `` weak spots . '' to confirm that the nanoparticles actually impart additional strength , we visually examined the location of fracture in our systems at @xmath109 . to find empty spaces that are at least the size of a particle ,
we discretize the system into a cubic lattice of overlapping spheres , each with a radius @xmath119 and a nearest neighbor separation @xmath120 .
we then identify spheres that do not contain any system force site . by adjusting the parameters @xmath119 and @xmath120
we can ensure that the voids we find are at least large enough for a single particle .
since the force sites of both the nanoparticles and polymers have diameter @xmath121 , we must choose @xmath122 to have physical relevance .
we find that values of @xmath123 and @xmath124 work best for visualizing voids .
figure [ fig : voids ] shows that the voids in the system occur in regions of pure polymer for the icosahedral system .
the same is also true for the rods and square sheets .
quantitative analysis of the force sites within nearest - neighbor distance @xmath125 of the voids confirmed this suggestion , with over 99 % of the resulting force sites belonging to polymer chains .
thus , while the nanoparticles impart an increased strength to the nanocomposite , the correlation to @xmath90 is not apparently causal .
evidently , there is something more subtle controlling the magnitude of @xmath1 in the sheet filled nanocomposite that we have not yet identified . to complete the structural analysis
we calculate @xmath99 for the is .
figure [ fig : fraction_minimized ] shows that the sheet nanoparticles are not in contact with many polymers , consistent with the fact that @xmath118 . in fact , after minimization the difference in @xmath99 between the rods and the sheets is even larger than observed in the equilibrium configurations shown in fig .
[ fig : fraction ] . figure [ fig : fraction_minimized ] shows that the relative ordering of @xmath99 is _ opposite _ to that of @xmath1 , hence the number of nanoparticle - polymer contacts alone also does not provide a good indicator of @xmath1 .
a potential clue to the increase of @xmath1 of the sheet nanocomposite is the fact that @xmath109 is _ significantly smaller _ than for the other systems .
this allows the sheet composites to undergo a greater deformation before fracture .
this large increase in both the strength and toughness of the polymer matrix with incorporation of nanoparticles is reminiscent of the changes in the properties of natural and synthetic rubbers with the inclusion of carbon black and nanofiller additives @xcite .
extraordinarily large increases in both the strength and toughness of materials have recently been observed in the case of exfoliated clay sheets dispersed in the polymer polyvinylidene fluoride ( pvdf ) @xcite .
although some of this change is associated with the modification of the crystallization morphology by the clay nanoparticles , this does not explain in itself the observed toughening mechanism .
it is known that nanofiller particles can behave as temporary cross - linking agents that impart viscoelastic characteristics to the fluid .
gersappe @xcite further emphasizes the role of this transient network formation in impeding cavitation events that initiate material rupture and the potential significance of this phenomenon in understanding the nature of biological adhesives ( abalone ) , fibers ( spider silk ) , and shell material ( nacre ) @xcite .
gersappe further suggests that the relative mobility of nanoparticles has a role on the toughening of materials .
however , such reasoning is inconsistent with the fact that the sheet nanoparticles lead to the strongest material while being the _ least mobile _ of our additives .
recent work indicates that flexible sheet - like structures , such as studied here , are characterized by a _
negative _ poisson ratio @xmath126 @xcite , _ i.e. _ the material expands normal the the direction of stretching , as opposed to normal materials which expand in the same direction as the stretching . the formation of a composite material with negative poisson ratio ( `` auxetic '' materials ) is expected to lead to a reduction of the poisson ratio of the composite as a whole @xcite .
a large reduction in @xmath126 is expected to give rise to materials that are strong and fracture resistant @xcite .
related to this , we observe in our simulations that the voids that form in the sheet - filled nanocomposites tend to be fewer and smaller than those for the other nanoparticles at @xmath109 .
as we further reduce @xmath103 for the sheet system , the voids grow more numerous as opposed to growing larger as in the other nanocomposites .
microvoid formation has also been established as a mechanism for toughening polymer materials @xcite .
previous work devoted to understanding the high impact strength of polycarbonate and other glassy polymers has likewise emphasized the importance of large `` free volume '' within the polymer material as a necessary condition for large toughness @xcite .
based on our simulation results and previous observations , we tentatively suggest that the addition of these `` springy '' sheet materials reduces the effective poisson ratio of our nanocomposites , and that the microvoid formation process that we observe is a manifestation of the non - uniformities in the elastic constants within the nanocomposite . in principle , the elastic constants can be determined by the pel approach @xcite , but this analysis will be deferred to a future work since it is rather involved .
our tentative interpretation of the predicted @xmath1 variation in the sheet nanocomposites also suggests the need for better characterization of the geometric rigidity properties of the sheet nanoparticles , since these variables may be important for understating how such particles modify the stiffness and toughness of nanocomposites .
in this work we have focused on how nanoparticle shape influences the viscosity of polymer - nanoparticle melt mixtures at high temperature and the ultimate tensile strength ( estimated from the pel formalism ) of polymer nanocomposites in the ideal glass state .
our results suggest that chain bridging between the nanoparticles can have a large effect on @xmath0 of the mixture when the polymer - particle interactions are attractive , so that the nanoparticles disperse readily .
in addition , there is a relatively weak increase in @xmath0 for sheet nanoparticle composites , which tend to cluster in our simulations , supporting the expectation that nanoparticle clustering diminishes the viscosity enhancement .
( however , the formation of `` open '' or percolating fractal clusters may have the opposite effect on viscosity ) .
curiously , the tensile strength of the sheet nanocomposite is greatest , in spite of the sheet stacking .
one of the most intriguing effects of the nanoparticles is that , regardless of shape , the dependence of the tensile strength on chain length for the nanocomposites is _ opposite _ to that for the pure polymer melt .
we reiterate that our results for @xmath1 rely on the pel approach , and this approach should be carefully compared with experimental measurements to test the expected relationship between @xmath1 from simulations and tensile strength found experimentally .
the trends observed for the @xmath1 are more difficult to understand than those for @xmath0 .
there is evidently no clear - cut correlation of @xmath1 with the formation of bridging chains or with the attractive particle - particle interactions . in the absence of such a correlation
, we suggest that the nanoparticles effect the mechanical properties by modifying the ratio of the bulk and shear moduli in the low temperature nanocomposite state ( i.e. the nanocomposite poisson ratio ) .
recent work has noted that molecularly thin sheets have a negative poisson ratio when they are flexible enough to crumple by thermal fluctuations .
such additives could reduce the poison ratio of the material as a whole and provide a potential rationale for interpreting the increases in both the strength and toughness observed in our simulation of sheets dispersed in a polymer matrix , as well as in experiments on clay nanocomposites .
the authors thank v. ganesan , s. kumar , and k. schweizer for helpful discussions .
we thank the nsf for support under grant number dmr-0427239 .
.the table details all system variants simulated ; listing chainlength @xmath3 , number of chains @xmath63 , loading fraction @xmath60 , number of force sites per nanoparticle @xmath113 , and number of nanoparticles @xmath128 . [ cols="^,^,^,^,^,^ " , ] as a function of shear rate @xmath58 for chain length @xmath108 at @xmath76 .
the rods show the largest @xmath0 , followed by the icosahedra , and lastly the square sheets .
we show the statistical uncertainty of @xmath0 for each system at @xmath70 , where the fluctuations is largest , and hence represents an upper bound for the relative uncertainty for all @xmath0 calculations .
the uncertainty is the result of `` block averaging '' @xcite .
the inset includes the bulk polymer and shows that the pure system has much lower viscosity than any of the nanocomposites by roughly an order of magnitude .
the lines in this and subsequent figures are drawn only as guide for the reader s eyes.,width=325 ] and ( b ) reduced viscosity @xmath79 as a function of chain length @xmath3 .
all values are from systems at @xmath76 and @xmath77 .
chain length appears to have no effect on the ordering of @xmath0 amongst the systems.,width=325 ] on the fraction of bridging chains @xmath100 in each of the three nanoparticle systems at @xmath93 for ( a ) @xmath129 and ( b ) @xmath77 .
the order of @xmath100 with respect to nanoparticle type matches the trend in fig .
[ fig : shear_figure_1 ] for @xmath0.,width=325 ] as a function of @xmath103 for the pure polymer and three nanocomposite systems at @xmath75 .
the pure , icosahedron , and rod systems all have a sastry density @xmath111 , while for the sheets @xmath130 . the uncertainty intervals ( `` error bars '' ) represent the statistical uncertainty in our average for @xmath131 from block averaging @xcite . , width=325 ] as a function of chain length @xmath3 for the pure polymer and three nanocomposite systems .
we find that as @xmath1 increases as @xmath3 increases .
this is in contrast to the behavior of the pure polymer where @xmath1 decreases with increasing @xmath3 . only for systems with the sheet nanoparticles
or the system with icosahedral nanoparticles at @xmath132 does the addition of nanoparticles produce a net benefit relative to the bulk polymer.,width=325 ] | nanoparticles can influence the properties of polymer materials by a variety of mechanisms .
with fullerene , carbon nanotube , and clay or graphene sheet nanocomposites in mind , we investigate how particle shape influences the melt shear viscosity @xmath0 and the tensile strength @xmath1 , which we determine via molecular dynamics simulations .
our simulations of compact ( icosahedral ) , tube or rod - like , and sheet - like model nanoparticles , all at a volume fraction @xmath2 , indicate an order of magnitude increase in the viscosity @xmath0 relative to the pure melt .
this finding evidently can not be explained by continuum hydrodynamics and we provide evidence that the @xmath0 increase in our model nanocomposites has its origin in chain bridging between the nanoparticles .
we find that this increase is the largest for the rod - like nanoparticles and least for the sheet - like nanoparticles .
curiously , the enhancements of @xmath0 and @xmath1 exhibit _ opposite trends _ with increasing chain length @xmath3 and with particle shape anisotropy . evidently , the concept of bridging chains alone can not account for the increase in @xmath1 and we suggest that the deformability or flexibility of the sheet nanoparticles contributes to nanocomposite strength and toughness by reducing the relative value of the poisson ratio of the composite .
we note the molecular dynamics simulations in the present work focus on the reference case where the modification of the melt structure associated with glass - formation and entanglement interactions should not be an issue .
since many applications require good particle dispersion , we also focus on the case where the polymer - particle interactions favor nanoparticle dispersion .
our simulations point to a substantial contribution of nanoparticle shape to both mechanical and processing properties of polymer nanocomposites . |
we report the results of the optical interferometry of the interacting system of galaxies ngc 7769 , 7770 , 7771 and 7771a and analyze their kinematics .
a detailed description of the morphological features of the galaxies as well as photometry and color analysis of ngc 7769 are presented in the complete version of the study : @xcite .
we also discuss the influence of interaction on the kinematics , dynamics and star formation in the system .
known models of galaxy interactions are based mostly on statistical observational data .
we try to illustrate how and to what extend these models can be applied to explain the features of the galaxies in this system . in order to study the velocity fields of the galaxies ,
the observations were carried out at the 2.6 m telescope of the byurakan astrophysical observatory ( bao , armenia ) on 8 november 1996 , with the byufosc ( byurakan faint object spectral camera ) in the interferometric mode , attached at the prime focus of the telescope . based on the h@xmath0 velocity fields ( the right - hand panels of figure [ vrot_velr ] ) , we calculated the rotation curves of the galaxies ( the left - hand panels of figure [ vrot_velr ] ) by using data points within sectors along the maximal gradient direction , see isovelocity contours in the right - hand panels of figure [ vrot_velr ] .
maximal rotational velocity of ngc 7769 is observed at the radius of around 15 arcsec from the galaxy nucleus .
the rotational velocities in figure [ vrot_velr ] are in good agreement with the hi measurements ( @xmath1 ) in @xcite .
our measurements of velocities , having a better spatial resolution compared with those of the previous studies ( @xcite ) , reveal weak perturbations of the rotation curve of ngc 7769 , which may be caused by interaction with ngc 7771 .
the same can not be said about the velocity field of ngc 7771 .
figure [ vrot_velr ] shows that there are perturbations and large dispersion in radial velocities at the distances larger than about 10 - 15 arcsec from the nuclei .
this distance is about half radius of the bar .
evidently , this scatter of radial velocities can be explained by the fact that part of the arms are included in the sector used to calculate radial velocities ( sector angle is @xmath2 ) .
however the asymmetric profile along the major axis suggests that northern and southern arms do not have the same radial velocity profiles .
the asymmetric tidal forces of ngc 7769 and ngc 7770 affecting on ngc 7771 , seem to be a natural cause of that . the rotation curve of ngc 7770 is significantly skewed .
this is probably because of the strong harassing interaction with the more massive ngc 7771 , see @xcite .
the rotation curve of ngc 7771a is typical for a late type sm galaxy .
@xmath3{vrot_7769.eps } & \includegraphics[width=0.52\hsize]{ngc7769.eps}\\ \includegraphics[width=0.47\hsize]{vrot_7770.eps } & \includegraphics[width=0.52\hsize]{ngc7770.eps}\\ \includegraphics[width=0.47\hsize]{vrot_7771.eps } & \includegraphics[width=0.52\hsize]{ngc7771.eps}\\ \includegraphics[width=0.47\hsize]{vrot_7771a.eps } & \includegraphics[width=0.52\hsize]{ngc7771a.eps } \end{array}$ ] by analyzing velocity fields , sizes , and shapes of spiral arms of ngc 7771 and ngc 7769 , in @xcite it has been suggested that ngc 7771 and ngc 7769 , which have a 2:1 mass ratio , appear to be having a prograde - retrograde interaction , with ngc 7769 being the retrograde one .
our better data support this conclusion .
this conclusion is in agreement with the latest models of galaxy collisions ( ) showing that during direct collisions tidally induced spiral arms are much longer and brighter than those during retrograde collisions .
we can conclude that galaxies ngc 7769 and ngc 7771 already have passed the first pericenter stage , however , probably the second encounter has not happened yet .
the first pericenter distance should have been large enough ( around few sizes of the galaxies ) , so that large disturbances in rotation curves have not appeared yet .
the quartet of galaxies ngc 7769 , 7770 , 7771 and 7771a is a system of interacting galaxies . here
, we present a fabry - perot imaging study of the system in h@xmath0 line .
we came to the following main conclusions : * close interaction between the component galaxies of the system has produced morphological features that are characteristic of the interactions .
we have detected features such as tidal arms , spiral arms induced by close interaction , bars and induced star formation . * from the results of our interferometric observations , we obtained the radial velocity profiles of galaxies .
the rotation curve of ngc 7769 is weakly distorted .
the rotation curve of ngc 7771 is strongly distorted by the tidal arms caused by direct flyby of ngc 7769 and flyby of a smaller neighbor ngc 7770 .
the rotation curve of ngc 7770 is significantly skewed because of the interaction with much massive ngc 7771 . *
the radial velocity profiles and morphological disturbances suggest that the ngc 7769 and ngc 7771 have passed the first pericenter stage , however , probably the second encounter has not happened yet .
study of such systems with methods combining photometric and visual analysis is an effective way to clarify features of star formation in different stages of interaction .
ongoing and future surveys using integral field spectroscopy will allow also to explore the spatial distribution of star formation in interacting systems .
alonso - herrero , a. , rosales - ortega , f. f. , snchez , s. f. , et al .
2012 , mnras , 425 , l46 chengalur , j. n. , salpeter , e. e. , & terzian , y. 1993 , apj , 419 , 30 di matteo , p. , combes , f. , melchior , a .-
semelin , b. 2007 , a&a , 468 , 61 nordgren , t. e. , chengalur , j. n. , salpeter , e. e. , & terzian , y. 1997 , aj , 114 , 77 yeghiazaryan , a. a. , nazaryan , t. a. , & hakobyan , a. a. 2015 , arxiv:1510.00193 | we performed the fabry - perot scanning interferometry of the quartet of galaxies ngc 7769 , 7770 , 7771 and 7771a in h@xmath0 line and studied their velocity fields .
we found that the rotation curve of ngc 7769 is weakly distorted .
the rotation curve of ngc 7771 is strongly distorted with the tidal arms caused by direct flyby of ngc 7769 and flyby of a smaller neighbor ngc 7770 .
the rotation curve of ngc 7770 is significantly skewed because of the interaction with much massive ngc 7771 .
the rotation curves and morphological disturbances suggest that the ngc 7769 and ngc 7771 have passed the first pericenter stage , however , probably the second encounter has not happened yet . |
direct or prompt photons are photons that emerge directly from the hard scattering of the partons and not from secondary decays or as the radiation product of initial or final state partons . since these photons come directly from parton - parton hard scattering , they provide a clean signature of the underlying hard scattering dynamics .
prompt photon production in hadronic interactions provides a precision test of pqcd predictions as well as information on the gluon density inside the colliding hadrons .
theoretical investigations have shown@xcite that the cross - section for direct photon production has a pseudorapidity ( @xmath7 ) dependence which is sensitive to the parameterization of the gluon distributions@xcite .
the lhc energy will provide an opportunity to determine gluon density in a proton in new kinematic region of @xmath8 and @xmath9 ( @xmath10)@xcite . the analysis of the lhc data in combination with results from the tevatron and hera would allow to extend the qcd analysis in @xmath9 region @xmath11@xcite .
the study of physics beyond the sm at the lhc also requires a complete understanding of the qcd processes .
some of the qcd processes constitute major background to other sm processes , and also to signals of new physics .
thus these processes need to be well understood precisely in the new energy regime . in this work ,
we have concentrated on the qcd process pp @xmath12 + 1 jet + x , where x can be anything .
this process is a major background to other standard model processes such as h @xmath13@xcite , @xmath14@xcite , and signatures of physics beyond the sm such as large extra dimensions@xcite and susy@xcite .
inclusive direct photon production has been studied extensively by various experiments from fixed target energies@xcite to tevatron collider@xcite .
a recent study by the d0 experiment at tevatron has reported on the measurment of the triple differential cross - section for @xmath15 + jet at @xmath1=1.96 tev@xcite . in this work ,
the photon + jet cross - sections has been compared to the next - to leading order ( nlo ) based on the program jetphox@xcite with cteq6.1 m pdf set of parton distribution functions@xcite . however , both the completeness of theoretical calculations and consistency of the available data have been subject of intense discussion .
the results from various experiments have shown a definite pattern of deviation from theory at low @xmath16 where the measured cross sections have a steeper slope .
the origin of the disagreement has been attributed to the effect of initial - state soft gluon radiation which is not properly accounted for in the theoretical calculations .
it has been shown that the inclusion of some additional transverse momentum `` @xmath17 '' smearing of the partonic system due to soft - gluon emission yields better description of the data@xcite .
these effects are however expected to be negligible in the kinematic range explored in the present analysis . at the leading order ,
the direct photon production is defined by two qcd processes ; quark - gluon compton scattering , @xmath18 and quark - antiquark annihilation , @xmath19 . due to abundance of low momentum fraction gluons in the proton at the lhc energy ,
compton scattering becomes the dominant process contributing to the prompt photon production over most of the kinematical region .
unfortunately , the advantages of photon as a clean probe of parton distributions are offset by large qcd backgrounds which are @xmath20 to @xmath21 times larger than that of direct photon signal .
the background contribution to direct photon is mainly caused by the events where high @xmath16 photons are produced in the decays of neutral mesons such as @xmath22 , @xmath7 , @xmath23 and @xmath24 ; and from events where the photons are radiated from the quark ( i.e , bremsstrahlung photons in the next - to - leading order qcd subprocesses such as @xmath25 , @xmath26 and @xmath27 ) .
any analysis must separate true direct photons ( those coming from the hard scattering ) from those copiously produced in the decays of @xmath22 , @xmath7 , @xmath23 , @xmath24 and bremsstrahlung photons emerging from high @xmath28 jets .
isolation cuts imposed on reconstructed photon candidates effectively suppress these backgrounds@xcite .
the purpose of this work is to study the @xmath0 + jet events and its background processes with full detector simulation and reconstruction , and optimize the signal over background ( s / b ) ratio at the lhc energy . in this work
, we present the geant4 simulated results of level-3 triggered photons ( also called high level trigger ( hlt ) photons ) using cms software packages .
the event generation for @xmath29 + jet signal and background processes has been done using cmkin@xcite .
the passage of particles through the detector geometry , showering , energy loss in the calorimeters and reconstruction of the events are modeled in the cms simulation and reconstruction packages namely , oscar@xcite and orca@xcite respectively .
the simulated result has been compared with the theoretical calculations .
the rest of the paper is organised as follows . in the next section
, we discuss the physics of @xmath0 + jet .
section 3 gives a brief description of the cms detector relevant for the present analysis . in section 4 ,
we discuss event generation .
section 5 describes photon isolation and its effect on signal and background . in section 6
, we discuss the effect of @xmath3 cut between the photon and the jet on s / b .
statistical and systematic uncertainties are discussed in section 7 . in the last section
, we summarize this analysis with the conclusions .
[ fig : lo ] shows the two leading order ( lo ) processes namely , `` quark - antiquark annihilation '' and `` quark - gluon compton scattering '' which contributes to @xmath0 + jet events . in quark - antiquark annihilation ,
a direct photon is produced along with an outgoing gluon while in quark - gluon compton scattering , a gluon scatters from a quark producing a photon in the final state along with the scattered quark .
an example of a next - to - leading order ( nlo ) real correction and corresponding virtual corrections ( which cancel the soft and collinear divergences arising from the real radiation ) are shown in fig .
[ fig : nlo ] .
if the initial parton energies are equal , the photon and the jet are produced back to back ( @xmath30 ) . in case of large imbalance between initial parton energies , the photon and the jet tend to be produced at small angles , either both forward or backward .
since the gluon momentum distribution increases monotonically at lower momentum fraction values@xcite , high @xmath7 photons are likely to come from gluon compton scattering process rather than annihilation process .
the main background to the direct photon signal comes from jet fluctuations .
while most of the jets consist of many particles , thus easily distinguishable from a single photon , a small fraction ( @xmath31 - @xmath32 ) of jets fragments in such a way that a single particle carries most of the momentum of the parent parton .
the lightest and therefore most commonly produced neutral meson , the @xmath33 decays into two photons with a branching ratio of @xmath34 .
photons produced in decays of high @xmath35 @xmath33 ( and other neutral mesons ) are very close to each other in the electromagnetic calorimeter forming a single shower thus mimicking a direct photon . therefore , non - direct photons are common in hadronic collisions .
since jet production rates are @xmath20 times larger than that of direct photons ( depends on @xmath36 and @xmath16 ) , the number of jets which fragment to a single meson and then decay to photons , referred as electromagnetic ( em ) jets , are comparable to the level of direct photons . the background to direct photon + jet can also come from photons produced in initial and final state radiation . in these cases ,
the photon is not produced directly from the interaction vertex , and therefore , is not really a `` direct '' photon . in order to reduce the background contribution , an isolation criterion is imposed on the photon : typically a cone of size r ( in pseudorapidity @xmath37 and azimuthal angle @xmath38 ) around the photon is required to have less than a certain amount of energy . since a photon produced through bremsstrahlung tends to be collinear with the parent quark ( and thus the jet resulting from the quark s subsequent fragmentation ) , the requirement that the photon be isolated removes all but very large angle radiations .
[ fig : dirph - bkjets2 ] and [ fig : dirph - bkanam1 ] show major backgrounds to direct photons , namely , the em jets and the bremsstrahlung .
the compact muon solenoid ( cms ) is a general purpose detector with 4@xmath39 steridian coverage for efficient detection and precise energy / momentum measurement of electrons , photons , muons , hadrons ( jets ) and missing transverse energy . from inside to outside
, it comprises of a pixel detector , a silicon microstrip tracker system , a lead tungstate crystal electromagnetic calorimeter ( ecal ) , a sampling hadronic calorimeter ( hcal ) , a 4 tesla soleniodal magnet and the muon detectors .
the innermost component of the tracking system is the highly segmented silicon pixel detector which consists of 3 layers of barrel complemented with 2 disks on each side in the endcaps .
it consists of 66 million pixels of size 100 @xmath40 m @xmath41 150 @xmath40 m yielding a three dimensional spatial resolution of @xmath4210 @xmath40 m in the r-@xmath43 and @xmath4220 @xmath40 m in the z - coordinate , which will play a crucial role in the vertex identification and track reconstruction .
the following part of the tracking system consists of silicon microstrip detector modules arranged in ten concentric layers in barrel and nine disks in the endcaps providing a single point resolution between 23 - 52 @xmath40 m in r-@xmath43 and 230 - 530 @xmath40 m in z. it provides the necessary lever arm for accurate momentum measurement and improves the precision of vertex measurement .
the tracker has geometrical coverage upto @xmath44 .
the momentum resolution @xmath45 of high @xmath16 charged particles in the tracker is @xmath46 for @xmath47 and becomes @xmath48 as @xmath36 approaches 2.5@xcite .
the electromagnetic calorimeter is a hermetic , homogeneous calorimeter consisting of 61,200 lead tungstate @xmath49 crystals mounted in central part corresponding to @xmath50 , and 7324 crystals mounted in each of the two endcaps in the range of @xmath51 .
the crystals are 23 cm long , have short radiation length(@xmath52 = 0.89 cm ) of moliere radius(21.9 mm ) and correspond to @xmath4226@xmath52 , thereby containing more than 99@xmath53 of the shower energy .
the granularity of the calorimeter is @xmath54 x @xmath55 x @xmath56 in the barrel and upto @xmath54 x @xmath57 x @xmath58 in the endcaps .
the resolution for single photons with energies of 25 - 500 gev can be parameterized as @xmath59 = @xmath60 .
an additional pb - si preshower detector would be installed in front of the endcaps covering @xmath61 for efficient rejection of @xmath62@xcite .
the sampling hadronic calorimeter is made up of plastic scintillator tiles inserted between brass absorber plates and extends upto @xmath63 .
the barrel ( endcaps ) consists of 5 cm ( 8 cm ) thick brass plates interleaved with 3.8 mm thick scintillator tiles .
the barrel calorimeter has a depth of 79 cm corresponding to 5.15 hadronic interaction lengths @xmath64 and a segmentation of @xmath54 x @xmath65 x @xmath66 . outside of the soleniod
, an additional layer of 10 mm thick scintillator called outer hadron calorimeter covering the @xmath67 region , is placed to ensure sampling of hadronic shower with @xmath42 11 @xmath68 .
hermecity of the hcal is extended to @xmath69 by adding forward detector made up of steel absorbers and embedded radiation hard quartz fibers , which provide a fast collection of cerenkov light .
the resolution of the hcal ranges from @xmath59 = @xmath70 @xmath71 in the barrel ( at @xmath72 ) and @xmath59 = @xmath73 in the endcaps , to @xmath59 = @xmath74 in the forward region@xcite .
the muon detector consists of drift tubes in the barrel @xmath75 and cathode strip chambers in the endcaps @xmath76 for precision track measurement .
a set of resistive plate chambers are used in both parts of the system for resolving individual bunch crossing time .
the spatial resolution of the muon system varies between 50 to 200 @xmath40 m and standalone momentum resolution is at most 15@xmath53 for a 10 gev @xmath16 muon and 40@xmath53 at 1 tev@xcite .
at the lhc center of mass energy , the cross - sections for the channels pp @xmath77 + jet and pp @xmath78 jets are very large , thus making it nearly impossible to simulate direct photon signal and its background processes equivalent to an integrated luminosity of @xmath79 . in order to achieve this , a selection based on kinematic information of the particle
is performed at the generator level which allows only those events to be fully simulated that are likely to pass the analysis criteria . in our study
we have used @xmath80 to match the preshower coverage .
the cms experiment has optimized the l1 trigger @xmath16 cut for a l1 isolated single photon to be 23 gev , while that for the hlt photon it is 80 gev at a luminosity of @xmath81@xcite . for the signal , a pre - selection has been used at the generator level which retains only those events having a direct photon with @xmath82 gev and @xmath83 , which are well below the final analysis cut of @xmath84 gev and @xmath85 .
this preselection helps in discarding those events before simulation which are unlikely to pass geometrical acceptance cut .
the l1 trigger efficiency for the signal has been estimated to be above @xmath86 over the entire kinematic range@xcite . for the backgrounds , we have used
pythia generated qcd dijet events .
the samples @xmath87 have been preselected at the generator level with an enhanced l1 trigger efficiency@xcite . in this pre - selection ,
the first step is to look for the seed particles of electromagnetic objects like photons , electrons and positrons , which have @xmath88 gev and @xmath89 .
candidate electromagnetic calorimeter trigger tower energies are then estimated by adding energies of all electromagnetic particles found within a cone of @xmath900.09 and @xmath910.09 from the seed .
trigger tower candidates that lie within @xmath92 and @xmath93 from each other are suppressed and only those with the highest @xmath16 are retained . the l1 single photon electromagnetic trigger is simulated by requiring that one such candidate has transverse energy larger than 20 gev .
signal and background events have been reconstructed using orca version 8.7.3 . in the electromagnetic calorimeter
, the hybrid and island clustering algorithms@xcite were used to construct basic clusters in the barrel and the endcap regions respectively .
a supercluster is a cluster of the basic clusters .
a photon candidate is associated with each reconstructed supercluster .
events were generated in two different @xmath16 bins : @xmath94 gev @xmath95 gev , and @xmath96 gev .
the cross - sections , number of events and corresponding integrated luminosity for signal and background samples are shown in table [ table : cross - section ] .
the samples have been generated using pythia@xcite , simulated and reconstructed using oscar and orca .
we require that the leading photon should have @xmath97 80 gev and @xmath85 .
[ fig : numberofeventswithoutisolation ] shows the number of events / gev after @xmath16 and @xmath7 requirements calculated as a function of @xmath98 for the @xmath0 + jet signal and its background for an integrated luminosity of 1 @xmath99 . [ cols="^,^,^,^,^,^ " , ] fig .
[ fig : diff_crossx ] shows the the predictions for the differential cross - section of the @xmath0 + jet events as a function of @xmath100 compared with theoretical lo and nlo calculations@xcite .
the error bars on the x - axis represents the bin width .
the differential cross - section calculated from the simulated data is represented as lo ( calculated after full simulation ) in the above mentioned figure . to evaluate it ,
the cross - section from pythia was generated with cteq5l parton distribution function and renormalization scale @xmath101 .
various factors like geometrical acceptance , efficiency of the analysis cuts and luminosity have been taken into account .
theoretical lo and nlo calculations are provided by j. owens@xcite .
the results are found to be in good agreement with the theoretical lo calculation . as expected
the nlo qcd contribution is higher than the lo in the whole @xmath16 range under analysis .
in @xmath0 + jets events , the jets have been reconstructed using iterative cone type algorithm with cone size r = 0.5@xcite .
since the hcal extends up to @xmath102 the selection cuts used on the highest @xmath16 jet are : @xmath103 gev and @xmath104 . usually @xmath0 + 1 jet events
are produced in a
back - to - back " fashion , but this topology is disturbed by initial - state and final - state radiation .
the events with the vector @xmath105 being back - to - back " to the vector @xmath100 within @xmath3 can be defined by the equation ; @xmath106 = @xmath107 @xmath108 @xmath3 where @xmath106 is the angle between @xmath100 and @xmath105 vectors . to further optimize s / b ratio
while retaining a very large signal efficiency , the effect of @xmath3 cut has been studied after applying all photon isolation requirements .
[ fig : deltaphi ] shows the signal efficiency as a function of @xmath3 cut .
[ fig : deltaphi1 ] shows the performance of @xmath3 as a function of signal efficiency vs. s / b ratio for all selection cuts noted in table [ table : cmp_selection ] , when @xmath3 is varied in steps of @xmath109 .
it is found that the limit @xmath110 can be used as an additional cut to improve the s / b ratio .
requiring @xmath110 improves the s / b ratio by @xmath5 with a bare reduction of @xmath111 in signal efficiency .
the statistical uncertainty on the signal and the background rates are found to be @xmath112 and @xmath113 respectively .
we have studied some of the potential sources of theoretical and experimental uncertainty . the uncertainty on the signal ( background ) rate arising from the choice of the pdf used in the event generation is found to be @xmath114 ( @xmath113 ) .
the variation of the hard scale from default choice of @xmath9 in pythia to @xmath115 ( mstp 32=4 ) results in @xmath116 uncertainty on the signal rate and @xmath117 on the background rate .
the estimated uncertainty due to photon trigger is @xmath112@xcite . for @xmath118 gev , the uncertainty due to jet energy scale is @xmath119 .
experimental measurement uncertainties such as photon energy scale and jet energy resolution effect the rates within @xmath112 .
the systematic uncertainty for the signal and the background rate arising from the preselection efficiency is less than @xmath120 .
these uncertainties when added in quadrature leads to @xmath121 uncertainty on the signal rate and @xmath122 on the background rate . in the initial phase of the lhc operation , for an integrated luminosity of @xmath123 , the error on the measured luminosity at cms
is expected to be @xmath124@xcite .
studies from other collider experiments@xcite have reported uncertainties due to photon purity , inefficiency in event vertex determination and conversion of photons in the detector
. however , the understanding of such sources of systematic uncertainty in the context of this analysis would be better and clear after the availablity of the lhc data .
in this analysis , we have explored the direct photon + jet production at the cms in the region of photon transverse momentum ( @xmath100 ) from hlt @xmath16 threshold of 80 gev to a few hundred gev .
we have studied the tracker , the ecal and the hcal based isolation conditions for the direct photon physics . for low luminosity phase of the lhc operation , it is found that with optimized isolation conditions the background can be reduced by two orders of magnitude while the signal efficiency remains between @xmath125 .
isolation requirement within a larger cone around a photon provides a clean and pure signal .
it is found that for the entire @xmath16 range in the analysis , carefully chosen values of the isolation threshold parameters can yield appreciable increase in the s / b ratio as compared to previous studies while retaining very high signal efficiency .
we have found selection cut c ( table [ table : cmp_selection ] ) to be optimum .
we have matched our results from the pythia based simulation with an independent theoretical calculation and they are in excellent agreement .
the topology of the @xmath0 + jet events is such that at high @xmath16 most of the direct photon and jet events are back to back " within certain @xmath3 .
the inclusion of a very wide @xmath3 cut at @xmath4 in the analysis leads to a further increase of @xmath5 thus leading to an overall gain @xmath6 in s / b ratio .
a better understanding of fake photon rejection will naturally lead to improved efficiency for detecting low mass higgs decaying to photon pairs .
we are grateful to j. owens for providing theoretical calculations .
we thank m.pieri for helping with the background samples .
we also wish to express thanks to p. demin , d. futyan , a. kumar , m. mangano , c. seez and r. y. zhu .
the work has been supported by a grant from the department of science and technology , govt . of india .
we gratefully acknowledge the facilities provided by the center for detector and related software technology , university of delhi .
pg would like to express gratitude to the council of scientific and industrial research , govt.of india , for financial assistance and to prof .
shivpuri and prof .
raghuvir singh for support and encouragement .
99 p. aurenche et al . , `` prompt photon production at colliders '' , _ phys .
d _ * 42 * ( 1990 ) 1440 - 1449 .
reid at al .
, `` direct photon in cms '' , _
cms note-2000/063_. d.v .
bandourin , v.f .
konoplyanikov , `` on the application of photon+jet events for setting the absolute jet energy scale and determining the gluon distribution at the lhc '' , _ hep - ex/0207028_. d.v .
bandourin , n.b .
skachkov , `` on the possibilities of measuring the gluon distribution using photon/@xmath126+jet events at tevatron run ii and lhc '' , contributed to proc .
of xi international workshop on deep inelastic scattering `` dis 2003 '' , st .
petersburg , april 2327 , 2003 _ hep - ex/0403024_. m. pieri et al .
, `` inclusive search for the higgs boson in the h@xmath13 channel '' , _
cms note-2006/112_. f. abe et al .
, `` search for the rare decay @xmath14 '' , _ phys .
lett . _ * 76 * ( 1996 ) 2852 .
n. arkani - hamed et al . , `` the hierarchy problem and the new dimensions at a millimeter '' , _
b _ * 429 * ( 1998 ) 263 - 272 .
j. weng et al . , `` search for add direct graviton emission in photon plus missing transverse energy final state at cms '' , _
cms note-2006/129_. cdf collaboration , `` search for heavy , long - lived particles that decay to photons at cdf ii '' , _ phys .
* 99 * ( 2007 ) 1217801 .
l. apanasevich et al . , `` evidence for parton @xmath17 effects in high- @xmath16 particle production '' , _ phys .
* 81 * ( 1998 ) 2642 - 2645 .
abazov et al . , `` measurement of the isolated photon cross section in p - pbar collisions at @xmath1=1.96 tev '' , _ phys .
b _ * 639 * ( 2006 ) 151 - 158 .
b. abbott et al . , `` comparison of the isolated direct photon cross sections in p - pbar collisions at @xmath1=1.8 tev and @xmath1=0.63 tev '' , _ phys .
* 84 * ( 2000 ) 2786 - 2791 .
d. ascota et al .
, `` isolated photon cross section in p - pbar collisions at @xmath1=1.8 tev '' , _ phys .
* 65 * ( 2002 ) 112003 .
d. ascota et al .
, `` ratio of isolated photon cross section in p - pbar collisions at @xmath1=630 and @xmath1 = 1800 gev '' , _ phys .
lett . _ * 87 * ( 2001 ) 251807 .
d. ascota et al . , `` direct photon cross - section with conversions at cdf '' , _ phys.rev .
d _ * 70 * ( 2004 ) 074008 .
a. kumar for the d0 collaboration , `` measurement of triple differential photon plus jet cross - section by d0 '' , _
arxiv:0710.0415v1[hep - ex]_. t. binoth at al .
, `` a full next - to - leading order study of direct photon pair production in hadronic collisions '' , _ eur .
j. c _ * 16 * ( 2000 ) 311 .
s. catani at al .
, `` cross section of isolated prompt photons in hadron - hadron collisions '' , _ jhep _ * 0205 * ( 2002 ) 028 .
j. pumplin et al . , `` new generation of parton distributions with uncertainties from global qcd analysis '' , _ jhep _ * 0207 * ( 2002 ) 012 .
d. stump et al .
, `` inclusive jet production , parton distribution and search for new physics '' , _ jhep _ * 0310 * ( 2003 ) 046 .
a. kumar et al .
, `` study of parton @xmath17 smearing effects in direct photon production at the fermilab tevatron '' , _ phys . rev .
* 68 * ( 2003 ) 014017 .
l. apanasevich et al .
, `` @xmath17 effects in direct photon production '' , _ phys .
* 59 * ( 1999 ) 074007 .
l. agostino , m. pieri , `` high level trigger selection of electrons and photons '' , _
cms note-2006/078_. cmkin , `` cms interface for event generator '' , _ http://cmsdoc.cern.ch / cmsoo / projects / cmkin_. oscar,``object - oriented simulation for cms analysis and reconstruction '' , _
http://cmsdoc.cern.ch / oscar_. orca , `` object - oriented reconstruction for cms analysis '' , _
http://cmsdoc.cern.ch / orca_. j.p .
ralston , `` pocket partonometer '' , _ phys .
b _ * 172 * ( 1986 ) 430 - 434 .
cms collaboration `` cms tracker technical design report '' , _ cern / lhcc 98 - 6_. cms collaboration `` addendum to the cms tracker tdr '' , _ cern / lhcc 2000 - 016_. cms collaboration , `` cms physics technical design report volume i '' , _ cern / lhcc 2006 - 001_. cms collaboration `` cms ecal technical design report '' , _ cern / lhcc 97 - 33_. cms collaboration `` cms hcal technical design report '' , _ cern / lhcc 97 - 31_. cms collaboration `` cms muon technical design report '' , _ cern / lhcc 97 - 32_. e. meschi et al . , `` electron reconstruction at cms electromagnetic calorimeter '' , _
cms note-2001/034_. t. sjostrand and p. z.
skands , `` transverse - momentum - ordered showers and interleaved multiple interactions '' , _ eur.phys.j .
c _ * 39 * ( 2005 ) 129 - 154 .
h. baer at al .
, , `` next - to - leading - logarithm calculation of direct photon production '' , _ phys .
* 42 * ( 1990 ) 61 .
jeff owens _ private communications_. cms collaboration , `` the tridas project design report , volume ii : data acquisition and high - level trigger '' , _ cern / lhcc 2002 - 26_. s. baffioni et al .
, `` discovery potential for the sm higgs boson in the h@xmath127 decay channel '' , _
cms note-2005/115_. cms collaboration , `` cms physics technical design report volume ii '' , _ jour . of phys g _
* 34 * ( 2007 ) 995 - 1580 .
h1 collaboration , `` measurement of prompt photon cross - sections in photoproduction at hera '' ,
_ eur.phys.j .
c _ * 38 * ( 2005 ) 437 - 445 . | we present simulation results of @xmath0 + jet analysis using cms ( compact muon solenoid ) object - oriented software at the large hadron collider ( lhc ) center of mass energy @xmath1=14 tev .
the study of direct photon production helps in validating the perturbative quantum chromodynamics ( pqcd ) and providing information on the gluon distribution in the nucleons .
direct photon processes also constitute a major background to several other standard model ( sm ) processes and signals of new physics .
thus these processes need to be understood precisely in the new energy regime . in this work ,
we have done a detailed study of the geant4 simulated @xmath0 + jet events generated with pythia , and the related background processes .
isolation cuts have been optimized for direct photon which improves the signal over background ratio by @xmath2 as compared to previous studies done in cms .
the inclusion of a large @xmath3 cut between the photon and the leading jet at @xmath4 in the analysis leads to a further increase of @xmath5 in s / b , thus giving an overall gain of @xmath6 in s / b ratio .
= 100000 # 1(#1 ) 1t_1 cms - note-2007/004 + accepted in european journal of physics c * study of direct photon plus jet production in cms experiment at @xmath1=14 tev * + pooja gupta , b.c.choudhary , s.chatterji,and s.bhattacharya + department of physics and astrophysics , university of delhi , + delhi , 110 007 , india + 1.0 cm |
nuclear lattice effective field theory ( nleft ) is a first - principles approach , in which chiral eft for nucleons is combined with numerical auxiliary - field quantum monte carlo ( afqmc ) lattice calculations .
nleft differs from other _ ab initio _
methods @xcite in that it is an unconstrained monte carlo calculation , which does not rely on truncated basis expansions or many - body perturbation theory , nor on prior information about the structure of the nuclear wave function . as in chiral eft ,
our calculations are organized in powers of a generic soft scale @xmath5 associated with factors of momenta and the pion mass @xcite .
we denote @xmath6 as leading order ( lo ) , @xmath7 as next - to - leading order ( nlo ) , and @xmath8 as next - to - next - to - leading order ( nnlo ) contributions .
the present calculations are performed up to nnlo .
we define @xmath9 as the lo lattice hamiltonian , and @xmath10 as the equivalent hamiltonian with the pion - nucleon coupling @xmath11 and contact interactions that respect wigner s su(4 ) symmetry . in our nleft calculations ,
@xmath9 is treated non - perturbatively ( see ref .
@xcite for a review ) .
the nlo contribution to the two - nucleon force ( 2nf ) , the electromagnetic and strong isospin - breaking contributions ( emib ) , and the three - nucleon force ( 3nf ) which first enters at nnlo , are all treated as perturbations .
it should be noted that our `` lo '' calculations use smeared short - range interactions that capture much of the corrections usually treated at nlo and higher orders @xcite . at nnlo ,
the 3nf overbinds nuclei with @xmath12 due to a clustering instability which involves four nucleons on the same lattice site .
a long - term objective of nleft is to remedy this problem by decreasing the lattice spacing and including the next - to - next - to - next - to - leading order ( n3lo ) corrections in chiral eft . in the mean time , the overbinding problem has been rectified by means of a 4n contact interaction , tuned to the empirical binding energy of either @xmath13he or @xmath14be @xcite . while this provides a good description of the alpha nuclei up to @xmath15 including the hoyle state @xcite , the overbinding is found to increase more rapidly for @xmath16 . in ref .
@xcite , a non - local 4n interaction which accounts for all possible configurations of four nucleons on adjacent lattice sites was introduced , and adjusted to the empirical binding energy of @xmath3 mg .
a detailed study of the spectrum and electromagnetic properties of @xmath1o ( with inclusion of the effective 4n interaction ) has been reported in ref .
the hoyle state is a resonance with spin - parity quantum numbers @xmath17 in the spectrum of @xmath0c , which plays an important role in resonantly enhancing the reaction rate for the so - called triple - alpha process , which is responsible for the production of carbon in massive stars that have reached the red giant stage in their evolution .
this reaction represents a significant bottleneck in the stellar nucleosynthesis , as @xmath14be is an unstable ( though relatively long - lived ) resonance .
for @xmath0c to form , a third alpha particle must combine with the @xmath14be resonance to form the hoyle state , which subsequently decays electromagnetically to the ground state of @xmath0c .
this reaction may then proceed further ( non - resonantly ) to form @xmath1o through addition of a fourth alpha particle .
however , the temperature of the stellar plasma at which the triple - alpha process takes place depends exponentially on the energy @xmath18 of the hoyle state above the triple - alpha threshold , which is experimentally known to be @xmath19 kev .
stellar model calculations @xcite have shown that only a narrow window of @xmath20 kev exists in @xmath18 where sizable amounts of carbon and oxygen can be produced simultaneously . .lattice results at leading order ( lo ) and available experimental values for the root - mean - square charge radii and quadrupole moments of the @xmath0c states .
[ hoyle1 ] [ cols="^,^,^",options="header " , ] in table [ oxygen2 ] , we note that the lo charge radius @xmath21 of the ground state of @xmath1o is smaller than the empirical value @xmath22 .
this leads to a systematic deviation , which arises from the overall size of the second moment of the charge distribution . to compensate for this overall scaling mismatch
, we have also calculated `` rescaled '' quantities multiplied by powers of the ratio @xmath23 , according to the length dimension of each observable . with such a scaling factor included , we find that the nleft predictions for the @xmath24 and @xmath25 transitions are in good agreement with available experimental values .
we have presented an overview of the central nleft results for the low - lying even - parity spectra of @xmath0c and @xmath1o .
this includes the hoyle state of @xmath0c which plays a central role in the stellar nucleosynthesis of life - essential elements .
we have also shown that the electromagnetic properties and transition rates of @xmath0c and @xmath1o are in agreement with available experimental data .
while the long - term objectives of nleft are to decrease the lattice spacing and include higher orders in the eft expansion , we also find that the missing physics up to @xmath4si can be approximated by an `` effective '' 4n interaction .
these results represent an important step towards more comprehensive nleft calculations of medium - mass nuclei in the near future .
we are grateful for the help in automated data collection by thomas luu .
partial financial support from the deutsche forschungsgemeinschaft ( sino - german crc 110 ) , the helmholtz association ( contract no .
vh - vi-417 ) , bmbf ( grant no .
05p12pdfte ) , and the u.s .
department of energy ( de - fg02 - 03er41260 ) is acknowledged .
this work was further supported by the eu hadronphysics3 project , and funds provided by the erc project no .
259218 nucleareft .
the computational resources were provided by the jlich supercomputing centre at the forschungszentrum jlich and by rwth aachen .
b. borasoy , e. epelbaum , h. krebs , d. lee , and ulf - g .
meiner , eur .
j. a * 31 * , 105 ( 2007 ) .
e. epelbaum , h. krebs , d. lee , and ulf - g .
meiner , phys .
lett . *
104 * , 142501 ( 2010 ) ; _ ibid .
_ , eur .
j. a * 45 * , 335 ( 2010 ) .
e. epelbaum , h. krebs , d. lee , and ulf .- g .
meiner , phys .
lett . * 106 * , 192501 ( 2011 ) .
e. epelbaum , h. krebs , t. a. lhde , d. lee , and ulf - g .
meiner , phys .
* 109 * , 252501 ( 2012 ) ; _ ibid . _ ,
lett . * 110 * , 112502 ( 2013 ) ; _ ibid .
_ , eur .
phys . j. a * 49 * , 82 ( 2013 ) .
t. a. lhde , e. epelbaum , h. krebs , d. lee , ulf - g .
meiner , and g. rupak , arxiv:1311.0477 [ nucl - th ] .
e. epelbaum , h. krebs , t. a. lhde , d. lee , ulf - g .
meiner , and g. rupak , _ to appear in physical review letters _ , arxiv:1312.7703 [ nucl - th ] | we review the calculation of the hoyle state of @xmath0c in nuclear lattice effective field theory ( nleft ) and its anthropic implications for the nucleosynthesis of @xmath0c and @xmath1o in red giant stars .
we also review the extension of nleft to the regime of medium - mass nuclei , with emphasis on the determination of the ground - state energies of the alpha nuclei @xmath1o , @xmath2ne , @xmath3 mg and @xmath4si by means of euclidean time projection . finally , we review recent nleft results for the spectrum , electromagnetic properties , and alpha - cluster structure of @xmath1o . |
the intrinsic non - gaussianity of primordial curvature perturbations is a valuable tool to distinguish among different models for the origin of structure in the very early universe @xcite .
even if the primordial curvature perturbation has a gaussian distribution , the initial density perturbation is non - gaussian due to the non - linearity of the einstein field equations @xcite .
non - gaussianity ( ng ) may be characterized by a variety of different parameters , which control the departure of the underlying probability distribution function ( pdf ) of primordial fluctuations from a purely gaussian distribution .
these parameters can be related to the amplitude of connected @xmath2-point ( @xmath2-pt ) correlation functions of primordial curvature fluctuations that vanish for @xmath3 in the gaussian case .
the connected @xmath2-pt functions may be scale- and shape - dependent , in a manner determined by the particular model that generates them . in this work
, we will focus on the so - called local shape of ng , in which the primordial gravitational potential in coordinate space can be expressed as an expansion in powers of one or more gaussian random fields that determine the primordial density fluctuations . in this case
3-pt functions of curvature fluctuations are determined by a parameter , @xmath1 , that at lowest order in a perturbative expansion characterizes the skewness of the pdf of primordial fluctuations .
4pt functions are determined by parameters , @xmath0 and @xmath4 , controlling the kurtosis of the pdf at lowest order in perturbations .
the recent release of planck data allows one to set the best constraints yet on primordial non - gaussian parameters from cosmic microwave background ( cmb ) data alone @xcite .
the quantity @xmath1 ( local ) is now constrained to be @xmath5 at 2@xmath6 level , while @xmath7 ; an analysis of the constraints on @xmath0 local from planck data has yet to be completed , but the forecast @xmath8 error bar with planck data has been estimated to be @xmath9 by @xcite and @xmath10 by @xcite
. it may be unlikely that we live in a universe with a large hierarchy between @xmath1 and @xmath0 local @xcite , but this possibility can not be excluded _ a priori _ and there are theoretical models able to predict this pattern ( see for example @xcite or the discussion in @xcite ) . since
@xmath1 and @xmath0 control distinct features of the pdf ( skewness and kurtosis ) it is very important to have the best observational constraints on each of them .
complementary observables have to be considered to set more stringent bounds on non - gaussian parameters , in particular on @xmath0 , and one possibility is to use the statistics of large - scale structure .
pioneering papers by dalal et al @xcite , slosar et al @xcite , matarrese et al @xcite , afshordi and tolley @xcite highlighted an interesting feature of primordial ng of local shape : it introduces a specific correlation between modes of different wavelengths which leads to a characteristic scale dependence of the halo bias .
much work has been done so far to explore this interesting topic ( see @xcite and @xcite for reviews ) .
however it turns out to be challenging to disentangle the contributions from different non - gaussian parameters using only halo and matter power spectra , since the characteristic scale dependence of the bias is primarily sensitive to a particular combination of the @xmath1 and @xmath0 parameters @xcite .
mild corrections associated with the red - shift dependence of the halo mass function have been used to set the first lss constraints on @xmath0 in @xcite , but it is difficult to convincingly distinguish the effect of @xmath1 from that of @xmath0 using only galaxy power spectra .
independent constraints on @xmath1 from cmb are useful , but not conclusive because @xmath1 could be characterized by a significant scale dependence @xcite , that makes its value probed at lss scales different with respect to the one tested at cmb scales .
a promising method that may break the degeneracy between @xmath1 and @xmath0 is to study the bispectra of halo and matter densities , which are sensitive to a non - linear bias parameter that depends specifically on @xmath0 , and allows us to break the aforementioned degeneracy .
jeong and komatsu @xcite ( see also @xcite ) were the first among various groups to include the scale dependence of halo bias when studying bispectra , by considering the non - linear evolution of the halo overdensity in a local , univariate bias expansion , where the halo abundance is taken to be a function of the local density .
they have shown that galaxy bispectra are sensitive to non - gaussian parameters beyond @xmath1 , as confirmed by @xmath11-body simulations @xcite . in this work , we elaborate on this subject .
we will show that halo and matter bispectra have interesting qualitative features that may allow us to distinguish between the effects of different primordial ng parameters .
we implement a peak - background split method within a barrier crossing approach to re - derive in part the results of @xcite in a physically transparent way , and to extend their analysis in various directions with the main aim of understanding how the parameters @xmath1 and @xmath0 can be possibly distinguished when analyzing the statistics of halo and matter bispectra . in order to focus only on the consequences of primordial non - gaussian initial conditions on the properties of bispectra
, we will not include the effects of non - linear gravitational clustering in our analysis ; that is , we work in lagrangian space and linearly transform to eulerian space .
we also neglect non - linearity in the halo mass function , i.e. , non - linear local bias , which implies that any non - vanishing bispectra will be solely due to primordial ng .
our work contains several results , and we present these in a modular way to render the paper easier to read .
we begin with some theoretical sections discussing how to implement the peak - background split and barrier crossing approaches to determine concise analytical expressions for halo and matter bispectra , when the gravitational potential is a function of either single or multiple sources ( e.g. , due to fluctuations of single or multiple fields during inflation in the very early universe ) .
our formulae are obtained by applying to bispectra the methods developed in @xcite for power spectra .
see also @xcite for an interesting paper discussing how to implement the peak background splitting approach to bispectra .
we use a generalization of the univariate bias approach , making the hypothesis that the local number of halos forming at a given position is a function of the matter contrast , _ and _ of its local correlation functions evaluated at that position .
this is an alternative to the bivariate approach @xcite where the halo abundance is assumed to be a function of both the density field and the gravitational potential . in practice ,
the physical effect of long - wavelength modes of the potential is to modulate the small - scale variance of the density field in local models of ng .
our extended univariate approach is physically transparent , and allows us to recover some of the results of the bivariate approach , in appropriate limits .
in addition we can include local variations of effective non - gaussianity parameters on halo abundance @xcite .
the resulting bispectra contain several contributions scaling with different powers of the scale @xmath12 , weighted by coefficients depending on primordial non - gaussian parameters .
in the limit of large scales , in which analytic expressions for the transfer functions can be derived , we will discuss how bispectra have qualitative features that make it possible , at least in principle , to distinguish the effects of @xmath1 and @xmath0 .
for example , by studying the properties of the bispectra as a function of the scale , we determine a connection between the sign of the halo bispectrum and ng parameters , in particular the value of @xmath0 .
moreover , we show that there exists a combination of halo and matter bispectra that is sensitive to @xmath1 only , allowing us to probe @xmath1 without contaminations from @xmath0 .
we then extend all these results , obtained in the case of a single source for the primordial gravitational potential , to a two - field example that allows us to understand how multiple sources contribute to galaxy bispectra .
in @xcite it was pointed out that if multiple fields source the primordial gravitational potential , then halo overdensities need not be fully correlated with matter overdensities .
we show that a combination of halo and matter power spectra is particularly convenient for studying the phenomenon of _ stochastic halo bias_. our analysis enable us to extend the concept of stochastic halo bias in the presence of primordial ng to the cases of both power spectra and bispectra , including the effect of @xmath0 . the fact that this stochastic bias is non - vanishing can be interpreted in terms of inequalities satisfied by primordial non - gaussian parameters , and these provide information on the number of fundamental fields sourcing ng . we determine a new combination of halo and matter bispectra
whose value can be used to distinguish the effect of single and multiple sources on the primordial gravitational potential , and discuss its physical interpretation . in the second part of our work
, we apply our theoretical findings to a systematic numerical analysis of bispectra in appropriate squeezed limits , by using press - schechter halo mass functions supplemented by non - gaussian corrections controlled by an edgeworth expansion .
this discussion enables us to apply our results to scales and redshifts that can be probed in present or future surveys . by plotting the resulting halo bispectra
, we explore the different roles played by @xmath1 and @xmath0 .
the numerical analysis confirms our analytical results .
we show that qualitative features of the profiles of halo and matter bispectra as a function of the scale are sensitive to different ng parameters , possibly allowing us to test them individually .
these results clearly demonstrate how galaxy bispectra have features that could distinguish between different non - gaussian parameters , in a way that is not possible by studying only the scale dependence of galaxy bias in power spectra of halos and matter .
we conclude with a discussion on possible improvements and future directions for our investigation .
we start with an introductory section outlining the ideas that we will use to treat the statistics of dark matter halos @xcite .
the primordial gravitational potential @xmath13 is related to the linearly evolved matter density contrast @xmath14 in terms of the poisson equation in fourier space _ k(z)=|(k , z)_k , where |(k , z)=. @xmath15 is the matter transfer function normalized such that @xmath16 at large scales @xmath17 .
@xmath18 is the linear growth factor , as a function of redshift @xmath19 .
it is normalized such that @xmath20 in matter domination . at very large scales
@xmath21 , the function @xmath22 is proportional to @xmath23 @xcite:[formulal ] |(k , z ) = _ 0(z)k^2 _ 0(z)2.1610 ^ 7()()^2
. this asymptotic behavior will be useful to get analytic results in the discussion of section [ subsec : squeezed ] .
to render the formulae less cumbersome , from now on we will suppress the @xmath19 dependence .
the field @xmath24 denotes the linear density contrast smoothed with a top - hat filter of radius @xmath25 where @xmath26 is the halo mass , and @xmath27 the background matter density . in fourier space ,
we write _ m()=w_m(k)_[smrel ] with @xmath28 the fourier transform of top - hat filter : w_m(k)=3 .
we also define @xmath29 and @xmath30 . at large scales @xmath31 . in
what follows , we will investigate the statistical properties of the spatial distribution of galaxies . in the widely accepted model of galaxy formation , known as the _ halo model _
, galaxies form in the potential wells of gravitationally collapsed dark matter _ halos _ @xcite .
the distribution of galaxies naturally follow the distribution of halos .
there are various approaches to characterize their formation , and how they are distributed in space . in the barrier crossing approach , initiated by press and schechter @xcite ,
halos are thought to form when the matter linear overdensity field overcomes a certain threshold @xmath32 .
we denote the number of halos of mass @xmath26 at time @xmath33 by @xmath34 : the press - schechter mass function gives the mean number density @xmath35 of halos as a function of the properties of the underlying linear matter overdensity field @xmath36 . in the following
, we will drop the dependence on @xmath26 and @xmath33 ( or @xmath19 ) to simplify the notation . in our work , we assume that @xmath37 depends on local physics only , i.e. it is a local function of @xmath38 and of its @xmath2-pt correlation functions @xmath39_v ( \vec{x})\,\equiv\ , \int_{{k}_i\ge \ell^{-1 } } \,\frac{d^3 k_1}{(2\pi)^3 } \dots \frac{d^3 k_n}{(2\pi)^3 } \,e^{-i \left ( \vec{k}_1+\dots+\vec{k}_n \right)\,\vec{x}}\,\langle \delta_{\vec{k}_1}\dots \delta_{\vec{k}_n } \rangle\ ] ] evaluated within a region @xmath40 around the position @xmath41 ( see section [ sec - ss ] for an additional discussion on this assumption ) : n_h ( ) = |n_h ( ( ) , _ v ( ) ) .
the autocorrelations @xmath42_v$ ] depend on the primordial intrinsic ng of the fundamental fields that seed the gravitational potential @xmath13 . by measuring how the statistics of the halo displacement @xmath43 is related to the statistics of the matter density displacement @xmath14
, one can extract information about primordial ng .
the local statistics of the autocorrelations @xmath42_v$ ] can be calculated using a heuristic but powerful method , the so called peak - background split approach @xcite .
consider a large subvolume @xmath40 of the universe containing a large number of halos .
take for concreteness a fiducial scale @xmath44 , say @xmath45 , still much larger than the scale associated with halos of mass @xmath26 .
the linearized matter overdensity field @xmath38 can be decomposed into a short wavelength part @xmath46 and a long wavelength background @xmath47 with respect to the scale @xmath44 : [ sldec ] ( t,)=_short(t,)+_long(t , ) where the short and long wavelength contributions to @xmath48 are defined as _
short(t,)&=&_k ^-1 e^-i _ ( t ) + _
long(t,)&=&_k ^-1 e^-i _ ( t ) .
the short mode @xmath46 at scales @xmath49 can be considered as the source of halos within the subvolume @xmath50 we are focussing on . as we mentioned above ,
locally @xmath35 depends on the quantity @xmath46 within the subvolume @xmath50 , and on its correlation functions @xmath51_v$ ] evaluated on the same subvolume .
instead , distances larger than @xmath52 , associated with long mode @xmath47 , are the scales over which we measure the statistics of halo displacements @xmath53 . within the subvolume @xmath50 , @xmath47 acts as a smooth background density field in the linear regime .
as we will discuss in the next sections , in the presence of primordial ng , @xmath47 has the important feature to modulate in a specific way the correlation functions @xmath51_v$ ] controlling the halo displacement @xmath53 .
this implies that primordial ng of the gravitational potential controls the statistics of @xmath53 in such a way that it makes possible to extract from the latter unambiguous information on the former .
hence , the barrier crossing approach and the peak - background split method allow us to analyze how primordial ng affects the statistics of lss .
the connections between the two methods have been carefully discussed in @xcite : the statistics of @xmath47 has the effect to perturb the critical threshold that the linear part of @xmath46 has to reach to start to collapse , and it is this effect that renders the barrier crossing approach sensitive to primordial ng . in the next sections , we will develop and apply these ideas to find quantitative results for the statistics of halo and matter displacements , both at the level of power spectrum and bispectrum , for single and multiple sources .
we start analyzing the case of single source , extending to the case of the bispectrum the results of @xcite .
we consider a large subvolume of the universe containing many halos , over which the long mode is reasonably constant , and we denote with @xmath54_{\rm v}$ ] a spatial average over this subvolume while @xmath55 is the spatial average over the entire universe . in this section
we consider a single gaussian field @xmath56 as an unique source for the primordial gravitational potential .
we split such a field in long and short modes with respect to a fiducial scale @xmath44 : [ phisplit ] = _ s+_l .
this gaussian field has zero average over the entire space @xmath57 , while non - zero average over the subvolume @xmath58_v$ ] , with @xmath40 : & = & 0 , + _ v&=&_l .
we make a local expansion of the primordial gravitational potential up to third order in terms of the gaussian field @xmath56 : & = & + ( ^2-^2)+ ( ^3 -3 ^2 ) . after implementing the long / short splitting we introduced in eq .
( [ phisplit ] ) , we get in coordinate space & = & + ( ^2-^2 ) + ( ^3 -3 ^2 ) + & & + ( 1 + 2 + 3 ( ^2-^2 ) ) _ s + & & + ( + 3 ) ( _ s^2- [ _ s^2]_v ) + & & + ( _ s^3 -3 [ _ s^2 ] _ s ) .
the previous expression demonstrates how the long wavelength mode modulates the gravitational potential .
the coefficients of the different powers of @xmath59 , which control the statistics of this quantity within the subvolume @xmath50 , receive contributions depending on the long mode @xmath60 , and this acts as a background quantity from the point of view of the subvolume @xmath50 . in light of this
, we can define the small scale effective power @xmath61_v^{1/2}$ ] , @xmath62 and @xmath63 as [ siges1 ] ( ^e ) & = & _ v^1/2 ( 1 + 2 + 3 ( ^2-^2 ) ) , + [ fnles1 ] ^e&=&+3 , + [ gnles1 ] ^e&= & , so we learn that long wavelength modes and primordial ng affect the two point statistics of the short modes , as well as the non - gaussian parameters inside the small box .
the long mode dependence of eqs ( [ siges1])-([gnles1 ] ) will be used to estimate the number of halos at a given position .
there are various approaches to the problem of quantifying how long modes contribute to the process of halo formation .
the conceptually simplest one is to use an univariate approach , in which the number of halos is expressed in power series of @xmath14 , the matter density contrast .
an univariate approach is used by jeong and komatsu @xcite .
this method includes the effects of the long mode @xmath60 of the field sourcing the gravitational potential in the halo distribution , as much as @xmath60 modulates @xmath14 .
a physical issue with this approach is that , strictly speaking , the number of halos does not depend only on the local value of the matter density contrast at a given position @xmath41 . indeed
, it also depends on how the matter density contrast is _ distributed _ around @xmath41 that is , on its correlation functions evaluated at @xmath41 . in the presence of local ng ,
these correlation functions are affected by the long mode @xmath60 of the primordial field that sources the gravitational potential ( see eqs ( [ siges1 ] ) and ( [ fnles1 ] ) ) , in a way that is not completely described by the dependence of @xmath14 on @xmath60 . a way to deal with this issue
was adopted by giannantonio and porciani @xcite : they propose a bivariate approach , in which the halo number density is expressed as power series in the long modes of both the matter density contrast _ and _ the gravitational potential . following this method
, they show that the results fit better with n - body simulations .
this approach has been then used by baldauf et al in @xcite for studying the features of galaxy bispectra . here
, we implement another method that allows us to overcome the conceptual difficulty outlined above .
it is a generalization of the univariate approach that was introduced in @xcite in the context halo power spectra : we extend it to the study of bispectra .
we assume that the number of halos at a given position , @xmath64 does not depend on the matter density contrast only , but also on all its correlation functions [ ansnhc ]
n_h = n_h ( ( x ) , [ ^n]_v ( x ) ) .
where @xmath65_v ( \vec x ) $ ] denotes the correlation function @xmath66 evaluated over a volume of size @xmath50 around the point @xmath41 . expressing the halo number as a function of the correlation functions of matter density contrast , we faithfully capture the contributions of the long mode @xmath60 on the process of halo formation . in a sense , the method that we adopt is a generalization of the univariate approach ; on the other hand , it is also able to include features associated with the long modes and primordial ng , like in the bivariate approach . in order to analyze the properties induced by primordial ng
only , we perform a taylor expansion of the halo density contrast at first order in each of the arguments of @xmath35 in eq ( [ ansnhc ] ) .
further terms in the expansion would induce further non - linearities on the long mode dependence of @xmath35 , and thus rely on the non - linear dependence of this function on the matter density contrast and its correlations .
since we intend to focus our attention on the specific non - linearities associated with primordial ng only , and we do not have full theoretical control over higher derivatives of @xmath35 along the matter contrast correlation functions , we discard these contributions , although it would be interesting to include them in a more complete analysis ( see however section [ sec : gnllss ] for some discussion on this point ) . for the same reason
, we do not include the effects of non - linear gravitational clustering in our analysis : that is , we work in lagrangian space and linearly transform to eulerian space .
as we will see , our procedure leads to manageable expressions for the bispectra , that we use to derive interesting physical consequences ; we will also compare our results with others in the literature .
let us now concretely apply the method we explained .
we are assuming that the halo number density @xmath67 within the subvolume is a function of the one - point pdf of the underlying dark matter field @xmath36 , linearly evolved until today and smoothed over the halo scale .
this matter field is proportional , via the @xmath28 and @xmath68 functions , to the gravitational potential @xmath13 within the halo region : @xmath69 .
we define the quantity @xmath70 corresponding to the primordial gaussian long mode as _ l= _ l , that is , in terms of the linearly evolved long mode for the gaussian scalar field @xmath60 ( defined by integrating over small momenta the fourier transform of @xmath56 ) .
we emphasize that , consistently with the barrier crossing approach , @xmath70 and @xmath36 are proportional to the linearly evolved primordial fluctuations . at very large scales
, we can apply the transfer function to the poisson equation to express @xmath71 as should be intended as convolutions .
we do this in order to keep a compact notation . ]
[ expdm ] ^(l)_m=_l+ .
notice that @xmath71 is made by a combination of powers of long wavelength ( gaussian ) modes . for weakly non - gaussian states ,
the one - point pdf in each subvolume can be characterized by its mean , variance , and higher connected cumulants .
therefore , at very large scales , we can write an expansion for the halo displacement field [ expfdh ] _ h&=&b_g + & & + ( - 1 ) + ( ^e-)+ ( ^e- ) with the standard bias evaluated at ( @xmath72 ) by taking derivatives along the long wave - length mode @xmath71 [ eq : bg ] b_g=. moreover , we define at the same point [ eq : beta ] _
2&=&2 0.5cm,0.5cm_3=3 0.5cm,0.5 cm _
a treatment similar to the previous one has been carried on also in @xcite , where the derivatives are done directly in terms of correlation functions instead of ng parameters .
the quantities ( [ eq : bg])-([eq : beta ] ) depend on the specific halo mass function one considers : we will discuss theoretical expressions for the @xmath73 coefficients in section [ sec : gnllss ] using an extended press - schechter approach based on an edgeworth expansion . in the following analysis
, we leave these parameters free . using formulae ( [ siges1])-([gnles1 ] ) , eq .
( [ expfdh ] ) becomes [ eqn : deltah ] _
h&= & _ l(b_g+ + ) + & & + ( ^2-^2)(b_g+ _ 2 ) + & & + ( ^3 - 3 ^2 ) ( b_g ) . we can collect these results in the convenient expression [ coldh ] _ h= b_1_l+(_l^2-_l^2 ) + ( _ l^3 - 3 _ l_l^2 ) , where we define the new bias parameters @xmath74 as b_1 & & b_g+ + [ eq : b1 ] , + b_2 & & 2 b_g + 3 _ 2 [ eq : b2 ] , + b_3 & & 6 b_g [ eq : b3 ] .
the quantity @xmath75 corresponds to the linear bias .
it receives a contribution due to primordial ng , that scales as @xmath76 at large scales : this is the well - known scale dependence of halo bias due to primordial ng . in this large scale limit ,
the linear bias function @xmath75 depends on a particular combination of @xmath1 and @xmath0 , and can not distinguish between these two quantities . on the other hand ,
the bias @xmath77 depends specifically on @xmath0 . in what follows ,
we study the bispectrum of halos and matter , which depend on @xmath77 : this can provide unambiguous information on @xmath0 allowing us to distinguish it from @xmath1 . before continuing
, we re - emphasize that in our approach we define the linear and non - linear bias in eq .
( [ coldh ] ) as coefficients of powers of the primordial , linear gaussian field @xmath60 : thus , @xmath53 acquires a non - gaussian statistics of purely local form , and this allows straightforward computation of @xmath2-point functions .
the primordial quantity @xmath78 is then linearly evolved in terms of the function @xmath79 , consistently with the barrier crossing approach in which it is the linearly evolved gravitational potential that affects the formation of halos . within this framework , as we will see in what follows , one can easily derive analytic expressions for the coefficients @xmath73 used above , using an edgeworth expansion . on the other hand , our approach differs from the univariate method of jeong and komatsu @xcite ( see also @xcite ) , in which @xmath53 is expressed in terms of powers of the evolved matter field with its own non - gaussian features associated with non - linear evolution : we will discuss more in details in appendix [ app : comparison ] the differences among the results . nevertheless , our set - up has the virtue of being particularly intuitive and simple to deal with , and is fully consistent with the hypothesis at the basis of the barrier crossing approach that we adopt , that it is the linearly evolved matter density contrast that determines the formation of halos .
as discussed above , our generalized univariate approach and results are more similar in spirit to the bivariate approach to bispectra adopted by baldauf et al @xcite : we will also compare in appendix [ app : comparison ] their results with our own .
we adopt the following definitions for the power spectra associated with the two point functions of halo and matter densities p_mm(2)^3(+)&=&_m _ m , + p_hm(2)^3(+)&= & , + p_hh(2)^3(+)&=&_h _ h .
we define the halo - halo power spectrum assuming that the shot - noise contribution @xmath80 has been subtracted , and analogously for the matter - halo and matter - matter power spectrum . using the formulae ( [ expdm ] ) , ( [ coldh ] ) and the linear , primordial power spectrum . ] @xmath81
, we find p_mm&= & , + p_hm&= & , + p_hh&= & . in the limit of very large scales
, we can express the previous expressions for the power spectra as p_mm&=&_0_0 ^ 2 k , + p_hm&=&k(_0b_g++ ) _ 0 _ 0 , + p_hh&=&k(_0b_g++ ) ^2_0 , neglecting the terms that are subleading in powers of @xmath12 , and using eq .
( [ formulal ] ) which also holds specifically at large scales .
notice the famous scale dependence of halo bias induced by primordial ng in the halo - halo power spectrum .
on the other hand , the primordial non - gaussian parameters @xmath1 and @xmath0 appear in the same footing in the previous expressions , rendering them difficult to disentangle . in order to overcome this degeneracy , in the following we will consider the statistics of three point function and study the bispectra of halo and matter density contrasts .
the three point functions are defined as stands for @xmath82 .
the same for @xmath83 . as for the power spectra
, we assume that shot - noise contributions are removed . ]
b_mmm(2)^3(_1+_2+_3)&=&_m _ m _ m , + b_hmm(2)^3(_1+_2+_3)&= & , + b_hhm(2)^3(_1+_2+_3)&= & , + b_hhh(2)^3(_1+_2+_3)&=&_h _ h _ h , obtaining at leading order ( tree - level ) b_mmm&=&2 _ 0 ^ 2 ^(1)^(2 ) ^(3)(++)[bsmmm ] , + b_hmm&=&\{_0 ^ 2^(2)^(3)+}[bshmm ] , + b_hhm&=&\{_0 ^ 2^(3)+}[bshhm ] , + b_hhh&=&_0 ^ 2(b_2^(1)+b_2^(2)+b_2^(3))[bshhh ] , where @xmath84 and @xmath85 .
notice that the scale dependence of @xmath86 is due to primordial ng , and this induces a dependence on @xmath87 for these quantities ( see eqs .
( [ eq : b1])-([eq : b3 ] ) ) . in the previous formulae , we focussed only on tree - level contributions to the bispectra , neglecting so - called `` loop '' effects .
for this reason , the bias parameter @xmath88 does not appear .
focussing on the halo bispectrum , eq .
( [ bshhh ] ) , recall that the bias parameters @xmath75 and @xmath77 depend on primordial ng , see eqs ( [ eq : b1],[eq : b2 ] ) .
we are thus able to recover the trispectrum dependent contributions identified by jeong and komatsu @xcite .
we will explain in appendix [ app : comparison ] how to compare ( in appropriate limits ) our results with @xcite : for the moment , we will elaborate the physical consequences of our findings .
we focus on isosceles triangle configurations , considering a squeezed limit in which primordial local ng contributions yield the largest signal @xcite .
we now follow the convention of jeong and komatsu by setting @xmath89 : in other words , we consider isosceles triangles in momentum space .
the squeezed limit , then , corresponds to configurations in which @xmath90 .
the previous expressions ( [ bsmmm])-([bshhh ] ) become b_mmm&= & [ bsqmmmms ] , + b_hmm&=&\ { 4 + } + [ bsqhmmms ] , + b_hhm&= & \ { 2 + + & & + } [ bsqhhmms ] , + b_hhh&= & \ { 2 ^3 ( ) b_1 ( ) b_2(k)+ b_2 ( ) ( k)b_1(k ) } , [ bsqhhhms ] where we keep all contributions .
these results , which use our generalized univariate approach , are similar to the ones obtained by @xcite using a bivariate approach : see appendix [ app : comparison ] for more details . expanding the previous functions , one finds at large scales @xmath21 and in squeezed limit @xmath90 , using @xmath91 and the approximation @xmath92 , b_mmm&&4 _ 0 ^ 2 _ 0 ^ 3 , + b_hmm&&4 _ 0 ^ 2 _ 0 ^ 3 b_g + , + b_hhm&&4 _ 0 ^ 2 _ 0 ^ 3 b_g^2 + , + & & + , + [ eq : bhhhk ] b_hhh&&4 _ 0 ^ 2 _ 0 ^ 3 b_g^3 + , + & & + , + & & + ( _ 2 + _ 3 ) ^2 .
the expansion of the various expressions for the bispectra at large scales makes manifest how the non - gaussian parameters @xmath1 and @xmath0 characterize different bispectra , and suggests that the study of bispectra allows to distinguish the effects of each of them .
indeed , several contributions appear with different scale dependences and different coefficients : these can be used to distinguish the effects of @xmath1 and @xmath0 . here
we discuss how our results allow one to break the degeneracy between @xmath0 and @xmath1 that affects the power spectra .
we will discuss two possible methods to use bispectra to distinguish these two quantities .
the previous results are obtained in the limit of large scales .
we use this limit in order to allow simple analytical approximations for the transfer function @xmath15 and thus allow us to analytically understand physically interesting features of the bispectra as a function of the scale @xmath12 , for configurations of squeezed isosceles triangles in momentum space @xmath89 . as a specific example , let us study in more detail the properties of @xmath93 , using the eq .
( [ eq : bhhhk ] ) valid at large scales .
the aim is to determine distinctive signatures of @xmath0 in the profile of halo bispectra .
( [ eq : bhhhk ] ) has been written in a form aimed to emphasize the different scale dependences of the various contributions , it can also be expressed in a more concise form as [ bhhhz ] b_hhh=(3_2 + 2_0 b_gk^2 ) , which makes the zeroes of @xmath93 clearly identifiable .
recall that the bispectrum ( [ bhhhz ] ) is defined for isosceles configurations in momentum space , in which the equal sides of the triangle have length @xmath12 , while the small side length @xmath94 .
hence , we are studying @xmath93 as a function of the size of the long sides of an isosceles triangle in momentum space . in the limit of large @xmath95 ,
( [ bhhhz ] ) has various roots : the ones that can be real are k_root^(1)&=&[kroot1 ] , + k_root^(2)&=&[kroot2 ] , hence the existence and positions of the zeroes of galaxy bispectrum depend on the values of the non - gaussian parameters @xmath1 and @xmath0 .
working in regimes in which the parameters @xmath96 , @xmath97 and @xmath98 are positive ( see the discussion in section [ sec : gnllss ] ) , the quantity @xmath99 is real only if @xmath100 and @xmath0 are both non - vanishing and have opposite signs .
hence , if the profile of @xmath93 changes sign as a function of the scale , it would be a tantalizing hint of the presence of non - vanishing @xmath0 with an opposite sign to @xmath1 . while these roots have been derived in the limit of large scales where we can neglect the scale dependence of the transfer function
, we will show that these features are accurately reproduced by a full numerical analysis in section [ sec : gnllss ] .
the bispectrum @xmath93 is not the only quantity that allows to distinguish among different ng parameters . combining different bispectra , indeed
, one can more directly probe the individual effects of @xmath1 with negligible contamination from @xmath0 .
for example , let us consider the combination [ intcomb ] c_fnl+3 -3 .
making any approximation , using equations ( [ bsqmmmms])-([bsqhhhms ] ) one finds that the previous quantity reads , for isosceles triangles , [ dcfnl ] c_fnl = .
hence it depends specifically on @xmath1 ( with only a minor dependence on @xmath0 in the denominator ) .
hence , a measurement of non - vanishing @xmath101 would provide a very clean probe of the quantity @xmath1 that is ( almost ) independent on the size of @xmath0 .
nevertheless , the combination @xmath101 is challenging to probe observationally , since it is harder to observe bispectra and power spectra involving dark matter densities ( although optimistically this might be realized in the future using gravitational lensing ) .
the bottom - line of this section is that the bispectra of halos and matter have distinctive qualitative features that allow one to distinguish the effects of different non - gaussian parameters , thereby providing observables that may allow one to break the degeneracies between @xmath102 and @xmath103 that are present in the study of power spectra only .
let us emphasize once more that the analytical results obtained so far include the effects of primordial ng only , without taking into account non - linear effects due to gravity .
this approximation allowed us to analytically identify more directly the effects of different ng parameters in the halo distribution .
a systematic analysis is needed to understand how much the features we pointed out survive when including the effects of gravity , and is left for future work .
see however section [ sec : gnllss ] for a numerical study of our analytic results , with some preliminary discussion on possible effects of non - linear gravitational clustering .
the next two sections are more theoretical , and investigate the implications of our findings when applied to the multiple source extension of our treatment .
in this section , we discuss how to extend the previous analysis to the case of more than one source field for the primordial gravitational potential .
we focus for definiteness on the case of two fields , that contribute to the gravitational potential as in the following ansatz [ mansatz ] & = & + + ( 1+)^2 ( ^2-^2)+(1+)^3 ( ^3 -3 ^2 ) , with = .
while @xmath56 can be thought as the inflaton fluctuation , @xmath104 can be seen as a spectator field that is responsible for introducing ng of local form in the gravitational potential .
the two fields @xmath56 and @xmath104 are by themselves gaussian : we split them as long and short mode with respect to a fiducial scale @xmath44 , as done in the previous section : & = & _ s+_l , + & = & _ s+_l .
this set - up can be seen as an extension of the curvaton - like model analyzed in section of 4.1 of @xcite , where we include the contribution of @xmath0 . from now on , @xmath105 and @xmath106 .
starting from ansatz ( [ mansatz ] ) , the three and four pt functions read ( k_1 ) ( k_2 ) ( k_3 ) & = & , + ( k_1 ) ( k_2 ) ( k_3 ) ( k_4 ) & = & 2(56 ) ^2 + , where = ( 65 ) ^2(1 + ) . the usual equality @xmath107 is recovered , as expected , in the single field limit @xmath108 .
after implementing the long / short splitting , one gets in coordinate space & = & _ l+_l + ( 1+)^2 ( _ l^2-_l^2)+6 ( 1+)^3 ( _ l^3 -3 _ l^2_l ) + & & + _ s+ ( 1 + 2 ( 1+)^2 _ l + 3 ( 1+)^3(_l^2-_l^2 ) ) _ s + & & + ( 1+)^2 ( + 3(1+)_l ) ( _ s^2- [ _ s^2]_v ) + & & + ( 1+)^3 ( _ s^3 -3 [ _ s^2 ] _ s ) .
the previous expression demonstrates how long wavelength modes modulate the gravitational potential .
the coefficients of the different powers of @xmath59 and @xmath109 , which control the statistics of this quantity within the subvolume @xmath50 , receive contributions depending on the long mode @xmath110 , and thus acts as a background quantity from the point of view of the subvolume @xmath50 . in light of this , we can define the small scales effective power @xmath61_v^{1/2}$ ] , @xmath62 and @xmath63 as [ siges2 ] ( ^e ) & = & _ v^1/2 \ { 1 + 2 ( 1 + ) _
l + ( 1+)^2(2 ^ 2_l^2 + 3 ( _ l^2-_l^2 ) ) + & & + 12(1+)^3 ( -8 ^ 3_l^3 + 12 _ l ( _ l^2-_l^2 ) ) } , + ( 1+)^2 ^e&=&(1+)^2 ( + 3(1+)_l ) , + [ gnles2 ] ( 1+)^3 ^e&=&(1+)^3 , so we learn that long wavelength modes affect the two - point statistics of the short modes , as well as the non - gaussian parameters inside the small box . following our treatment of the single source case , we can define the quantity @xmath70 corresponding to the linearly evolved sum of primordial gaussian perturbation as _ l= ( _ l+_l ) . using the poisson equation ,
the long modes in the smoothed matter density contrast reads ^(l)_m=_l+ .
we express the halo density contrast as the expansion _
+ & & + ( -1 ) + ( ^e-)+ ( ^e- ) , with the standard bias evaluated at @xmath111 b_g= , and , moreover , we define at the same point _ 2&=&2 0.5cm,0.5cm_3=3 0.5cm,0.5cm_4=4 .
substituting the formulae above , one finds [ eqn : deltah2 ] _ h&= & b_g _
l + 2_2 ( 1 + ) _
l + 6 _ 3 ( 1 + ) _
l + & & + + & & + + & = & b_g _ l+b_1_l+(b_2 _ l^2-b_3 _ l^2 ) + ( b_4_l^3 - 3 b_5 _
we define the bias @xmath74 as [ mb1 ] b_1 & & b_g + _ 2 ( 1 + ) + _ 3 ( 1 + ) , + [ mb2 ] b_2 & & ( 1+)^2 , + b_3 & & ( 1+)^2 , + b_4 & & ( 1+)^3 , + b_5 & & ( 1+)^3 .
we adopt the following definitions for the two point functions ( the same as in the single source case we discussed in previous sections ) p_mm(2)^3(+)&=&_m _ m , + p_hm(2)^3(+)&= & , + p_hh(2)^3(+)&=&_h _ h , where p_mm&=&= , + p_hm&=&= , + p_hh&=&= . the three point functions are defined as b_mmm(2)^3(_1+_2+_3)&=&_m _ m _ m , + b_hmm(2)^3(_1+_2+_3)&= & , + b_hhm(2)^3(_1+_2+_3)&= & , + b_hhh(2)^3(_1+_2+_3)&=&_h _ h _ h , obtaining b_mmm&=&2 ( 1+)^2_0 ^ 2 ^(1)^(2 ) ^(3)(++)[bmmmm ] , + b_hmm&=&\{_0 ^ 2^(2)^(3)+}[bmhmm ] , + b_hhm&=&\{_0 ^ 2^(3)+}[bmhhm ] , + b_hhh&=&_0 ^ 2(b_2^(1)+b_2^(2)+b_2^(3))[bmhhh ] , where @xmath84 and @xmath85 . in the previous formulae
, we focussed only on tree - level contributions , neglecting loop effects .
remarkably , only the bias parameter @xmath75 and @xmath77 appear .
we now set @xmath89 .
that is , we consider isosceles triangles .
the squeezed limit , then , corresponds to configurations in which @xmath90 .
the previous expressions ( [ bmmmm])-([bmhhh ] ) become b_mmm&= & [ bsqmmm ] , + b_hmm&=&\ { 4 ( 1+)^2 + + & & + } [ bmsqhmm ] , + b_hhm&= & \ { 2 ( 1+)^2 + + & & + } [ bmsqhhm ] , + b_hhh&= & \ { 2 ^3 ( k/)b_1 ( ) b_2(k)+ b_2 ( ) ( k)b_1(k ) } , [ bmsqhhh ] where we keep all contributions . notice that the structure of the bispectra is very similar to the case of single source , apart from coefficients depending on @xmath112 , the ratio of the power spectra of the gaussian fields .
this implies that when expanded using eqs .
( [ mb1],[mb2 ] ) , one finds various contributions depending both on bispectrum ( @xmath1 ) and trispectrum ( @xmath4 and @xmath0 ) parameters .
the qualitative features of the bispectra as a function of the scale remain the same as the ones discussed in the single source case .
for example , let us write the multi - source version of the quantity @xmath101 that we wrote in eq .
( [ dcfnl ] ) : it reads c_fnl = . also in the multiple source case , this quantity represents a clean probe of the ng parameter @xmath1 ( almost ) independent from the value of @xmath0 . in the next section ,
we investigate the contributions depending on @xmath112 that play an important role in defining the properties of the stochastic halo bias .
stochastic halo bias arises when halo overdensities are not fully correlated with matter overdensities @xcite .
an example is the inequality @xmath113 where @xmath114 is the halo - matter bias .
as we have seen , halo and matter overdensities depend on primordial ng : neglecting the effect of shot - noise , the aforementioned stochasticity is associated with inequalities between non - gaussian parameters .
such inequalities have been well studied in the analysis of ng from inflation , and are typically ( but not only @xcite ) associated with the presence of multiple sources for the primordial gravitational potential .
a famous inequality is the suyama - yamaguchi inequality @xmath115 . in this section
, we will investigate the role of @xmath0 for characterizing the stochasticity of halo bias , and we will extend the notion of stochasticity to the bispectra .
we will make use of some technical results on inequalities among primordial non - gaussian parameters , which we relegate to appendix [ app : ine ] . in this section , as in the previous ones , we will not include loop effects .
a convenient quantity to quantify the stochasticity of halo bias using power spectra is @xmath116 , defined as @xcite r_p=- ( ) ^2 . using the results of section [ sec : npoint ]
, we find at large scales @xmath117 [ psin ] r_p=0 .
hence @xmath118 , and the equality @xmath119 can be obtained when ng is absent , or in the single field limit @xmath120 . notice that @xmath1 and @xmath0 appear in a combination that renders the identification of their individual effects difficult a feature that we already discussed by studying power spectra in the previous sections .
let us provide a heuristic understanding for this result , extending the arguments of @xcite to include @xmath0 . in that paper ,
setting @xmath121 , the inequality @xmath122 was associated to the suyama - yamaguchi inequality @xmath123 . on the other hand , when including @xmath0 , we learn that eq .
( [ psin ] ) reads [ psin2 ] _ 0k^4 r_p=()^2 + 2_2_3+_3 ^ 20 .
hence we find two additional terms proportional to @xmath0 besides the first corresponding to the suyama - yamaguchi inequality .
as we have seen in the previous sections , the halo displacement can be schematically expressed as an expansion in terms of local correlation functions of the linearly evolved matter density field _
h = b_g + _ 2 [ ^2]+_3 [ ^3]+ using this expansion , we can schematically express the quantity @xmath116 as r_p&= & - + & = & _ 2 ^ 2(- ) + 2 _ 2 _ 3 ( - ) + _ 3 ^ 2 ( - ) [ exprp ] . to write the previous formula
, we used the schematic notation discussed in appendix [ app : ine ] to express squeezed and collapsed limits of @xmath2-pt functions . with @xmath124\rangle$ ] we denote the squeezed configuration of a 3-pt function , in which one of the momenta is sent to zero : this notation clearly demonstrates that a long mode @xmath14 modulates the 2-pt function @xmath125 $ ] . with @xmath126\ , [ \delta^2]\rangle$ ]
we denote the collapsed configuration of a 4-pt function , in which the momentum connecting the two @xmath127 $ ] is going to zero . the combination in the first term in the rhs of equation ( [ exprp ] )
relates the collapsed limit of a four point function with the squeezed limit of the three point function , and is associated with the suyama - yamaguchi inequality ( see eq .
( [ nongi1b ] ) ) .
it corresponds to the first term in ( [ psin2 ] ) .
the remaining terms are instead associated to the inequalities ( [ nongi2 ] ) , ( [ nongi3 ] ) , that involve collapsed limits of higher point functions , and are at the origin of the additional contributions in ( [ psin2 ] ) depending on @xmath0 .
the conclusion is that @xmath122 does not measure just the sy inequality , but a combination of contributions associated with various inequalities : thus , it can not clearly distinguish between @xmath1 and @xmath0 .
stochasticity can be defined also using bispectra : this allows one to break the degeneracy between @xmath1 and @xmath0 also at the level of stochastic halo bias .
consider the bispectra defined for isosceles triangles , both for the case of single and multiple sources as discussed in the previous sections .
( squeezed configurations are special cases in which one of the sides of the triangle has vanishing size , i.e. the parameter @xmath95 defined in section [ sec : isomulti ] is large ) .
we define the following quantity @xmath128 that uses bispectra to measure stochasticity r_b= + - - .
a non - vanishing @xmath128 is associated with stochasticity : indeed , using the same arguments of the previous subsection , one can expand the previous quantity @xmath128 in terms of @xmath73 , @xmath1 and @xmath0 finding a sum of several contributions that vanish in the single source limit .
these contributions , among other things , depend on appropriated collapsed limits of six point functions ; an example of contributions to @xmath128 is the following combination , which appears in the expansion for @xmath128 multiplied by suitable powers of the available parameters . using the inequalities among non - gaussian parameters discussed in appendix [ app : ine ]
, one can check that @xmath129 vanishes in the single source case ( neglecting loop effects ) , while it is generally non - vanishing for multiple sources .
for the specific two - field model of section [ sec : npoint ] , neglecting loop corrections , one can straightforwardly check that @xmath128 is non - vanishing only when @xmath130 , i.e. in presence of multiple sources .
we find the following value for @xmath128 at large scales and in the limit of squeezed configurations @xmath90 r_b= - + o(k^4 ) .
hence , at very large scales @xmath17 , @xmath128 is sensitive to @xmath1 and its sign : if @xmath1 vanishes , then @xmath128 vanishes no matter of the size of @xmath0 .
hence @xmath128 can be an useful complementary observable besides @xmath116 to study the stochasticity of halo bias .
the analytical results of the previous sections indicate that the effects of @xmath1 and @xmath0 can be distinguished by measuring bispectra of halo and matter densities , in a way that is not possible by studying power spectra only .
indeed , in the scale - dependent bias of the halo power spectrum the parameters @xmath1 and @xmath0 are weighted by the same power of the scale @xmath12 , while the bispectra contain different contributions depending on non - gaussian parameters that scale with different powers of @xmath12 , implying it is possible to overcome the degeneracy between @xmath1 and @xmath0 without having to study galaxy @xmath131-point functions .
we theoretically analyzed this subject in the previous sections , making simplifying assumptions aimed to concentrate on the effects of primordial ng on the halo bias . while in our theoretical discussions , in order to obtain analytic expressions for our quantities , we focussed on the limit of large scales where simple approximations for the transfer functions hold , in this section we numerically investigate the features of the bispectrum also for smaller scales , in order to determine properties of this quantity that allow us to distinguish the effects of @xmath0 from the ones of @xmath1 . in particular , we investigate two properties of halo and matter bispectra that we found when analytically studying them in the previous theoretical sections : 1 ) the dependence on @xmath0 of the sign of halo bispectra as a function of the scale . and 2 ) the fact that a particular combination of halo and matter bispectra , @xmath101 ( see eq .
( [ dcfnl ] ) ) depends on @xmath1 only , allowing one to cleanly distinguish @xmath1 from @xmath0 . to start with
, we focus on the single source case and investigate the slope of the halo bispectrum @xmath93 associated with squeezed isosceles triangles in momentum space .
we consider highly biased halos for all the possible combinations of @xmath132 and @xmath133 , in the squeezed configuration @xmath134 ( that is , the ratio between the long and short sides of an isosceles triangle in momentum space is 100 ) .
the aim of our analysis is to confirm the analytical results we determined in the previous sections , and to investigate at which scales , redshifts and masses a distinct signature of the effects of @xmath0 might be observed by future surveys , like euclid @xcite . on the other hand , our analysis intends to concentrate on the effects of primordial ng , without including the non - linear evolution of gravity , nor non - linearities associated with the particular dependence of the mass function on the matter density contast .
these contributions will be addressed in a future publication .
the starting point of our arguments is eq . for @xmath93 . in order to obtain the quantities @xmath75 , @xmath77 defined in eqs .
, , we compute @xmath135 by using the transfer function from camb @xcite with the state parameters consistent with the wmap@xmath136bao@xmath137sn cosmology @xcite ( the reason of this choice will be clear later )
. then , to evaluate the bias coefficients @xmath98 , @xmath97 , @xmath138 defined in eqs .
, , a specific mass function must be chosen at the expense of making additional assumptions .
the mass function gives the number density of halos at a given redshift @xmath19 with mass within @xmath26 and @xmath139 ; under the assumption of universality , the mass function depends only on @xmath140 .
it takes the form @xcite [ expnh ] n_h(m , z)= f ( ) , where the variance @xmath141 of the smoothed linear density contrast @xmath36 is given by _
m(z)^2w_m^2(k)|^2(k , z)p_(k ) and we replace the spherical collapse threshold @xmath142 with @xmath143 , to improve the agreement between the barrier model and simulations @xcite .
when the primordial gravitational potential is exactly gaussian , the mass function @xmath144 appearing in ( [ expnh ] ) assumes the press - schechter form [ psdis ] f_ps()=e^- . for non - gaussian primordial fluctuations , @xmath145 must be replaced by a more accurate mass function .
several mass functions that account for non - gaussian initial conditions have been proposed in the literature ( see for example @xcite and the reviews @xcite ) . for our analysis
, we will adopt the results of smith , ferraro , and loverde @xcite , based on the edgeworth expansion ; these provide a useful , theoretically motivated expression for halo bias which agrees with @xmath11-body simulations .
the edgeworth expansion can be applied within the barrier crossing model by building up the pdf as a series of higher - order cumulants ( defined as @xmath146 ) times the press - schechter gaussian distribution ( [ psdis ] ) . in order to have a positive definite distribution ,
it is important to specify the order of the cumulant at which the series is truncated ( see @xcite for further details , and also @xcite for additional papers that include the effects of kurtosis ) .
fortunately , for the weakly ng regime we are interested in , the edgeworth expansion converges rapidly , providing a useful approximation as long as @xmath147 . within this regime ,
the edgeworth expansion corresponds simply the press - schechter mass function plus first - order corrections in @xmath1 and @xmath0 : [ eq : edgeworth ] n(m)= , where @xmath148 is @xmath2-th hermite polynomial and we defined the @xmath2-th ng cumulant _ 3(m)&=_3^(1)(m)(6.610 ^ -4)[eq : kappa_3 ] , + _ 4(m)&=_4^(1)(m)(1.610 ^ -7)[eq : kappa_4 ] .
the approximated results above are taken from @xcite . under these approximations universality
is recovered , and within the validity of the edgeworth expansion , the bias coefficients are obtained by taking the derivatives of eq .. decomposing b_g = b_g0+b_0+b_0 , we have b_g0&=&1 + [ eq : bg0 ] , + b_0&=&-^(1)_3(m)()- ( ) , + b_0&=&-^(1)_4(m)()- ( ) .
moreover , _
2&=&2 ^ 2 - 2 [ eq : beta2 ] , + _ 3&=&^(1)_3(m)()- ( ) [ eq : beta3 ] .
the physical interpretation of the results above is the following : @xmath149 is the usual gaussian bias in eulerian coordinates , while the terms @xmath150 and @xmath151 describe the scale - independent shift to @xmath152 .
unfortunately they can not be used to constrain primordial ng , since the bias of a real tracer population is not known a priori .
then , @xmath97 is the well - known scale - dependent bias for an @xmath1 cosmology . by neglecting the explicit mass dependence of the edgeworth expansion
, we have forced eq . to satisfy the relation @xmath153 found in @xcite
; this is valid for a universal mass function .
finally , @xmath138 describes the scale - dependent bias introduced by the @xmath0 parameter .
recall that the mass function here is initially computed in lagrangian coordinates , therefore the @xmath154 in eq
. takes into account the dynamical effects produced by linear gravity and bring us to eulerian coordinates @xcite .
let us emphasize again that here we are only including the linear effects of gravity .
non - linear effects would add further non - gaussian features to bispectra not specifically due uniquely to primordial ng , the focus of this paper that we intend to analyze in a future publication .
smith et al .
compared the edgeworth prediction for @xmath138 with @xmath11-body simulations , finding that it breaks down at lower halo mass , @xmath155 . to fix this problem
, they first noted that eq .
can be written as _ 3=_3- ( ) , then , they found a good agreement with simulations by changing the coefficients of the polynomial in the brackets as follows [ eq : beta3ast ] _
3^=_3- ( ) .
our numerical analysis can be easily performed by using the following fitting functions @xcite _ 3&=&0.000329(1 + 0.09 z ) b_g0
^ -0.09[eq : kappa_32 ] , + & = & -0.000061(1 + 0.22 z ) b_g0 ^ -0.25[eq : dkappa_3 ] .
the quantities - , provide the necessary terms to describe highly biased tracers .
all the results provided by smith et al . have been tested against @xmath131 simulations with gaussian initial conditions , @xmath156 simulations with @xmath157 and @xmath158 simulations with @xmath159 , for a total of @xmath160 simulations in the wmap@xmath136bao@xmath137sn cosmology .
given that these results have been well tested on several simulations , we assume the same dataset , and we do not expect to see any radical change in our order - of - magnitude analysis for a slightly different @xmath161cdm cosmology .
however , further simulations should be carried out to test the validity of @xmath162 for the low values of @xmath1 and @xmath0 we are interested in .
let us now present our numerical results and their interpretation .
1 displays the absolute value of the halo bispectrum ( using a @xmath163-scaling ) , for halos of mass @xmath164 , at redshift @xmath165 , and @xmath12-space triangles with squeezing parameter @xmath134 .
each of the three plots in this figure shows the results for a different value of @xmath1 : @xmath166 for case 1 , @xmath167 for case 2 , and @xmath168 for case 3 .
the lines on each plot correspond to different values of @xmath0 .
each point on the @xmath12-axis actually involves three different values of the wavenumber specifying the three sides of a squeezed triangle in momentum space : two at @xmath169 ( the long equal sides ) and one at @xmath170 ( the small side ) .
the plots span an interval of @xmath12 between @xmath171 and @xmath172 , corresponding to scales within ranges that will be probed by future observations made by euclid , under the hypothesis that this survey will gain one order of magnitude in scale with respect to boss @xcite .
since negative values of the bispectrum are possible , we plot the absolute value of @xmath93 .
the presence of cusps indicate the changing from positive to negative values ( or vice - versa ) . for the sake of clarity ,
we use dashed lines for negative values of the bispectrum and solid lines for positive ones .
[ fig : first ] + + _ top _ : the absolute value of the halo bispectrum @xmath93 in the squeezed configuration @xmath173 , for halo mass @xmath164 , redshift @xmath165 , @xmath166 .
different lines correspond to the values @xmath174 .
we use solid lines to describe positive values and dotted lines for neagative values ; the presence of a cusp indicates a change of sign . the pattern of zeros in the plotted range of wavenumbers can be used to distinguish @xmath1 from @xmath0 : see the text for further details .
+ _ bottom left _ : same as the top panel , but for @xmath167 .
for @xmath175 a zero is present .
+ _ bottom right _ : same as top panel , but for @xmath168 .
notice that a single zero appears , for @xmath176 .
an inspection of these plots suggests an important qualitative feature of these results . if the parameters are contained within appropriate intervals , the halo bispectrum as a function of @xmath12 changes sign more times if @xmath0 has opposite sign with respect to @xmath1 .
for example , in fig .
1 , case 1 corresponding to @xmath166 , we see that the green curve , corresponding to the logarithm of the bispectrum for @xmath177 , changes sign twice instead of once as the other curves .
moreover , the positions of the zero at smaller scales , that exists for all the curves , depends on the value of @xmath0 . in fig .
1 , case 2 with @xmath167 , the bispectra for different values of @xmath0 never change sign in the interval of scales we consider . finally , in fig .
1 , case 3 with @xmath178 , the red curve with @xmath179 changes sign while all the other curves do not .
the qualitative features in the bispectra profiles as a function of scale become even more pronounced when considering higher redshifts or higher mass objects . to show this behaviour , we present plots displaying the halo bispectrum for two additional cases .
the first set , displayed in fig .
2 , is for @xmath164 , @xmath180 , @xmath134 , while the second one is for @xmath181 , @xmath165 , @xmath134 and is displayed in fig .
each plot uses the same conventions as fig .
the qualitative considerations we made above are still valid .
in particular , we see , at higher mass or redshift than our fiducial choice ( @xmath164 , @xmath165 , @xmath134 ) , an increase in the absolute value of the halo bispectrum .
interestingly , the various contributions to the bispectrum weighted by different powers of @xmath12 make the magenta curve for @xmath175 in the plots with @xmath178 change sign , in both cases presented here .
this new feature is due to the dependence of the bias parameters @xmath75 , @xmath77 on mass and redshift .
[ fig : first2 ] + same as fig . 1 , but for @xmath182 .
the amplitude of the bispectra have increased considerably over the case with @xmath183 ( shown in fig .
1 ) , the positions of the zeros have changed for @xmath184 ( top panel , and the number of zeros has increased for @xmath185 ( bottom right panel ) . [
fig : first3 ] + same as fig . 1 ,
but for @xmath180 .
the amplitude of the bispectra have increased significantly over the case of @xmath165 ( shown in fig . 1 ) and
the positions of the zeros differ from those observed both in fig . 1 and fig
. 2 . hence ,
the previous plots show that the presence or absence of @xmath0 affects the bispectra profiles : in particular it governs how many times the halo bispectrum changes sign as a function of the scale .
we already discussed this qualitative behavior in section [ subsec : squeezed ] , when analytically studying the roots of the bispectrum in the large scale limit .
the numerical results of this section confirm those analytical findings , indicating that the qualitative profile of the halo bispectrum for isosceles triangles as a function of the scale can provide an interesting signature of the presence of @xmath0 .
hence , we learn that the fact that the bispectrum changes sign in multiple locations as a function of the scale is a distinctive signal of the presence of @xmath0 . optimistically , this can be used to constraint values of @xmath0 smaller than the ones that can be tested by cmb .
these features might be observed in cases in which large values of @xmath98 are realized , since in this case the bispectrum generated by primordial ng is more significant .
thus , the bispectrum is greater for larger halo masses @xmath26 and , at constant @xmath26 , greater at higher redshift .
for example , in fig .
4 we provide an example of halos at redshift @xmath180 , for @xmath166 . in this case , we see that the green curve corresponding to the halo bispectrum changes sign twice for @xmath186 .
an observation of this phenomenon can then probe quite small values for @xmath0 .
we can also compare the results of fig .
4 with the analytical formulae for the zeros of the bispectrum , eqs ( [ kroot1 ] ) and ( [ kroot2 ] ) . for the halo mass ,
redshift and @xmath95 parameter we are considering , we find @xmath187 , @xmath188 , @xmath189 . computing the positions of the zeros of the bispectrum for the case @xmath186 with the analytical formulae of eqs ( [ kroot1 ] ) and ( [ kroot2 ] ) , one finds the zeros at @xmath190 and @xmath191 h / mpc , in agreement with the numerical results within one order of magnitude .
this implies that our analytical findings , although obtained in the approximation in which we neglect the scale dependence of the transfer function , provide reasonably accurate predictions for the positions of the zeroes . + the plot shows the absolute value of the halo bispectrum @xmath93 in the squeezed configuration @xmath173 , for halo mass @xmath164 and redshift @xmath180 .
@xmath1 is kept constant and equal to @xmath192 , while different lines correspond to the values @xmath193 .
we use solid lines to describe positive values and dotted lines for negative values ; the presence of a cusp indicate a change of sign .
@xmath166 produces a specific pattern of zeros in the line with @xmath186 .
if this feature can be observed in the halo bispectrum , it will improve the constrain on @xmath0 to few hundreds .
[ fig : fourth ] we now present a plot that represents the second theoretical observation we made in the previous sections : if it is possible to probe and observe bispectra correlating halo and matter densities ( for example using gravitational lensing ) then by analyzing combinations such as @xmath101 of eq .
( [ dcfnl ] ) we can have clean measurements of @xmath1 only , with a negligible contamination of @xmath0 . in fig .
5 we represent this quantity for different values of @xmath1 , halo masses and redshifts ; the lines are practically insensitive on @xmath0 , and depend only on @xmath1 and on the halo mass .
this feature , as we discussed around eq .
( [ dcfnl ] ) , is valid at all scales ( although at smaller scales one should take into more proper account the non - linear effects of gravity ) .
this concretely shows that an observation of a non - vanishing value for the quantity @xmath101 would be a clean indication of a non - vanishing @xmath1 .
combined with other measurements ( for example associated with power spectra ) this fact could then be used to obtain independent constraints on @xmath0 .
+ the plot shows the quantity @xmath101 in the squeezed configuration @xmath173 , for different halo masses ( @xmath194 ) and redshifts ( @xmath195 ) .
@xmath101 is positive for @xmath178 and negative for @xmath166 . remarkably , the lines are insensitive to the values of @xmath0 . [
fig : five ] we conclude that the scale dependence of bispectra of isosceles configurations in momentum space have qualitative features that might allow us to distinguish the effects of different ng parameters , in a way that is not possible by studying power spectra only . while in the previous discussion we neglected the effects of non - linear gravity in the results for the bispectra ,
let us end this section with a preliminary analysis of its contributions .
focussing on squeezed configurations , the next plot shows a comparison between the amplitude of bispectra associated with primordial ng only ( color lines ) and bispectra due uniquely to the non - linear effects of gravity ( black lines ) .
we divide the bispectrum due to gravity in two contributions .
the black dotted line corresponds to the amplitude of bispectra associated only with linear bias @xmath152 , and is controlled by the so - called @xmath196 function .
this corresponds to the three point function for first order fluctuations .
the black continuous line , instead , is the contribution associated with the quadratic coefficient as @xmath197 , although in the literature it is usually referred to as @xmath77 . ]
@xmath197 of the local nonlinear biasing model of @xcite .
we make the choice @xmath198 in order to be more conservative respect to @xcite ( see the review @xcite for more details . ) + the plot shows the amplitude of halo bispectra as a function of scale .
the colored lines are bispectra associated with primordial ng only ; the black lines are contributions to bispectra due to non - linear gravitational clustering .
the black dotted line corresponds to the gravitational clustering contributions to @xmath93 depending on the linear bias @xmath152 only .
the continuous line is the gravity bispectrum associated with non - linear bias @xmath77 ( we take @xmath199 ) .
[ fig : six ] we notice that the amplitude of the bispectrum due to primordial ng tends to dominate at large scales ( small @xmath12 ) .
however , if the primordial ng parameters are chosen to be small , the bispectrum induced by gravity is more important at smaller scales ( see green line ) . for somewhat larger values of the ng parameters ( with the magnitude of @xmath0 still below cmb constraints )
the amplitude of the primordial bispectrum can dominate over the gravity contributions also at relatively small scales ( see violet line ) .
the primordial bispectrum represented in the violet line has non - trivial profile due to competing effects of primordial @xmath1 and @xmath0 , in particular it has zeroes associated with the presence of @xmath0 : if such features can be detected in the bispectrum profile , they can be used to help distinguishing among the effects of @xmath1 and @xmath0 .
the plot above is only indicative , in the sense that one should study in more detail the _ combined _ effects of non - linear gravity and primordial ng in shaping the bispectra , and not only comparing their relative amplitude . on the other hand ,
it gives the idea that for interesting values of ng parameters the primordial contribution can lead to a rich profile for the bispectrum , controlled by the primordial ng parameters , with an amplitude larger or comparable to the one due to gravitational clustering only .
primordial ng characterizes the interactions of the fields sourcing the density fluctuations that seed the large - scale structure of our universe . in this work , using a generalized univariate approach that we compared with other methods in the literature , we analytically and numerically investigated how the statistics of lss , in particular the study of bispectra of halos and matter , allow one to probe primordial ng .
our aim was to find ways to determine distinctive features of the local non - gaussian parameters @xmath1 and @xmath0 , controlling respectively skewness and kurtosis of the probability distribution function , and whose effects in the halo and matter power spectra are nearly degenerate .
this investigation is important since although cmb constraints on @xmath1 have greatly improved with planck , it is not clear whether cmb measurements only can set stringent constraints on @xmath0 . in the first part of this work
, we showed analytically that the profiles of halo and matter bispectra have qualitative features that if measurable would clearly distinguish the effects of each non - gaussian parameter .
we exploited the scale dependence of the halo bias induced by primordial ng .
we worked in a simplified situation in which only linear bias and linear effects of gravitational clustering are taken into account , to single out the effects of primordial ng in the statistics of lss .
we have shown that the profile of the halo bispectrum as a function of the scale , for isosceles configurations of triangles in momentum space , has properties that provide simple , qualitative information on each non - gaussian parameter .
for example , the number and the position of the zeros of this function depend on the size of @xmath0 , and can be analytically computed with our formulae . if @xmath0 is present , and its sign is opposite with respect to the sign of @xmath1 , the number of zeros of the bispectra increases as a function of the scale .
then , we determined a particular combination of halo and matter bispectra and power spectra , denoted @xmath101 , that is proportional to @xmath1 only and is ( nearly ) independent on the value of @xmath0 .
a detection of this combination would be a clean probe of @xmath1 with no contaminations associated with @xmath0 .
we then generalized our findings to the case in which multiple sources contribute to form the primordial density fluctuations , and studied how stochastic halo bias is affected by the presence of @xmath0 .
we also provided new ways to study stochastic halo bias in terms of combinations of bispectra and power spectra . in the second part of this work ,
we numerically confirmed and further developed our analytical results , focussing on a press - schechter approach to halo formation supplemented by appropriate non - gaussian initial conditions .
this analysis allowed us to apply our theoretical formulae for the bispectra focussing on scales and redshifts that might be probed by future surveys .
we confirmed our theoretical considerations , showing plots representing how the qualitative profiles of halo bispectra in suitable configurations depend on non - gaussian parameters , and might be used to set constraints on them .
we ended our discussion with a preliminary analysis of the contributions of non - linear gravitational clustering , indicating that non - linear gravity tends to contaminate the bispectrum profiles , but that the qualitative features we investigated survive for sufficiently large primordial non - gaussian parameters .
our results can be extended in various directions . from the theoretical side , the non - linear effects associated with features of the mass function , bias and gravitational clustering ( the non - linear transformation from lagrangian to eulerian coordinates ) , as well as loop effects ,
should be properly included in the context of perturbation theory , in order to understand whether and how primordial ng controls the qualitative features of bispectra also after taking these non - linearities into account . having a good control of gravitational effects would also allow one to reliably study the bispectra for smaller values of the scale and for configurations that are less squeezed than the ones we considered here , and extend the validity of our results in these regimes .
we plan to explore this interesting issue in a future publication .
another direction of investigation is a more systematic study of observational applications of our results , trying to quantify whether the methods we propose are sufficiently efficient to set constraints on @xmath0 that are more stringent than those available from the cmb .
an important extension of our work is thus to forecast the signal - to - noise of galaxy bispectra as a function of scale , for future surveys such as euclid , and thereby test whether the features we predict are observable .
such study will exploit our theoretical inputs to develop optimised methods to distinguish the effects of @xmath1 and @xmath0 using galaxy bispectra and thus quantify their size and sign independently .
we thank tommaso giannantonio and kazuya koyama for useful discussions .
gt is supported by an stfc advanced fellowship st / h005498/1 .
dw is supported by stfc grant st / k00090x/1 .
gt and dw thank the participants of the workshop constraining the early universe in the post planck era " , funded by the royal society , for useful feedback on this work .
in this appendix , we compare our results with the ones obtained in two previous papers that studied galaxy bispectra .
we start discussing a comparison with the paper of jeong and komatsu @xcite .
jeong and komatsu were the first to investigate the halo bispectrum up to the ng parameter @xmath0 including the scale dependence of bias .
it is important to point out the connection between their results and our findings .
we address the issue in this appendix .
first of all , let us briefly sketch the formalism used in @xcite .
they assume that the halo density contrast can be expressed as a local bias expansion of the matter overdensity in eulerian coordinates , whose fourier transform is _
@xmath200 , while @xmath201 is the @xmath2-th order quantity of the linear density contrast , given by perturbation theory @xcite .
the advantage of this approach is that non - linear evolution is automatically included when building the @xmath158-point function , even though calculations become harder . on the other hand , the assumptions that led us to the halo density contrast of eq .
are somewhat different .
as we will show , the peak - background split method within a barrier crossing approach allows to re - obtain part of the results of @xcite in a simple way ( at least in the squeezed limit ) and extend them up to higher powers of ng parameters .
however , since we have not yet accounted for the non - linear effects of gravity , a comparison is possible only between primordial ng effects at this stage .
moreover , it is known ( see for example @xcite ) that the local lagrangian and eulerian bias are not compatible in non - linear regimes . by plugging eqs .
, , into eq . and collecting all the terms proportional to the same ng parameter
, we find : b_hhh^&=&_0 ^ 2 b_g^3 ( k_1 ^ 3+k_2 ^ 3+k_3 ^ 3)4_0 ^ 2 b_g^3 ^ 3_0 , + + b_hhh^^2&=&^2_0 ^ 2 b_g^2 _ 2 4 ^ 2_0 ^ 2 b_g^2 _
2_0 ^ 2 , + + b_hhh^&=&_0 ^ 2 b_g^2 _ 2 ( + + ) 6_0 ^ 2 b_g^2 _
2_0 ^ 2 , where the approximated results are obtained by using @xmath91 and @xmath92 , in the squeezed configuration @xmath202 with large @xmath95 .
the term @xmath203 is the same primordial ng contribution found in @xcite , while the approximated results for @xmath204 and @xmath205 match respectively those of eq.@xmath206 and eq.@xmath207 of @xcite on large scales .
however , a subtle difference between our results and those of @xcite arise , due to the different initial assumptions .
the matching is truly correct only if the equalities @xmath208 , @xmath209 hold : it would be interesting to understand the physical meaning of this fact .
some interesting consideration follow from the ratios of the previous quantities : & = & , + + & = & _ 0 k^2 , + + & = & ^2 .
it is clear from the first ratio that the @xmath210 term can dominate over the @xmath1 one in the squeezed limit , as pointed out in @xcite .
moreover , the bispectrum distinguishes between @xmath205 and @xmath203 ( second ratio ) but , surprisingly , the @xmath0 term shows the same scaling as the @xmath210 one . at this point ,
all the primordial ng contributions found in @xcite have been recovered but new terms come out from eq .. we collect them explicitly here : b_hhh^^3&=&^3 _ 0 ^ 2 b_g _ 2 ^ 2 ( + + ) 4 ^3 _ 0 ^ 2 b_g _ 2 ^ 2 _ 0 , + + b_hhh^&=&_0 ^ 2 b_g\{b_g_3+_2 ^ 2 } + & & 2_0 ^ 2 b_g^3(2
+ 3 ) , + + b_hhh^^2&=&^2_0 ^ 2 b_g_2_3(++)6 ^ 2_0 ^ 2 b_g_2_3_0 , + + b_hhh^^2&=&^2_0 ^ 2 + & & 2 ^ 2_0 ^ 2 ^ 3(4 + 3 ) , + + b_hhh^^2&=&^2_0 ^ 2 + & & 4 ^ 2_0 ^ 2 ^ 3(+3 ) , + + b_hhh^^3&=&^3 _ 0 ^ 2 _ 2 _ 3 ^ 2 ( + + ) 6 ^ 3 _ 0 ^ 2 _ 2 _ 3 ^ 2 .
these additional terms appear naturally within our framework and suggest that higher powers of ng parameters can in principle dominate over the lower ones , extending what jeong and komatsu found for the ratio @xmath211 .
all the contributions above are important for the specific features of the halo bispectrum discussed in section [ sec : gnllss ] .
let us now compare our approach with the one by baldauf et al in @xcite .
the easiest way to proceed is to reconsider our equation ( [ coldh ] ) : by substituting our results for the bias coefficients @xmath74 ( see eqs ( [ eq : b1])-([eq : b3 ] ) ) , and using the relation @xmath212 , we can schematically write our expression for @xmath53 as an expansion [ bivarus ] _
h = b_g_l+_2_l + _ 3 _ l+b_g_l _ l+32_2_l^2 + b_g _ l_l^2 , where to render the expression more compact , we understand the subtraction of the averages contained in ( [ coldh ] ) .
this ` bivariate ' way of re - express our formula for the halo bias makes clear the connection with the bivariate approach of @xcite , in which the lagrangian halo bias is expanded in powers of @xmath70 and @xmath60 : [ bivar ] _
h = b_10 _ l+b_01 _ l+ _ l^2 + _ l^2+b_11_l_l+ , where the bias coefficient @xmath213 is defined as b_ij=()|__l=0 , _ l=0 .
the values we find for the @xmath213 coefficients are obtained by comparing eqs .
( [ bivarus ] ) with ( [ bivar ] ) corresponds to our @xmath214 .
] : b_10&=&b_g + [ eqb01 ] b_01&=&_2 + _
3 + b_11&=&b_g + b_02&=&3 _ 2 + b_20&=&0 they differ from the ones of @xcite for two reasons .
the first reason is that , as explained in the main text , in our approach we decided to perform a taylor expansion at _ linear order _ in @xmath14 and on all its correlation functions in langrangian space . in this way we consider only non - linear effects induced by primordial ng , and we exclude non - linearities associated with higher derivatives of the mass function .
this approximation is also justified by the fact that higher derivatives of the mass function along non - gaussian parameters are under less reliable theoretical control .
the second reason is that we include also the effects of ( first ) derivatives of the mass function along @xmath1 ( that provide contributions proportional to @xmath138 ) that are instead neglected in a bivariate approach .
hence , for example , the coefficient @xmath215 in eq ( [ eqb01 ] ) , using our definition of @xmath97 , @xmath138 reads b_01&=&+while the first term can be associated with the analogous result reported in @xcite , the second term is absent in their case .
indeed , the bivariate approach produces the @xmath216 dependence in the bias of the halo - halo and halo - matter power spectra as a contribution from the trispectrum @xcite . on the other hand , our expression for @xmath217 is vanishing because we decided not to consider the second derivatives of @xmath35 along @xmath70 .
for the same reason of not considering such second derivatives , various contributions in the remaining coefficients @xmath213 proportional to @xmath210 , found in @xcite are absent in our formulae .
in this appendix we collect some technical results on inequalities among primordial non - gaussian parameters , that we use in section [ sec : stochastic ] .
the most famous of these inequalities is dubbed suyama - yamaguchi inequality , and relates the collapsed limit of the 4-pt function with the square of the squeezed limit of a 3-pt function of a given fluctuation @xmath14 in fourier space @xcite [ nongi1 ] _ q0_k_1 _ -k_1-q _ k_2 _ -k_2+q _ q0 using the definitions for @xmath4 and @xmath1 and specializing to the case of primordial curvature fluctuations , this inequality corresponds to the suyama - yamaguchi inequality .
it is convenient to simplify the notation , expressing the squeezed limit of the 3-pt function in ( [ nongi1 ] ) in a synthetic way as @xmath124\rangle$ ] , meaning that the momentum @xmath218 connecting @xmath219 and @xmath127 $ ] is sent to zero : this notation emphasizes that the long mode @xmath220 with @xmath221 modulates the 2-pt function @xmath127 $ ] .
analogously , the collapsed limit of the 4-pt function can be expressed as @xmath222 [ \delta^2 ] \rangle$ ] . with the help of this notation , inequality ( [ nongi1 ] )
succinctly reads [ nongi1b ] in @xcite a simple proof of this inequality has been provided , by inserting a complete set of normalized momentum eigenstates @xmath223 into the quantity @xmath224 [ \delta^2 ] \rangle$ ] of the left hand side of ( [ nongi1b ] ) .
the zero - momentum , long wavelength eigenstate provides the square of the squeezed 3pt function in the right hand side , to which one has to add the additional ( positive definite ) contributions that lead to the inequality above . in the collapsed limit and single source case , the additional contributions vanish and one saturates the inequality .
although less explored in the literature , it is also possible to build further inequalities involving collapsed and squeezed limits of higher point functions ( see for example @xcite ) : the virtue of the method of @xcite is that the proofs of such inequalities can be straightforwardly generalized to those cases . as specific examples , we can write the following inequalities satisfied by five and six point functions , in appropriate collapsed limits [ ^3 ] [ ^2 ] & & [ nongi2 ] + [ nongi3 ] [ ^3 ] [ ^3 ] & & using the methods of @xcite , it is relatively straightforward to prove that the inequalities saturate to equalities in the single source case .
these inequalities have been used in section [ sec : stocp ] for physically interpreting the notion of stochastic halo bias applied to power spectra of halos and matter .
when considering stochastic halo bias for the bispectra , section [ sec : stocb ] other inequalities are needed , that involve subtler collapsed limits among higher order point functions . we collect them here & & + [ ^2 ] [ ^2 ] & & + [ ^2 ] & & they can be analyzed and proved again with the methods of @xcite .
n. bartolo , e. komatsu , s. matarrese and a. riotto , `` non - gaussianity from inflation : theory and observations , '' phys .
* 402 * ( 2004 ) 103 [ astro - ph/0406398 ] .
l. verde and s. matarrese , `` detectability of the effect of inflationary non - gaussianity on halo bias , '' astrophys . j. * 706 * , l91 ( 2009 ) [ arxiv:0909.3224 [ astro-ph.co ] ] .
n. bartolo , s. matarrese , o. pantano and a. riotto , `` second - order matter perturbations in a lambdacdm cosmology and non - gaussianity , '' class .
* 27 * , 124009 ( 2010 ) [ arxiv:1002.3759 [ astro-ph.co ] ] .
m. bruni , j. c. hidalgo , n. meures and d. wands , `` non - gaussian initial conditions in @xmath161cdm : newtonian , relativistic and primordial contributions , '' arxiv:1307.1478 [ astro-ph.co ] .
p. a. r. ade _ et al .
[ planck collaboration ] , `` planck 2013 results . i. overview of products and scientific results , '' arxiv:1303.5062 [ astro-ph.co ] .
t. sekiguchi and n. sugiyama , `` optimal constraint on @xmath225 from cmb , '' arxiv:1303.4626 [ astro-ph.co ] .
j. smidt , a. amblard , c. t. byrnes , a. cooray , a. heavens and d. munshi , `` cmb constraints on primordial non - gaussianity from the bispectrum ( @xmath226 ) and trispectrum ( @xmath227 and @xmath228 ) and a new consistency test of single - field inflation , '' phys .
d * 81 * ( 2010 ) 123007 [ arxiv:1004.1409 [ astro-ph.co ] ] .
s. nurmi , c. t. byrnes and g. tasinato , `` a non - gaussian landscape , '' jcap * 1306 * ( 2013 ) 004 [ arxiv:1301.3128 [ astro-ph.co ] ] ; c. t. byrnes and g. tasinato , `` non - gaussianity beyond slow roll in multi - field inflation , '' jcap * 0908 * ( 2009 ) 016 [ arxiv:0906.0767 [ astro-ph.co ] ] ; j. emery , g. tasinato and d. wands , `` local non - gaussianity from rapidly varying sound speeds , '' jcap * 1208 * ( 2012 ) 005 [ arxiv:1203.6625 [ hep - th ] ] .
t. suyama , t. takahashi , m. yamaguchi and s. yokoyama , `` implications of planck results for models with local type non - gaussianity , '' jcap * 1306 * ( 2013 ) 012 [ arxiv:1303.5374 [ astro-ph.co ] ] .
n. dalal , o. dore , d. huterer and a. shirokov , `` the imprints of primordial non - gaussianities on large - scale structure : scale dependent bias and abundance of virialized objects , '' phys . rev .
d * 77 * ( 2008 ) 123514 [ arxiv:0710.4560 [ astro - ph ] ] .
a. slosar , c. hirata , u. seljak , s. ho and n. padmanabhan , `` constraints on local primordial non - gaussianity from large scale structure , '' jcap * 0808 * ( 2008 ) 031 [ arxiv:0805.3580 [ astro - ph ] ] ; s. matarrese and l. verde , `` the effect of primordial non - gaussianity on halo bias , '' astrophys .
j. * 677 * ( 2008 ) l77 [ arxiv:0801.4826 [ astro - ph ] ] . n. afshordi and a. j. tolley , `` primordial non - gaussianity , statistics of collapsed objects , and the integrated sachs - wolfe effect , '' phys .
d * 78 * ( 2008 ) 123507 [ arxiv:0806.1046 [ astro - ph ] ] .
j. -o .
gong and s. yokoyama , `` scale dependent bias from primordial non - gaussianity with trispectrum , ''
arxiv:1106.4404 [ astro-ph.co ] . v. desjacques , u. seljak and i. iliev , `` scale - dependent bias induced by local non - gaussianity : a comparison to n - body simulations , '' mon . not .
roy . astron .
* 396 * ( 2009 ) 85 [ arxiv:0811.2748 [ astro - ph ] ] .
v. desjacques , d. jeong and f. schmidt , `` accurate predictions for the scale - dependent galaxy bias from primordial non - gaussianity , '' phys .
d * 84 * ( 2011 ) 061301 [ arxiv:1105.3476 [ astro-ph.co ] ] .
v. desjacques , d. jeong and f. schmidt , `` non - gaussian halo bias re - examined : mass - dependent amplitude from the peak - background split and thresholding , '' phys . rev .
d * 84 * ( 2011 ) 063512 [ arxiv:1105.3628 [ astro-ph.co ] ] .
v. desjacques and u. seljak , `` primordial non - gaussianity from the large scale structure , '' class .
* 27 * ( 2010 ) 124011 [ arxiv:1003.5020 [ astro-ph.co ] ] .
l. verde , `` non - gaussianity from large - scale structure surveys , '' adv .
astron .
* 2010 * ( 2010 ) 768675 [ arxiv:1001.5217 [ astro-ph.co ] ] .
n. roth and c. porciani , `` can we really measure fnl from the galaxy power spectrum ?
, '' arxiv:1205.3165 [ astro-ph.co ] .
v. desjacques and u. seljak , `` signature of primordial non - gaussianity of @xmath229-type in the mass function and bias of dark matter haloes , '' phys .
d * 81 * ( 2010 ) 023006 [ arxiv:0907.2257 [ astro-ph.co ] ] .
t. giannantonio , a. j. ross , w. j. percival , r. crittenden , d. bacher , m. kilbinger , r. nichol and j. weller , `` improved primordial non - gaussianity constraints from measurements of galaxy clustering and the integrated sachs - wolfe effect , '' arxiv:1303.1349 [ astro-ph.co ] .
m. loverde , a. miller , s. shandera and l. verde , `` effects of scale - dependent non - gaussianity on cosmological structures , '' jcap * 0804 * ( 2008 ) 014 [ arxiv:0711.4126 [ astro - ph ] ] .
c. t. byrnes , m. gerstenlauer , s. nurmi , g. tasinato and d. wands , `` scale - dependent non - gaussianity probes inflationary physics , '' jcap * 1010 * ( 2010 ) 004 [ arxiv:1007.4277 [ astro-ph.co ] ] ; d. jeong and e. komatsu , `` primordial non - gaussianity , scale - dependent bias , and the bispectrum of galaxies , '' astrophys . j. * 703 * ( 2009 ) 1230 [ arxiv:0904.0497 [ astro-ph.co ] ] .
e. sefusatti , `` 1-loop perturbative corrections to the matter and galaxy bispectrum with non - gaussian initial conditions , '' phys .
d * 80 * ( 2009 ) 123002 [ arxiv:0905.0717 [ astro-ph.co ] ] .
t. nishimichi , a. taruya , k. koyama and c. sabiu , `` scale dependence of halo bispectrum from non - gaussian initial conditions in cosmological n - body simulations , '' jcap * 1007 * ( 2010 ) 002 [ arxiv:0911.4768 [ astro-ph.co ] ] .
e. sefusatti , m. crocce and v. desjacques , `` the halo bispectrum in n - body simulations with non - gaussian initial conditions , '' arxiv:1111.6966 [ astro-ph.co ] .
d. baumann , s. ferraro , d. green and k. m. smith , `` stochastic bias from non - gaussian initial conditions , '' arxiv:1209.2173 [ astro-ph.co ] .
r. scoccimarro , l. hui , m. manera and k. c. chan , `` large - scale bias and efficient generation of initial conditions for non - local primordial non - gaussianity , '' phys .
d * 85 * ( 2012 ) 083002 [ arxiv:1108.5512 [ astro-ph.co ] ] .
t. giannantonio and c. porciani , `` structure formation from non - gaussian initial conditions : multivariate biasing , statistics , and comparison with n - body simulations , '' phys .
d * 81 * ( 2010 ) 063530 [ arxiv:0911.0017 [ astro-ph.co ] ] . c. t. byrnes , s. nurmi , g. tasinato and d. wands , `` inhomogeneous non - gaussianity , '' jcap * 1203 * ( 2012 ) 012 [ arxiv:1111.2721 [ astro-ph.co ] ] .
d. tseliakhovich , c. hirata and a. slosar , `` non - gaussianity and large - scale structure in a two - field inflationary model , '' phys .
d * 82 * ( 2010 ) 043531 [ arxiv:1004.3302 [ astro-ph.co ] ] .
f. bernardeau , s. colombi , e. gaztanaga and r. scoccimarro , `` large scale structure of the universe and cosmological perturbation theory , '' phys .
* 367 * ( 2002 ) 1 [ astro - ph/0112551 ] .
s. d. m. white and m. j. rees , `` core condensation in heavy halos : a two stage theory for galaxy formation and clusters , '' mon . not .
* 183 * ( 1978 ) 341 .
w. h. press and p. schechter , `` formation of galaxies and clusters of galaxies by selfsimilar gravitational condensation , '' astrophys .
j. * 187 * ( 1974 ) 425 .
s. cole and n. kaiser , `` biased clustering in the cold dark matter cosmogony , '' mon . not .
soc . * 237 * , 1127 ( 1989 ) .
j. m. bardeen , j. r. bond , n. kaiser and a. s. szalay , `` the statistics of peaks of gaussian random fields , '' astrophys .
j. * 304 * ( 1986 ) 15 .
s. ferraro , k. m. smith , d. green and d. baumann , `` on the equivalence of barrier crossing , peak - background split , and local biasing , '' arxiv:1209.2175 [ astro-ph.co ] .
t. baldauf , u. seljak and l. senatore , `` primordial non - gaussianity in the bispectrum of the halo density field , '' jcap * 1104 * ( 2011 ) 006 [ arxiv:1011.1513 [ astro-ph.co ] ] .
s. yokoyama , t. matsubara and a. taruya , `` halo / galaxy bispectrum with primordial non - gaussianity from integrated perturbation theory ( ipt ) , '' arxiv:1310.4925 [ astro-ph.co ] .
g. tasinato , c. t. byrnes , s. nurmi and d. wands , `` loop corrections and a new test of inflation , '' phys .
d * 87 * ( 2013 ) 4 , 043512 [ arxiv:1207.1772 [ hep - th ] ] .
r. laureijs _ et al . _
[ euclid collaboration ] , `` euclid definition study report , '' arxiv:1110.3193 [ astro-ph.co ] .
e. komatsu _ et al .
_ [ wmap collaboration ] , `` five - year wilkinson microwave anisotropy probe ( wmap ) observations : cosmological interpretation , '' astrophys . j. suppl .
* 180 * ( 2009 ) 330 [ arxiv:0803.0547 [ astro - ph ] ] .
a. pillepich , c. porciani and o. hahn , `` universal halo mass function and scale - dependent bias from n - body simulations with non - gaussian initial conditions , '' arxiv:0811.4176 [ astro - ph ] .
m. grossi , l. verde , c. carbone , k. dolag , e. branchini , f. iannuzzi , s. matarrese and l. moscardini , `` large - scale non - gaussian mass function and halo bias : tests on n - body simulations , '' mon . not .
* 398 * ( 2009 ) 321 [ arxiv:0902.2013 [ astro-ph.co ] ] .
s. matarrese , l. verde and r. jimenez , `` the abundance of high - redshift objects as a probe of non - gaussian initial conditions , '' astrophys .
j. * 541 * ( 2000 ) 10 [ astro - ph/0001366 ] . k. m. smith , s. ferraro and m. loverde , `` halo clustering and @xmath103-type primordial non - gaussianity , '' jcap * 1203 * ( 2012 ) 032 [ arxiv:1106.0503 [ astro-ph.co ] ] .
m. maggiore and a. riotto , `` the halo mass function from excursion set theory with a non - gaussian trispectrum , '' mon . not .
astron . soc .
* 405 * ( 2010 ) 1244 [ arxiv:0910.5125 [ astro-ph.co ] ] .
m. loverde and k. m. smith , `` the non - gaussian halo mass function with @xmath1 , @xmath0 and @xmath228 , '' jcap * 1108 * ( 2011 ) 003 [ arxiv:1102.1439 [ astro-ph.co ] ] . h. j. mo and s. d. m. white , `` an analytic model for the spatial clustering of dark matter halos , '' mon . not .
* 282 * ( 1996 ) 347 [ astro - ph/9512127 ] .
a. j. ross , w. j. percival , a. carnero , g. -b .
zhao , m. manera , a. raccanelli , e. aubourg and d. bizyaev _ et al .
_ , `` the clustering of galaxies in sdss - iii dr9 baryon oscillation spectroscopic survey : constraints on primordial non - gaussianity , '' arxiv:1208.1491 [ astro-ph.co ] .
l. anderson , e. aubourg , s. bailey , d. bizyaev , m. blanton , a. s. bolton , j. brinkmann and j. r. brownstein _ et al .
_ , `` the clustering of galaxies in the sdss - iii baryon oscillation spectroscopic survey : baryon acoustic oscillations in the data release 9 spectroscopic galaxy sample , '' mon . not .
* 427 * ( 2013 ) 4 , 3435 [ arxiv:1203.6594 [ astro-ph.co ] ] . j. n. fry and e. gaztanaga , `` biasing and hierarchical statistics in large scale structure , '' astrophys . j. * 413 * ( 1993 ) 447 [ astro - ph/9302009 ] .
t. matsubara , `` nonlinear perturbation theory integrated with nonlocal bias , redshift - space distortions , and primordial non - gaussianity , '' phys .
d * 83 * ( 2011 ) 083518 [ arxiv:1102.4619 [ astro-ph.co ] ] .
r. scoccimarro , e. sefusatti and m. zaldarriaga , `` probing primordial non - gaussianity with large - scale structure , '' phys . rev .
d * 69 * ( 2004 ) 103513 [ astro - ph/0312286 ] .
e. sefusatti and e. komatsu , `` the bispectrum of galaxies from high - redshift galaxy surveys : primordial non - gaussianity and non - linear galaxy bias , '' phys .
d * 76 * ( 2007 ) 083004 [ arxiv:0705.0343 [ astro - ph ] ] .
t. suyama and m. yamaguchi , `` non - gaussianity in the modulated reheating scenario , '' phys .
d * 77 * ( 2008 ) 023505 [ arxiv:0709.2545 [ astro - ph ] ] .
t. suyama and s. yokoyama , `` extension of local - type inequality for the higher order correlation functions , '' jcap * 1107 * ( 2011 ) 033 [ arxiv:1105.5851 [ astro-ph.co ] ] .
v. assassi , d. baumann and d. green , `` on soft limits of inflationary correlation functions , '' jcap * 1211 * ( 2012 ) 047 [ arxiv:1204.4207 [ hep - th ] ] . | the statistics of large - scale structure in the universe can be used to probe non - gaussianity of the primordial density field , complementary to existing constraints from the cosmic microwave background . in particular , the scale dependence of halo bias , which affects the halo distribution at large scales , represents a promising tool for analyzing primordial non - gaussianity of local form .
future observations , for example , may be able to constrain the trispectrum parameter @xmath0 that is difficult to study and constrain using the cmb alone .
we investigate how galaxy and matter bispectra can distinguish between the two non - gaussian parameters @xmath1 and @xmath0 , whose effects give nearly degenerate contributions to the power spectra .
we use a generalization of the univariate bias approach , making the hypothesis that the number density of halos forming at a given position is a function of the local matter density contrast and of its local higher - order statistics . using this approach ,
we calculate the halo - matter bispectra and analyze their properties .
we determine a connection between the sign of the halo bispectrum on large scales and the parameter @xmath0 .
we also construct a combination of halo and matter bispectra that is sensitive to @xmath1 , with little contamination from @xmath0 .
we study both the case of single and multiple sources to the primordial gravitational potential , discussing how to extend the concept of stochastic halo bias to the case of bispectra .
we use a specific halo mass - function to calculate numerically the bispectra in appropriate squeezed limits , confirming our theoretical findings . |
resonant magnetic perturbations ( rmps ) are one of the promising methods of edge - localized mode ( elm ) control in future tokamak reactors @xcite , which is necessary as already for iter it is predicted that at full plasma current uncontrolled elms can not be tolerated @xcite . the rmps either suppress the elms completely or cause a change of the elm character towards smaller energy loss and higher frequency @xcite . the latter effect is called elm mitigation and may be a viable option for elm control on iter if the reduction in peak heat load at the divertor is sufficient to protect the plasma - facing components .
one major unknown parameter in the predictions of tolerable elm sizes is the elm wetted area , which together with the total elm loss determines the peak heat load on the plasma - facing components @xcite .
the compatibility of mitigated elms with plasma - facing components is therefore not only determined by the elm energy loss reduction achieved , but also by any changes of the wetted area accompanying the mitigation @xcite . as an example , it was observed on mast that mitigation achieves a less significant reduction of peak elm heat load than of the total elm loss due to a concurrent reduction of the wetted area @xcite .
similar results were reported on jet with the carbon wall @xcite . on the contrary on jet with the iter - like wall
the reduction of the peak heat load during elm mitigation was partly due to an increase of the wetted area @xcite .
the study of spatial structure of elm loads and its changes due to mitigation is thus important for extrapolating elm mitigation scenarios and still an open area of research .
one significant change of the spatial structure of elm loads on divertor during mitigation was observed on diii - d @xcite , mast @xcite and nstx @xcite : the power is deposited predominantly in the footprint structures @xcite formed by the magnetic perturbation . on mast it is observed only at the early stage of the elm , though .
this effect may be one of the contributing factors to the wetted area changes .
indeed on jet footprints of heat flux are also seen during the mitigated elm crash and were proposed as the mechanism of the wetted area increase @xcite .
it also can provide a valuable insight in the physics of elms and their mitigation .
the effect suggests that the phase of the elms becomes locked to the phase of the applied rmp .
the rmp has usually a low toroidal mode number though ( @xmath2 , up to @xmath3 and @xmath4 on mast ) while the elm may have a higher toroidal mode number . on mast
the toroidal mode number of the elm filaments was found to be between 10 and 20 and , remarkably , was not affected by the mitigation by @xmath3 or @xmath5 rmp @xcite . although a high-@xmath1 elm can nonlinearly couple to a low @xmath1 perturbation @xcite , it is not a priori clear how this can lead to the elm producing a structure on the divertor which corresponds to a perturbation with a much lower @xmath1 .
the goal of this paper is to present a simple mechanism which may in some cases explain the observed peaking of elm loads .
it is assumed that at least a part of the elm energy loss is caused by parallel transport along open , chaotic field lines in the homoclinic tangle formed by the magnetic component of the elm @xcite .
the simplest explanation would be thus based on the analysis of the structure of the homoclinic tangle , without considering mhd interactions of the elms with the applied rmp .
we will show that the structure of the homoclinic tangle itself can indeed provide such an explanation .
it is then needed to demonstrate that the explanation remains valid when the mhd interaction of the elm with the rmp is taken into account . for this we will use the results of simulations of rmp - mitigated elms with the nonlinear mhd code jorek .
finally we will discuss the relation of the mitigated elm features predicted by our model to the observations on mast .
the homoclinic tangle is formed by two intersecting surfaces the stable and unstable invariant manifolds of the x - point @xcite .
the manifolds themselves are formed by field lines which approach the x - point asymptotically when followed in the direction parallel or antiparallel to the field vector . in the axisymmetric case
they coincide and form the separatrix . under the influence of the magnetic perturbation they split and
their distance is in the first order approximation given by the melnikov integral @xcite .
the usefulness of the concept of invariant manifolds lies in the fact that they form the boundary of the region of stochastic field lines which connect the divertor plates with the plasma core .
the intersection of the stochastic region with the divertor plates manifests itself as a footprint receiving high parallel heat fluxes and having long connection length .
the shape of the footprints can be calculated by tracing the connection length of field lines starting at the divertor plates and representing the result in a so - called laminar plot @xcite .
alternatively the melnikov integral can be used to approximate the shape of the manifolds .
the melnikov integral is a function of the field line label at or near the separatrix , which we call the homoclinic coordinate @xmath6 @xcite and define it to be equal to the toroidal angle at the outboard midplane and constant on field lines .
as noted in @xcite , a radially displaced bundle of field lines at the outboard midplane will show up as a spiralling structure on the divertor , along a line of constant @xmath6 .
the displacement of the field line is caused by the magnetic perturbation and the boundary of the bundle is the invariant manifold .
the melnikov integral @xmath7 quantifies the displacement in flux coordinate of the invariant manifold with respect to the original separatrix at the divertor plates .
the shape of the footprint boundary in the @xmath8 space is therefore given by the equation @xmath9 .
@xmath10 means that the manifold is inside the separatrix and therefore does not intersect the divertor . to know the footprint shape on the divertor we perform the transform from the @xmath11 space to the divertor coordinates
@xmath12 , where @xmath13 is the radial distance along the divertor surface and @xmath14 the toroidal angle .
the manifold will then take an elongated shape along a spiralling curve of constant @xmath6 .
more details can be found in @xcite .
the melnikov integral can be calculated by integrating the radial component of the perturbation along field lines on the unperturbed separatrix . for our application
it is crucial that the melnikov integral for a perturbation with a single dominant toroidal mode @xmath1 will also have a simple harmonic form : @xmath15 which can be characterized only by the amplitude @xmath16 , phase @xmath17 and a toroidal mode number @xmath1 .
there is thus very little freedom in the form of divertor footprints in a given equilibrium magnetic field .
the melnikov integral is linear : for a linear combination of perturbation fields , the resulting melnikov integral is the corresponding linear combination of their melnikov integrals . we will use the melnikov integral formalism to calculate the footprints in the case of the elm , in the case of the applied rmp and in the case where the elm is present together with the rmp field , representing an mitigated elm .
we will represent here the elm as a magnetic perturbation with a higher toroidal mode number than the rmp .
for concreteness we will use @xmath18 for the elm and @xmath0 for the rmp . in the @xmath8 space
the melnikov integrals are simple sinusoids : @xmath19 , @xmath20 .
the footprints of the elm on the divertor ( with coordinates @xmath13 the radial distance along the plate and @xmath14 the toroidal angle ) form a similar pattern as the footprints of the rmp , only with a finer structure of spirals .
figure [ fig : footprintn8 ] shows the invariant manifold ( red ) with the curve of constant @xmath6 ( black , compare with fiures in @xcite ) .
indeed , field line tracing in a simple model of the elm magnetic perturbation indicates @xcite and observations on asdex upgrade confirm @xcite that the elm create a spiral - shaped heat load footprint which lies on a curve of a constant toroidal angle when mapped to the midplane ( called @xmath6 here ) , in agreement with our picture
. for the combined rmp and elm field we start with the simplest assumption that the rmp does not have an influence on the elm and the total field is merely a linear superposition of the rmp and a natural elm : @xmath21 ( figure [ fig : footprintn2n8 ] , left ) .
the elm will then act as a carrier wave modulated by the rmp and its fine footprints will be concentrated in the areas of the footprints of the rmp ( figure [ fig : footprintn2n8 ] , right ) .
if we average over multiple elms or over time while one elm is rotating ( this can be represented by varying the phase @xmath22 ) , the average footprint pattern of heat deposition will be similar to the one formed by the rmp alone , in agreement with experimental observations .
this effect relies on the elm perturbation amplitude to be comparable or smaller than the rmp amplitude : @xmath23 , otherwise the rmp structure would be obscured by the larger elm structure . in practice
we may expect the rmp to interact with the elm , in particular to decrease its amplitude ( mitigation ) .
this does not significantly affect our reasoning and may help to satisfy the relation @xmath24 even in cases where it is not satisfied for unmitigated elms .
in the previous sections we developed an understanding of the form of mitigated elm divertor footprints based on a very simple representation of the perturbed magnetic field .
we shall now verify the model in a more realistic magnetic field and a full 3-d geometry .
such calculations were already performed in @xcite where the elm field was modelled using a source formed of a set of equidistant current filaments . here
we will use an even more realistic model : mhd simulations of the elm crash with the code jorek @xcite , including rmps .
the simulations were run for a jet - like equilibrium .
the jorek simulations use a fourier representation in the toroidal direction , harmonics @xmath25 were used in the present simulations .
the applied rmp is here produced by the error field correction coils ( efccs ) of jet , but only the dominant @xmath0 toroidal harmonic of their field is present ( as the boundary condition for the @xmath0 toroidal harmonic of the simulated field ) , the higher @xmath1 sidebands that occur in realistic coil designs are omitted for simplicity .
more details on the simulations can be found in @xcite .
the evolutions of magnetic energies of the toroidal modes are shown in figure [ fig : modes ] , together with the time instants where the following figures are taken . figure [ fig : n8elmonly ] shows laminar plots on the divertor calculated for three time instants during the elm crash in a jorek simulation .
the dominant toroidal mode number is @xmath18 here , corresponding to the 8 visible footprint lobes .
we may see that the whole pattern is rotating .
the next simulations were run with a @xmath0 rmp similar to the one produced by jet efccs .
figure [ fig : elmplusrmp ] shows the resulting laminar plot without elms ( produced by running the simulation with @xmath26 and @xmath0 harmonics only ) and with an elm crash ( using the harmonics @xmath27 with a step of 2 ) .
the primary strike point is distorted into an undulating pattern due to the perturbation being non - negligible in the x - point area .
the footprint pattern with an elm shows a similar @xmath0 structure as without an elm on a large scale , but its boundary has a more complex shape with multiple spirals on a smaller scale .
the fine scale structures evolve over time while the large scale @xmath0 footprints remain mostly the same .
it should be noted that while without the rmp the dominant toroidal mode number of the elms is @xmath18 , with the rmp the energy `` cascades '' to lower modes @xmath28 due to their nonlinear coupling @xcite as detailed in @xcite .
we are also interested in cases where the toroidal mode number is unchanged by the rmp , corresponding to the experimental observations on mast .
such a case is provided by a jorek simulation with an increased diamagnetic rotation , where the @xmath5 mode is the most unstable and is not mitigated by rmp corresponding to a current of 40 kat in the efccs @xcite .
the results for three time instants are shown in figure [ fig : elmplusrmpn6 ] .
the laminar plot has a @xmath0 pattern , split into a structure of finer footprints which evolve over time while conserving the basic @xmath0 pattern .
we may observe that without the magnetic perturbation the footprints rotate in the direction of the increasing toroidal angle @xmath14 .
this is the clockwise direction when viewed from top and is also the direction of the electron diamagnetic and @xmath29 drifts , in which the ballooning instability rotates @xcite . with the magnetic perturbation ,
the fine details of the boundary fluctuate , which may be due to the rotation of the higher @xmath1 modes due to the elms .
the superposition with the dominant @xmath0 pattern from the rmp does not allow to distinguish the rotation clearly , though .
for this reason we performed the field line tracing again in a field constructed by subtracting the @xmath0 mode from the total field , while the higher modes are being kept .
the @xmath0 mode is still present in the mhd simulation and thus modifies the other modes via nonlinear coupling , it is removed only in the field line tracing step .
the footprints calculated in this way are somewhat artificial they do not correspond to power loads on the divertor in any way but allow us to determine the rotation of the higher modes ( corresponding to elms ) .
the laminar plots corresponding to figure [ fig : elmplusrmp ] ( with one plot added at an intermediate time instant ) are shown in figure [ fig : elmplusrmpskiprmp ] .
the footprint pattern still rotates in the direction of increasing @xmath14 , albeit more slowly than in the unmitigated phase .
due to the presence of multiple modes , the pattern does not only rotate , but also progressively changes its shape as the relative amplitudes of the modes vary .
we may also note that the periodic distortion of the primary strike point due to the rmp has vanished .
for comparison , figure [ fig : elmplusrmpskiprmp ] shows the poloidal profiles of density at the same time instants ( in this case with all the harmonics including @xmath0 ) .
the pattern rotates in the electron diamagnetic direction ( counterclockwise ) . for the case when the rmp does not mitigate the elms ( figure [ fig : elmplusrmpn6 ] ) the corresponding laminar plot without the @xmath0 mode
is shown in figure [ fig : elmplusrmpn6skiprmp ] .
again , the footprint pattern of the elm rotates and changes its shape .
we have shown that a simple linear superposition of the field of an elm and an applied rmp field leads to concentration of open field lines in the footprints of the rmp field .
( remarkably , our model does not rely on any nonlinear interaction between the elm and the rmp , although such an interaction is known to occur in the mhd simulations @xcite . ) assuming that at least a part of the elm loss is caused by parallel transport along field lines , this would lead to an apparent `` locking '' of the divertor heat flux pattern to the rmp even though the elm has a different toroidal mode number and its rotation may not actually lock .
this was indeed observed on mast , where though the elm heat flux follows the rmp footprints only when averaged over the start of the elm pulse .
this result can be reconciled with our model by using the fact that a part of the elm energy is carried away by filaments and only the rest may be caused by the transport in the homoclinic tangle assumed here .
furthermore , the filaments arrive at the target with a delay @xcite . if the structure of the filament losses is unaffected by the rmp , they will in a later stage of the elm crash blur the pattern caused by the stochastic transport in the beginning , explaining the observed effect .
we shall note that only the structure of the stochastic losses was studied here and further research is needed to determine if the rmp has any impact on the filament losses or not .
we also studied only the connection length in the jorek simulations , as a too high perpendicular diffusion needed for numerical stability reasons prevented us from extracting meaningful target power profiles .
solving this problem will allow us to tackle also the filament question .
while our theory predicts that the footprints will have generally the shape of the rmp footprints , the elm field causes an additional distortion of their edge , depending on the relative amplitude of the elm compared to the rmp .
if the footprints due to the elm field are longer than those due to the rmp , the rmp structure may even stop being distinguishable .
this may explain the nstx results where the rmp structure was seen only in smaller elms @xcite , assuming that larger elms have longer footprints .
the distortion of the footprints was seen especially in the @xmath5 jorek runs , where the elms are not mitigated .
this fine distortion is moving across the static rmp pattern in the jorek simulations .
the modification of the static rmp pattern was proposed in @xcite as the explanation of mast results where the elm load follows the footprints only in some elms .
if the phase of the elm is such that its melnikov integral has a constructive interference with the one of the rmp at the footprint tip , the footprint will become longer and thus better observable . on the other hand ,
a destructive interference at the tip will reduce it . in @xcite it was noted that this explanation requires that the elm phase locks ( i. e. , does not rotate ) .
this is a different regime than observed in the jorek results where the phase always rotates .
the rotation was observed to slow down though in our simulations , so if the change of phase is negligible during the infra - red camera exposure time ( 35 @xmath30 on mast ) , the result may be indistinguishable from complete locking of the phase . in this way
the experimental observations may be reconciled with the simulations .
our model relies only on the general assumption that an elm has a magnetic component which can form a homoclinic tangle , and does not depend on a particular origin of this field .
we have seen that jorek simulations predict the existence of such a field during the elm crash .
another suggestion was that such a field is generated by thermoelectric currents flowing through the homoclinic tangle @xcite .
we may say that this theory is not needed for a qualitative explanation of the experimental results , as jorek simulations do not include the thermoelectric currents and still show the effect .
it may be the case though that a quantitative analysis will show the need for the thermoelectric currents in order to explain the data .
such a claim was made in @xcite , where it was found that the vacuum rmp footprints do not match those observed during the mitigated elms on jet .
the model used in @xcite did not apparently include the field of the elm itself , so the possibility that the vacuum rmp field superposed with the elm field will be sufficient to match the data can not be completely ruled out yet .
further work shall concentrate on determining the changes of the elm wetted area due to the effect described here .
an increase of the wetted area due to rmps was reported @xcite on jet with the iter - like wall ( contrary to the carbon wall results ) and attributed to the observed footprint pattern , while on mast the wetted area was found to decrease @xcite .
the latter effect is probably not due to the footprint pattern , as it follows the same scaling as the unmitigated elms .
other possible mechanisms behind the footprint pattern in mitigated elms should be investigated .
for instance , if an elm has a toroidal mode number equal to the one of the rmp , it may directly lock to it , as suggested in @xcite , and produce a large pattern due to constructive interference .
the influence of rmps on the filament pattern shall be modeled , as filaments are responsible at least for a part of the elm loss .
this work was part - funded by the czech science foundation under grants ga14 - 35260s and gap205/11/2341 .
this work was granted access to the hpc resources of aix - marseille universit financed by the project equip@meso
( anr-10-eqpx-29 - 01 ) of the program investissements
davenir supervised by the agence nationale pour la recherche .
this work has been carried out within the framework of the eurofusion consortium and has received funding from the european union s horizon 2020 research and innovation programme under grant agreement number 633053 .
the views and opinions expressed herein do not necessarily reflect those of the european commission .
the views and opinions expressed herein do not necessarily reflect those of the iter organization . | resonant magnetic perturbations ( rmps ) can mitigate the edge - localized modes ( elms ) , i.e. cause a change of the elm character towards smaller energy loss and higher frequency . during mitigation
a change of the spatial structure of elm loads on divertor was observed on diii - d and mast : the power is deposited predominantly in the footprint structures formed by the magnetic perturbation . in the present contribution
we develop a theory explaining this effect , based on the idea that part of the elm loss is caused by parallel transport in the homoclinic tangle formed by the magnetic perturbation of the elm .
the modified tangle resulting from the combination of the elm perturbation and the applied rmp has the expected property of bringing open field lines in the same areas as the tangle from the rmp alone .
we show that this explanation is consistent with features of the mitigated elms on mast .
we in addition validated our theory by an analysis of simulations of mitigated elms using the code jorek .
we produced detailed laminar plots of field lines on the divertor in the jorek runs with an elm , an applied rmp , and an elm mitigated by the presence of the rmp .
the results for an elm clearly show a high - n rotating footprint structure appearing during the nonlinear stage of the elm , which is not present in the early stage of the elm .
the results for a @xmath0 rmp from the elm control coils show the expected @xmath0 footprint structure .
the results for the mitigated elm show a similar structure , modulated by a higher @xmath1 perturbation of the elm , consistent with our theory . |
the range of surface density profiles for galaxies in a cluster is a telltale of its dynamical history .
for instance , an excess of cuspy profiles over , say , the mean field galaxy distribution , is indicative of recent merger activity .
conversely , extended exponential profiles are representative of quiescent evolution over long timescales ( e.g. , toth & ostriker 1992 ) .
the study of galaxy light profiles has a rich history with seminal early contributions from de vaucouleurs ( 1948 ; 1959 ) , srsic ( 1968 ) , and freeman ( 1970 ) .
de vaucouleurs ( 1948 ) established that the surface brightness ( hereafter sb ) profiles of early type stellar systems may follow a strongly concentrated distribution now referred to as the `` de vaucouleurs '' or `` @xmath7 '' profile .
de vaucouleurs ( 1959 ) later reported the ubiquity of exponential sb profiles in disk galaxies while srsic ( 1968 ) proposed a generalized fitting function that encompasses the exponential and de vaucouleurs profiles .
based on photographic images for 36 disk and s0 galaxies , freeman ( 1970 ) postulated that the distribution of disk central surface brightness ( hereafter csb ) , defined as the intercept of an exponential disk fit at @xmath8 , peaks at @xmath9=21.65 b mag arcsec@xmath6 .
deep , wide - field ccd , galaxy surveys have since revealed a rather continuous distributions of csbs from high surface brightness ( hsb ) to low surface brightness ( lsb ) galaxies ( de jong & lacey 2000 ) . simulations and observations of galaxy structure have thus far suggested a continuous range of properties over , rather than fundamental departures from , the hsb and lsb regimes .
however , based on near - infrared ( nir ) observations of the ursa major ( uma ) cluster , tully & verheijen ( 1997 ; hereafter tv97 ) inferred that the distribution of disk central surface brightnesses could be bimodal . based on a sample of 62 ursa major cluster galaxies observed at optical and nir bands , tv97 argued that the distribution of disk central surface brightnesses , @xmath9 , was not continuous as expected from simple structure formation models , but rather bimodal .
the bimodality of the @xmath9 distribution was not convincingly observed at optical ( bri ) bands , which tv97 argued might be due to extinction effects at optical bands . at k@xmath10 filter
was described by wainscoat & cowie ( 1992 ) .
it resembles the 2mass ks , with the primary difference being that it excludes the longest wavelength part of the k atmospheric window in order to reduce thermal emission . ]
, tv97 could assert the existence of two peaks in the sb distribution at 17.28 k@xmath10 @xmath11 for hsb galaxies ( corresponding to `` freeman s law '' ) and at 19.69 k@xmath10 @xmath11 in the lsb regime .
the surface brightness of the gap was centered at @xmath518.5 k@xmath10 @xmath11 .
inclination and extinction corrections to the optical surface brightness profiles would bring a csb bimodality to light in these bands as well .
tv97 stressed , however , that the sb bimodality at optical bands could plausibly be an artifact of their extinction correction which applies only to hsb galaxies .
still , the fact that the k@xmath10-band is nearly insensitive to dust and that the distribution of @xmath12 values is clearly bimodal is a strong case for tv97 s argument of structural bimodality in spiral galaxies .
tv97 divided their sample into galaxies with and without a significant near - neighbour , to identify an environmental effect .
the small uma membership however means that the isolated and crowded sub - samples contain only roughly 30 galaxies each . despite the statistical limitations , tv97 reported evidence for an enhancement of the bimodality in the isolated sub - sample
the majority of galaxies with intermediate @xmath9 measurements were found to have near neighbours .
this led tv97 to suggest the existence of two stable dynamical configurations of baryonic and dark matter , likely induced by the environment , leading to the distinct lsb and hsb populations . comparing light distributions with total mass distributions inferred from synthesis maps allowed
tv97 to further classify these two states . in lsb galaxies ,
dark matter dominates the potential at all radii ; gas and stars have sufficient angular momentum that they have not settled to the core and are orbiting in response to the dark matter distribution . in hsb galaxies ,
baryonic matter has dissipated its energy and transferred angular momentum sufficiently to become self - gravitating over the central @xmath13 exponential disk scale lengths ( courteau & rix 1999 ; dutton et al .
2007 , hereafter d07 ) .
a gap between lsb and hsb galaxies may suggest that galaxies avoid a situation where baryonic and dark matter have comparable dynamical influence in the inner disk .
the higher incidence of intermediate surface brightness ( hereafter isb ) galaxies with near neighbors suggests that tidal influences might drive a galaxy from the lsb to the hsb state .
although tv97 s study provided several valid arguments for a structural bimodality of cluster galaxies , it still suffered a few shortcomings .
first , the uma cluster membership is small . with only 62 galaxies ,
a structural study is plagued by statistical uncertainty ( bell & de blok 2000 ) .
tv97 estimated the disk central surface brightness , @xmath9 , by fitting an exponential profile to the outer disk in order to avoid the bulge component .
the uncertainty in this `` marking - the - disk '' technique depends on the subjective interpretation of the bulge size and the disk fit baseline .
this caveat was however examined carefully by mcdonald et al .
( 2008 ) and found not to be the cause of the observed bimodality .
finally , tv97 s result could be unique to the uma cluster and not representative of the true nature of disk galaxies in clusters or the field ( de jong & lacey 2000 ) .
because the uma cluster lacks early type galaxies , the observed bimodality could simply be due to a missing morphological class of galaxies .
it is , however , our impression that none of these objections are stronger or more convincing than the claim of bimodality itself , as we shall verify below for virgo cluster galaxies . following the release of the sloan digital sky survey ( york et al .
2000 ; hereafter sdss ) , there have been many reports of bimodality in the distributions of galaxy color , star formation and clustering properties ( strateva et al . 2001 ; blanton et al .
2003 ; kauffmann et al . 2003
; baldry et al . 2004 ; brinchmann et al . 2004 ; balogh et al . 2004 ) and even reports of trimodality in galaxy concentrations ( bailin & harris 2008 ) .
colors , star formation and clustering of galaxies are all intimately linked with a transition baryonic ( light - weighted ) mass of 3@xmath1410@xmath15m@xmath16 ( kauffmann et al .
red galaxies , at the upper end of the mass scale typically have low current star formation rates and are found primarily in cluster or high - density environments , while the less massive blue galaxies have high star formation rates and are found primarily in the field ( dressler 1980 ) .
these results have catalyzed several new variants of galaxy formation models and evolution .
for instance , an appealing explanation for any star formation bimodality involves two phases of gas accretion in galaxies via hot and cold modes ( birnboim & dekel 2003 ; dekel & birnboim 2006 ) . the hot mode is the standard picture of gas that is shock heated to the virial temperature during gravitational collapse of the gas and halo .
this shock - heated gas eventually cools and settles into a galaxy ( rees & ostriker 1977 ; white & rees 1978 ; white & frenk 1991 ) .
the cold mode consists of cold filaments penetrating far inside the halo ( birnboim & dekel 2003 ) .
these can occur in low - mass halos where virial shocks are absent and cold gas is accreted quasi - spherically . for halos with larger masses ,
the gas is shock heated to the virial temperature everywhere except where cold , dense gas filaments penetrate the virial radius . as the mass increases , only filaments of increasingly dense cold gas can penetrate the halo without being shock - heated .
kere et al .
( 2005 ) used semi - analytic models to predict that the transition _ halo _ mass where hot accretion becomes more efficient than cold accretion is @xmath5@xmath17m@xmath16 .
this concords with the transition _ baryonic _ mass , @xmath18 m@xmath16 , in the color / sfr bimodality , assuming a universal baryonic to dark matter ratio of @xmath510% ( zavala et al .
kere et al .
( 2005 ) also found that low - mass halos in dense environments have enhanced hot accretion .
the cold mode would thus apply mostly to late - type , disk galaxies while the hot mode pertains mostly to early - type , spheroidal , galaxies .
this simplistic picture appears to match observations well , but realistic star formation prescriptions are needed for more robust data - model comparisons .
we now wish to test tv97 s claim of structural bimodality in a different environment and using a larger sample .
thanks to its proximity , large size , and broad morphological coverage , the virgo cluster is the next logical cosmic structure to consider as we revisit tv97 s analysis . if confirmed , we can also assess whether the reported bimodality of surface brightnesses , for spiral galaxies at least , is in any way related to the sdss color bimodality discussed above and whether the global trimodality of galaxy concentrations for all hubble types ( bailin & harris 2008 ) is also observed .
the organisation of this paper is as follows : in , we discuss briefly our database for the 742 virgo cluster catalog ( binggeli et al . 1985 ;
hereafter vcc ) that are found in the sdss 6th data release ( adelman - mccarthy et al . 2008 ; hereafter dr6 ) . out of these
, we have defined a complete , magnitude - limited , sample of 286 vcc galaxies with @xmath19 for which deep h - band photometry was obtained .
the full description of our data set , and details about the data acquisition and reduction methods are given in mcdonald et al .
( 2009 ; hereafter `` data paper '' ) .
we introduce in specific parametric and non - parametric quantities that will permit a comprehensive understanding of the structure of virgo cluster galaxies of all types . in appendix a , we determine the effects of various errors on the distributions of various non - parametric quantities in .
we present final results in , and examine their possible interpretations in .
we assume in this paper a distance to all virgo cluster galaxies of 16.5 mpc or m - m=31.18 ( mei et al .
2007 ) . at that distance , @xmath20 = 80 pc .
grey points are galaxies in our vcc / sdss sample , while black points are the remaining galaxies in the vcc .
the size of the points scales with the total luminosity .
the concentric dashed circles are projected galactocentric distances ( 1 , 2 and 3 mpc ) around the center of the virgo cluster at m87 .
the dashed red circle corresponds to a distance of @xmath21 from m87.,scaledwidth=48.0% ] the study of the unbiased distribution of galaxy surface brightness requires volume completeness .
a solution to this issue is to study galaxies at a common distance in a cluster .
our sample is drawn from the virgo cluster catalogue ( binggeli et al . 1985 ;
hereafter vcc ) .
the vcc catalogue contains 2096 galaxies within an area of @xmath5140 deg@xmath22 on the sky , centered on the galaxy m87 at @xmath23@xmath512@xmath2425@xmath25 and @xmath26@xmath513@xmath27 .
the vcc is asserted to be complete down to a limiting absolute magnitude of @xmath28 and to contain many objects as faint as @xmath29 ( binggeli et al .
1985 ) .
our first task was to acquire deep , optical , photometry for as many vcc galaxies as could be found in the sdss / dr6 .
a further spatial cut was made following trentham & tully ( 2002 ) to reject 40% of the cluster that is contaminated by the w , w and m background groups .
the w , w and m groups were identified by de vaucouleurs ( 1961 ) and ftaclas et al . ( 1984 ) , respectively . w and m lie about twice as far away as virgo and w lies about 50% further than virgo . these and related structures in the virgo southern extension all lie in a flattened plane close to the supergalactic equator , contaminating the western edge of the virgo cluster .
trentham & tully ( 2002 ) discuss this problem and describe how the projected area of the cluster can be separated between zones with negligible and serious contamination .
images from the sdss / dr6 were extracted for a total of 742 vcc galaxies obeying these spatial cuts .
we will refer to this sub - sample as the `` vcc / sdss '' sample . in order to ensure that our virgo catalog reaches below the lsb peak of uma ( tv97 )
, we must achieve completeness at least down to @xmath30 mag ( assuming a distance of 15.5 mpc to uma ) . in order to ensure that no intermediate surface brightness galaxies are missed
, we can define a complete sub - sample that includes all vcc galaxies with @xmath31 mag ( i.e. @xmath19 ) .
we also make an additional two spatial cuts to ensure that we only include bound cluster members : 1 ) all galaxies identified as `` background '' by binggeli et al .
( 1985 ) are removed and 2 ) all galaxies further than @xmath21 from the center of the cluster , defined as the position of m87 , are removed .
these magnitude and spatial cuts leave us with a total of 303 vcc galaxies centered on m87 . a further 8 vcc galaxies ( 1068,1217,1258,1355,1665,1768,1889 and 2096 ) with recessional velocity measurements , @xmath32 3000 kms^-1km@xmath33s@xmath34 , and the galaxies vcc0723 and vcc0991 , with significant foreground stars , were also excluded . the following seven vcc galaxies ( 530,950,1052,1287,1571,1822 and 1992 )
could also not be detected at h - band ( see ) , and were left out of the sample ( this has no effect on our final conclusions ) .
we are left with a final , complete , sub - sample of 286 vcc galaxies with @xmath19 , covering a wide range of luminosities and morphologies . for reasons that will soon be clear
, we will refer to this complete sub - sample as our `` h - band '' sample .
shows the distribution of all vcc galaxies in black , as well as the 742 vcc / sdss galaxies in grey , and the h - band sample in multi - colors . shows the distribution of galaxy morphologies , as taken from the nasa extragalactic database ( hereafter ned ) .
the broad morphological coverage is important to ensure that no distribution of galaxy structural parameter is biased by morphological segregation .
shows that virgo is intrinsically rich in de , s0 and i m galaxies ( see also mei et al .
this is not due to a bias in our sample but simply to the nature of the virgo cluster ( as represented by the vcc ) .
the gas - rich vcc galaxies are represented at approximately the same levels in our vcc / sdss and h - band samples relative to the full vcc with completeness between 11% and 42% .
the bright magnitude limit of the h - band sample also implies a higher number of early - type galaxies relative to the later types .
however , we will show in that the bimodality of surface brightnesses is detected in each morphological bin ( early or late type ) and that it is thus unlikely due to the morphological make - up of this specific cluster or our sampling of it .
following tv97 s study of the uma cluster , we have sought to obtain optical and especially near - ir photometry as well as dynamical measurements for all of our h - band sample galaxies .
the sdss @xmath35 imaging for this sample will yield a distribution of luminosities , surface brightnesses , scale lengths and concentrations , as a function of optical wavelength .
the optical colors will enable a comparison of any surface brightness bimodality , if present , with the observed sdss galaxy color bimodality ( e.g. , strateva et al .
2001 ) and the ability to determine if one is simply a consequence of the other .
the near - ir images are however essential to uncover the true distribution of galaxy surface brightnesses , luminosities , and other galaxy parameters largely free of extinction by dust .
we describe the extraction of the optical and nir imaging data below . while most critical , our collection of dynamical parameters for vcc galaxies is still in progress and will be reported elsewhere .
we have extracted calibrated @xmath35 images from the sdss / dr6 for 742 vcc galaxies , including the 286 galaxies in our `` h - band '' sample .
surface brightness profiles and total luminosities were obtained for these galaxies in all five sdss bands by , first , performing isophotal ellipse fitting to the @xmath3-band images according to the methods of courteau ( 1996 ) , and then , applying the @xmath3-band isophotal solutions to the images in the other sdss bands .
the latter ensures that color gradients extracted from all sdss images are computed from the same matching isophotes ( macarthur et al .
@xmath36-band images were consistently shallower than the @xmath37 bands and were thus discarded .
sky levels for background subtraction and the photometric zero - points for calibration were obtained from the sdss image headers and the sdss archives , respectively .
the remainder of the profile extraction technique is identical to that used for the near - ir photometry , as described below .
due to practical constraints , new , deep , h - band imaging that would sample well below the putative surface brightness bimodality scale of tv97 could only be obtained for a smaller sample of vcc galaxies .
this is the magnitude - limited `` h - band '' sample of 286 vcc galaxies with sdss imaging described above .
deep h - band imaging for some vcc galaxies is already available from the two micron all - sky survey ( skrutskie et al .
2006 ; hereafter 2mass[multiblock footnote omitted ] ) and from the goldmine[multiblock footnote omitted ] database ( gavazzi et al . 2003 ) .
we were able to secure h - band imaging for the remainder of the h - band sample with the detectors ulbcam at the uh 2.2-m telescope , wfcam at ukirt and wircam at cfht over the period 2005 - 2008 .
some deep k - band imaging was also available for a few galaxies in 2mass and goldmine .
h - band images from goldmine were kindly provided by g. gavazzi .
calibrated 2mass galaxy images were extracted from the online database .
many of the 2mass and goldmine images were not deep enough for our purposes . whilst adequate for large hsb galaxies , the high 2mass brightness threshold ( typically @xmath38 mag arcsec@xmath6 .
bell et al .
2003 ; courteau et al .
2007 ; kirby et al .
2008 ) limits the use of those data bases for deep extragalactic studies .
likewise , just a handful of goldmine images were deep enough to properly separate the bulge and disk light .
we have defined the relative depth criterion , @xmath39 , as the ratio of the maximum extent of the h - band surface brightness profile divided by that of the sdss @xmath3-band profile : @xmath40 , where @xmath41 is the radius where the surface brightness error exceeds 0.15 @xmath11 ( see mcdonald et al .
2008 for details ) .
wherever possible , we impose @xmath42 for our nir data . ultimately , 20 2mass and 79 goldmine galaxy profiles were deemed useable for our study .
a final 187 virgo cluster galaxies needed new observations .
these new and existing observations are summarized in table [ obstable ] .
cccc tel - camera & targets & collected & avg seeing + uh88@xmath43 - ulbcam & 52 & 04/2005 & 1.2 @xmath44 0.2 + uh88@xmath43 - ulbcam & 16 & 04/2006 & 1.5 @xmath44 0.3 + uh88@xmath43 - ulbcam & 31 & 04/2007 & 1.3 @xmath44 0.2 + uh88@xmath43 - ulbcam & 23 & 03/2008 & 1.0 @xmath44 0.2 + + ukirt - wfcam & 31 & 07/2008 & 1.1 @xmath44 0.1 + cfht - wircam & 34 & 02 - 06/2008 & 1.1 @xmath44 0.1 + goldmine & 79 & - & 2.2 @xmath44 0.9 + 2mass & 20 & - & 2.6 @xmath44 0.1 + [ obstable ] the surface brightness profiles for the deep goldmine and 2mass images were measured using the same techniques as our new nir images to ensure uniformity for the entire database .
further details regarding the data reduction process and quality are presented in mcdonald et al .
( 2008 ; 2009 ) .
surface brightness profiles were extracted for the 742 vcc / sdss galaxies at @xmath37 bands and for the 286 h - band galaxies . in this section ,
we examine the various parametric and non - parametric properties derives from these profiles .
mcdonald et al .
( 2008 ) already considered the parametric ( i.e. model dependent ) analysis of uma spiral galaxies in order to test , and ultimately confirm , tv97 s claim of bimodality .
the current , larger , virgo cluster sample however includes galaxies of all morphologies .
while a parametric approach to model the shape of complex light profiles involves a multi - component decomposition ( e.g. , macarthur et al .
2003 ; mcdonald et al .
2008 ) , a non - parametric approach is free of model assumptions and reveals different aspects of galaxy structure .
we explore both approaches below and it will be shown later that surface brightness bimodality is essentially independent of the method of light profile analysis .
spheroidal and flattened galaxy systems have traditionally been modeled as the sum of a bulge and disk components ( see macarthur et al .
2003 ; mcdonald et al .
2008 for more details ) . the 1d light profile of a galaxy disk is typically parametrized as an exponential function : i_d(r)=i_0\{-r } , where @xmath45 and @xmath46 are the disk central surface brightness and scale length , respectively .
meanwhile , the projected 1d bulge light profile is better modeled as a srsic function ( srsic 1968 ) : i_b(r)=i_e\{-b_n } , [ eq : sersic ] where @xmath47 is the half - light radius , @xmath48 is the surface brightness at that radius , and @xmath49 is the srsic shape parameter . with @xmath50 ,
the srsic function reduces to the exponential function .
in addition to the bulge and disk , we consider other components that may affect the light profile such as compact nuclei , spiral arms and disk truncations . ignoring these may result in large errors in the bulge and disk parameters ( mcdonald et al .
2008 ) .
b / d decompositions were performed only for the h - band and optical @xmath37 light profiles of the 286 galaxies in the `` h - band '' sample ( mcdonald et al .
2009 ) .
non - parametric quantities are measured directly from the surface brightness profile , with no prejudice for any assumed model .
the only assumptions inherent to the non - parametric measurements below are : ( i ) that the total galaxy light is an extrapolation of the light profile to infinity , and ( ii ) that inclination estimates and extinction corrections are valid .
non - parametric quantities allow a direct , unbiased comparison of galaxies across the full hubble sequence .
the galaxy light concentration is a measure of the relative light fraction between the inner and outer parts of the galaxy . unlike the b / d ratio , which relies on a model for the light distribution
, @xmath51 is a straightforward , model - independent morphological indicator .
the concentration , @xmath51 is defined as ( e.g. , kent 1985 ; courteau 1996 ) : c_28 5(r_80 ) [ eq : c28 ] where @xmath52 and @xmath53 are the radii enclosing 80 and 20 percent of the total light . for a pure exponential function ( @xmath50 in ) , @xmath51=2.8 .
concentration indices are a function of wavelength and while fractional radii depend on inclination and extinction corrections , we expect the ratio @xmath52/@xmath53 to be roughly independent of projection effects .
we justify this assumption below .
fractional parameters refer to specific quantities measured at radii that contain specific fractions of the total galaxy light .
the effective radius , @xmath47 , is the radius that encloses half of the total light .
while @xmath47 was already introduced in the context of a parametric profile function ( ) , it is formally defined in a non - parametric way as : ^_0 i(r)2rdr 2^r_e_0 i(r)2rdr .
[ eq : reff ] the effective surface brightness , @xmath48 , is the surface brightness at @xmath47 ; @xmath54 ( @xmath55log@xmath48 ) . unlike the extrapolated central surface brightness of the disk , @xmath45 ,
the effective surface brightness @xmath48 is non - parametric , making it ideal for the comparison of mean surface brightness levels for galaxies of varying morphology ; @xmath48 ( or @xmath56 ) applies to all galaxy types rather than @xmath45 ( or @xmath9 ) which is restricted to spiral disks ( @xmath57 for pure disks ) .
our discussion about the variations in galaxy surface brightness profiles will indeed rely on @xmath47 and @xmath56 .
another fiducial light marker is the fractional radius defined as : ^r_x_0 i(r)2rdr ^_0 i(r)2rdr where @xmath58 contains some percentage ( @xmath59 ) of the total light .
@xmath60 is the surface brightness in @xmath11 at @xmath58 . a final , non - parametric , measure of surface brightness is the average surface brightness interior to some radius @xmath58 , @xmath61 , defined as : i _ x .
@xmath62 is the magnitude equivalent of @xmath63 .
shows the dependence of the h - band @xmath64 and @xmath51 on the @xmath3-band axial ratio , @xmath65 for the h - band sample .
there appears to be a correlation between @xmath64 and @xmath65 , but it is driven largely by the spheroidal nature ( high axial ratios ) of compact galaxies ( small @xmath47 ) .
any correlation between @xmath47 and @xmath65 is significantly weakened if compact galaxies are excluded .
the insensitivity of @xmath51 to projection effects ( top right window ) is even clearer .
the respective corrections for projection on @xmath53 and @xmath52 roughly cancel out .
we also see in that there is a good linear correlation between @xmath3- and h - band scale radii and concentrations . , and concentration , @xmath51 for the virgo h - band sample as a function of @xmath3-band axial ratio ( top ) , and against similar @xmath3-band measurements in the middle section .
the point / line types represent : de - ds ( red , open / dashed ) , e - s0 ( red , closed / solid ) , sa - sb ( green , closed / solid ) , sc - sd ( blue , closed / solid ) and irr ( blue , open / dashed ) . the black , dotted , line in the middle windows is the slope unity correlation .
the bottom windows are population histograms.,scaledwidth=48.0% ] ultimately , concentrations , fractional radii and surface brightnesses were computed for all 742 vcc / sdss at @xmath66 wavelengths and for all 286 `` h - band '' galaxies at h - band .
we now examine the distribution of structural parameters of virgo cluster galaxies , starting with an analysis of the gas - rich systems in the h - band sample for comparison with tv97 s similar analysis .
we then derive the distribution of effective surface brightnesses and the optical and nir luminosity functions for the full h - band and vcc / sdss samples , with all morphologies considered .
we will find for the virgo gas - rich galaxies in the h - band sample , a bimodal distribution of the extrapolated disk central surface brightness , @xmath67 , and thus bolster the similar claim for uma galaxies by tv97 .
for the complete sample of gas - rich and gas - poor virgo galaxies , we determine the distribution of effective surface brightness , @xmath56 ( independent of b / d decompositions ) and find compelling evidence for brightness bimodality in each morphological group .
we present in the distributions of @xmath9 and , in , various fractional and average surface brightness measures .
we compute in the optical and nir luminosity functions and compare these to the recently published optical luminosity function of virgo ( rines and geller , 2008 ) and field nir luminosity function from ukidss ( smith et al .
2008 ) . in
, we present the distributions of various scale radii and the galaxy concentration , @xmath51 .
the latter is compared to the distribution of concentrations for sdss galaxies by bailin and harris ( 2008 ) . in
we examine the bivariate distributions for most of the structural parameters addressed in this section . in order to compare with the study of uma cluster galaxies by tv97 and mcdonald et al .
2008 , we restrict our h - band sample to disk galaxies .
b / d decompositions , as in mcdonald et al . ( 2008 ) , were performed for this sub - sample of 166 vcc disk galaxies .
the b / d fits , which include optional bulge , nucleus and spiral arm components , were applied to each @xmath37 and h - band sb profiles .
the derived @xmath9 values were corrected for projection effects to their face - on value in the absence of extinction .
thus , we write : @xmath68 , where @xmath65 is the measured axial ratio of the outermost isophote at a given band in each galaxy . we take @xmath69=1 for the dust transparent case .
this is likely a fair assumption at h - band , the focus of our analysis , but less adequate for the @xmath37 bands .
we keep our analysis free of dust correction for now . for 166 vcc spiral and irregular galaxies ( hatched ) .
the red line shows the results from the analyses of uma galaxies by tv97 and ourselves ( mcdonald et al .
2008 ) , while the green line shows the two - gaussian fit to the vcc data .
the f - test value demonstrates that a single - gaussian fit can be rejected with a confidence of 85% .
the transition between the two brightness peaks is at @xmath70.,scaledwidth=48.0% ] , with morphological type for 166 vcc galaxies . filled and open black squares represent hsb and lsb galaxies , respectively.,scaledwidth=48.0% ] the distribution of inferred @xmath71 values
is shown in .
there is a clear dearth of galaxies at @xmath7219 mag arcsec@xmath6 .
this result for later - type vcc galaxies matches very well that determined by tv97 and ourselves for uma disk galaxies with a minimum in the number of galaxies at @xmath73 mag arcsec@xmath6 .
similarly , we find excellent agreement with tv97 in the location of the hsb and lsb peaks at @[email protected] mag arcsec@xmath6 and @[email protected] mag arcsec@xmath6 , respectively , with a peak separation of @xmath75=2.4 mag arcsec@xmath6 . considering a typical h - [email protected] for galaxy disks , the distributions for @xmath74 in the virgo and uma clusters
are very similar . using a statistical f - test
, we can reject the hypothesis of a normal distribution for @xmath74 in favor of a bimodal distribution with a confidence level of 85% .
shows which galaxies contribute mostly to different surface brightness levels . as expected
, early - type disk galaxies dominate the hsb peak , while late - type disk galaxies and irregulars are present in both the hsb and lsb peaks . , from optical ( sdss ) to near - ir bands for 166 spiral and irregular virgo cluster galaxies from the h - band sample .
the red histogram in the @xmath3-band panel refers to the 272 disk galaxies in the vcc / sdss sample .
the numbers on the right are the statistical f - test probability for a 2-component gaussian .
, scaledwidth=48.0% ] let us now examine the distribution of inclination - corrected @xmath9 with wavelength
. shows the different @xmath71 distributions from @xmath1 ( top ) to k ( bottom ) .
the gap between the two sb peaks grows as a function of wavelength since the hsb peak brightens at a faster rate with wavelength than the lsb peak .
this is explained by the higher dust content in hsb galaxies , in contrast with their relatively transparent ( @xmath76 ) lsb counterparts ( tv97 ) , as well as the relative colors of the constituent stars ( hsb galaxies tend to be redder than lsb galaxies ) .
dust obscuration is less effective at longer wavelengths and , as a result , we see more deeply along any given line - of - sight thus increasing the observed surface brightness . recall that we have not applied any correction for obscuration ( c@xmath77=1 ) to surface brightnesses and the bimodality disappears as we consider shorter wavelengths .
it is possible that the correct choice of c@xmath78 would restore the bimodality at optical wavelengths , as suggested by tv97 , but any such complications can be avoided by restricting our analysis to nir wavelengths only , as we do here .
for 166 spiral and irregular vcc galaxies .
the red , hatched histogram includes only those galaxies with very little contribution from the bulge component ( bulge - to - total ratio , b / t , @xmath79 25% ) .
this shows that bimodality is not the result of bulge - dominated vs bulgeless systems.,scaledwidth=48.0% ] as a function of disk scale length , @xmath46 , for the sub - sample of 166 spiral and irregular vcc galaxies .
red dashed lines represent constant luminosity ; our detection limit is @xmath510@xmath80l@xmath81 at h - band . filled and empty squares correspond to hsb and lsb galaxies respectively , as in .,scaledwidth=48.0% ] , and non - parametric brightnesses @xmath56 and @xmath82 for the 166 gas - rich vcc galaxies .
the point types are as follows : open red circle - s0 , green circles - sa & sb , blue circles - sc & sd , open blue circles - irregulars .
the dashed lines are @xmath83 and @xmath84 , representing a pure exponential disk.,scaledwidth=48.0% ] we can ask if the bimodality in the disk csb is correlated with any bulge structure . fig .
[ mu0_bulge ] shows the same distribution as in fig .
[ virgo_csb ] but now with a second @xmath74 distribution including only galaxies with small bulges .
we define a `` small '' bulge as one which contributes @xmath85% of the total galaxy light .
indeed , we see that if we remove those galaxies with bulges which contribute more than 25% of the total luminosity , bimodality is preserved .
though the f - test confidence level drops , this is primarily due to the reduction in sample size - the isb gap is very well preserved
. shows the relationship between @xmath9 and disk scale length , @xmath46 , at h - band . at intermediate luminosity
, there can be both hsb and lsb galaxies , depending on the disk surface density ( faint / high @xmath9 & large @xmath46 or bright / low @xmath9 & short @xmath46 ) .
the fact that two galaxies with the same luminosity can have wildly different surface brightnesses leads to the belief that some mechanism likely related to the initial halo angular momentum ( dalcanton et al .
1997 ) prevents lsb systems from collapsing to the same densities as the hsbs .
this figure makes clear that for a given luminosity the lsb galaxy must be more extended than the hsb galaxy and that no lsb galaxy exceeds 10@xmath86l@xmath81 .
the distribution of disk scale lengths also shows some evidence of multiple peaks , though the statistical significance is low ( 68% ) , none of which are correlated with the structure in @xmath74 .
we have shown thus far that the nir central surface brightnesses , @xmath74 of disk galaxies in the virgo and uma clusters are distributed bimodally .
our b / d decompositions confirm a result that is already well - known - that galaxy bulges come in two types : cuspy ( high concentration ) and cored ( low concentration ) .
hsb disks harbour both types of bulges , while lsb disks only harbour low - concentration bulges .
we now expand our analysis to take full advantage of the diverse virgo cluster population by considering early - type and dwarf galaxies as well . and @xmath87 residuals as a function of concentration.,scaledwidth=48.0% ] unlike the uma cluster of galaxies ,
virgo is rich in giant and dwarf early - type galaxies .
the mean density of virgo is @xmath55 times that of uma .
many of the vcc galaxies have no observable disk component and therefore our analysis of galaxy properties based on @xmath9 would be moot .
we consider instead two more versatile quantities : @xmath56 , the galaxy effective surface brightness and @xmath82 , the mean surface brightness , defined as the average surface brightness within @xmath47 ( 4.2.2 ) .
these non - parametric quantities can be measured for all galaxies .
shows the comparison for the 166 gas - rich vcc galaxies of @xmath9 with @xmath56 and @xmath82 . there is a strong correlation between @xmath9 and @xmath56 with the expected zero - point offset ( @xmath88 ) for pure exponential disks ) .
the scatter increases for earlier - type galaxies .
this scatter is further examined in fig.[mu - resid ] , where we confirm that the differences between @xmath9,@xmath56 , and @xmath82 are a function of morphology or , for simplicity , concentration .
that is , for higher concentration the scatter in the @xmath89 and @xmath87 relations increases .
we have also considered the average surface brightness within @xmath47 , @xmath82 , as a measure of a galaxy s characteristic surface brightness .
use of this parameter is motivated by a study of late - type field galaxies by de jong & lacey ( 2000 ; hereafter dl00 ) .
dl00 studied the distribution of @xmath82 for 1000 sb - sdm field and cluster galaxies which , they reported , is not bimodal . the right side of shows that @xmath82 correlates well with @xmath9 , though with slightly larger scatter than @xmath56 . from this point on
, we will adopt @xmath56 as our standard measure of surface brightness for three reasons : 1 ) for spiral galaxies , @xmath56 scales directly with @xmath9 , 2 ) @xmath56 can be measured for any galaxy morphology , and 3 ) @xmath56 is independent of any assumptions about the shapes of the galaxy bulges ( if any ) and disks .
we also adopt a geometric correction for @xmath56 in all galaxies : @xmath90 , where @xmath91 is the axial ratio at the effective radius .
regardless of what surface brightness measure we use , and how we correct for projection effects , we will show in 6 that the overall shape of the distribution of surface brightnesses is preserved .
, in 5 bands for all 286 vcc galaxies in our sample .
the red histogram in the @xmath3-band window is the distribution of @xmath92 for all 742 vcc / sdss galaxies ; that distribution is strongly bimodal with an f - test of 95%
. troughs are significant in the brightness distributions from @xmath3 to @xmath93 bands.,scaledwidth=48.0% ] we show in the distribution of @xmath92 with wavelength for all 286 h - band vcc galaxies .
the differential effects of extinction with wavelength are well known ( though poorly understood ) and likely the cause for the smoother contributions at shorter wavelengths .
the @xmath56 distribution of the nearly dust - free lsb galaxies should not change drastically from @xmath1 to h , modulo a color term due to stellar populations .
the hsb peak from @xmath1 to h will , however , be altered by effective dust obscuration .
the faint tail of the hsb peak will thus be stretched to fainter values , effectively filling any intrinsic trough between the hsb and lsb peaks ( as shown by tv97 for uma galaxies ) .
for this reason , we shall now rely solely on nir surface brightness measurements for the remainder of our analysis .
, for 286 vcc galaxies of varying morphology .
the red vertical dotted lines correspond to the locations of the esb ( extreme surface brightness ) , hsb and lsb peaks .
the upper - right panel shows all gas - poor galaxies , the middle - right panel is the sum of all the gas - rich galaxies , and the bottom - right is the sum of all morphological .
the sum of the two _ bimodal _ distributions for the gas - poor and gas - rich galaxy types respectively , results in a _ trimodal _ distribution of all the vcc galaxies in our sample .
the numbers in the upper and middle right panels refer to the f - test confidence for bimodality.,scaledwidth=48.0% ] , for galaxies of varying morphology .
the numbers in the upper and middle right panels refer to the f - test confidence for bimodality.,scaledwidth=48.0% ] shows the distribution of @xmath94 measurements for our sample of 286 h - band vcc galaxies .
characteristic structure can be seen in most of the morphological bins , as well as in the sample as a whole .
the separate distributions of e and s0 galaxies are each weakly bimodal .
the combination of these morphological types ( top right panel of ) gives a stronger bimodality for @xmath94 .
the bimodality of @xmath94 for gas - rich galaxies emerges clearly from the hsb and lsb peaks for the early - type ( sa - sbc ) and irregular galaxies ; the distribution of sc - sd galaxies shows the least features and is least abundant of all vcc types .
this results in a strong bimodality for all the gas rich galaxies ( middle - right panel of ) . if we compare the distributions of @xmath94 for gas - poor and gas - rich galaxies , the lower surface brightness spheroids line up with the hsb peak for late - type and dwarf galaxies .
this results in an _ apparent _ trimodality of @xmath56 for the complete h - band sample ( lower right panel of ) . however , we stress that the trimodality is in fact a superposition of an early- and late - type sb bimodality .
the distribution of @xmath95 in , shows similar trends .
however , illustrates the sensitive nature of binning for this sort of argument .
for instance , the distribution of @xmath96 for sc - sd is here roughly bimodal when it was flat in .
this illustrates the need for large samples .
we are still reassured that the distributions for @xmath56 or @xmath97 for the sum of both the gas - rich and gas - poor galaxies show very similar ( bimodal ) features .
[ virgo_esb_bymorph]-[virgo_avesb_bymorph ] show that a structural bimodality is present in both early and late - type galaxies with a tail towards the brightest galaxies .
a statistical f - test confirms the rejection of a single peak in favor of two distinct peaks with 71% and 82% confidence for early- and late - type galaxies , respectively . we can consider another significance test . in order to fill the early and late - type brightness gaps , an additional 15 and 29 intermediate sb galaxies would be needed , respectively
. however , the current number of galaxies in these troughs is 24 and 39 , respectively , so the required additions to achieve flat sb distributions would correspond to 3 and 4.5 standard deviations .
this argument , based only on poisson statistics , complements nicely the f - test ( which assumes a gaussian shape )
. figs .
[ virgo_esb_bymorph]-[virgo_avesb_bymorph ] tell an exciting story ; not only is a structural bimodality present in spiral galaxies , but also in spheroids .
equally intriguing is that spheroids of lower surface brightness have identical surface brightnesses as hsb spirals , possibly pointing to a common formation and/or structural mechanisms . in order to better understand this result
, we now examine the correlation of surface brightness with various other parameters and what role it plays in fundamental structural relations . but
first , let us introduce another unexpected result , this time in the nir luminosity function , to shed more light on the structural bimodality in both gas - poor and gas - rich galaxies .
mcdonald et al .
( 2008 ) showed that the k - band luminosity function for ursa major has a noticeable dip at intermediate luminosities .
we now examine this claim for virgo galaxies with both optical and near - ir luminosities .
shows the luminosity functions of our 286 vcc galaxies in the @xmath1,@xmath2,@xmath3,@xmath4 and h bands .
there appears to be a dip in the luminosity function at optical wavelengths , which develops into a broad trough at h - band .
this trough clearly separates two distinct luminosity classes .
-band panel , the open triangles correspond to petrosian magnitudes taken directly from the sdss for comparison with rines & geller ( 2008 ) . in the @xmath3-band panel , the closed triangles correspond to magnitudes computed for all of the 742 vcc / sdss galaxies .
open circles refer to magnitude bins suffering from incompleteness at the low - luminosity end .
, scaledwidth=48.0% ] rines and geller ( 2008 ) obtained the same result for their study of the faint end of the virgo lf ( see black triangles in @xmath2-band window of ) , but dismissed it on grounds that sdss data may be flawed at luminosities where the dip is observed .
it would however be challenging to think of such a systematic bias in the sdss pipeline reductions .
furthermore , our luminosities are measured from sdss images but fully independent of any other sdss data product ( e.g. , petrosian luminosities and radii ) .
the dip in the virgo luminosity function is also observed at h - band which further confirms that it is not an artifact of the sdss pipeline . our luminosity function and that of rines & geller are a striking confirmation of each other , establishing the reality of the dip in the luminosity function of virgo cluster galaxies .
the dip in the nir luminosity is not reproduced in the ukidss luminosity function for field galaxies ( smith et al .
this could indeed mean that there is no trough in the luminosity function of field galaxies and that the lack of intermediate - luminosity galaxies in uma and virgo is purely a cluster - driven phenomenon .
however , the ukidss luminosity function suffers incompleteness at luminosities fainter than @xmath98 , while the low - luminosity peak in the uma luminosity function is at @xmath99 .
this fact , along with the difficulty of achieving volume completeness in field galaxy samples , could conspire to hide any putative trough in the nir luminosity function if it is , in fact , independent of environment .
we will see in 4.5 and fig .
[ lum - mue ] that the high - luminosity peak is comprised of `` extreme '' surface brightness ( hereafter esb ) galaxies and the low - luminosity peak is comprised of lsb galaxies .
however , the hsb galaxies are distributed bimodally in luminosity , implying a bimodality in sizes for these galaxies since they all have nearly the same surface brightness .
the combination of figs [ virgo_esb_bymorph ] and [ lumfunc ] shows that the majority of the low - luminosity hsb galaxies are compact de galaxies , while the majority of the high - luminosity hsb galaxies correspond to traditional hsb spirals . finally , fig . [ everything ]
shows the multi - dimensional variation of various structural parameters with optical and nir luminosity for the full sample of 286 vcc galaxies .
the dumbbell - shaped distribution of points in all plots of luminosity versus @xmath56 or @xmath100 shows clearly the separate galaxy populations .
the l-@xmath101 relation appears to be very similar whether we use @xmath56 or @xmath102 , although the scatter is somewhat reduced at @xmath56 .
finally , there is a strong evolution of most of these scaling relations with wavelength that will be studied elsewhere . in appendix
a , we have attempted to predict the observed distribution of @xmath51 based on normal distributions of @xmath103 , @xmath104 , @xmath105 , @xmath106 and @xmath49 . in fig .
[ c28_bymorph ] we show the distribution of measured concentrations for the 286 vcc galaxies binned by morphology as well as for the complete h - band sample . for comparison ,
we show the predicted distribution from our models ( appendix a ) . , for 286 vcc galaxies of different morphologies .
the red histogram in the sa - sd , irr window refers to the distribution of @xmath51 for uma galaxies from mcdonald et al .
( 2008 ) for galaxies of the same morphology .
the red histogram in the bottom - right window is based on 100,000 model surface brightness profiles , discussed in
5.4.,scaledwidth=48.0% ] the predicted and observed distributions are remarkably similar , though the observation of a second peak at high concentration with c@xmath107 was not predicted by our simulations of isolated galaxies . the overall distribution peaks at c@xmath108 , corresponding to pure exponential disks .
the distribution terminates rather sharply at c@xmath109 , suggesting a lower limit for how centrally concentrated the baryons in a galaxy can be . in order to understand the high-@xmath51 peak ,
let us look at the distribution of @xmath51 for different morphological types
. shows that the sc - sd , irr and de galaxies populate a single peak at @xmath110 , while e , s0 and sa - sbc galaxies occupy two distinct peaks near @xmath111 and @xmath112 .
thus , there is a concentration bimodality only in early - type ( bulge - dominated ) galaxies .
the difference between the model and observed distributions is due in large part to the fact that our model distribution of bulge surface brightness is normal , while the one we observe is bimodal .
the transition at @xmath113 represents the transition from lsb to hsb bulges .
we combine figs .
[ virgo_esb_bymorph ] and [ c28_bymorph ] in to examine whether the early - type bimodality in concentration corresponds to the bimodality in surface brightness . , and effective surface brightness , @xmath94 .
the dashed vertical line is an arbitrary transition at @xmath114 from low- to high - concentration , while the two dashed horizontal lines delineate the lsb , hsb and esb classes.,scaledwidth=48.0% ] we see again in that the two hsb peaks are comprised primarily of e , s0 and sa galaxies , while the lsb peak is comprised primarily of dwarf , irregular and late - type spiral galaxies .
most importantly , we see two isolated clouds : one containing high - concentration hsb and esb galaxies and the other containing low - concentration hsb and lsb galaxies . again , lsb galaxies never have high concentration and esb galaxies never have low concentration .
this provides some insight into the structural bimodality in spheroids .
it is possible that galaxies with c@xmath115 can only form through major mergers , which naturally explains the lack of both lsb and star - forming galaxies in the high concentration peak .
we can also verify that the trimodality observed in sdss concentrations by bailin & harris ( 2008 ) is not detected here . in order to assess
if the brightness bimodality extends to , or even stems from , other galaxy parameters , we now construct a number of bivariate distributions for @xmath56 vs other fundamental structural parameters such as color , luminosity and dynamical mass .
the color - surface brightness bivariate distribution is certainly of great interest in light of the color bimodality of sdss galaxies ( e.g. , strateva et al .
it is of obvious interest to ask if the transition in galaxy colors matches the transition in galaxy surface brightnesses .
a careful examination of fig [ mue_gi ] defuses this idea .
the fact that the blue and red peaks each include all three sb peaks demonstrates that the structural bimodality is indeed different from the color bimodality . the same distribution but for @xmath1-h colors ( fig .
[ mue_gh ] ) reveals a different color bimodality ( resulting from stellar population effects that will be investigated elsewhere ) .
the fact that the red peak alone is comprised of all three sb classes again confirms that surface brightness trends and color bimodality are independent phenomena .
we summarize our findings in fig .
[ mue_col ] with the distributions of @xmath116 , @xmath117h and @xmath94 for individual morphologies .
there is a strong correlation between color and surface brightness for elliptical galaxies ( and weaker for s0s ) , with redder ellipticals having higher surface brightness than blue ellipticals .
there appears to be no correlation between color and surface brightness for the other types , suggesting again that the surface brightness bimodality is independent of stellar populations .
color for 286 vcc galaxies .
point colors and types are the same as for fig .
[ c28_mue ] .
dotted lines represent individual gaussian fits while dashed lines are the sum of these fits.,scaledwidth=48.0% ] -h color for 286 vcc galaxies .
point colors and types are the same as for fig .
[ c28_mue ] .
dotted lines represent individual gaussian fits while dashed lines are the sum of these fits.,scaledwidth=48.0% ] for 102 vcc spiral galaxies .
the gaussian fits in the upper panel correspond to the full distribution of 286 vcc galaxies .
dotted lines represent individual gaussian fits while dashed lines are the sum of these fits.,scaledwidth=48.0% ] the correlation of @xmath94 with rotational velocity gives greater hope of establishing a link with the observed bimodality in @xmath56 .
indeed , we note in hints of bimodality for virgo galaxies in the distribution of the maximum galaxy rotational velocity , v@xmath118 . with high - quality , heterogeneous ( combination of linewidths and h@xmath23 rotation curves ) dynamical information available from hyperleda ( paturel et al .
2003 ) for only 102 of the 286 vcc galaxies , the observed weak correlation ( r=0.53 ) between v@xmath118 and @xmath94 may not be surprising . a homogeneous collection of resolved rotation curves for the complete sample of galaxies will be needed to establish a dynamical extension of the galaxy structural bimodality . in a similar vein ,
tv97 found for a subsample of 35 uma galaxies that the surface brightness correlated strongly with the slope of the rotation curve , a subject we address in appendix 2 . yet again , a much larger sample is needed for a conclusive analysis .
finally , we observe the distribution of luminosities vs @xmath94 in . there is a clear trend between luminosity and surface brightness , as expected . although there is scatter in both variables , any bimodality in @xmath119 , as observed in ought to be reflected in @xmath120 by virtue of the velocity - luminosity relation of galaxies .
furthermore , since @xmath121 , and @xmath122 we would expect the @xmath56 bimodality to be reflected in the distribution of luminosities , unless the sb bimodality is due to the relation @xmath123 .
shows a mild bimodality in l , but which is uncorrelated with the stronger bimodality in @xmath56 .
we have analyzed bivariate distributions involving color , rotation velocity , luminosity and surface brightness . from figs .
[ mue_gi]-[lum - mue ] , we find no strong correlation between the structure seen in the distribution of surface brightnesses and that of any other structural parameter
. there is no correlation between surface brightness and color , with both hsb and lsb galaxies occupying the red and blue peaks .
figs [ vel - mue ] & [ vratio ] suggest that the source of the bimodality in surface brightness may be linked to galaxy dynamics and not stellar populations .
tv97 also proposed a dynamical explanation for the structural bimodality . in the following section we offer a tentative explanation for this exciting result in a more global context
in 4 , we have verified that the distribution of surface brightnesses is bimodal in uma and virgo cluster galaxies .
before we discuss possible explanations for this effect , let us investigate any possible biases in the data measurements and calibrations that could conspire to produce the observed bimodalities .
environment , which we discuss below , may also play a role in the structural bimodality .
various angles to address any cause for bimodality - field vs cluster galaxies or optical vs ir data , or both - will then be explored .
we conclude with plausible explanations for the observed galaxy surface brightness bimodality
. for 286 nir vcc galaxies with an over - subtraction of the sky of 0.05% ( red ) , and under - subtraction of 0.05% ( blue ) and the nominal sky subtraction ( black).,scaledwidth=48.0% ] we here investigate any potential source of bias from our measurement and calibration techniques . in our data paper ( mcdonald et al .
2009 ) , we quote estimates for the error induced by our photometric calibration of @xmath50.1 h-@xmath11 . in figs .
[ virgo_csb]a and [ virgo_esb_bymorph ] , we observed that the separation between the lsb and hsb peaks is @xmath52 h - mag arcsec@xmath6 . in order to artificially cause this gap ,
a surface brightness error greater than 0.5 h-@xmath11 on both lsb and hsb peaks must be invoked .
this error would also be systematic in order to `` evacuate '' isb galaxies : bright lsb galaxies would have to be made fainter and/or faint hsb galaxies would have to be made brighter .
we will examine below three possible processes that could induce such a large error : inclination correction , sky subtraction , and surface brightness measurements .
we will attempt to show that , although far from negligible , these errors can be controlled and should not bias the shape of surface brightness distributions .
finally , we will examine the possibility that either of the sb bimodalities are due to regions with clustered hsb or lsb galaxies ( i.e. an infalling group ) . in appendix
a , we demonstrated how a sky measurement error of @xmath440.01% can result in an artificial truncation or upturn in a deep surface brightness profile .
however , while lsb galaxies are highly sensitive to sky errors , a systematic over- or under- subtraction of the sky could hardly enhance the isb gap . in
, we show the distribution of @xmath56 if the images for all the sample galaxies are systematically over- or under - subtracted by 0.05% ( a huge error at h - band ) .
it is clear that , the net effect is to shift all galaxies towards lsb , not to create a gap . in 4.1
, we discussed deprojection corrections to measured surface brightness quantities . at nir wavelengths ,
galaxies are nearly transparent and the correction for deprojection is purely geometric .
since @xmath9 is measured from the disk fit , the geometric correction is trivial , provided the inclination of the disk is known .
however , the quantity @xmath56 , which depends on both bulge and disk light , is less easily corrected .
we thus make use of two different methods to correct @xmath56 for projection .
first , we can write the corrected effective surface brightness of the whole galaxy , @xmath92 , as : @xmath124 where @xmath125 is the axial ratio at the effective radius , @xmath47 .
this correction ( method a ) is straightforward , since we know the ellipticity as a function of radius from isophotal fits ( 3.2.3 ) .
the second inclination correction ( method b ) requires the decomposition of the sb profile into bulge and disk components and deprojection of the two components separately . knowing the inclination of the disk , we can deproject the disk fit ( assuming full transparency ) .
we further assume that the bulge requires no deprojection .
this yields the correction for the global @xmath92 : @xmath126 where @xmath127 is the axial ratio of the outermost isophote .
the effective radius is measured from the total light profile and the contributions from the individual bulge and disk profiles at this radius are used .
this correction is intuitively appealing as the bulge should require less of a deprojection correction than the disk .
the assumption of a spherical bulge may however be inadequate , especially in the presence of a bar .
however , we find that the isophotes in the central regions of most of our galaxies that exhibit bulges are , in fact , fairly round
. shows the result of applying the two aforementioned inclination corrections .
we find very little difference between methods a & b. as expected , the corrected values to face - on orientation for @xmath94 are systematically fainter .
most importantly , regardless of the correction used , bimodality is always preserved .
as there is little difference in the resulting distributions , we prefer method a for its simplicity and empirical , rather than parametric , quality .
furthermore , method b is less desirable since it may yield large errors in cases when the disk fit is unreliable or if the assumption of spherical bulge is unwarranted .
ideally , we wish to avoid any sensitivity to the bulge light as well as to localized radial structure .
such a parameter for pure disks is @xmath9 , but we do not have an analog for both disks and spheroids .
indeed , @xmath56 is affected by local structure while @xmath82 encompasses bulge light . of interest to us
is to investigate if the use of a local vs global estimator of surface brightness at different galactocentric radii should cause bimodality .
we can compute @xmath128 and @xmath129 as in 3.2.2 . at short radii
, we expect both @xmath128 and @xmath130 to be biased by any bulge or bar , while at large radii @xmath128 should suffer mostly from sb fluctuations and sky subtraction errors . and
table [ sbtable ] show the results of our exercise for all 286 vcc galaxies in our sample at h - band .
the result is interesting : regardless of the specific measure of surface brightness , whether local or global , computed at small or large radii , the three sb peaks are always preserved .
cccccc sb measured & esb & hsb & lsb & e - h & h - l + @xmath131&16.61 & 18.42 & 20.14 & 1.81 & 1.72 + @xmath132&16.44 & 18.65 & 20.71 & 2.16 & 2.06 + @xmath133&17.61 & 19.23 & 20.83 & 1.62 & 1.60 + @xmath134&17.93 & 19.54 & 21.16 & 1.61 & 1.62 + @xmath135&18.59 & 20.16 & 21.74 & 1.57 & 1.58 + @xmath136&19.15 & 20.36 & 21.76 & 1.21 & 1.41 + @xmath137&19.46 & 20.92 & 22.54 & 1.46 & 1.62 + + @xmath138 & 15.76 & 17.86 & 19.68 & 2.10 & 1.82 + @xmath139 & 16.61 & 18.42 & 20.14 & 1.81 & 1.72 + @xmath140 & 16.96 & 18.65 & 20.40 & 1.69 & 1.75 + @xmath141 & 17.43 & 19.08 & 20.86 & 1.65 & 1.78 + @xmath142 & 17.80 & 19.79 & 21.66 & 1.99 & 1.87 + @xmath143 & 19.19 & 20.18 & 21.77 & 0.99 & 1.59 + @xmath144 & 19.56 & 20.66 & 22.13 & 1.10 & 1.47 + [ sbtable ] as can be seen from table [ sbtable ] , the separations between the esb and hsb peaks , and the hsb and lsb peaks are roughly equal .
the remarkable result that the peak separations @xmath145esb - hsb@xmath146 and @xmath145hsb - lsb@xmath146 are nearly identical at most sb levels suggests that there may be a single mechanism responsible for all three peaks .
thus , we have shown that both the early- and late - type structural bimodality persist regardless of the surface brightness measurement , inclination correction , and sky subtraction error .
let us then address possible causes for the observed galaxy bimodalities , convinced that they are not an artifact of our methodology .
it is conceivable that one of the three surface brightness peaks could be due to the infall of a small group of galaxies with a narrow surface brightness distribution . in order to investigate this issue
we turn to , which shows the spatial distribution of galaxies in the four different sb peaks : hsb & lsb early - type galaxies , and hsb & lsb late - type galaxies .
it is clear from that both the hsb and lsb spiral and irregular galaxies are spread uniformly throughout the cluster .
likewise , most of the spheroids are distributed uniformly throughout the cluster , with an overabundance of hsb spheroids near the cluster center .
thus , although there is a constant infall of galaxies into the virgo cluster , it does not appear that this is causing an artificial peak in the surface brightness distribution .
freeman s law ( 1970 ) in its modern form is understood as the maximum surface brightness any galaxy disk can have , but it does not preclude the existence of fainter systems .
the general picture today is that the surface brightness function of galaxies has a tail at low surface brightnesses that extends to arbitrarily low values ( de jong & lacey 2000 ) .
previous studies that have attempted to quantify the distribution of galaxy surface brightnesses have focused mostly on field systems ( e.g. , mcgaugh & de blok 1995 ; de jong & lacey 2000 ; driver et al .
the effect of environment on galaxy surface brightnesses , if any , would thus be missed . in their pioneering study
, tv97 observed that the structural bimodality is strengthened if their uma sample is restricted to isolated galaxies .
for uma , an isolated galaxy was defined as a galaxy that has not encountered another galaxy during the lifetime of the cluster .
since uma is dynamically young , this definition is a reasonable one .
however , since the crossing time of the virgo cluster is less than one - tenth of a hubble time ( trentham & tully 2002 ) , it is unreasonable to perform a similar analysis for virgo cluster galaxies .
every galaxy in the cluster should have had at least one significant encounter within the last gyr , which means that we can not divide the sample into isolated and non - isolated galaxies .
various studies have reported that lsb galaxies form primarily in isolated environments ( e.g. , rosenbaum & bomans 2004 ) likely from 1 - 2@xmath147 peaks in the initial density fluctuation spectrum .
however , these authors found that local lsbs are almost exclusively found in clusters and filaments .
if lsb galaxies form from 1 - 2@xmath147 peaks , they would migrate from voids to high - density regions .
this does not explain how dynamically young galaxy clusters , such as uma , could have a high abundance of lsb galaxies .
instead , perhaps the cluster environment enhances lsb formation which would explain the current location of the lsb population as well as the sb bimodality seen in uma and virgo . for the uma and virgo clusters ,
there is a bimodality of disk surface brightness in spiral galaxies as well as bimodality in effective surface brightness for early - type galaxies . since
previous studies of field galaxies have not shown any evidence for bimodal surface brightness distributions , we might therefore think of structural bimodality as a cluster effect .
however , it should be noted that most of the studies of the surface brightness distribution of field galaxies have focused strictly on optical imaging for lack of available complete similar data bases at ir wavelengths .
we showed in figs .
[ virgo_sdss_csb ] & [ virgo_esb_sdss ] that the bimodality in disk and total surface brightnesses is enhanced in the nir , due to uncertain and significant corrections for dust at optical wavelengths .
none of the major nir data bases available today ( e.g. , de jong 1996 ; jarrett et al . 2003 ; grauer et al .
2003 ; macarthur et al . 2003 ; smith et al .
2008 ) cover the full range of surface brightness to sample the lsb peak .
for this reason , it is unclear whether the structural bimodality observed at nir wavelengths is indeed a cluster effect since it has only been observed in galaxy clusters thus far , or whether it is inherent to galaxies in all environments .
since we see structural bimodality in the most isolated galaxies of the uma cluster as well as in the denser environment of virgo , it appears that this effect is independent of environment .
new data from field nir surveys such as panstarrs and ukidss ( e.g. , smith et al .
2008 ) will be needed to revisit this question for field galaxies ( provided careful assessment of completeness for field samples ) .
we now entertain possible physical interpretations for structural bimodality .
dressler ( 1980 ) first demonstrated the existence of strong correlations between various galaxy properties and environment .
red sequence galaxies live preferentially in dense environments , whilst blue , star - forming galaxies are predominantly found in isolation .
this dependence on environment was reinforced as a major cause for the observed sdss color and star formation bimodality .
not surprisingly , the local density of galaxies causes a strong morphological bias . merging and tidal interactions , which are more common in dense environments ( moss & whittle 1993 ; gnedin 2003 ) , can deplete the gas content of spiral and irregular galaxies , turning them into ellipticals and possibly s0s ( toomre & toomre 1972 ; farouki & shapiro 1981 ; barnes 1999 ) .
an intracluster gas can also strip gas from late - type galaxies , reducing star formation drastically ( gunn & gott 1972 ) .
processes such as galaxy - galaxy harassment , tidal stripping , and starvation , which are more efficient in dense environments , will all reduce the amount of active star formation .
it is then not surprising that there should be a correlation between color , star formation and environment in this context .
there are also several properties for the dark halo that appear to be dependent on environment .
the mass , shape , spin and formation time of dark matter halos formed in denser environments that become clusters or filaments appear to be significantly different than those formed in isolation ( hahn et al .
simulations suggest that galaxy halos with masses m @xmath148 m@xmath149 , where m@xmath149 is the galaxy halo mass at which the mass function turns over , have parameters that correlate strongly with environment , while those with m @xmath150 m@xmath149 do not .
hahn et al .
( 2007 ) find that low mass halos are older and have higher angular momenta when formed in proto - cluster environments , while low mass halos in the field form later and have lower angular momenta .
the dependence of halo properties on environment could explain part of the observed differences in cluster and field galaxies . however , using a sample of 329 nearby cluster and field star forming galaxies , vogt et al . ( 2004 ) failed to find any correlation between stellar mass - to - light ratios or circular velocities with environment .
more observational and theoretical studies of environmental effects on galaxies will be needed .
we have found so far that the gas - poor and the gas - rich uma and virgo cluster galaxies each exhibit a surface brightness bimodality .
it is unclear whether this result is unique to clusters , since there is no complimentary nir survey of field galaxies that reaches the same depth in surface brightness .
however , uma very much resembles the field due to its low density ( see tv97 ) , whilst virgo is a much richer cluster , suggesting that environment may not play a strong role .
this surprising result requires an explanation which current models of galaxy formation fail to provide .
however , our results may suggest routes for further investigation and allow the rejection of certain hypotheses .
we now address various explanations for the structural bimodality in light of the available data and current theoretical ideas . 1 .
_ stellar populations _ : previous studies based on sdss imaging for very large number of galaxies have shown bimodal distributions of color and star formation rate ( e.g. , strateva et al . 2001 ) , which is largely the result of two galaxy populations : blue , star - forming spiral galaxies and red , quiescent spheroidal galaxies .
we showed in 4.5 that color is uncorrelated with the observed surface brightness bimodalities of cluster galaxies ; stellar populations are thus unlikely to be a factor .
lsb formation enhanced in clusters _ : if the structural bimodality is only present in clusters , it could result from more efficient tidal torquing in dense environments ( barnes & efstathiou 1987 ; bett et al . 2007 ;
hahn et al .
if dark matter halos acquire more angular momentum in a proto - cluster environment , then lsb galaxies ( which halt their collapse early due to high angular momentum ) would preferentially be formed in these dense environments . in this scenario
, higher surface brightness galaxies could form from the merging of lower sb galaxies if they contain gas that can dissipate energy before forming stars .
one would thus expect a `` smearing '' of the lsb peak towards hsb , but a discrete gap is still unaccounted for .
it has been well documented that there is a maximum surface brightness that galaxy disks will not exceed ( freeman 1970 ) .
if this maximal disk surface density is obeyed , then galaxies which evolve at various rates from low to high surface brightnesses via merging may ultimately halt their evolution at a disk surface density equal to the freeman value , thus enhancing the hsb peak .
this scenario does not preclude the formation of isb galaxies via the merging of two lsb galaxies or other such events , but simply suggests that the isb case might be a rapid transition stage for galaxies as they merge and evolve towards the freeman limit .
indeed , we do not see an absence of isb galaxies in clusters , but rather a relative excess of hsb and lsb galaxies .
+ our group is currently studying the velocity function ( and thus the distribution of angular momenta ) of virgo cluster galaxies which will provide further insight into this problem . more studies like that of hahn et al .
( 2007 ) would help elucidate the role of environment on the distribution of various dynamical properties .
3 . _ specific stable radial configurations _ : expanding upon the ideas of mestel ( 1963 ) , tv97 suggested that galaxy disks could settle into one of two stable radial configurations : disks with a dynamically important baryonic component in the center and those without ( also known as maximal and minimal disks ) .
they suggested that intermediate surface brightness galaxies are transient phenomena in high - density environments where interactions re - arrange the baryons .
this is supported by the fact that , in uma , most intermediate surface brightness galaxies have significant neighbors and thus the bimodality is enhanced for isolated systems .
however , the lack of isb galaxies in the densest regions of the virgo cluster suggests the opposite result .
+ if the structural bimodality is unaffected by environment then there may be a more fundamental mechanism preventing the stability of isb galaxies .
the isb regime may correspond to an unstable dynamical configuration , possibly implying a bimodality in the spin parameter , @xmath151 , for disk galaxies .
4 . _ gas depletion in clusters halts evolution _ :
the amount of gas retained by cluster and field galaxies differ due to various factors such as high frequency of tidal interactions in clusters ( moss & whittle 1993 ; gnedin 2003 ) and the dense intracluster medium which can cause gas stripping ( gunn & gott 1972 ) .
if isb galaxies form primarily via passive evolution and hsb galaxies form primarily via mergers , then the cluster environment would act to prevent lsb - to - isb evolution and enhance lsb - to - hsb evolution . without being able to pinpoint the role that environment plays in the observed structural bimodalities , it is difficult to provide an explanation for their existence .
it is likely that no single hypothesis above fully explains the surface brightness bimodality and that the separate bimodalities in early - type and late - type galaxies will require their own explanations ( although their alignment and equal spacing , as seen in table [ sbtable ] , suggests something fundamentally similar ! ) .
it is difficult to speculate further without a proper analysis of the distribution of @xmath56 for field galaxies , a complimentary dynamical analysis of the virgo cluster , as well as detailed numerical simulations for the effects of external processes , such as tidal stripping and mergers , or internal processes , such as outflows , on surface brightness profiles of cluster and field galaxies .
we have constructed the deepest complete sample of near - ir photometry of virgo cluster galaxies to date and with it , we were able to achieve a detailed analysis of the distribution of surface brightness profiles and luminosities for 286 virgo cluster galaxies .
we also extracted @xmath37 photometry from the sdss for comparative analysis with a larger sample of 742 vcc galaxies .
this virgo cluster study also follows from our re - analysis of similar data for the smaller uma cluster ( mcdonald et al .
our main findings are : * there exists a structural bimodality in each of the early and late type galaxy populations of the uma and virgo clusters . from nir images of these galaxies ,
we observe surface brightness peaks at @xmath94=21.7 @xmath11 and @xmath94=19.7 @xmath11 for late - type galaxies , and @xmath94=19.7 @xmath11 and @xmath94=17.8 mag arcsec@xmath6 for early - type galaxies .
these three peaks are independent at a confidence level @xmath15290% . * the observed structural bimodalities in each galaxy class ( early or late ) is independent of the observed color bimodality . *
the nir luminosity function for virgo has a prominent dip at intermediate luminosities which is also seen , to a lesser extent , in the optical sdss data .
the fact that the dichotomy between hsb and lsb galaxies is independent of color and morphology implies a possible dynamical connection .
we have detected a weak , but non - negligible , correlation between @xmath153 and @xmath56 , but more data will be needed to strengthen any dependence .
the range of densities from uma to virgo also suggests that the mechanism for producing hsb and lsb galaxies is independent of environment . without a sample of true field galaxies it would , however , be unwise to speculate about any environmental effect on @xmath56 .
unfortunately , current studies of the distribution of surface brightnesses of field galaxies are either hampered by dust extinction in the optical or have not achieved deep and complete enough coverage at nir wavelengths .
thus , most reasonable scenarios for field / cluster galaxy evolution are still plausible . the most obvious and relevant test for the hypothesis of different @xmath151-distributions in cluster and field environments , as proposed by hahn et al .
( 2007 ) , is to assemble a complete sample of resolved rotation curves for cluster galaxies .
this would enable new constraints on the distribution of the spin parameter , @xmath151 , which may be responsible for setting @xmath56 ( e.g. , dutton et al .
hahn et al .
( 2007 ) predict a distribution that is skewed towards high values of @xmath151 in clusters and filaments . a complete sample of resolved dynamical data for cluster and field galaxies should allow a full assessment of the observed structural bimodality in terms of unbiased parameters such as dynamical mass and the slope of the rotation curve .
it has been over ten years since the discovery of the surface brightness dichotomy in disk galaxies by tv97 .
we have now extended that study to a broader environment to establish the presence of three separate peaks in the distribution of effective surface brightnesses in all galaxies with @xmath15290% confidence , and two peaks in the distribution of disk central surface brightnesses in disk galaxies with 95% confidence .
the next few years should see significant improvements in the size and scope ( e.g. , range of surface brightness ) of galaxy nir surveys and in studies into the effect of environment on @xmath56 .
the theoretical interpretation for these results however awaits a better determination of @xmath151 for galaxies in the field and in clusters , and a better interpretation of the combined effects of merging and stripping to the overall surface brightness of galaxies .
we acknowledge valuable conversations with jeremy bailin , aaron dutton , julianne dalcanton , roelof de jong , lauren macarthur , hans - walter rix , frank van den bosch , and sylvain veilleux .
many thanks also to guiseppe gavazzi for providing precious goldmine galaxy images , to yucong zhu for providing sdss luminosities for the vcc / sdss sample , and to joel roediger for his stellar mass - to - light ratios for the virgo `` h - band '' galaxies .
we are also grateful to jon loveday for his careful referee report which led to an improved presentation of our paper .
sc would like to acknowledge financial support via a discovery grant from the national science and engineering council of canada .
rbt acknowledges support from us national science foundation award ast 03 - 07706 .
this research has made use of @xmath154 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 , as well as nasa s astrophysics data system ; @xmath155 the @xmath156 @xmath157 @xmath158 @xmath159 ( sdss ) . funding for the creation and distribution of the sdss archive has been provided by the alfred p. sloan foundation , the participating institutions , the national aeronautics and space administration , the national science foundation , the u.s .
department of energy , the japanese monbukagakusho , and the max planck society .
the sdss web site is http://www.sdss.org/. the sdss is managed by the astrophysical research consortium ( arc ) for the participating institutions . @xmath160
the @xmath161 database ( http://leda.univ-lyon1.fr ) . 99 adelman - mccarthy , j. k. , et al .
2008 , , 175 , 297 andredakis , y. c. , peletier , r. f. , & balcells , m. 1995 , , 275 , 874 smith , a. , loveday . , j , cross . , n. 2008 , arxiv:0806.0343v1 bailin , j. , & harris , w. e. 2008 , , 385 , 1835 baldry , i. k. , glazebrook , k. , brinkmann , j. , ivezi , . ,
lupton , r. h. , nichol , r. c. , & szalay , a. s. 2004 , , 600 , 681 balogh , m. l. , baldry , i. k. , nichol , r. , miller , c. , bower , r. , & glazebrook , k. 2004 , , 615 , l101 barnes , j. , & efstathiou , g. 1987 , , 319 , 575 barnes , j. e. 1999 , the evolution of galaxies on cosmological timescales , 187 , 293 bell , e. f. , & de blok , w. j. g. 2000 , , 311 , 668 bell , e. f. , mcintosh , d. h. , katz , n. , & weinberg , m. d. 2003 , , 149 , 289 bett , p. , eke , v. , frenk , c. s. , jenkins , a. , helly , j. , & navarro , j. 2007 , , 376 , 215 binggeli , b. , sandage , a. , & tammann , g. a. 1985 , , 90 , 1681 [ vcc ] birnboim , y. , & dekel , a. 2003 , , 345 , 349 blanton , m. r. , et al .
2003 , , 594 , 186 brinchmann , j. , charlot , s. , white , s. d. m. , tremonti , c. , kauffmann , g. , heckman , t. , & brinkmann , j. 2004 , , 351 , 1151 courteau , s. 1996 , , 103 , 363 courteau , s. , de jong , r. s. , & broeils , a. h. 1996 , , 457 , l73 courteau , s. 1997 , , 114 , 2402 courteau , s. , & rix , h .- w .
1999 , , 513 , 561 courteau , s. , dutton , a. a. , van den bosch , f. c. , macarthur , l. a. , dekel , a. , mcintosh , d. h. , & dale , d. a. 2007 , , 671 , 203 dalcanton , j. j. , spergel , d. n. , & summers , f. j. 1997 , , 482 , 659 de jong , r. s. 1996 , , 313 , 45 de jong , r. s. , & lacey , c. 2000 , , 545 , 781 de vaucouleurs , g. 1948 , annales dastrophysique , 11 , 247 de vaucouleurs , g. 1959 , handbuch der physik , 53 , 311 de vaucouleurs , g. 1961 , , 6 , 213 dekel , a. , & birnboim , y. 2006 , , 368 , 2 dressler , a. 1980 , , 236 , 351 driver , s. p. , liske , j. , cross , n. j. g. , de propris , r. , & allen , p. d. 2005 , , 360 , 81 dutton , a. a. , van den bosch , f. c. , dekel , a. , & courteau , s. 2007 , , 654 , 27 efstathiou , g. 2000 , , 317 , 697 efstathiou , g. , lake , g. , & negroponte , j. 1982 , , 199 , 1069 fall , s. m. , & efstathiou , g. 1980 , , 193 , 189 farouki , r. , & shapiro , s. l. 1981 , , 243 , 32 ferguson , a. m. n. , & clarke , c. j. 2001 , , 325 , 781 foyle , k. , courteau , s. , r. thacker 2008 , mnras , in print freeman , k. c. 1970 , , 160 , 811 ftaclas , c. , struble , m. f. , & fanelli , m. n. 1984 , , 282 , 19 gavazzi , g. , pierini , d. , baffa , c. , lisi , f. , hunt , l. k. , randone , i. , & boselli , a. 1996 , , 120 , 521 gavazzi , g. , boselli , a. , donati , a. , franzetti , p. , & scodeggio , m. 2003 , , 400 , 451 gnedin , o. y. 2003 , , 589 , 752 grauer , a. d. , rieke , m. j. , & quillen , a. c. 2003 , , 149 , 327 gunn , j. e. , & gott , j. r. i. 1972 , , 176 , 1 gunn , j. e. 1982 , astrophysical cosmology proceedings , 233 hahn , o. , porciani , c. , carollo , c. m. , & dekel , a. 2007 , , 375 , 489 hinz , j. l. , rieke , g. h. , & caldwell , n. 2003 , , 126 , 2622 huchtmeier , w. k. , & richter , o .- g .
1986 , , 64 , 111 jarrett , t. h. , chester , t. , cutri , r. , schneider , s. e. , & huchra , j. p. 2003
, , 125 , 525 jenkins , l. p. , hornschemeier , a. e. , mobasher , b. , alexander , d. m. , & bauer , f. e. 2007 , , 666 , 846 kambas , a. , davies , j. i. , smith , r. m. , bianchi , s. , & haynes , j. a. 2000 , , 120 , 1316 kauffmann , g. , et al .
2003 , , 341 , 33 kent , s. m. 1985 , , 59 , 115 kere , d. , katz , n. , weinberg , d. h. , & dav , r. 2005 , , 363 , 2 kirby , e. , jerjen , h. , ryder , s. , & driver , s. 2008 , arxiv:0808.2529 kormendy , j. , & kennicutt , r. c. , jr . 2004 , , 42 , 603 lin , d. n. c. , & pringle , j. e. 1987 , , 320 , l87 macarthur , l. a. , courteau , s. , & holtzman , j. a. 2003 , , 582 , 689 mcdonald , m. , courteau , s. , & tully , r. b. 2008 , arxiv:0810.5530 mcgaugh , s. s. , & de blok , w. j. g. 1997 , , 481 , 689 mei , s. , et al .
2007 , , 655 , 144 mestel , l. 1963 , , 126 , 553 meurer , g. r. , et al .
2002 , the 2002 hst calibration workshop : hubble after the installation of the acs and the nicmos cooling system , 65 mo , h. j. , mao , s. , & white , s. d. m. 1998 , , 295 , 319 moffat , a. f. j. 1969 , , 3 , 455 moss , c. , & whittle , m. 1993 , , 407 , l17 paturel , g. , petit , c. , prugniel , p. , theureau , g. , rousseau , j. , brouty , m. , dubois , p. , & cambrsy , l. 2003 , , 412 , 45 pohlen , m. , & trujillo , i. 2006 , , 454 , 759 rees , m. j. , & ostriker , j. p. 1977
, , 179 , 541 rines , k. , & geller , m. j. 2008 , , 135 , 1837 rosenbaum , s. d. , & bomans , d. j. 2004 , , 422 , l5 rubin , v. c. , waterman , a. h. , & kenney , j. d. p. 1999
, , 118 , 236 sabatini , s. , davies , j. , scaramella , r. , smith , r. , baes , m. , linder , s. m. , roberts , s. , & testa , v. 2003 , , 341 , 9 sersic , j. l. 1968 , cordoba , argentina : observatorio astronomico , 1968 , skrutskie , m. f. , et al .
2006 , , 131 , 1163 smith , a. j. , loveday , j. , and cross , n. j. g. 2008 , mnras ( astro - ph/0806.0343 ) strateva , i. , et al .
2001 , , 122 , 1861 toth , g. , & ostriker , j. p. 1992
, , 389 , 5 toomre , a. , & toomre , j. 1972 , , 178 , 623 trentham , n. , & tully , r. b. 2002 , , 335 , 712 tully , r. b. , & verheijen , m. a. w. 1997 , , 484 , 145 tully , r. b. , & fouque , p. 1985
, , 58 , 67 van den bosch , f. c. 2001 , , 327 , 1334 vogt , n. p. , haynes , m. p. , giovanelli , r. , & herter , t. 2004 , , 127 , 3325 wainscoat , r. j. , & cowie , l. l. 1992 , , 103 , 332 white , s. d. m. , & rees , m. j. 1978 , , 183 , 341 white , s. d. m. , & frenk , c. s. 1991 , , 379 , 52 york , d. g. , et al .
2000 , , 120 , 1579 zavala , j. , avila - reese , v. , hernndez - toledo , h. , & firmani , c. 2003 , , 412 , 633
in order to characterize the intrinsic distribution of structural parameters such as @xmath56 , @xmath47 and @xmath51 , free from evolutionary effects , we have generated a monte carlo suite of idealized galaxy models using combinations of a large number of model parameters . a single galaxy light profile can be determined with 8 basic parameters . for the bulge and disk ,
the parameters @xmath162 , r@xmath163 , @xmath164 , r@xmath165 and srsic @xmath49 are selected randomly from gaussian distributions whose widths match roughly observed galaxy distributions .
another 2 parameters govern the location and brightness of spiral arms .
the location of the arms varies from 1 to 2 disk scale lengths and the peak arm brightnesses ranges from 0.3 to 0.5 @xmath11 above the disk brightness .
finally , all model profiles were convolved with a gaussian function to simulate the effect of atmospheric blur ; the seeing fwhm is our 8th model parameter .
we have also forced 1 in every 5 profiles to be a pure - sersic function to reflect the presence of systems with no disk .
the gaussian distribution of input parameters for our models is shown in the top 6 panels of . in order to infuse more realism into this experiment ,
two different types of error were added to the model profiles : photon ( galaxy + sky ) random noise and systematic sky subtraction error .
we determined the appropriate photon noise in the h - band from the distribution of measured surface brightness errors for our h - band images ( mcdonald et al .
2009 ) . from this distribution ,
an error function was determined to add cosmic error to our model surface brightness profiles : @xmath166 this result is very similar to that produced by courteau ( 1996 ) for @xmath2-band surface brightness profiles . a typical noise value @xmath167 now is added randomly to each model profile at each surface brightness level . in order to simulate sky - subtraction errors , a systematic `` measurement error ''
was first determined .
this value is @xmath168 of the sky brightness at h-band.this random sky error can artificially produce truncated ( over - subtracted sky ) or upbending ( under - subtracted sky ) surface brightness profiles , depending on its sign .
we have generated model galaxies by combining the 8 model parameters , each selected randomly from a realistic range of values , and accounting for random and systematic errors .
the recovered parameter distribution for the 100,000 simulated surface brightness profiles can be seen in the lower 3 panels of for the total galaxy light .
the @xmath56 distribution is nearly gaussian with a peak slightly brighter than the disk @xmath56 peak due to the inclusion of substantial bulges in most profiles .
an hsb tail can be seen in the @xmath56 distribution due to severely bulge - dominated systems .
these systems have higher @xmath56 values since bulges typically have higher surface brightnesses than disks .
the distribution of the @xmath47 parameter is a combination of bulge and disk components ; the peak is controlled by the bulge @xmath47 for bulge - dominated systems and the tail is due to disk - dominated systems .
finally , there is a strong peak in the @xmath51 distribution near @xmath51=2.8 , the expectation value for pure - disk systems .
profiles with @xmath51@xmath169 arise from cored ( @xmath170 ) systems or galaxies with spiral arms .
profiles with @xmath51@xmath171 or higher are bulge - dominated .
@xmath51 values cover the full range from 1.5 to 6.5 , in excellent agreement with our data ( see ) . with our realistic model surface brightness profiles
, we can trace the effects of sky measurement errors .
this is important since such errors can drastically affect the shape of the density profile , causing an artificial truncation or anti - truncation at large radii .
the distribution of total @xmath56 could thus be biased as a result . in
, we show the effect of increasing the sky measurement error from @xmath440.001% to @xmath440.05% .
the net effect of a positive sky error is to pull lsb galaxies below the sky noise yielding a faint tail , but the bright sb peak never shifts .
shows the variations in @xmath56 due to over- and under - subtracted sky errors .
an under - subtraction error ( upper half ) rarely affects the measured @xmath56 by more than 0.2 @xmath11 .
an over - subtraction error ( lower half ) has a more drastic effect as it can shift a significant part of the outer disk below observable levels .
recall that , while these tests about sky errors can be illuminating , the measured h - band sky error is typically @xmath440.005% and the effect of a systematic sky error is thus minimal .
our simulations confirm that if the input structural parameters are distributed normally , we should expect a nearly normal distribution of effective surface brightnesses . under no conditions
can we generate a distribution of @xmath56 that deviates significantly from the normal function .
this result can be used as an unbiased comparison against galaxy structural parameters extracted from real data .
the study of the distribution of galaxy structural parameters from light profiles can benefit greatly from the availability of dynamical information .
for instance , tv97 found a clear relationship between a galaxy s disk central surface brightness and the inner slope of its rotation curve . in order to test
this claim , and other correlations between dynamical and photometric parameters , resolved rotation curves would be ideal . failing any information about the shape of the rotation curves , we resort to 1d measures of maximum rotational velocity .
provided sufficient accuracy , they can , at least , enable global comparisons with light profile parameters for the spiral galaxies in our sample .
we retrieved data for 58 vcc galaxies from the extragalactic distance database ( edd ; http://edd.ifa.hawaii.edu/ ) which overlap with our h - band sample .
these data have been corrected for aperture and turbulent effects following tully & fouqu ( 1985 ) who defined the linewidth parameter @xmath172 .
we have also verified that the linear ( almost 1-to-1 ) relation between radio and optically measured line widths observed by courteau ( 1997 ) and others is also observed in this sample .
all linewidths were deprojected as @xmath173 and @xmath174 is given by : in order to duplicate fig .
14 from tv97 , we are interested in computing the ratio @xmath178/@xmath179 . here
@xmath178 is the maximum rotation velocity resulting from a purely baryonic system , while @xmath179 is the observed rotation speed ( which includes dark matter ) at a fiducial radius . tv97 postulate that lsb galaxies tend to be dominated by dark matter all the way into the innermost regions ( @xmath180 ) while hsb galaxies are baryon - dominated within roughly two disk scale lengths ( @xmath181 ; e.g. , courteau & rix 1997 ; d07 ) . where @xmath183 refers to the luminosity within the radius , @xmath184 , and @xmath185 is the stellar mass - to - light ratio computed using sdss colors and bell et al .
the peak value of v@xmath186 is the value we call v@xmath149 .
we do not see the strong dynamical bimodality that is apparent in tv97 , however there does appear to be a correlation between v@xmath149/v@xmath187 and @xmath74 , as well as hints of a bimodality in @xmath74 .
lsb galaxies tend to be dark matter - dominated ( v@xmath149/v@xmath188 ) , while hsb galaxies would be baryon - dominated ( v@xmath149/v@xmath189 ) .
the galaxies with v@xmath149/v@xmath190 tend to have bright bulges or compact nucleii which cause a spike in the light - weighted rotation curve at short radii
. however , with linewidths for only 20% of our sample , it is impossible to make these claims with statistical significance .
to do this analysis correctly , we need to either obtain reliable linewidths or deep , resolved , optical rotation curves for the full sample .
our team is currently pursuing a complete characterization of the velocity function of virgo cluster galaxies in order to address these issues . /v@xmath187 and
@xmath94 for 58 disk galaxies in the virgo cluster for two different mass - to - light ratios .
the dashed histogram corresponds to the bell et al .
( 2003 ) mass - to - light ratios , while the solid histogram corresponds to a constant mass - to - light ratio of 1.0.,scaledwidth=48.0% ] | we have acquired deep , h - band , imaging for a sample of 286 virgo cluster galaxies with b@xmath0 mag and extracted surface photometry from optical @xmath1,@xmath2,@xmath3,@xmath4 sloan digital sky survey images of 742 virgo cluster catalog galaxies , including those with h - band images .
we confirm the detection of a dip in the luminosity function indicative of a discontinuity in the cluster galaxy population ; the dip is more pronounced at redder wavelengths .
we find , in agreement with earlier works of tully & verheijen and ours for ursa major cluster galaxies , a clear dichotomy between high and low surface brightness galaxy disks .
the difference between the low and high brightness peaks of virgo disk galaxies is @xmath52 h - mag arcsec@xmath6 , significantly larger than any systematic errors .
the high surface brightness disk galaxies have two distinct classes of high and low concentration bulges , while low surface brightness galaxies have only low concentration bulges .
early - type galaxies exhibit a similar structural bimodality though offset from that of the spiral galaxies towards higher surface brightnesses .
both the early- and late - type structural bimodalities are uncorrelated with colour or any other structural parameter except , possibly , circular velocity .
random realizations of realistic surface brightness profiles suggest that a bimodal distribution of effective surface brightness is unexpected based on normal distributions of bulge and disk parameters .
rather , the structural bimodality may be linked to dynamical properties of galaxies .
low angular momentum systems may collapse to form dynamically important disks with high surface brightness , while high angular momentum systems would end up as low surface brightness galaxies dominated by the dark halo at all radii .
the confirmation of structural bimodality for gas - rich and gas - poor galaxies in the high - density virgo cluster as well as the low - density uma cluster suggests that this phenomenon is independent of environment .
[ firstpage ] |
we consider the diffusion processes @xmath4 pertaining to the following distributed control system , with small random perturbations ( see fig .
[ fig - dcs ] ) @xmath5 where * @xmath6 is an @xmath7-valued diffusion process that corresponds to the @xmath8th - subsystem ( with @xmath9 ) , * the functions @xmath10 are uniformly lipschitz , with bounded first derivatives , @xmath11 is a small positive number ( which is related to the random perturbation level in the system ) , * @xmath12 is lipschitz with the least eigenvalue of @xmath13 uniformly bounded away from zero , i.e. , @xmath14 for some @xmath15 , * @xmath16 ( with @xmath17 ) is a @xmath18-dimensional standard wiener process , * @xmath19 is a @xmath20-valued measurable control process to the @xmath8th - subsystem , i.e. , an admissible control from the measurable set @xmath21 . in this paper , we identify two admissible controls @xmath22 , for @xmath23 , being the same on @xmath24 $ ] if @xmath25 \bigr\ } = 1 $ ] .
if @xmath26 , then , for every @xmath27 , there exists a borel measurable function @xmath28 , \mathbb{r}^m \bigr ) \rightarrow \mathcal{u}_i$ ] ( with respect to some underlying borel @xmath29-algebra ) such that @xmath30 , \label{eq2}\end{aligned}\ ] ] with probability one ( w.p.1 ) .
the functions @xmath31 , for @xmath2 , in equation , with any progressively measurable control @xmath32 , depend only on @xmath33 .
furthermore , we assume that the distributed control system , which is formed by the first @xmath3 subsystems , satisfies an appropriate hrmander condition , i.e. , a hypoellipticity assumption on the diffusion processes @xmath34 ( e.g. , see @xcite or ( * ? ? ?
* section 3 ) ) .
notice that the random perturbation has to pass through the second subsystem , the third subsystem , and so on to the @xmath3th - subsystem .
hence , such a distributed control system is described by an @xmath35 dimensional diffusion process , which is degenerate in the sense that the backward operator associated with it is a degenerate parabolic equation .
( 50,13 ) ( 1,9.5)(1,0)45 ( 1,13)(0,-1)3.5 ( 46,13)(0,-1)3.5 ( 18,11)@xmath36 ( 1,13)(1,0)45 ( 18.75,4)(1,0)0.2 ( 19.50,4)(1,0)0.2 ( 20.25,4)(1,0)0.2 ( 33.75,4)(1,0)0.2 ( 34.50,4)(1,0)0.2 ( 35.25,4)(1,0)0.2 @xmath37 @xmath38 + @xmath39 , ~ \forall s \ge 0 , ~ j = 1 , 2 , \ldots n,\\ & i_i^{in}(x^1 , x^2 , \ldots , x^{i-1 } ) ~ \text{and } ~ i_i^{out}(x^1 , x^2 , \ldots , x^{i } ) ~ \text{are information for the expanding construction . } \end{array}$ ] [ r1 ] in general , the hypoellipticity is related to a strong accessibility property of controllable nonlinear systems that are driven by white noise ( e.g. , see @xcite concerning the controllability of nonlinear systems , which is closely related to @xcite and @xcite ) .
that is , the hypoellipticity assumption implies that the diffusion process @xmath40 has a transition probability density @xmath41 , which is @xmath42 on @xmath43 , with a strong feller property .
let @xmath44 , for @xmath45 , be bounded open domains with smooth boundaries ( i.e. , @xmath46 is a manifold of class @xmath47 ) .
moreover , let @xmath48 be the open sets that are given by @xmath49 suppose that , for a fixed @xmath50 , the distributed control system , which is compatible with expanding construction , is formed by the first @xmath3 subsystems ( i.e. , obtained by adding one after the other , until all @xmath3th subsystems are included ) .
furthermore , assume that the newly constructed distributed control system is composed with some admissible controls @xmath51 , @xmath52 , for @xmath53 .
let @xmath54 be the exit - time for the diffusion process @xmath40 ( corresponding to the @xmath3th - subsystem ) , for a fixed @xmath55 , with @xmath56 , from the given domain @xmath57 , i.e. , @xmath58 which depends on the behavior of the following ( deterministic ) distributed control system @xmath59 in this paper , we specifically consider a risk - sensitive version of the mean escape time criterion with respect to the @xmath3th - subsystem , i.e. , @xmath60 where @xmath61 , for each @xmath50 , are positive design parameters and the expectation @xmath62 is conditioned on the initial point @xmath63 as well as on the admissible controls @xmath64 .
notice that @xmath65 in the exit - time for the diffusion process @xmath66 ( which corresponds to the @xmath67st - subsystem ) from the domain @xmath68 with respect to the admissible ( optimal ) control @xmath69 , @xmath70 , with @xmath71 . ] [ r2 ] here we remark that the criterion in equation makes sense only if we have the following conditions @xmath72 moreover , such conditions depend on the constituting subsystems , the admissible controls from the measurable sets @xmath73 , as well as on the given bounded open domains @xmath74 , for @xmath75 ( see section [ s3(2 ) ] for further discussion ) .
then , the problem of risk - sensitive escape control ( with respect to the @xmath3th - subsystem ) will amount to obtaining a supremum value for @xmath76 , i.e. , @xmath77 with respect to some progressively measurable control @xmath56 , for each @xmath50 .
notice that , for a fixed admissible control @xmath78 from the measurable set @xmath79 , if we obtain a representation for equation as a minimal cost for an associated stochastic optimal control problem , then we will be able to obtain a representation for @xmath80 as a value function for a stochastic differential game .
this further allow us to link this progressively measurable control @xmath78 in the original control problem with a strategy for the maximizing player of the associated stochastic differential game .
furthermore , such a connection between the risk - sensitive value function and a deterministic differential game can be made immediately , when the small random perturbation vanishes in the limit . before concluding this section
, it is worth mentioning that some interesting studies on risk - sensitive control problem for dynamical systems with small random perturbations have been reported in literature ( for example , see @xcite using pde viscosity solution techniques ; see @xcite using the probabilistic argumentation and the variational representation for degenerate diffusion processes ; see also @xcite , @xcite or @xcite for some connections between the risk - sensitive stochastic control and dynamic games ) .
an outline of the paper is as follows . in section [ s2 ]
, we introduce a family of two - player differential games where _ player_-@xmath67 will attempt to maximize the mean escape time criterion corresponding to each of the subsystems ; while _ player_-@xmath81 will attempt to minimize it . in this section
, we also provide some preliminary results that are useful for proving our main results . in section [ s3 ] , we present our main results where we consider a risk - sensitive version of the mean escape time criterion with respect to each of the subsystems . using the variational representation ,
we characterize the risk - sensitive escape control for the diffusion process as the lower and upper values of the associated stochastic differential game .
finally , we comment on the implication of our results , where one is also interested in evaluating the performance of the risk - sensitive escape control for the diffusion process , when there is some norm - bounded modeling error in the distributed control system .
in this subsection , we consider a family of two - player differential games . for a fixed @xmath50 , at each time @xmath82 , _
player_-@xmath67 picks a strategy @xmath83 from the admissible control space @xmath79 , and _
player_-@xmath81 picks a control @xmath84 from @xmath85 in such a way that the functions @xmath86 and @xmath87 belong to the strategy sets @xmath88 and @xmath89 respectively . here , we also identify that @xmath90 and @xmath91 , for any @xmath92 , as metric spaces under any metric which is equivalent to convergence in @xmath93 , \mathbb{r}^{r_i } \bigr)$ ] and @xmath93 , \mathbb{r}^m \bigr)$ ] .
suppose that both players have played the game up to the @xmath94th - stage ( see footnote [ fn ] ) .
let @xmath51 , @xmath52 , for @xmath53 , be the admissible control strategies picked by the maximizing _ player_-@xmath67 .
then , at the @xmath3th - stage , the dynamics of the game is given by the following differential equations @xmath95 with an associated cost criterion @xmath96 where @xmath97 .
note that the goal of _ player_-@xmath67 is to maximize @xmath98 with respect to @xmath78 and while that of _ player_-@xmath81 is to minimize it with respect to @xmath99 , for each @xmath92 . here , we remark that _ player_-@xmath67 will attempt preventing the diffusion process @xmath40 from leaving the given domain @xmath57 ( i.e. , representing the exact control in risk - sensitive problem ) ; while _ player_-@xmath81 will attempt forcing out the diffusion process from the domain ( i.e. , acting the role of the disturbance in the distributed control system ) .
picked by _
player_-@xmath67 affects only the dynamics of the game , not directly the cost criterion . ]
furthermore , a mapping @xmath100 is said to be a strategy for the maximizing player if it is measurable and , for @xmath101 , @xmath102\end{aligned}\ ] ] implies @xmath103(t ) = \alpha_{\ell}[\hat{v}_{\ell}](t ) , \quad \forall t \in [ 0,\ , s],\end{aligned}\ ] ] almost everywhere , for every @xmath27 . similarly , a mapping @xmath104 is a strategy for the minimizing player if it is measurable and , for @xmath105 , @xmath106\end{aligned}\ ] ] implies @xmath107(t ) = \beta_{\ell}[\hat{u}_{\ell}](t ) , \quad \forall t \in [ 0,\ , s],\end{aligned}\ ] ] almost everywhere , for every @xmath27.note that during each expanding construction ( i.e. , when a new subsystem is added to the existing distributed control system ) , we assume that both players play a differential game .
for example , for @xmath108 , the dynamics of the game is given by @xmath109 with an associated cost criterion @xmath110 and an exit - time @xmath111 such that _ player_-@xmath67 optimally picks a strategy ( in the sense of best - response correspondence ) to the stagey of _ player_-@xmath81 .
then , the game advances to the next stage , i.e. , @xmath112 , and continues until @xmath0 . ]
let us denote the set of all maximizing strategies by @xmath113 and the set of all minimizing strategies by @xmath114 .
furthermore , let us define the lower and the upper values of the differential game at the @xmath3th - stage by @xmath115 and @xmath116 for each @xmath117 , respectively .
moreover , if @xmath118 then the differential game has a value .
[ r3 ] note that the greatest payoff that _ player_-@xmath67 ( i.e. , the maximizing player ) can force is called a lower value of the game and , similarly , the least value that _ player_-@xmath81 ( i.e. , the minimizing player ) can force is termed an upper value of the game . in section [ s3 ] , we provide conditions under which these values coincide . in this subsection , we provide additional results that will be useful for proving our main results in section [ s3 ] . [ d1 ]
we define @xmath119 to be the set of all @xmath85-valued @xmath120-progressively measurable processes @xmath121 , that satisfies @xmath122 for each @xmath123 . [ l1 ] [ variational representation formula ( cf .
* proposition 2.5 or theorem 5.1 ) ) ] for a fixed @xmath124 , let @xmath125 , \mathbb{r}^m \bigr ) \rightarrow \mathbb{r}$ ] be any borel measurable bounded function
. then @xmath126 for any @xmath117 .
[ l2 ] for any @xmath127 , with @xmath117 , and @xmath124 , let @xmath128 be a measure induced on @xmath129 , \mathbb{r}^m\bigr)$ ] by @xmath130 under @xmath131 .
then , the relative entropy of @xmath128 with respect to @xmath131 satisfies the following @xmath132 let @xmath133 a polish space ( i.e. , a complete separable metric space ) , with a borel @xmath29-algebra , and let @xmath134 be the set of measures defined on @xmath133 that satisfies the usual hypotheses ( e.g. , see @xcite ) .
[ l3 ] [ cf .
* theorem 2.1 ) ] consider a sequence of measures @xmath135 in @xmath134 satisfying @xmath136 where @xmath137 .
let @xmath138 be a borel - measurable function .
then , the followings hold a. if @xmath139 weakly converges to another measure @xmath140 as @xmath141 , then @xmath142 b. if @xmath143 is a sequence of uniformly bounded functions that almost surely converges to @xmath144 , then @xmath145
in this subsection , we relate the lower and upper values of the associated differential game with the risk - sensitive escape control problem for the diffusion process . in particular
, using the variational representation ( e.g , see @xcite or @xcite ) , we present our main results , i.e. , proposition [ p1 ] and proposition [ p2 ] . for each fixed
admissible control @xmath56 , the following proposition ( which is a direct consequence of lemma [ l1 ] ) characterizes the risk - sensitive escape control problem ( cf .
equations and ) with an associated stochastic differential game ( cf .
equations below ) .
[ p1 ] suppose that , for a fixed @xmath117 , the admissible optimal controls @xmath51 , @xmath52 , for @xmath53 , are given .
consider any admissible control @xmath56 .
further , for every @xmath27 , let @xmath146 be a borel measurable function such that @xmath147 for @xmath148 $ ] , w.p.1 .
let the exit - time @xmath149 be given by @xmath150 which is associated with the following diffusion processes @xmath151 , i.e. , @xmath152 then , the following variational representation holds @xmath153 where the exit - time @xmath154 is given by @xmath155 which is associated with the following diffusion processes @xmath156 , i.e. , @xmath157 moreover , the admissible control @xmath158 satisfies @xmath159 , \quad \text{w.p.1 } , \label{eq23}\end{aligned}\ ] ] for any @xmath160 .
[ p2 ] for a fixed @xmath117 , let @xmath161 and @xmath162 be the lower and upper values of the associated differential game given in equations and .
suppose that @xmath51 , @xmath52 , for @xmath53 , are admissible optimal controls .
for any @xmath56 , let @xmath151 be the unique solution to equation .
then , a. the lower value of the game satisfies @xmath163 b. for a given @xmath164 , there exists a measurable function @xmath165 , \mathbb{r}^m\bigr ) \rightarrow m_{\ell}$ ] such that the upper value of the game satisfies @xmath166 with @xmath167 $ ] ( i.e. , when the maximizing player picks such a strategy ) , c. if the lower and upper values of the game coincides , i.e. , @xmath168 then the game has a value . in this subsection
, we briefly remark on the implication of our main results where one is also interested in evaluating the robust performance of the risk - sensitive escape control , when there is some norm - bounded modeling error in the distributed control system ( see @xcite for related discussion , but in different context ) . in
what follows , we assume that that the statements in propositions [ p1 ] and [ p2 ] are true .
suppose that , for a fixed @xmath117 , @xmath51 , @xmath52 , for @xmath169 , are the admissible optimal control strategies picked by _
further , we consider the following distributed control system ( which contains @xmath3 subsystems ) @xmath170 where @xmath171 .
define the value function @xmath172 as @xmath173 where @xmath174 , with @xmath175 .
then , for any @xmath176 , with @xmath177 , we have the following inequalities @xmath178 and @xmath179 suppose that @xmath180 , where @xmath181 is interpreted as a modeling error in equation .
further , assume that the value @xmath182 is used as a qualitative measure on the performance of the distributed control system . for a given specification @xmath183 , with @xmath184 , if there exists a design parameter @xmath185 such that @xmath186 then , we obtain an upper bound on the norm of the modeling error , which guarantees the desired performance against all modeling errors satisfying such a norm bound .
that is , if @xmath187 , \label{eq32}\end{aligned}\ ] ] then we have @xmath188 moreover , the above equation ( together with equation ) further implies the following @xmath189 for @xmath53 , with @xmath50 .
[ r4 ] note that , for each @xmath123 , the norm on the modeling error is inversely proportional to the design specification @xmath190 , and , therefore , the robustness of the distributed control system increases as the bound on the performance measure decreases .
befekadu gk , antsaklis pj ( 2014 ) on the minimum exit rate for a diffusion process pertaining to a chain of distributed control systems with random perturbations . http://arxiv.org/abs/1409.2751[arxiv:1408.6260 ] http://arxiv.org/archive/math.ct[[math.ct ] ] | in this paper , we consider an expanding construction of a distributed control system , which is obtained by adding a new subsystem one after the other , until all @xmath0 subsystems , where @xmath1 , are included in the distributed control system .
it is assumed that a small random perturbation enters only into the first subsystem and is then subsequently transmitted to the other subsystems .
moreover , for any @xmath2 , the distributed control system , compatible with the expanding construction , which is obtained from the first @xmath3 subsystems , satisfies an appropriate hrmander condition . as a result of this ,
the diffusion process is degenerate , i.e. , the backward operator associated with it is a degenerate parabolic equation .
our main interest here is to prevent the diffusion process ( that corresponds to a particular subsystem ) from leaving a given bounded open domain .
in particular , we consider a risk - sensitive version of the mean escape time criterion with respect to each of the subsystems .
using a variational representation , we characterize the risk - sensitive escape control for the diffusion process as the lower and upper values of an associated stochastic differential game .
finally , we comment on the implication of our results , where one is also interested in evaluating the performance of the risk - sensitive escape control , when there is some modeling error in the distributed control system . |
the milagro observatory@xcite has made long term observations of the cygnus arm .
they report an excess of over 5.5@xmath1 over a 5.9@xmath2 square bin in ra and dec.@xcite .
this excess is inconsistent with a point source and may be due to a giant molecular cloud(gmc ) located in the same region as the excess .
this cloud has been reported by dame et .
al.@xcite to be at a distance of 1.7 pc with a estimated mass of @xmath3 .
the angular extent of the cloud is 44 square degrees .
diffuse emission of @xmath0 rays at tev energies have long been speculated to be the result of cosmic ray interactions with giant molecular clouds@xcite . in this scenario , galactic cosmic rays
interact with hydrogen and produce neutral pions .
these pions quickly decay and produce @xmath0 rays .
predictions by aharonian and atoyan @xcite have indicated that the flux from these gmc should follow the galactic cosmic ray flux ( excluding enhancements by local sources ) and would be proportional to the gmc mass over the square of the distance to the gmc .
the cygx cloud is a good target since it is close and very massive .
the whipple 10 meter atmospheric cherenkov telescope utilizes the well proven imaging technique to reject cosmic ray background events and to determine source geometry@xcite .
this method uses the shape of the shower image ( fitted to an ellipse ) to determine if the shower was initiated by a @xmath0 primary or a cosmic ray primary .
additionally , if the source is assumed to be at the center of the field of view ( fov ) , the angle between the major axis of the ellipse and the line formed by the centroid of the image and the center of the fov(@xmath4 angle ) , can be used to eliminate events not coming from the source location .
the energy threshold for the whipple 10 meter is 390 gev for a crab like spectrum@xcite extensions of this method have been made to make observations for objects that may not be in the center of the fov .
this is often the case when searching for new sources , diffuse emission , or sources that have been identified by other experiments with relatively low angular resolution . in this two dimensional analysis @xcite , the source location
is geometrically constrained to lie along the major axis of the shower image ( as it the case with the one dimensional analysis ) , but no requirement is made of the @xmath4 angle with respect to the center of the camera .
the distance from the image centroid to the source location along the major axis is estimated using @xmath5 where the _ width _ refers to the size of the minor axis , _ length _ refers to the size of the major axis , _ d _ is the distance along the major axis , and @xmath6 is a scaling parameter that must be determined . to break the ambiguity as to which direction along the major axis the source lies , the skewness in the image is used .
the @xmath6 parameter was determined by examining the crab supernova remnant @xcite .
the two dimensional analysis was applied to on - source crab data . to optimize the @xmath6 parameter ,
the value of @xmath6 was varied in steps of @xmath7 .
the optimal value was determined by the maximum signal at the source location the optimal value was determined to be @xmath8 .
once the @xmath6 parameter has been determined the data can binned and the point spread function ( psf ) for the method can be determined . here
we have used a 0.36@xmath9 square bin in ra and dec .
this bin size was found to optimize the significance of the on source crab observations .
the binning of the data is shifted six times in ra and dec . in steps of 0.06@xmath2 in order to compensate for edge effects in the binning . applying this analysis to the on source crab data we get a maximum significance of 11.6@xmath1 from 5.6 hours of on source data ( 4.9@xmath1/@xmath10 ) .
the psf of the excess in ra and dec . is fit to a gaussian distribution with a @xmath11 = 0.18@xmath2 for points source off axis ( that is to say , not in the center of the field ) the psf becomes broader as the source moves further away from the center of the fov . while the radial spread to the psf stays roughly the same , the azimuthal spread increases slightly from 0.18@xmath2 to 0.21@xmath2 at one degree offset .
the behavior of the psf as function off offset was determined by analyzing crab data taken at 0.3 , 0.5 , 0.8 and 1.0 degree offsets from the center of the field .
data used in this work was taken during the months of august 2004 through november 2004 .
the observation window for this object is small as the whipple 10 meter generally suspends observations in the summer months due to poor weather conditions in southern arizona . in this analysis
we have used 12 on / off pairs of 28 minutes each .
the total number of events in the on / off field after shape cuts is 14406/14594 ( on / off ) .
the coordinates of the observations are ra = 20:40:7.9 ( 310.03@xmath2 ) and dec = 42:39:51.12 ( 42.66@xmath2 ) in j2000 coordinates .
these coordinates were chosen to overlap with the morphology of the milagro excess @xcite as well as overlap with large values of neutral hydrogen column densities in the region @xcite .
the above analysis fails to find strong evidence for a point source of @xmath0-rays within the 2-d fov of the observations .
figure 2 shows the excess map and sigma map from the field .
the significance was calculated using the standard li and ma method @xcite .
the most significant bin in the map ( figure 2 ) is located at ra=310.8@xmath2 and dec=41.3@xmath2 .
the pretrial significance is 3.8@xmath1 in this bin . to account for trials factors associated with the binning and the oversampling we simulated 30,000 data sets for this field .
we find the chance probability of getting one bin with a sigma of 3.8 or higher is 12% as no compelling point source was found within this field of view , we must conclude that the milagro source@xcite must be rather diffuse in this region , or must be at a flux level below the sensitivity of this analysis .
diffuse sources are rather difficult with the whipple 10 meter , particularly if size of the regions of emission is extends beyond the fov
. such emission would be difficult to detect with this technique , which is optimized for point - source detection .
upper limits have been calculated in the manner prescribed by helene@xcite .
once the upper limit is known , the values are adjusted to correct for the changing sensitivity across the fov .
this changing sensitivity has been determined by the analysis of offset crab data .
figure 3 shows the upper limits for each bin in ra and dec .
the field has been limited to the inner 1@xmath2 of the fov . beyond this
the sensitivity decreases such that it is difficult to set meaningful upper limits .
we find no point source within in the inner half degree of the fov above 3.0% , 3.3% , and 4.0% of the crab at the 90% , 95% and 99% confidence level . in the outer half degree
we find no point source above 15.0%,16.7% and 20% of the crab at the 90% , 95% , 99% confidence level .
these upper limits are obtained by comparing the total number of excess counts at a given offset ( within the 2@xmath1 ellipse of the psf ) with the expected excess counts from the crab at the same offset .
this method assumes a crab like spectrum and point source distribution .
further work is being done to convolute the psf with the expected source distribution , obtained from neutral hydrogen column density maps of the region .
this will allow us to set stronger upper limits for the region and may reveal a sources that the point source analysis failed to detect .
we are also continuing to collect data on this target and are also observing other fields within the region of the milagro excess .
this work supported by the us national science foundation grant ( # phy 0099580 ) , the u.s .
department of energy , the smithsonian institute , the nserc in canada , the science foundation of ireland and by pparc in the united kingdom . | the whipple 10 meter atmospheric cherenkov telescope has made observations of the region known as the cygnus arm .
this region has been recently reported by the milagro experiment to contain a diffuse tev @xmath0-ray source centered at ra=308 and dec=42 .
we report upper limits ( using the whipple 10 m telescope ) obtained during the fall 2004 observing season centered on ra=310 and dec=42.65 . |
frequently problems arise in science which involve both additive and multiplicative noise .
the first type is relatively easy to handle with the help of the central limit theorem .
the situation changes dramatically with the appearance of multiplicative noise .
famous examples are the anderson localization , turbulence , and the kicked quantum rotator among others . in this
field results of an importance comparable to the central limit theorem are still lacking .
moreover , the approaches are in general numerical ones and analytical tools are the rare exception .
we present such an analytic approach which permits to deal in great generality with processes involving multiplicative and additive noise even in the limit of strong disorder . in this paper
we apply the formalism to the famous anderson localization in a two - dimensional ( 2-d ) disordered system which is one of the paradigms of solid state theory .
the quantum mechanical consequences of disorder in solids have first been revealed by anderson @xcite .
the anderson model provides a standard framework for discussing the electronic properties of disordered systems , see reviews @xcite .
the nature of electronic states in the anderson model depends strongly on the spatial dimension @xmath0 .
it has been shown rigorously that in one dimension ( 1-d ) all states are localized at any level of disorder @xcite .
the shape of these localized wave functions is characterized by an asymptotic exponential decay described by the lyapunov exponent @xmath1 .
the most important results for dimensions higher than one follow from the famous scaling theory of localization @xcite , which assumes a single scaling parameter for the dimensionless conductance @xmath2 or , equivalently , the localization length @xmath3 .
the conclusion of the scaling theory is that for @xmath4 all states are localized at any level of disorder , while a delocalization ( metal - insulator ) transition occurs for @xmath5 if the disorder is sufficiently strong .
a detailed review of the scaling theory for disordered systems can be found in @xcite .
the 2-d case still presents a problem , since there is no exact analytical solution to the anderson problem , and all numerical results published so far rely on finite - size scaling @xcite .
recent studies @xcite have questioned the validity of the single parameter scaling theory , including the existence of a finite asymptotic localization length for @xmath6 .
additional boost of interest in the @xmath7 anderson model has been triggered by experimental observations of kravchenko et al .
@xcite of a metal - insulator transition in thin semiconductor films , which contradicts the conventional scaling theory .
moreover , recent experiments of ilani et al .
@xcite can be interpreted in terms of the coexistence of localized and delocalized states .
these experiments are still being discussed controversially .
the experimental reality is certainly more complex than the simple tight - binding schemes used in the theoretical treatment so far and in particular the electronic iteractions could play a role in the above mentioned experimental situations .
but nevertheless these results add doubts to the status of the localization theory in 2-d . before embarking on computational schemes beyond the tight - binding approach , which necessarily lead to more restricted system sizes and other approximations
, it appears advisable to try to solve as rigourously as possible the problem in the tight - binding scheme . in the present controversial situation
the first step in resolving the conflict is thus in our opinion to consider exact results that do not rely on the scaling theory or small parameter expansions .
the starting point for the method presented in this paper is found in the work of molinari @xcite , in which the anderson problem for the 1-d system is dealt with as a statistical stability problem for the solutions @xmath8 of the tight binding hamiltonian in a semi - infinite system , @xmath9 .
it was shown in ref .
@xcite that the equations for the statistical moments of the type @xmath10 can be obtained analytically ( explicit solutions are given for @xmath11 ) , which enabled the author to derive exact generalized lyapunov exponents .
we will show in the following that this approach can be further generalized for systems of higher spatial dimensions .
but it turns out to be unavoidable to change again the mathematical tools for the treatment . in the present investigation
we use both for the 1-d and the 2-d case the tool of signal theory abundantly used in electrical engineering , see e.g.@xcite .
the basic idea in applying signal theory to the problem of anderson localization is to interpret certain moments of the wave function as signals .
there is then in signal theory a qualitative difference between localized and extended states : the first ones correspond to unbounded signals and the latter ones to bounded signals . in the case of a metal - insulator transition extended states ( bounded signals )
transform into localized states ( unbounded signals ) .
signal theory shows that it is possible in this case to find a function ( the system function or filter ) , which is responsible for this transformation .
the advantage of working with filters instead of the signals themselves lies in the fact that the filters do not depend on initial conditions in contrast to the signals .
the existence of this transformation in a certain region of disorder and energy simply means that the filter looses its stability in this region .
the meaning of an unstable filter is defined by a specific pole diagram in the complex plane .
these poles also define a quantitative measure of localization .
thus it is possible here to determine the socalled generalized lyapunov exponents as a function of disorder and energy . the outline of the present article is as follows . in chapter 2
we treat the 1-d case in detail describing also essential elements of signal theory .
the theory for the 2-d problem is presented in chapter 3 and the results are given in chapter 4 . in the latter chapter also
the implications of the present approach for perturbation theory , the order of the phase transition and the problem of self - averaging is discussed as well as numerical scaling .
we start for pedagogical and methodical reasons with the treatment of the 1-d case .
the aim of this section is to apply mathematical tools which are new in the field of anderson localization but well - known from other fields and which may and do in fact prove useful also for the higher dimensional cases .
we start from the hamiltonian of the standard 1-d anderson model in the tight - binding representation @xmath12,\ ] ] where @xmath13 is the hopping matrix element ( units with t=1 are used below ) and the @xmath14 are the on - site potentials which are random variables to simulate the disorder .
the @xmath15 are independently and identically distributed with existing first two moments , @xmath16 and @xmath17 .
this model is well investigated @xcite and it is known that for any degree of disorder the eigenfunctions become localized .
this in turn means that @xmath18 with @xmath19 is growing exponentially in the mean on a semi - infinite chain where the rate of growth is described by the lyapunov exponent . in the one - dimensional case
one can compute the lyapunov exponent by different methods and under different assumptions @xcite or numerically by means of the transfer matrix method which results in solving a recurrence equation @xcite .
the disadvantage of these analytical methods is that they can not be easily extended to the 2-d and 3-d cases . in this paper
we want to show that one can construct an approach which is able to handle all cases from a universal point of view .
it is well - known that the two - point boundary problem for a finite difference analog to the stationary schrdinger equation with homogeneous boundary conditions can be reduced to a cauchy problem with the one - side ( initial ) condition @xcite @xmath20 due to the linearity of the equation the non - zero value of @xmath21 serves only as a normalization parameter .
the asymptotic behaviour of the solution is completely determined by the energy @xmath22 and the level of disorder @xmath23 .
the presence of a boundary permits one to rewrite the schrdinger equation @xmath24 in the form of a recursion relation for the grid amplitude of the wave function @xmath25 ( @xmath26 ) : @xmath27 statistical correlations in this equation are separated : it is easy to see that in a formal solution of the recursion relation the amplitude @xmath28 depends only on the random variables @xmath29 with @xmath30 , which is the causality principle
. this observation has been extensively used by molinari@xcite in the solution for the 1-d case .
we emphasize that this property is fundamental as a key for obtaining an exact solution to the anderson problem for all dimensions .
note that both amplitudes @xmath8 and @xmath31 on the r.h.s . of eq .
( [ recursion ] ) are statistically independent of @xmath32 . the equation for the first moment of the random variable @xmath8 does not include the parameter @xmath33 : @xmath34 the ansatz @xmath35 results in the @xmath36 which satisfy the equation @xmath37 .
energies with @xmath38 , where @xmath39 and @xmath40 , obviously coincide with the band of delocalized states in a perfect chain ( @xmath41 ) .
a bounded asymptotical behaviour of the first moment at any level of disorder indicates that there are always physical solutions inside the band @xmath38 .
further information about the character of these states ( localized / delocalized ) can be gained by considering the other moments . for one - dimensional models with random potential
the eigenstates are always exponentially localized @xcite .
the natural quantities to investigate is therefore the lyapunov exponent @xmath42 @xmath43 the generalized exponents have been studied extensively by pendry et al .
( see ref . @xcite and references given there ) , in a systematic approach based on the symmetric group .
the method involves the construction of a generalized transfer matrix by means of direct products of the transfer matrices , followed by a reduction of the matrix size .
this generalized transfer matrix produces the average values of the required power of the quantity under consideration .
the concept of lyapunov exponent @xmath1 and the corresponding localization length @xmath44 describe a statistical stability of solutions of the tight - binding equations .
anderson localization as a metal - insulator transition is a typical critical phenomenon . to determine the phase - diagram of the system we need only make use of the qualitative aspect of the lyapunov exponent .
two phases differ qualitatively : @xmath45 for conducting states ( metallic phase ) and @xmath46 for localized states ( insulating phase ) .
the lyapunov exponent @xmath1 is a typical order parameter @xcite .
it is useful to check whether this kind of relation is valid for other systems like ferromagnets and ferroelectrics too . here
the order parameters are the magnetization @xmath47 and the polarization @xmath48 , respectively . at high temperatures and zero external field
these values are @xmath49 ( paramagnetic or paraelectric phase ) . at lower temperatures , however ,
spontaneous magnetization and polarization arise , @xmath50 .
the order parameter is zero for one phase and becomes nonzero for the other phase .
it is well - known that there is no unambiguous definition of an order parameter @xcite .
this holds also in our case , many different definitions are possible ( see e.g. ref .
@xcite and literature cited there ) .
every exact lyapunov exponent either via the log - definition ( @xmath42 ) or via the @xmath51-definition ( @xmath52 ) gives this property .
this permits us to consider a transition from the quality @xmath45 to the quality @xmath46 as a critical point , where all moments @xmath53 diverge for @xmath54 simultaneously .
if on the other hand we are interested in the values themselves of the lyapunov exponents @xmath55 , they always depend on the definition .
if one prefers a particular definition , this can only be by agreement and always remains quite arbitrary . in the 1-d case
we find the metallic phase ( @xmath56 ) only for @xmath41 ; even for infinitesimally small disorder one has @xmath57 and only the insulating phase exists .
the critical point is therefore @xmath58 independently of the value of the energy @xmath22 , and the phase - diagram is trivial .
this situation is typical for a critical phenomenon in a 1-d system with short range interaction @xcite : all 1-d systems ( e.g. 1-d ising - model ) do not possess a true phase transition at a finite value of the parameter ( temperature in the ising model or disorder @xmath23 in the anderson model ) ; one of the phases exists only at a point .
one also says @xcite , that 1-d systems do not possess a phase transition .
the advantage of the new @xmath51-definition rests in the possibility to play with the parameter @xmath51 which gives a new degree of freedom . in quantum theory
there are many examples , where a problem is insoluble in the schrdinger picture , but looks rather simple in the heisenberg picture .
the same holds for representations in coordinate or momentum space .
it is always useful to transform the mathematics in such a way that the problem becomes soluble . and nothing else is done here . for @xmath42 ( log - definition )
the problem remains analytically insoluble .
molinari @xcite was the first to show that with a generalisation of the definition , and for special values of @xmath59 , a simple analytical ( algebraic ) investigation becomes possible .
molinari@xcite has not only considered a standard definition via `` the log of the wave function '' i.e. @xmath42 , but also a set of so - called generalized lyapunov exponents @xmath60 , @xmath61 ( @xmath62 for the square of the wave function ) . in general
all the @xmath60 are quite different parameters .
however , it is important that a transition from the quality @xmath45 to the quality @xmath46 is simultaneous , all @xmath63 for @xmath41 and @xmath64 for a disordered system .
as has already been established @xcite , in the limit of small disorder different definitions of the lyapunov exponent or localization length give values differing only by an integer factor .
this property is again a signature for a critical phenomenon . in the vicinity of the critical point @xmath58 as common for all critical phenomena and only one scale dominates .
thus molinari has shown that for the 1-d anderson problem all @xmath65 for @xmath66 ( this is proven for @xmath67 ) , where @xmath68 is taken from other investigations@xcite , the only difference being the numerical cofactor .
this means that even in 1-d there are other quantities besides the logarithm of the wave function which can be used for the analysis . in order to obtain the phase diagram
it is ( based on the above discussion ) sufficient to choose a particular and convenient value of the parameter @xmath51 , e.g. @xmath69 , the second moments .
it is shown in ref .
@xcite that the calculation of the higher moments , @xmath70 with @xmath71 , is important for determining the shape of the distribution of @xmath72 , but at the same time the higher moments of the on - site potentials beyond the second one must be considered .
we will restrict ourselves here to the pair moments only .
then the initial full stochastic problem eq .
( [ recursion ] ) can be mapped onto an exactly solvable algebraic problem , in which the random potentials are characterized by a single parameter @xmath23 .
two points should be mentioned : ( i ) we are at present not interested in the shape of the distribution which is influenced by the higher moments of the on - site potentials but in the problem of localization ( phase - diagram ) ; ( ii ) in the analysis of the moments of the amplitudes the localization of states finds its expression in the simultaneous divergence of the even moments for @xmath73 .
because the second moment depends only on the parameter @xmath23 this means that the critical properties are completely determined by @xmath23 .
we are interested in the mean behavior of @xmath74 which follows from eq .
( [ recursion ] ) as : @xmath75 ^ 2 \rangle \nonumber \\ = ( e^2 + \sigma^2 ) \langle \psi_n^2 \rangle - 2 e \langle \psi_n \psi_{n-1 } \rangle + \langle \psi_{n-1}^2 \rangle.\end{aligned}\ ] ] in the derivation of eq .
( [ secondmoment_e2 ] ) the mean of the product of uncorrelated quantities was replaced by the product of the means .
the resulting equation is open ended , but the new type of the means @xmath76 can be easily calculated : @xmath77 \psi_{n-1 } \rangle \nonumber \\&= & e \langle \psi_{n-1}^2 \rangle - \langle \psi_{n-1 } \psi_{n-2 } \rangle.\end{aligned}\ ] ] let us rewrite these equations using @xmath78 and @xmath79 : @xmath80 the initial conditions are : @xmath81 let us summarize the intermediate results up to this point .
the causality principle has led to a set of linear algebraic equations for the second moments of the random field instead of an infinite hierarchy of equations that couple the second moments with the third ones etc .
the set of equations is closed , but it does not include all possible second moments of the type @xmath82 ( this fact is irrelevant in the search for localization criteria ) . at this level
the on - site potentials are also characterized by a single second moment @xmath83 only , which implies that the shape of the distribution does not matter for localization .
information on higher moments is a crucial step for other models , e.g. , in turbulence and econophysics because the third and the fourth moments are linked to skewness and kurtosis of the distribution which are interesting features in these fields .
equations for the higher moments can be easily constructed using the same causality property@xcite , but in the treatment of the anderson problem we restrict ourselves to the second moment only .
the set of equations is exact in the sense that no additional approximations were made in its derivation .
therefore , we conclude that at the given point the stochastic part of the solution to the anderson problem is completed and one has to deal further only with a purely algebraic problem .
the set of equations ( [ relations ] ) , ( [ relations2 ] ) can be solved by different methods of linear algebra .
we employ first a simple matrix technique .
the eqs.([relations ] ) , ( [ relations2 ] ) can be rewritten in the form @xmath84 where the vector @xmath85 and the transfer matrix @xmath86 are respectively @xmath87 initial conditions transform into @xmath88 .
an explicit formula for @xmath85 is derived by diagonalizing the transfer matrix @xmath86 .
the characteristic equation of the eigenvalue problem for the @xmath86 matrix is @xmath89 with function @xmath90 .
the solution of the cubic characteristic equation ( [ chareq ] ) can be given explicitly in radicals ( cardan s formula ) .
corresponding expressions are known and not presented here .
it is significant that one of the eigenvalues ( let it be @xmath91 ) is always real and @xmath92 .
other solutions , @xmath93 and @xmath94 , are either complex conjugate to each other or both real and satisfy @xmath95 . these properties result from the fact that the coefficients @xmath96 and @xmath83 are non - negative . the solution of eq .
( [ matform ] ) reads as follows : @xmath97 where @xmath98 is a diagonal matrix containing the n - th power of @xmath99 and @xmath100 is the eigenvector matrix .
the resulting exact formula for @xmath101 is @xmath102 thus we have the full algebraic solution for any energy @xmath22 and any degree of disorder @xmath103 . from the functional form of the eq .
( [ x ] ) one can see that the roots of eq .
( [ chareq ] ) , @xmath99 , give us the lyapunov exponents @xmath1 of the problem , @xmath104 , so the judgement on a localization transition can be done immediately after arriving at eq .
( [ chareq ] ) .
electronic states satisfying the inequality @xmath105 correspond to localization .
an alternative solution makes use of the so - called z - transform .
this is used mainly in electrical engineering for discrete - time systems and we suggest as publicly available source @xcite .
the z - transform of the quantities @xmath101 and @xmath106 to functions @xmath107 , @xmath108 is defined by @xmath109 the inverse z - transform is quite generally defined via countour integrals in the complex plane @xmath110 we further on need the following properties : for the z - transform @xmath111 and for the inverse z - transform @xmath112 in this way eqs .
( [ xy ] ) and ( [ shift ] ) translate eqs .
( [ relations ] ) , ( [ relations2 ] ) into a system of two coupled linear equations for the two unknowns @xmath113 and @xmath114
@xmath115 this is easily solved for @xmath107 @xmath116 where the function @xmath117 is defined above .
the inverse z - transform gives us again eq.([x ] ) .
it is easy to see that the solution of the characteristic equation ( [ chareq ] ) for the transfer matrix is equivalent to the determination of the poles of the @xmath107 function .
the eigenvalues @xmath99 , i.e. the poles , determine according to eq . (
[ lambda ] ) the asymptotic behaviour of the solution @xmath101 for @xmath118 .
( this is a simplified example for the general relation ) .
let us start with the following definitions .
let @xmath119 describe an ideal system ( @xmath120 for @xmath121 ) .
this function is independent of the parameter @xmath23 . for @xmath122 @xmath123 with @xmath124}{\mathcal{d}(z)}.\end{aligned}\ ] ]
note that the boundary conditions ( parameter @xmath21 ) influence only the function @xmath125 .
the function @xmath126 for @xmath41 .
( [ xh ] ) possesses quite a remarkable structure which is better interpreted in the context of signal theory @xcite , which makes intensive use of the z - transform .
let us define the system input as @xmath125 ( it characterizes the ideal system ) , the system output as @xmath107 ( the disordered system ) , then the function @xmath127 is the system function or filter .
the inverse z - transform gives @xcite
signal theory is not very crucial for the @xmath129 case , because the solution , eq.([xz ] ) , looks very simple and the inverse z - transform is always possible . for @xmath130
, however , the use of signal theory is exceedingly important because the corresponding solution @xmath107 can be a very complicated function and an inverse transform may be impossible to find . in the present case it is ,
however , completely sufficient to investigate the analytic properties of the filter function @xmath127 , i.e. the poles @xmath131 , because the poles determine uniquely the properties of the system .
the essential idea is very simple . in the band
@xmath132 all wave amplitudes ( and input signals @xmath133 ) are bounded .
output signals @xmath101 are unbounded only under the condition that the filter @xmath134 is unbounded for @xmath54 .
this property depends on the position of the poles @xmath99 in the function @xmath127 .
it is known @xcite that the filter @xmath127 can be characterized by a pole - zero diagram which is a plot of the locations of the poles @xmath99 and zeros in the complex z - plane . since the signals @xmath133 and @xmath101 are real
, @xmath127 will have poles and zeros that are either on the real axis , or come in conjugate pairs .
for the inverse z - transform @xmath135 one has to know the region of convergence ( roc ) . as follows from physical reasons we are interested only in causal filters ( @xmath136 for @xmath137 ) that have always rocs outside a circle that intersects the pole with @xmath138 .
a causal filter is stable ( bounded input yields a bounded output ) if the unit circle @xmath139 is in the roc . to give an example .
we start the analysis of the solution with the case of the ideal system , @xmath41 .
all solutions in the band @xmath140 ( defined as a region with asymptotically finite first moment ) are delocalized with @xmath141 , @xmath142 , where @xmath143 .
the inverse z - transform gives us @xmath144 . in this case the filter @xmath126 and the roc of this filter is the full commplex z - plane .
the roc includes the unit circle , the filter @xmath127 is thus stable , which means the delocalization of all states . for @xmath145
we always have @xmath146 .
the roc corresponds to the region @xmath147 , and the unit circle @xmath139 lies outside the roc .
a filter @xmath127 is unstable , in other words , this is simply the localization of all states .
consider a 2-d lattice with one boundary .
the layers of this system are enumerated by an index @xmath148 starting from the boundary , and the position of an arbitrary lattice site in a particular layer is characterized by an integer @xmath149 .
the presence of a boundary permits one to rewrite the schrdinger equation @xmath150 in the form of a recursion relation for the grid amplitude of the wave function @xmath151 ( @xmath152 ) : @xmath153 for the sake of a compact notation an operator @xmath154 is introduced which acts on the index @xmath155 according to the equation @xmath156 the lattice constant and the hopping matrix element are set equal to unity .
the on - site potentials @xmath157 are independently and identically distributed with existing first two moments , @xmath158 and @xmath159 .
eq.([recursion d ] ) is solved with an initial condition @xmath160 it turns out to be convenient to consider the index @xmath161 not as a spatial coordinate , but as discrete time
. then eq .
( [ recursion d ] ) describes the time evolution of a @xmath162 dimensional system .
it is easy to see that in a formal solution of this recursion relation the amplitude @xmath163 depends only on the random variables @xmath164 with @xmath165 ( causality )
. we encounter a very important feature in eq .
( [ recursion d ] ) : grid amplitudes on the r.h.s .
are statistically independent with respect to @xmath166 .
we are going to generalize further the result of ref .
@xcite that the set of equations for a certain combination of pair moments is self - contained and can be solved analytically .
the divergence of the moments , which is caused by the localization , is the basis for the existence of lyapunov exponents for the @xmath161-direction , which is generally a functional @xmath167 $ ] of the field @xmath168 .
the well - known idea to define the fundamental lyapunov exponent for the problem of anderson localization as a minimal one @xmath169 \}$ ] , is an algorithm but not a general definition , because a fundamental quantity is independent of the initial condition .
the proper definition is possible in the framework of signal theory @xcite .
we define the solutions of the equations for the second moments with disorder ( @xmath170 and without it ( @xmath41 ) as @xmath171 ( system output ) and @xmath172 ( system input ) . because these equations are linear
, there exist an abstract linear operator @xmath173 ( system function or filter ) , which transforms one solution into the other one , @xmath174 .
it is important that the initial conditions @xmath168 only determine the signals , the filter on the other hand is a function of the disorder @xmath23 only .
divergence of the moments ( unbounded output for bounded input ) simply means that the filter is unstable@xcite .
this approach utilizing the concept of the system function is a general and abstract description of the problem of localization . instead of analyzing the signals
@xmath171 which have restricted physical meaning in the present context because of the chosen normalization we study the filter @xmath173 with properties described by generalized lyapunov exponents
. then e.g. delocalized states ( bounded output ) are obtained by transforming the physical solutions inside the band @xmath175 ( bounded input ) provided that the filter @xmath173 is stable .
the transformation @xmath174 is not only valid for individual signals , but also for linear combinations of these signals .
let us regard an ensemble of initial conditions in eq .
( [ rand d ] ) which is obtained by trivial translation in @xmath155-space , @xmath176 .
translation generates _ physically _ equivalent signals with identical lyapunov exponents @xmath167 $ ] .
a linear combination of these signals also has this same value @xmath167 $ ] .
we construct a linear combination from all such signals with equal weights ( this corresponds simply to an average @xmath177 over all possible translations @xmath178 ) .
we here start from the basic fact that the determination of the phase - diagram ( the fundamental topic of the paper ) requires only the lyapunov exponents .
signals @xmath171 are only a means to arrive there .
consequently it is possible to make certain operations with the signals , but under the strict condition that these operations have no influence on the lyapunov exponents @xmath167 $ ] .
next we define a full averaging over random potentials and over the ensemble of translations in @xmath155-space .
this latter averaging correponds to the construction of the linear combinations discussed above .
full averaging restores the translational invariance along the @xmath179axis , which appears e.g. in @xmath179independent diagonal elements , @xmath180 , in the set of the second moments of the type @xmath181 and @xmath182 .
we further on regard @xmath101 as a one - dimensional signal .
if we succeed to solve the equations for @xmath101 then we also find not only the lyapunov exponent @xmath167 $ ] but simultaneously the projection of the abstract operator on the one - dimensional space , @xmath183 , because for one - dimensional signals the convolution property eq.([x_n ] ) exists @xcite .
the filter @xmath134 possesses the same fundamental information as the abstract filter @xmath173 and can easily be analyzed , because signal theory provides a definite mathematical language for this aim ( see above ) .
we emphasize here that we do not reduce the problem to a one - dimensional one ; it remains two - dimensional .
this is quite apparent from the equations below which contain two spatial variables , @xmath161 and @xmath184 , the distance along the m - axis . after the full averaging as defined above the moment @xmath185 ( where @xmath186 ) transforms the initial condition by replacing the field @xmath187 by its property correlation function @xmath188 ( it is assumed that the field itself and its correlation function are finite ) .
the initial condition for @xmath101 reads as @xmath189 .
non - diagonal elements , @xmath190 , depend on the difference @xmath191 . in the following
we will need only three types of moments @xmath192 with @xmath193 , @xmath194 .
let us denote them by @xmath195 , @xmath196 , and @xmath197 : @xmath198 the corresponding initial conditions are @xmath199 ; @xmath200 ; and @xmath201 .
since the definition of @xmath202 for @xmath203 coincides with @xmath101 , we have the boundary condition @xmath204 the moments @xmath205 can be calculated directly from the definition , eqs.([ma])-([mc ] ) .
let us consider e.g. @xmath206 . in order to calculate the average , the first factor , @xmath163 or @xmath207 ,
is expressed from the main relation eq .
( [ recursion d ] ) , while the second factor , @xmath208 or @xmath209 , is left unchanged .
analogously , the equation for the moments @xmath202 and @xmath210 are obtained .
we have ( @xmath211 ) @xmath212 the operator @xmath154 is introduced in eq.([l ] ) which acts on the index @xmath213 according to the equation @xmath214 in the derivation of eqs .
( [ ma1])-([mc1 ] ) the mean of the product of uncorrelated quantities was replaced by the product of the means .
the basic equation for the variable @xmath101 is obtained by squaring both sides of eq .
( [ recursion d ] ) and averaging over the ensemble .
one gets @xmath215 the expression for @xmath216 includes the seconds moments of the type @xmath217,@xmath218 introduced earlier ( moments @xmath219 are obviously absent ) . in the following derivations we utilize two types of algebraic transforms : the z - transform @xcite @xmath220 and the discrete fourier
transform @xmath221 z - transform of the eqs .
( [ ma1])-([mc1 ] ) gives @xmath222 for boundary condition , eq.([boundary ] ) , we have @xmath223 after simplification of the equation for the moment @xmath224 we have eq.([mb2 ] ) and new equation for moment @xmath225 : @xmath226 it turns out to be convenient to lift the constraint @xmath227 in the form : @xmath228 this equation requires expressing the parameter @xmath229 in a self - consistent way via the boundary condition , eq.([a0 ] ) . after an additional fourier
transform one gets @xmath230 here @xmath231 is the fourier transform of the operator @xmath232 .
z - transform of the basic equation ( [ basic ] ) gives @xmath233 the function @xmath234 can be represented with the help of the fourier transform @xmath235 after simplification of eq.([chi1 ] ) ( we use here eqs.([ma5]),([mb5 ] ) ) one gets @xmath236 or @xmath237 because for the boundary condition , eq.([a0 ] ) , we have @xmath238 > from eq.([basic z ] ) together with eq.([chi3 ] ) we obtain @xmath239 it is apparent , that the moment @xmath240 as a solution of the eqs .
( [ ma5]),([mb5 ] ) depends on the parameter @xmath229 .
the parameter @xmath229 in turn is determined by @xmath113 according to eq .
( [ recurs3 ] ) .
one gets @xmath241 the equation for the signal @xmath113 is obtained in a self - consistent way from eq.([iden ] ) .
then we finally obtain eq.([xh ] ) , where @xmath242 note that the boundary conditions ( field @xmath168 or correlation function @xmath188 ) influence only the function @xmath125 which is independent of the parameter @xmath23 and describes an ideal system , @xmath126 for @xmath41 .
this point needs some comments .
we have already stated that the initial conditions @xmath168 only determine the signals , the filter is a fundamental function of the disorder @xmath23 only .
this fact has a very simple and important logical consequence .
a filter @xmath127 is defined via the relation between input and output signals , eq.([xh ] ) .
it is therefore completely sufficient to determine only once this relation e.g. for a particular boundary condition , @xmath168 , where the calculation of input and output signals is trivial .
the simplest case is @xmath243 . for this condition
our system has ( after averaging over random potentials ) the translational invariance along the @xmath179axis and we need not do a further averaging over the ensemble of translations in @xmath155-space . for @xmath244
we get back to eq.([h ] ) .
the corresponding input signal @xmath125 follows from eq.([x0 ] ) , if one makes use of the simple relation @xmath245 .
it is appropriate to return to the topic of averaging over the ensemble of translations in @xmath155-space and the full averaging , section [ signal ] .
we clearly see now that this procedure is not at all obligatory .
one could avoid it altogether .
we have , however , used this procedure for pedagogical reasons to demonstrate clearly that the mentioned property of the filter ( the filter is a function of the disorder @xmath23 only ) really exists and is independent of the initial conditions .
for the sake of illustration we restrict ourselves first to the case of the band center @xmath246 . evaluating integrals by standard methods ( contour integrals )
one gets : @xmath247 where generally the complex parameter @xmath248 is defined in the upper half - plane , @xmath249 .
changing the complex variable @xmath250 to the parameter @xmath251 corresponds to the conformal mapping of the inner part ( @xmath252 , transform @xmath253 ) or the outer part ( @xmath254 , transform @xmath255 ) of the circle onto the upper half - plane , the circle itself maps onto the interval @xmath256 $ ] .
note also that if @xmath127 has complex conjugate poles , then on the upper @xmath251 half - plane they differ only by the sign of @xmath257 . to avoid complicated notations , we seek for poles in the sector @xmath258 and double their number if we find any .
the inverse function @xmath259 is double - valued , it has two single - valued branches that map the selected @xmath251 sector onto either the inner part of the half - circle ( @xmath252 , @xmath260 sign in the formula ) or the half - plane with the half - circle excluded ( @xmath254 , @xmath261 sign in the formula ) .
it is also easy to derive that @xmath262 substituting ( [ int ] ) and ( [ gpm ] ) into ( [ h ] ) gives the system function @xmath263 or @xmath264 we see that the filter @xmath127 is a non - analytic function of the complex variable @xmath250 .
the unit circle @xmath139 divides the complex plane into two analytic domains : the interior and exterior of the unit circle .
the inverse z - transform is quite generally defined via countour integrals in the complex plane , eq.([count ] ) , and this definition is only possible in an analytic domain . in this way in the formal analysis of the problem
multiple solutions result .
let us consider first the solution @xmath265 which is formally defined in the region @xmath254 .
the function @xmath265 has two poles @xmath266 and @xmath267 , where @xmath104 , @xmath268 and @xmath269 the first pole lies inside the region of definition and the second one is located outside of it ( virtual pole ) .
however , for the inverse z - transform this fact is irrelevant .
note also that in the case of general @xmath22 the pole @xmath91 which lies on the real axis and which can be found from the parametric representation , eq.([zw ] ) , in the @xmath251 sector defined earlier , has its virtual counterpart @xmath270 .
for @xmath271 the roc for a causal filter is given by the inequality @xmath272 .
hence , the unit circle does not belong to the roc , and therefore the filter @xmath273 is unstable .
the inverse z - transform gives @xmath274 which is an exponentially growing function .
this result can be generalized for all @xmath22 values , however the expression for the function @xmath273 will be more complex .
therefore , the solution given by the system function @xmath265 always gives unbounded sequences @xmath101 , in full analogy with the solution of the one - dimensional problem @xcite .
the natural interpretation of this result is that states are localized .
the parameter @xmath1 is nothing else but the generalized lyapunov exponent @xcite , defined for the second moment of the random amplitudes .
therefore , the localization length is @xmath275 .
the case of the filter @xmath276 is a little more complicated .
the filter is formally defined in the region @xmath252 . for @xmath246 and @xmath277
the poles are found at @xmath278 and @xmath279 ( virtual pole ) with @xmath280 , @xmath281 .
a roc which is consistent with the causality restriction corresponds to the inequality @xmath282 and this lies outside the region of definition of the solution @xmath252 .
therefore , any physically feasible solution is absent .
however , for @xmath283 the poles lie on the unit circle @xmath284 , @xmath285 .
the critical value @xmath286 corresponds to the equation @xmath287 let us consider this problem as a limiting case of a modified problem in which the poles are shifted into the unit circle , @xmath288 and @xmath289 .
the casual filter has roc @xmath290 , however , the poles move onto the unit circle . in the literature on electrical engineering @xcite the filter that has a pole on the unit circle , but none outside it is called marginally stable
. marginally stable means there is a bounded input signal @xmath133 that will cause the output @xmath101 to oscillate forever .
the oscillations will not grow or decrease in amplitude .
the inverse z - transform gives @xmath291 i.e. a bounded oscillating function . evaluating the required integrals
one gets @xmath293 a similar analysis as above shows that the marginally stable filter exists for @xmath294 , where @xmath295 and the value of the energy is limited by @xmath296 . for either @xmath297 ( @xmath145 ) or @xmath298 ( @xmath299 ) a physical solution of this filter
is absent .
the critical value of @xmath300 follows again from eq.([hw ] ) .
we conclude that the system function @xmath276 exists in a well - defined region of energies @xmath22 and disorder @xmath23 .
it always gives bounded sequences @xmath101 which can be naturally interpreted as delocalized states .
marginal stability corresponds to the existence of quite irregular wave functions . the first solution @xmath273 is defined outside the unit circle and always exists .
the filter @xmath273 describes localized states and it is possible to connect its properties with the notion of the localization length . in the energy range @xmath301
the lyapunov exponent is a non - analytical function at zero disorder : @xmath302 , because without disorder all states here are extended ones , @xmath303 .
the second solution @xmath276 is defined inside the unit circle and does not always represent a solution which can be physically interpreted ( this is the mathematical consequence that the filter be causal ) .
the filter @xmath276 describes delocalized states .
we also note that if both solutions @xmath304 and @xmath305 for the system function exist simultaneously , they both give solutions @xmath101 of the initial problem that satisfy all boundary conditions . in this sense ,
a general solution of the problem is @xmath306 , where the parameter @xmath307 is left undefined by the averaging procedure ( this is natural for the problem with particular solutions of different asymptotic behavior ) .
such a solution @xmath134 may be interpreted as representing two phases , considering that @xmath304 determines the properties of the insulating state , but @xmath305 the metallic state .
therefore the metal - insulator transition should be looked at from the basis of first - order phase transition theory .
this opinion differs from the traditional point of view , which considers this transition as continuous ( second - order ) . for first order phase transitions
the coexistence of phases is a general property .
there exists for the present case an experimental result which is at least consistent with the present non - trivial result .
ilani et al.@xcite have studied the spatial structure at the metal - insulator transition in two dimensions .
they found @xcite : the measurement show that as we approach the transition from the metallic side , a new phase emerges that consists of weakly coupled fragments of the two - dimensional system .
these fragments consist of localized charge that coexists with the surrounding metallic phase . as the density is lowered into the insulating phase , the number of fragments increases on account of the disapearing metallic phase. the results of the present investigation - if it proves to be correct - have certain consequences for the validity of many theoretical tools used for the investigation of anderson localization in the past .
our comments do not exclude the possibility that better theroretical tools in the traditional approaches can be found , which may better master the situation .
our comments in the subsequent paragraphs are intended to highlight the differences between the present and previous approaches and the basic problems encountered by previous approaches - all this under the premise that the present analytical theory is proven to be correct .
our first comment refers to the anderson localization as a critical phenomenon .
one has always believed that all states in a 1-d and 2-d systems are localized for infinitesimal disorder , whereas in 3-d a metal - insulator transition occurs . from our studies it , however , emerges that a metal - insulator transition occurs in 2-d system . because localization as a metal - insulator transition is a typical critical phenomenon , the parameters @xmath1 and @xmath44 are quite generally not analytical functions neither in the energy nor in the disorder parameter .
it is well - known that perturbation theory is not a suitable method for describing critical phenomena .
e.g. for @xmath246 and @xmath66 one has from eq.([gam ] ) @xmath308 . from this
follows that the function @xmath1 can not be represented as a series in powers of @xmath83 .
i.e. perturbation theory is not applicable to the anderson problem in 2-d ( as it is neither for other critical phenomena @xcite ) : the corresponding series expansions tend to diverge .
this should be quite generally valid ; that is why all estimations which result from first order perturbation theory ( e.g. for the mean free path ) are physically extremely doubtful , because they are the first term of a divergent series . in the case of 1-d systems
it is known@xcite that in the limit of small disorder all lyapunov exponents @xmath309 . here
perturbation theory is completely acceptable , because 1-d systems do not show a phase transition @xcite .
these results of perturbation theory tend , however , to become inapplicable for higher spatial dimensions . in a similar vein
it can be stated that results for the 1-d ising - model have no relevance for the 2-d model ( onsager solution ) @xcite .
physics of disorder associates experimental quantities with quantities obtained by averaging over random potentials ( statistical ensemble of macroscopically different systems ) .
this approach is well - known @xcite .
one expects that , although the results of measurement of this physical quantity are dependent on the realization of disorder , the statistical fluctuations of the result are small .
one assumes that physical quantities are only those which do not fluctuate within the statistical ensemble in the thermodynamic limit ( length of system @xmath310 ) .
one defines these as self - averaging quantities .
it is also known that in connection with localization this self - averaging property is not trivially fulfilled , e.g. the transport properties of the disordered systems are in general not self - averaging@xcite . here
the fluctuations are much larger than expected or they even diverge in the thermodynamic limit .
even though the existence of non - self - averaging quantities is known , the theory of the physics of disorder starts from the assumption that certain quantities will still be self - averaging and physical .
however , one has to the authors knowledge never analyzed the conditions for the validity of this statement from the point of view of phase transition theory . in a system without phase transitions one can prove the existence of certain self - averaging quantities .
their existence is also possible if there is a second order phase transition . here in the thermodynamic limit
there exists only one or the other phase , a coexistence of phases is impossible by principle . in each phase there
do exist certain average values which have physical relevance .
strong fluctuations are observed only as usual in the vicinity of a critical point . in a system with a first order phase transition one
finds certain parameter values for which two phases do coexist ; i.e. a macroscopic system is heterogeneous and consists of macroscopically homogeneous domains of the two phases . in this case a formal averaging over the statistical ensemble takes into consideration also an averaging over the phases , and the resulting averages have no physical meaning .
physically meaningful are only the properties of the pure homogeneous phases .
the trivial example in this respect is the coexistence of water and ice .
an average density of this system has no sense and depends on the relative fraction of the phases , whereas the density of pure water and pure ice are meaningful quantities .
formally systems with a first order phase transition do not possess self - averaging quantities .
the well - known idea @xcite , to analyze such non - self - averaging quantities via probability distribution functions or all of its moments , can not be realized in practice .
in the theory of phase transitions @xcite one has a general idea to treat such multi - component systems , but these exist only as approximations .
one writes down the equations for certain averages . if these equations possess a multiplicity of solutions one interprets the corresponding solutions as phases .
let us now consider the present solution of the anderson problem from this point of view .
it is well - known that the exact equations of statistical physics are always linear but form an infinite chain ( e.g. see the equations which form the basis of the bbgky theory ( bogoliubov , born , green , kirkwood , yvonne ) @xcite . via decoupling approximations which use multiplicative forms of distribution functions
( see e.g. the kirkwood approximation @xcite ) one derives from the insoluble infinite chain a finite set of equations which , however , are now nonlinear .
this latter property leads to the multiplicity of solutions and the possibility to describe the phase transition .
it remains , however , unclear in which way the original linear equations describe mathematically the phase transition .
the anderson localization problem is the extremely rare case where an analytic solution was found for the chain of equations . as a result one
can clearly see how mathematics produces a multiplicity of solutions in the form of non - analytic filter functions @xmath127 .
the z - transform lifts the phase degeneracy allowing one to study the properties of each phase independently .
particular solutions @xmath304 and @xmath305 can be naturally interpreted not as the result of full averaging over all realizations of random potentials , but as averaged over one of the two classes of filtered realizations . if a particular realization of random potentials for a given energy @xmath22 leads to unbounded sequences for all boundary conditions , we classify it as belonging to the first class .
otherwise the realization of random potentials belongs to the second class .
sorting of the realizations into two classes corresponds mathematically to an infinitely weak constraint on the on - site potentials , which are still characterized in the equations by a single parameter @xmath23 .
therefore the equations for the second moments given above will describe both one - phase solutions with full averaging ( if the metallic phase is impossible ) and each phase of the two - phase solution separately with partial averaging . for known technical reasons one treats d - dimensional systems as quasi - one - dimensional ones @xcite ; this is e.g. done for the recursive calculation of the lyapunov exponents .
one starts with a system of the size @xmath311 : it is infinite in the @xmath312direction and the length @xmath313 in the other directions is finite .
numerical scaling studies @xcite assume the scaling variable to be self - averaging .
a further crucial assumption is that there is one - parameter scale function .
one calculates the localization length @xmath314 as the average of the logarithm of the quantum mechanical transmission probability @xcite , and assumes that this quantity is self - averaging . for any finite length @xmath313
the localization length @xmath314 is finite too .
this fact is rather trivial .
every quasi - one - dimensional system is qualitatively similar to a true one - dimensional one .
because in the 1-d anderson model all states are localized , the same will hold for quasi - one - dimension systems .
it is known that a rigorous theoretical definition of the phases requires the investigation of the thermodynamic limit @xmath310 .
one assumes that in order to be extrapolate to infinite system size it is necessary to investigate the scaling behaviour of @xmath314@xcite .
it is possible to establish a scaling function @xmath315 that does not depend on the disorder .
the scaling parameter @xmath316 is a function of disorder .
this method yields complete localization in 2-d dimension , and an anderson transition in 3-d case . is the scaling parameter necessary to scale numerical data onto the same curve .
values of the disorder @xmath23 are indicated .
, width=302 ] one notices that the idea of numerical scaling has a weak point ; one starts from the assumption that self - averaging quantities exist .
the analytic solution presented in the present paper , section [ 2d ] , offers in principle the possibility to clarify the problems connected with the numerical scaling procedure .
the analytic equations treat from the very start a system in the thermodynamic limit ( @xmath317 ) , because the integer @xmath318 in eq.([schroedinger ] ) . a transition to a finite system ( @xmath319 ) in this context is rather trivial and corresponds simply to the transition from the fourier integral , eq.([ft ] ) to the fourier series .
thus we arrive from the filter - function in the form of eq.([h ] ) at another equation : @xmath320
the difference is large .
the filter ( [ h ] ) has a multiplicity of solutions : the unit circle @xmath139 divides the complex plane into two analytic domains . for a finite system one has only one solution , the unit circle @xmath139 does not play a role anymore .
the filter ( [ hl ] ) corresponds to only one phase , where all states are localized .
mathematically this means that between the possible poles @xmath131 of the function ( [ hl ] ) there always exist the pole @xmath321 , and the filter is unstable . if one defines the localization length as @xmath322 , where according to the definition by using the second moments @xmath323 , then this length is always finite .
the physical reason for this result is rather clear : in the @xmath161-direction the system remains infinite since the @xmath324-transform always `` feels '' the true asymptotic behavior .
fixing a finite size @xmath313 in the @xmath155-direction effectively renders the system a one - dimensional one at large scales , hence the corresponding behaviour ( full localization ) . from this point of view
, @xmath314 is essentially the crossover length form 2-d to 1-d behavior .
let us do now a typical scaling study for the localization length @xmath314 at the band centre , @xmath246 .
we have found that the scaling parameter @xmath316 is identical with the value @xmath44 , where @xmath325 and @xmath1 is defined in eq.([gam ] ) .
disregarding for the moment the fact that one commonly uses a different definition of the localization length @xmath314 via the log - definition , we arrive qualitatively at the same results as numerical scaling@xcite .
there exists a scaling function , fig.1 , which is typical for 2-d systems and whose behaviour one commonly interprets as complete localization in 2-d dimensions .
the corresponding scaling parameter is , however , identical with the localization length @xmath44 in the insulating phase .
i.e. numerical scaling is not capable to analyze a system consisting of two phases .
this approach to the problem with particular solutions of different asymptotic behaviour , @xmath326 or @xmath46 , always sees only one diverging solution ( @xmath46 ) .
the basic idea of the present work is to apply signal theory to the anderson localization problem .
we interpret certain moments of the wave function as signals .
there is then in signal theory a qualitative difference between localized and extended states : the first ones correspond to unbounded signals and the latter ones to bounded signals . in the case of a metal - insulator transition extended states ( bounded signals )
transform into localized states ( unbounded signals ) .
signal theory shows that it is possible in this case to find a function ( the system function or filter ) , which is responsible for this transformation .
the existence of this transformation in a certain region of disorder and energy simply means that the filter looses its stability in this region .
the meaning of an unstable filter is defined by a specific pole diagram in the complex plane .
these poles also define a quantitative measure of localization . thus it is possible here to determine the socalled generalized lyapunov exponents as a function of disorder and energy . for pedagogical reasons
we have analyzed in this paper first the 1-d case . here
no new results are obtained .
all states are localized for arbitrary disorder .
the aim of this section consists in showing to the uninitiated reader a possible alternative to the traditional mathematical tools of dealing with anderson localization and to interpret its content .
the power of the approach comes only to its full bearing when it proves possible to find analytically the filter function also for the 2-d case .
although the filter is quite in general defined by an integral , it is possible to give a representation in terms of radicals . as a consequence we have been able to find an exact analytical solution for the generalized lyapunov exponents for the well - known and notorious anderson problem in 2-d . in this way
the phase diagram is obtained .
the approach suggested in the present paper does not in any way represent the complete solution to the problem of anderson localization in 2-d .
the very important topic of the transport properties of disordered systems lies outside the presumed region of applicability of the method .
the results of the studies give us a few answers to existing questions but open up even more questions .
we have only considered the phase diagram of the system .
the theory permits us to ascertain to which class of problems anderson localization belongs . in the 1-d case
all states are localized for infinitesimally small disorder in complete agreement with the theoretical treatment of molinari @xcite . in the 2-d case
we have shown that in principle there is the possibility that the phase of delocalized states exists for a non - interacting electron system .
this phase has its proper existence region and it belongs to the marginal type of stability .
all states with energies @xmath327 are localized at arbitrarily weak disorder . for energies and disorder , where extended states may exist we find a coexistence of these localized and extended states .
thus the anderson transition should be regarded as a first order phase transition .
the argument of mott that extended and localized states can not be found at the same energy is probably not applicable in our case as the solutions for the two relevant filter functions do not mix .
this has to be worked out in detail in the future .
also the role of different disorder realizations has to be elaborated , which could be a source of the observed phenomena .
our findings can in principle explain the experimental results of kravchenko et al .
@xcite on films of a - si .
they are also in agreement with the experimental findings of ilani et al @xcite which showed the coexistence of phases .
although one has to be cautious in concluding from our results directly onto specific experimental situations our theory is the first one to be in accord with experiment for 2-d systems .
the qualitative diagnosis which we propose here is in several aspects different from the predictions of prevailing theories , scaling theory , mean field and perturbation theory .
it has to be stated here that much work still remains to be done both via the present approach as via other established approaches . 1
. the anderson problem belongs to the class of critical phenomena .
it is known from the scientific literature that in such a case one encounters in general a poor convergence behaviour of approximations @xcite , which in their basis contain a mean field approach . to cite an example : the exact solution of onsager for the 2-d ising model is in complete disagreement with the landau theory of phase transitions ( mean field theory ) @xcite .
all critical exponents differ in the two approaches .
yet there is at least a qualitative agreement between the two theories : the landau theory describes second order phase transitions and the exact onsager theory confirms this aspect .
2 . for the anderson problem the outlook on the theories
might also be far reaching , if our approach should prove to be correct .
all presently accepted theories derive here from the scaling idea which is in turn based on the assumption that the anderson transition belongs to the class of second order phase transitions .
this basic idea is up to now a hypothesis which can not be proved in the absence of the exact solution .
one has assumed that averaged or self - averaging quantities are always physical quantities .
and every theory had the direct aim to calculate such quantities .
our exact solution for the phase diagram , however , indicates that in this system the coexistence of phases is possible , i.e. that the anderson transition should be regarded as a phase transition of first order .
3 . for a phase transition of first order with a coexistence of phases there surely also exist self - averaging quantities , whose fluctuations are small .
the quantities , however , tend to be unphysical because the average over an ensemble includes an average over the phases . physically meaningful in this case
are only the properties of the pure phases .
consequently one requires a new idea to calculate the transport properties of the disordered systems .
we can at the present time only surmise that this future theory of transport properties will calculate the transport properties with the help of equations which have the property of a multiplicity of solutions
. this theory should also formulate in a new way the average over the ensemble of random potentials in order to arrive at the properties of the pure phases . | the anderson localization problem in one and two dimensions is solved analytically via the calculation of the generalized lyapunov exponents .
this is achieved by making use of signal theory .
the phase diagram can be analyzed in this way . in the one dimensional case
all states are localized for arbitrarily small disorder in agreement with existing theories . in the two dimensional case for larger energies and large disorder
all states are localized but for certain energies and small disorder extended and localized states coexist .
the phase of delocalized states is marginally stable .
we demonstrate that the metal - insulator transition should be interpreted as a first - order phase transition .
consequences for perturbation approaches , the problem of self - averaging quantities and numerical scaling are discussed . |
among the technologies for biometric identification and verification , face recognition has become a widely - used method , as it is a non - intrusive and reliable method .
however , the robustness of a facial recognition system is constrained by the degree of head pose that is involved and the recognition rates of state of the art systems drop significantly for large pose angles . especially for tasks where a great variation in head pose has to be expected , like in human - robot interaction , pose - invariant face recognition is crucial .
the rise of collaborative and assistive robots in industrial and home environments will increase the demand for algorithms that can adapt to changing settings and uncontrolled conditions .
a humanoid robot s cognitive ability to interpret non - verbal communication during conversations relies also heavily on the face of the human counterpart .
the impact of head pose on the face analysis performance can be minimised by using normalisation techniques that transform non - frontal faces into frontal representations .
this type of image pre - processing can loosely be classified into two categories : * 2d methods ( cylindrical warping , aam ( active appearance models ) @xcite , 2d warping @xcite ) * 3d methods ( 3dmm , gem ( generic elastic models ) @xcite ,
mean face shape @xcite ) as the 2d methods are trained on 2d data , the warping will only be an approximation of the underlying 3d rotation .
they also do not model imaging parameters like the pose of the face and illumination , and face parameters like expressions , explicitly . instead , these parameters are inherent and implicitly encoded in the model parameters .
when dealing with larger pose ranges than around @xmath040@xmath1 in yaw angle , a single 2d model is often no longer sufficient and a multi - model approach must be used ( e.g. @xcite , @xcite ) .
the use of a 3d face model has several advantages , especially when further analysis involving the shape of the face is required , like in emotion detection . to correct the pose for face recognition ,
a 3d representation of the face can be rotated easily into a frontal view . at close ranges ,
the use of a 3d or rgb - d sensor can provide the additional depth information for the model .
however , since the error of depth measurements increases quadratically with increasing distance when using rgb - d sensors like microsoft kinect @xcite , they are not suitable for acquiring image data from a distance . on the other hand ,
2d cameras are more widely - used in existing platforms and cost - efficient , so that a method combining the advantages of 2d image acquisition and the analysis with a 3d model is needed . the 3d morphable model ( 3dmm )
@xcite satifies this need by providing a parameterised principal component analysis ( pca ) model for shape and albedo , which can be used for the synthesis of 3d faces from 2d images . a morphable model can be used to reconstruct a 3d representation from a 2d image through fitting .
fitting is the process of adapting the 3d morphable model in such a way that the difference to a 2d input image is as small as possible .
depending on the implementation , the fitter s cost function uses different model parameters to iteratively minimise the difference between the modeled image and the original input image .
existing fitting algorithms solve a complex cost function , using the image information to perform shape from shading , edge detection , and solve for both shape and texture coefficients ( @xcite ) . due to the complexity of the cost functions , these algorithms require several minutes to fit a single image . while the execution time may not be crucial for offline applications , real - time robotics applications require faster solutions .
more recently , new methods were introduced that use local features @xcite or geometric features ( landmarks , edges ) with a fitting algorithm similar to the iterative closest point algorithm @xcite . in this paper
, we present and evaluate a complete , fully automatic face modelling framework consisting of landmark detection , landmark - based 3d morphable model fitting , and pose - normalisation for face recognition purposes .
the whole process is designed to be lightweight enough to run in real - time on a standard pc , but is also especially intended to be used for human - robot interaction in uncontrolled , in the wild environments .
the theory is explained in section [ sec : lm_3dmm ] . in section [ sec : exp ] , we show how the approach improves the recognition rates of a regular cots ( commercial off - the - shelf ) face recognition system when it is used as a pre - processing method for pose - normalisation on non - frontal and uncontrolled images from two image databases .
finally , we show how our approach is suitable for real - time robotics applications by presenting its integration into the hmi framework of our scitos robot platform in section [ sec : app ] .
section [ sec : conclusion ] concludes the paper and gives an outlook to future work .
all parts of the framework are publicly available to support other researchers : the c++ implementations of the landmark detection and the fitting algorithm and the face model , as well as a demo app that combines both .
in this section , we will give a brief introduction to the 3d morphable model and then introduce our algorithm to recover pose and shape from a given 2d image .
we will then present our pose - invariant texture representation in form of a so - called isomap .
a 3d morphable model consists of a shape and albedo ( colour ) pca model constructed from 169 3d scans of real faces .
these meshes first have to be brought in dense correspondence , that is , vertices with the same index in the mesh correspond to the same semantic point on each face . the model used for this implementation
consists of 3448 vertices .
a 3d shape is expressed in the form of @xmath2^\mathrm{t}$ ] , where @xmath3^\mathrm{t}$ ] are the coordinates of the @xmath4th vertex and @xmath5 is the number of mesh vertices .
the rgb colour values are stacked in a similar manner .
pca is then applied to both the vertex- and colour data matrices separately , each consisting of @xmath6 stacked 3d face meshes , resulting in @xmath7 shape eigenvectors @xmath8 , their variances @xmath9 , and a mean shape @xmath10 , and similarly for the colour model ( @xmath11 , @xmath12 and @xmath13 ) .
a face can then be approximated as a linear combination of the respective basis vectors : @xmath14 where @xmath15^\mathrm{t}$ ] and @xmath16^\mathrm{t}$ ] are vectors of shape and colour coefficients respectively .
given an image with a face , we would like fit the 3d morphable model to that face and obtain a pose invariant representation of the subject . at the core of our fitting algorithm
is an affine camera model , shape reconstruction from landmarks , and a pose invariant textural representation of the face .
we thus only use the shape pca model from the 3dmm and not the colour pca model , and instead use the original texture from the image to obtain the best possible image quality for subsequent face analysis steps .
similar to aldrian and smith @xcite , we obtain a linear solution by decomposing the problem into two steps which can be alternated . the first step in our framework
is to estimate the pose of the face . given a set of 2d landmark locations and their known correspondences in the 3d morphable model , we compute an affine camera matrix .
the detected 2d landmarks @xmath17 and the corresponding 3d model points @xmath18 ( both represented as homogeneous coordinates ) are normalised by similarity transforms that translate the centroid of the image and model points to the origin and scale them so that the root - mean - square distance from their origin is @xmath19 for the landmark and @xmath20 for the model points respectively : @xmath21 with @xmath22 , and @xmath23 with @xmath24 .
using @xmath25 landmark points , we then compute a normalised camera matrix @xmath26 using the _ gold standard algorithm _ @xcite and obtain the final camera matrix after denormalising : @xmath27 . the second step in our framework
consists of reconstructing the 3d shape using the estimated camera matrix .
we estimate the vector of pca shape coefficients @xmath28 : @xmath29 where @xmath30 is the number of landmarks and @xmath31 is a weighting parameter for the regularisation that is needed to only allow plausible shapes .
@xmath32 are the detected landmark locations and @xmath33 is the projection of the 3d morphable model shape to 2d using the estimated camera matrix .
subsequently , the camera matrix can then be re - estimated using the now obtained identity specific face shape , instead of only the mean face .
both steps are iterated for a few times - each of them only involves solving a small linear system of equations . in the experiments with automatically detected landmarks , we use a cascaded regression based approach similar to feng et al . @xcite .
after an initial estimate , i.e. placing the landmarks in the region found by a face detector , we extract local features ( hog @xcite ) to update the landmark locations until convergence towards their actual position .
the detected 2d landmarks have known corresponding points in the 3d morphable model , and these points are then used to fit the model .
the combination of the regression based landmark detection and the landmark - based 3d morphable model fitting results in a lightweight framework that is feasible to run on videos ( all components run in the order of milliseconds on a standard cpu ) . after fitting the pose and shape ,
a correspondence between the 3d mesh and the face in the 2d image is known for each point in the image .
we use this correspondence to remap the original face texture from the image onto a pose - invariant 2d surface that covers the face from all angles .
we create such a generic representation with the isomap algorithm @xcite : it finds a projection from the 3d vertices to a 2d plane that preserves the geodesic distance between the mesh vertices .
our mapping is computed with the algorithm from tena @xcite .
the isomap of different persons are in dense correspondence with each other , meaning each location in the map corresponds to the same physical point in the face of every subject ( for example , a hypothetical point @xmath34 $ ] is always the center of the right eye ) .
it can consequently show the accuracy of the fitting and is therefore a plausible representation of the result .
figure [ fig_isomap ] shows an isomap and a frontal rendering side by side .
the isomap captures the whole face , while the rendering only shows the frontal parts of the face . as the face in the input image
is frontal , with most parts of the face visible , there is little self occlusion . in case of large poses , a major part of the isomap can not be filled with face texture , and these regions will be marked black ( for example , a small black spot next to the nose can be observed in the figure ) .
to analyse the performance of our approach , we have run experiments on two different image databases ( muct and point and shoot challenge ( pasc ) ) .
our experiments are face verification experiments , which means that the system has to verify the identity of probe images compared to a set of gallery images via matching .
the results are then statistically analysed regarding the ratio of the two possible error types , the false acceptance rate ( far ) and false rejection rate ( frr ) and plotted into detection error tradeoff ( det ) curves @xcite , using the ground - truth subject - id information provided in the metadata of the image databases . in the experiments on the pasc database ( section [ sec : pasc ] ) , we filter the probe and gallery images according to different head poses . we used a market - leading , commercial face recognition engine for the process of enrolment and matching . as a reference measure ,
all experiments were also conducted using the original , unprocessed images from the databases ( denoted as _ org _ ) . with this reference , we are able to analyse the impact of using our approach compared to a conventional face recognition framework without 3d modelling .
the muct database consists of 3755 faces of 276 subjects and is published by the university of cape town @xcite . for our experiments only persons without glasses and their mouths closed were used . after this filtering
, 1221 pictures were left .
the pictures are taken from 5 different cameras that cause different pose angles and there are in total 10 different lighting schemes applied to the pictures . moreover , the size of the faces within the images is quite large and the subjects are placed in a controlled lab environment .
muct comes with 76 manually labeled landmarks given for every image , from which we used 16 for the initialisation of our landmark - based 3dmm fitting ( _ lm _ ) . in total
, we used the unprocessed images ( _ org _ ) and the pose normalised renderings coming from our fitting method ( _ lm _ ) .
we ran a verification experiment on these images , leading to similarity matrices with 1 490 841 scores for both methods . using these scores ,
a det curve was plotted , of which a relevant part is depicted in figure [ fig_muct ] .
it is clear to see that the verification errors of the face recognition drop significantly when our landmark - based modelling approach is used to correct the head pose .
this experiment shows that when pose is the major uncontrolled factor , like on muct , the usage of our _ lm _ aproach leads to a clearly improved face recognition performance . for our proposed use case in assistive or humanoid robotics ,
muct is not representative of typical in the wild images we encounter . to verify the performance in heavily uncontrolled settings
, we decided to do further investigations on a larger , more challenging database .
the point and shoot challenge ( pasc ) is a performance evaluation challenge for developers of face recognition systems initiated by the colorado state university @xcite .
as the name depicts , the images and videos used in the challenge are taken with consumer point and shoot cameras and not with the help of professional equipment .
the database offers 9376 still images of varying resolution up to 4000@xmath353000 pixels and labeled metadata .
the arrangement of the image sceneries raises some major challenges for the automated recognition of faces .
the pictures were taken in various locations , indoors and outdoors , with complex backgrounds and harsh lighting conditions , accompanied by a low image quality due to blur , noise or incorrect white balance . additionally , the regions of interest for face recognition are also relatively small and of low resolution because the photographed people are not the main subject of the scenery . in a direct comparison ,
the verification rate on pasc equals only 0.22 at 0.001 far , in contrast to the more controlled databases mbe ( 0.997 ) , gbu ( 0.8 ) and lfw ( 0.54 ) @xcite .
because of these harsh , uncontrolled conditions , which are similar to a real life scenario , we chose this image database for our experiments and applied our fitting algorithm ( _ lm _ ) to the images .
like the previous experiment on muct ( sec .
[ sec : muct ] ) , it was a verification experiment using the same cots face recognition algorithm .
our experiments on pasc are intended to be as near to a realistic scenario as possible .
therefore , we used the automatic landmark detection of @xcite instead of using manually annotated facial landmarks .
the detector and our fitting algorithm ( _ lm _ ) allow the use of different landmark schemes , which can vary in their amount .
we tested three different sets of 7 , 13 and 49 landmarks , which are visualised on the 3d morphable model in figure [ fig_det_landmarks ] ( a ) . to compare the different amounts of landmarks ,
their corresponding results for the face recognition experiment on pasc are shown in figure [ fig_det_landmarks ] ( b ) .
it can be observed that a higher number of facial landmarks leads to a higher quality 3dmm fitting , which then leads to a better verification performance .
one of the key questions of this work is whether our approach can improve the performance of face recognition under uncontrolled conditions , especially when dealing with head pose . to allow a more detailed interpretation of the results
, we generated head pose annotations for the pasc database .
we used opencv s implementation of the posit algorithm @xcite to calculate the yaw angles for each image in the database .
posit estimates the pose of an object from an image using a 3d reference model and corresponding feature points of the object in the image . in our case , the 3d model was the 3d morphable model and the feature points were facial landmarks .
we used this yaw angle annotation to filter the scores into different groups of yaw angles .
the results obtained with our method were then compared to the results without any face modelling . for comparison
, we also tested the performance of a commercial face modelling system ( _ cfm _ ) using the same workflow .
the _ cfm _ uses additional modelling capabilities like mirroring ( to reconstruct occluded areas of the face ) and deillumination .
figure [ fig : ce_lm ] depicts frontalised renderings of the commercial system and our fitter . in total , we compared all images of pasc against each other and the yaw angle filter separated the scores into 10@xmath36 bins that span from -70@xmath36 to + 70@xmath36 . to better see how the pose normalisation methods improve the performance , the resulting curves show the absolute difference of the false rejection rate to the baseline without face modelling ( _ org _ ) at a specific operating point .
this type of diagram allows a more simple and clearly arranged representation of the results at different head poses . in figure [ fig_curvedeltayaw_0 ] ,
such a curve from the experiment is shown .
a set of gallery images of 0@xmath36 yaw is compared to sets of probe images from 0@xmath36 to 70@xmath36 . as gallery databases in face recognition applications often contain frontal images , this experiment reflects a common use case .
at the operating point 0.01 far , we then plot the improvement for each group of yaw angles .
the commercial face modelling solution shows its first improvements over the conventional face recognition starting at 30@xmath36 yaw .
surprisingly , our 3d morphable model based method is even able to slightly improve the matching performance for frontal images .
although the cots face recognition system was used outside of the specification for the large yaw angles , we are able to improve the performance for the whole pose range . by using the original image s texture , our fitter ( _ lm _ ) does not alter the face characteristics and corrects only rotations in the pitch or roll angles . in the next two experiments ,
we take a look at how well the system operates when the gallery is also not frontal . in figure
[ fig_curvedeltayaw_20_40 ] , the results for the experiment with a 20@xmath36 and a 40@xmath36 gallery can be seen . in this case , matching the faces is a much more difficult task , especially when the subjects look in different directions and large areas of the faces are not visible due to self - occlusion . while we are still able to improve the face recognition capabilities in the experiment with a 20@xmath36 gallery , our method _ lm _ struggles to keep up with the commercial solution for a 40@xmath36 gallery .
the ability to reconstruct parts of the non - visible texture is a key component of the commercial modelling system ( _ cfm _ ) in such a use case which our system currently does nt offer .
although , for the robotics application , which we present in section [ sec : app ] of this paper , it is likely that frontal images are used for the gallery enrolment .
our fast and accurate 3d face modelling method is particularly suitable for face analysis tasks in assistive or humanoid robotics . in the following ,
we demonstrate our efforts in this field by illustrating how we integrate our approach into a mobile robot s software framework . a mobile robot platform combined with an hmi system
is the basis for our research on human - robot interaction .
the robot is based on a scitos g5 drive system with onboard industrial pc which is able to navigate autonomously in indoor buildings . for mapping and localisation ability , as well as collision avoidance ,
two laser scanners are attached to the base ( sick s300 and hokuyo urg-04lx ) . using this base platform ,
the robot is able to approach humans safely .
the hmi part of the robot consists of a touch - based display and a head with moveable , human - like eyes . on top of the robot
, we mounted a pan - tilt - unit ( ptu ; directed perception d46 - 17 ) , where additional sensors and cameras can be installed .
the ptu adds an additional degree of freedom which is used for camera positioning to enhance the tracking process . to align the cameras to the region of interest , we track people by applying an adaptive approach which uses a particle filter to track the position and size of the target and estimates the target motion using an optical flow based prediction model .
furthermore , an svm - based learning strategy is implemented to calculate the particle weights and to prevent bad updates while staying adaptive to changes in object pose @xcite . at this point , the 3dmm
pose normalisation of our approach can be used for the image of the tracked face .
it can be either applied continuously or if the pose of the subject is greater than a certain threshold ( using the knowledge obtained in the experiments of section [ sec : pasc ] ) .
key frames of an example video taken from our robot system are presented in the upper row of figure [ fig_filmstreifen_philipp ] .
the opencv object detection with a face cascade @xcite was used to initialise the regression based landmark detector , which then automatically found 15 facial landmarks .
the fitting results , represented in isomaps , are shown in the second row of figure [ fig_filmstreifen_philipp ] .
on the robot , we use the frontal renderings as an input for the same face recognition engine that was also used for the experiments . to distinguish the person s identity properly
, the system should generate a rather high score for this positive match . the pose normalised renderings allow the face recognition algorithm to achieve a clearly higher score compared to the original , unprocessed images .
possible use cases for a robot of this kind are industrial mobile robotics , shop or museum tour guides or a surveillance system ( e.g. in a supermarket ) . in all these cases ,
the robot has to interact naturally with technically unskilled people .
it is therefore convenient to be able to analyse people independent of their pose , without requiring them to look at the robot directly .
identification , age and gender estimation on pose normalised renderings offer the possibility to personalise the robot s behaviour .
we proposed a powerful method for head pose correction using a fully automatic landmark - based approach for fitting a 3d morphable model . by using an efficient fitting algorithm ,
our approach can be used for tasks which require a fast real - time solution that can be also used on live video streams .
in contrast to existing work , we focus on an evaluation of our approach with automatically found landmarks and in - the - wild images and publish the whole 3dmm fitting pipeline as open source software ( see section [ sec : intro ] )
. an experimental evaluation on the two commonly - used image databases muct and pasc showed the significance of our approach as a means of image processing for facial recognition in both controlled and heavily unconstrained settings .
for recognition algorithms that are mostly trained for frontal faces , a rotation of the head means a loss of information that results in a lower matching score .
we showed that our methodology is capable of improving the recognition rates for larger variations in head pose .
compared to a commercial face modelling solution , our approach keeps up well and even outperforms it in certain scenarios .
the addition of 3dmm pose normalisation certainly brings advantages compared to a conventional face recognition framework when the setting is uncontrolled - like in robotics , surveillance or consumer electronics .
furthermore , we presented our software framework on a mobile robot platform which uses our pose normalisation approach to allow a more reliable face analysis .
we are currently doing further research on age- and emotion - estimation systems to expand the abilities of our robot . in the future
, we also plan to statistically evaluate the effect on their performance when using our landmark - based pose normalisation , like we did for face recognition in this paper .
the authors would like to thank huan fui lee and the rt - lions robocup team of reutlingen university .
we would also like to thank cyberextruder , inc . for supporting our research . | face analysis techniques have become a crucial component of human - machine interaction in the fields of assistive and humanoid robotics .
however , the variations in head - pose that arise naturally in these environments are still a great challenge . in this paper , we present a real - time capable 3d face modelling framework for 2d in - the - wild images that is applicable for robotics .
the fitting of the 3d morphable model is based exclusively on automatically detected landmarks . after fitting
, the face can be corrected in pose and transformed back to a frontal 2d representation that is more suitable for face recognition .
we conduct face recognition experiments with non - frontal images from the muct database and uncontrolled , in the wild images from the pasc database , the most challenging face recognition database to date , showing an improved performance .
finally , we present our scitos g5 robot system , which incorporates our framework as a means of image pre - processing for face analysis . |
the origin of planets is a vast , complex and still quite mysterious subject . despite decades of space exploration , ground based observations and detailed analyses of meteorites and cometary grains ( the only space samples available in our laboratories ) it is still not clear how the planets of the solar system formed .
the discovery of extrasolar planets has added confusion to the problem , bringing to light evidence that planetary systems are very diverse , that our solar system is not a typical case and that categories of planets that do nt exist in our system are common elsewhere ( e.g. the super - earth planets ) .
there are several recent reviews on planet formation .
johansen et al .
( 2014 ) focused on planetesimal formation , morbidelli et al .
( 2012 ) and raymond et al .
( 2014 ) focused on terrestrial planets , and helled et al .
( 2014 ) focused on giant planets .
these reviews remain up to date .
therefore , the goal of this chapter is to focus instead on open problems and the main issues of debate .
accordingly , here we identify and discuss what we ( the authors of this review ) consider the top 5 bottlenecks in the field of planet formation .
these are : 1 . what is the structure of protoplanetary disks ?
( section 2 ) 2 .
how did the first planetesimals form ?
( section 3 ) 3 .
how do planets migrate ?
( section 4 ) 4 . what is the origin of the _ tri - modal _ structure of the solar system ?
( section 5 ) 5 .
what is the relationship between terrestrial planets and super - earths ?
( section 6 ) these issues cover a range of size scales and time scales , and illustrate the upcoming challenges in planet formation .
our choice of the `` top 5 '' is admittedly biased . nonetheless , these issues are clear and present obstacles to our understanding of planet formation .
we conclude this review in section 7 with a discussion .
circumstellar disks of gas and dust represent the cradles of planet formation . the problem is to understand the mechanism by which protoplanetary disks evolve and transport mass to the central star ( see , e.g. , armitage 2011 ) . depending on
which mechanism is dominant , the resulting global disk structures may differ significantly ( see figure 1 ) , with profound effects on planet formation and migration .
it has long been thought that transport of angular momentum in the disk is due to turbulent viscosity ( shakura and sunayev , 1973 ) and that the magneto - rotational instability is the main source of turbulence ( balbus and hawley , 1991 ) . with simple scaling arguments on the size of the vortices and their rotation period ( or , equivalently , assuming that the stress tensor scales linearly with the local gas pressure ) , shakura and sunayev ( 1973 ) modeled the viscosity @xmath0 as proportional to the square of disk s scale height @xmath1 and to rotation frequency @xmath2 : @xmath3 , where @xmath4 is the proportionality coefficient .
the assumption that @xmath4 is the same everywhere leads to the so - called @xmath4-disk model ( lyndel - bell and pringle , 1974 ; pringle , 1981 ; balbus and papaloizou , 1999 ) .
the structure of these disks basically depends only on three parameters : the accretion rate of gas onto the central star , the value of @xmath4 and the dust / gas ratio , the latter governing the disk s opacity .
the disk evolves under its own viscosity until the accretion rate on the star drops below a few times @xmath5/y . at this point
the photo - evaporation process becomes important , as it removes gas from @xmath6 au at a comparable rate ( see alexander et al .
, 2014 , for a review ) .
photo - evaporation divides the disk into an inner part , rapidly accreted onto the star , and an outer part , rapidly photo - evaporated inside - out ; it explains the rapid final clearance of protoplanetary disks .
stone et al .
( 1998 ) pointed out that the magneto rotational instability ( mri ) is unlikely to act uniformly across the disk .
the central part of the disk , which is opaque to radiation , should not be ionized and therefore the mri should not operate there .
this led to the concept of a `` dead zone '' , a low - viscosity region within the disk .
disks with a dead zone can still be modeled with the @xmath4 prescription , provided that @xmath4 is function of the radial and vertical coordinates .
because in the dead zone @xmath4 is smaller , the gas density is much larger than in the mri active zone to ensure the smooth transport of material through the disk . in disks modeled in 1d ( i.e. radial structure only )
the contrast in gas surface density @xmath7 is proportional to @xmath8 ( e.g. martin et al . , 2012 ) .
in 2d models ( @xmath9 ) , however , there is no simple proportionality , because the gas can flow near the surface of the disk ( bitsch et al .
, 2014b ) .
recent studies have shown that the mri is likely to be suppressed almost everywhere in the disk because of non - ideal mhd effects such as ambipolar diffusion ( bai and stone , 2013 ; lesur et al . , 2014 ; see turner et al . , 2014 for a review ) .
disk winds have been proposed to be the main process removing angular momentum from the disk , thus causing gas to accrete onto the central star ( bai and stone , 2013 ; bai , 2013 , 2014 , 2015 , 2016 ; gressel et al . , 2015 ; simon et al . , 2015 ) .
the resulting disk structure can be very different from that of an @xmath4-disk .
for instance , suzuki et al .
( 2010 , 2016 ) find that the inner part of the disk can be substantially depleted , although bai ( 2016 ) still finds a @xmath10 surface density profile in the innermost @xmath11 au , similar to that of an @xmath4-disk .
the actual surface density profile is very important for models of planetesimal formation and planet migration .
for instance , a depleted inner disk would slow down ( or even reverse ) the radial migration of protoplanets ( ogihara et al .
, 2015a , b ) . it would also have a much less sub - keplerian rotation than an @xmath4-disk , favoring particle accumulation and the formation of planetesimals . .
each curve shows the surface density as a function of heliocentric distance .
two models of low - viscosity disks where transport is dominated by winds are shown , from bai ( 2016 ) ( red color ; @xmath12 ) and suzukui et al ( 2016 ) ( green color ; @xmath13 ) .
they are labeled as `` wind - disks '' to contrast with the classic @xmath4-disk , whose density distribution is also shown in blue . in the @xmath4-disk model the surface density of the gas scales inversely to the assumed viscosity ( here @xmath14 is assumed ) .
note the depletion of gas in the inner part of the disk in suzuki et al . model . ]
an important question to address for a wind - dominated disk is how low the viscosity ( or the turbulent strength ) can be .
there are sources of turbulence other than the mri , such as the baroclynic instability ( klahr and bodenheimer , 2003 ) , the vertical shear instability ( nelson et al . , 2013 ) and the zombie - vortex instability ( marcus et al . , 2015 ) .
stoll and kley ( 2014 ) estimated that the vertical shear instability can sustain turbulence with a strength equivalent to @xmath15 a few times @xmath16 , but more recent work ( stoll and kley , 2016 ) argues that this is true only beyond @xmath175 au .
thus , the actual strength of the turbulence and the associated viscosity in the inner disk is still unknown .
the strength of turbulence is a key parameter for planetesimal growth , because it governs the settling of particles towards the midplane .
the viscosity is also the key to estimate the temperature of the disk .
in fact , in the absence of viscous heating , the only source of heat is irradiation from the central star which should generate heat .
the role of these currents on the thermal structure of the disk is still unexplored . ] .
a disk that is solely heated by the star is called a passive disk ( chiang and goldreich , 1997 ; dullemond et al . , 2001 , 2002 ) .
a passive disk is very cold throughout its evolutionary stages , until it becomes transparent to the stellar irradiation .
chiang and youdin ( 2010 ) estimated that the snowline ( the location where the temperature is 160170k so that water has a transition from vapor to solid state ) in such a disk should be around @xmath18 au .
if this were the case , all bodies in the inner solar system should be water rich , including main belt asteroids and the earth .
while these bodies contain some water , their bulk water abundance is much less than that expected from solar abundance ( about 50% in mass ; lodders , 2003 ) , suggesting that the snowline was , at least during the early phases of the disk , much father out than their location .
in fact , the mechanisms proposed to explain why the inner solar system bodies did not accrete a lot of water as the disk cooled off ( morbidelli et al . , 2016 ; sato et al . , 2016 )
require that the disk was initially too hot for the existence of water - ice at 12 au .
thus , we expect that there was some non - negligible turbulent viscosity in the inner disk , even if the origin of this viscosity is still not clear . for reference , for the snowline to be at about 3 au , as suggested by asteroid composition
, the viscous transport in the disk should have been of about @xmath19 ( bitsch et al . ,
if the disk had this viscous transport at some point during its evolution , its viscosity had to be large or the disk very massive .
see morbidelli et al .
( 2016 ) for a more in depth discussion . while it may seem tangential to planet formation
, angular momentum transport and heating in protoplanetary disks is essential in determining the disk s structure .
this feeds back into particle growth and orbital migration .
the solid building blocks of planets are a trace component within gas - dominated protoplanetary disks .
the processes that govern the growth of macroscopic particles remain poorly understood .
perhaps the most vexing problem is the growth of the first planetesimals .
once planetesimals are present in the disk , their further growth by accreting pebbles is modestly - well understood .
the first population of planetesimals may serve as a planetary system s blueprint . in the classic model of planet accretion ,
dust particles settle to the midplane of the disk , collide with each other and form aggregates held together by electrostatic forces ( e.g. , dominik & tielens 1997 ) .
little by little these aggregates grow and get compacted by collisions , forming small planetesimals ( e.g. , weidenschilling and cuzzi 1993 , blum and wurm 2008 ) . today we know that this classic picture of planetesimal growth has severe problems ( illustrated in figure 2 ) .
when silicate grains grow to a size of about a millimeter , they start to bounce off each other instead of accreting ( zsom and dullemond , 2008 ; gttler et al . , 2009 ) . in the icy part of the disk
, particles can grow up to a few decimeters in size before starting to bounce .
this is called the _ bouncing barrier_. at these sizes , particles migrate rapidly in the disk towards the star due to gas drag .
this radial drift produces large relative velocities among particles of different sizes and hence disruptive collisions .
even if collisions were not disruptive and particles could continue to grow , eventually meter - size boulders would migrate so rapidly to be lost into the star before they can grow significantly further ( weidenschilling , 1977 ) .
this is the well known _
meter - size barrier _
( referred to as the `` drift barrier '' in figure 2 ) . over the past decade
new ideas have been proposed to overcome this problem . on the one hand , it has been shown ( okuzumi et al . , 2012 ; kataoka et al . ,
2013 ) that the regime of collisional coagulation might nevertheless work beyond the snowline because of the tendency of icy particles to form very fluffy aggregates .
these aggregates , thanks to their extremely high porosity , are very resistant to collisional disruption and their radial drift in the disk is also very slow .
thus both the bouncing / breaking barrier and the drift barrier can be circumvented and km - size planetesimals can be formed after a final phase of contraction of the aggregates .
this alleged formation path for planetesimals , however , should not work within the snowline , due to the reduced sticking properties of the silicate grains . on the other hand
, it has been shown that `` small '' particles ( whatever their chemical nature ) tend to clump due to turbulence in the disk ( cuzzi et al . , 2001 ;
johansen et al . , 2007 ) .
once the clump is sufficiently dense , its self - gravity can take control , holding all particles together until they settle in their common potential well and form a large planetesimal ( johansen et al . , 2007 ; cuzzi et al .
this recipe , however , is not without problems .
first , it requires turbulence within the disk .
yet the very existence of turbulence and its properties are far from being firmly established , as discussed in the previous section .
second , in order to clump under the effect of turbulence , particles must be much larger than that allowed by the bouncing barrier , at least in the inner disk .
in fact , the critical parameter governing the dynamics of a particle undergoing gas drag is its stokes number @xmath20 , where @xmath2 is the keplerian frequency and @xmath21 is the coefficient relating the acceleration of the particle due to the drag to the velocity difference between the particle and the fluid medium , i.e. @xmath22 . in the simulations by johansen et al .
( 2007 ) , particles effciently clumped have stokes number ( i.e. the product of the friction time and the orbital frequency ) larger than 0.5 which , at 1 au , corresponds to sizes of @xmath1720 cm . the concentration mechanism in low - vorticity regions proposed by cuzzi et al .
( 2001 ) seemed to operate for chondrule - sized particles , but a new reconsideration of this mechanism ( cuzzi et al . , 2016 ) , adopting an improved turbulent cascade model , also raises the particle size to 10 cm or so .
how particles can grow to this size in the inner disk is an open problem .
( for inward drift ) and @xmath23 ( growth ) .
the green line shows the fragmentation barrier above which growth is inhibited .
the dashed / purple curves show the trajectories of particles starting at different orbital radii ( calculated neglecting fragmentation ) .
the red line is the corresponding drift barrier .
the solid black line denotes where the stokes number is equal to 1 . from birnstiel
et al ( 2016 ) . ] in order to relax the hypothesis that the disk is a priori turbulent , it has been proposed that particles themselves can generate turbulence and then clump in a feedback process .
two mechanism can produce this result .
the sedimentation of particles towards the midplane of the disk can drive a kelvin - helmoltz instability , because the particles force the midplane layer of the gas to rotate around the sun faster than the surface layers ( weidenschilling , 1995 ) . as shown in johansen
et al . ( 2006 ) , the turbulence generated by the kelvin - helmholtz instability can clump the particles enough to form self - gravitating clusters .
but , again , this effect depends on particle size .
small particles do not sediment very fast and therefore do nt create a dense midplane layer ; moreover , being more coupled with the gas , the shear effect that they generate is weaker .
effective clumping leading to local sold / gas density ratio of a few 10s is observed only for particles with stokes number larger than 0.1 , which corresponds to a size of 10 cm at 1 au .
the second mechanism is the streaming instability ( youdin and goodman , 2005 ) .
this is a linear instability generated by the radial drift of particles .
when the local density of particles exceeds the background density ( i.e. a small clump is formed due to some statistical density fluctuation ) , the collective drag force exerted by the clump on the gas makes the disk rotate faster ; as a consequence , the clump as a whole drifts inwards slightly more slowly than individual particles .
this creates a sort of traffic jam .
other particles drifting inward at the nominal speed catch up to the clump , enhance the local overdensity of solids , and further decelerate the radial drift of the clump .
thus , the local accumulation of particles speeds up , and so forth in a positive feedback . yet
the streaming instability is not without problems . if particles are relatively large ( stokes number larger than 0.1 ) , clumping is effective only if the solid / gas surface density ratio is larger than 3% ( so , enhanced compared to solar value of 1% ; johansen et al . , 2009 ) . for chondrule - size particles in the inner disk ,
the solid / gas ratio has to be increased above 8% ( carrera et al . , 2015 ) .
models of particle coagulation and radial drift ( lambrechts and johansen , 2014 ) show that both the size of available particles and the solid / gas ratio decrease in time even as the gas surface density decays .
it is unclear whether the conditions for the streaming instability can ever be reached . according to drazkowska and dullemond ( 2014 ) the streaming instability
may only be operational in the outer disk for two reasons : ( i ) beyond the snowline , the sticky properties of ice allow the formation of larger particles and , ( ii ) for a given particle size , the stokes number is larger than in the inner disk . in the inner disk
the streaming instability may only operate in a relatively narrow ring ( drazkowska et al . , 2016 )
where pebble pile - up if the radial surface density profile of the disk is sufficiently shallow .
the high solid / gas ratio required for chondrule - size particles to clump in the asteroid belt can only be reached in the late stage of the disk , after a substantial fraction of the gas has been removed ( throop and bally , 2005 ) and _ if _ particles are regenerated by some process , for instance in planetesimal - planetesimal collisions .
the analysis of meteorites shows indeed that there were at least two generations of planetesimals in the inner disk .
the parent bodies of iron meteorites formed early , in the first my ( kruijer et al . , 2012 ) after the formation of the first solids ( the calcium - alluminium inclusions -cais- dated to have formed 4.568 gy ago ; bouvier et al . 2007a ; burkhardt et al .
in contrast , the parent bodies of chondritic meteorites formed after 3 - 4 my ( villeneuve et al . , 2009 ) .
it is unclear whether there was a gap in time in planetesimal formation and even whether the first and last generations of planetesimals formed in the same place .
today all meteorites come from the asteroid belt , but the asteroids may have been trapped into the belt from different original reservoirs ( bottke et al . , 2006 ; levison et al 2009 ; walsh et al . , 2011 ; raymond et al 2016 ) .
some authors have proposed that chondrules are the outcome of collisions among massive objects ( libourel et al . , 2006 ; asphaug et al .
, 2011 ; johnson et al . , 2015 ) .
all this may be suggestive that the first planetesimals formed only at select locations ( near the disk s inner edge or beyond the snowline ) and that elsewhere planetesimals only formed later , as the protoplanetary disk evolved towards the debris disk phase . but
this is still far from a consolidated view .
it is not clear , for instance , why the second generation asteroids would have only accreted chondrules and not a spectrum of collisional debris with different physical and chemical properties .
once planetesimals are formed , as long as they are embedded in a disk of gas and drifting particles and they are big enough ( typically larger than 100 km ) , they can keep growing by accreting the particles via a combination of gravitational deflection and gas drag ( ormel and klahr , 2010 ; johansen and lacerda , 2010 ; lambrechts and johansen , 2012 , 2014 ; guillot et al . , 2014 ; johansen et al .
, 2015 ; levison et al . , 2015 ; visser and ormel , 2016 ) .
this process is called _ pebble accretion _ and it is now fairly well understood and recognized to play a fundamental role in planet formation .
unfortunately , by definition , pebble accretion operates only after that the first planetesimals are formed .
thus , our poor understanding of the primary planetesimal formation process , discussed above , blocks our understanding of when , where and in which numbers , protoplanets formed in the disk .
growing planets interact gravitationally with the protoplanetary disk . they create a spiral density wave in the gas distribution that is stationary in the reference frame rotating with the planet ( ogilvie and lubow , 2002 ) . in turn , this non - axisymmetric density distribution exerts a torque on the planet . as a result , the angular momentum of the planet s orbit changes and the orbit expands or contracts depending on the sign of the torque . in other words , the planet undergoes a radial migration .
it can be demonstrated that in disks whose surface density is a power law of @xmath10 the torque is negative and hence the planet migration is directed towards the central star ( ward , 1997 ) . for low - mass planets ,
the migration speed scales linearly with the planet s mass ( goldreich and tremaine , 1980 ; lin and papaloizou , 1986 ) .
thus , planet migration can be neglected up to , at most , mars - mass embryos , such as the precursors of terrestrial planets ( wetherill , 1990 ) . according to tanaka et al .
( 2002 ) , the migration timescale of a mars - mass body at 1 au in a mmsn disk is 1 my .
given that these bodies took themselves a few my to form ( dauphas and pourmand , 2011 ) it is likely that they emerged in a disk less massive than the mmsn , and therefore did not migrate significantly during the disk phase .
for this reason , terrestrial planet formation models typically neglect radial migration .
but for proto - planets of one to few earth masses , such as the growing cores of giant planets , migration can not be ignored .
the discovery of hot jupiters ( mayor and queloz , 1995 ) was welcomed as a proof of planet migration ( lin et al 1996 ) .
but it was soon understood that the hot jupiters are the exception rather than the rule , as most extrasolar giant planets have orbital radii of 1 - 2 au ( butler et al 2006 ; udry & santos 2007 ) . in order to reproduce the observed distribution of planets , planet synthesis models
( e.g. ida and lin , 2004a , b ; alibert et al . , 2005 ) had to artificially reduce the speed of migration by one to two orders of magnitude relative to theoretical expectations .
no mechanism to globally reduce planets migration speed across the disk has been found .
detailed studies have identified effects that can reduce , stop and even reverse migration , but only at specific locations .
masset et al .
( 2006 ) found that a positive surface density gradient in the disk can act as a trap for migrating planets .
however , the only place where such a gradient is expected is at the inner edge of the disk ( or at the boundary between the mri active region and the dead zone of the disk , which is also very close to the star ) .
the mechanism of masset et al ( 2006 ) can explain why planets do nt fall on the star but accumulate on small semi - major axis orbits ; however , it can not explain how the planets that are a few au away could have avoided migrating closer to the star .
paardekooper and mellema ( 2006 ) found a positive torque that arises in non - isothermal disks with steep temperature gradients .
this torque can exceed the negative one exerted by the spiral density wave mentioned above , thus forcing the outer migration of the planet .
several years of studies led to a quantitative analysis of this torque ( paardekooper et al .
, 2010 , 2011 ) . when applying these torque formulae to the structure of an @xmath4-disk ( hasegawa and pudritz , 2011 ; hellary and nelson , 2012 ; bitsch et al . , 2015 ) , it is found that the outward migration region is confined in radial range and planet mass ( see fig .
[ bitsch ] ) . in an massive young , warm disk ,
the main region of outward migration extends from 3 to 9 au and encompasses planets between 5 and 50 earth masses ( see figure 3 ) .
this is good news , because it means that a planet in this mass range , migrating from the outer disk , would stop at the outer border of this region .
but later on , when the disk becomes cooler and less massive , the region of outward migration shifts inwards and to smaller planet masses ( fig .
[ bitsch ] ) .
bodies more massive than 10 earth masses are released to rapidly migrate inward . for low - viscosity , mri inactive disks ( see section [ disk ] ) ,
the maximal planet mass of the outward migration region is even smaller . for multiple planets
the situation may be even worse .
mutual interactions among the planets may raise their orbital eccentricities and in this case the aforementioned positive torque is strongly reduced , promoting inward migration ( bitsch and kley , 2010 ; hellary and nelson , 2012 ) . prescription ( shakura & sunyaev 1973 ) with @xmath24 . from the disk model of bitsch
et al ( 2015 ) . ]
the currently - favored paradigm of giant planet formation proposes that 10 - 20 earth - mass cores serve as the seeds for gas giants ( pollack et al 1996 ) .
there is a delay between the core s growth and the onset of rapid gas accretion . during this interval
massive cores may migrate inward and , following current models , will reach the inner edge of the disk before becoming giants ( coleman and nelson , 2014 ) . with our current understanding of migration , for giant planets to form and still be at a few au from the star when the disk disappears , bitsch et al ( 2015b ) found that their cores must start their growth at 1520 au and relatively late during the disk lifetime .
yet issues remain regarding this result .
because the outer disk is flared , both the streaming instability and pebble accretion are more efficient closer in than farther out .
thus , it is difficult to understand why giant planet cores could not form between the original snowline location ( say 3 - 5 au ) and 15 au , if they could form beyond this threshold distance .
one possibility is that planets that formed closer - in or earlier were engulfed into the central star and that the planets that we see today are the `` last of the mohicans '' , namely those that formed the last and the farthest , so that they could not reach the star before disk removal ( lin , 1997 ; laughlin and adams , 1997 ) .
but this is difficult to believe : the abundance of numerous close - in super - earths argues that planets stop migrating at the inner edge of the disk ( masset et al . ,
2006 ) and do nt fall onto the star ; indeed , disks have inner edges that stop inward migration ( romanova et al 2003 ; bouvier et al 2007b ) . moreover , in our solar system it is hard to conceive that numerous planets migrated through the inner disk without leaving a trace of their passage .
the case of our solar system may be helped by the fact that the contemporary presence of jupiter and saturn in nearby orbits reverses the migration direction of these giant planets ( masset and snellgrove 2001 ) .
but even in this case bitsch et al .
( 2015b ) could not achieve the required jupiter - saturn configuration if these planets started growing within 23 au .
it is likely that we are still missing a fundamental piece of the migration story .
one idea is that , if disks are much less viscous than expected ( see sect . [ disk ] ) relatively small planets could open gaps in the gas distribution .
they would then migrate in a different regime , known as the type - ii regime ( lin and papaloizou , 1986 ) , in which the migration speed is equal to the radial speed of the gas due to its viscous evolution .
if the viscosity of the disk is small , this speed is also small .
however , this idea seems naive .
first , to open a gap in the gas distribution , it is not enough that the torque exerted by the planet overcomes the viscous torque within the disk ; it is also needed that the planet has a hill radius larger than the scale height of the disk ( crida et al .
, 2006 ; morbidelli et al . ,
2014 ) , otherwise the gas flows into the gap from the vertical direction .
second , type - ii migration is an ideal process , but in reality gap - forming planets do not migrate at the viscous radial - drift speed ( duffell et al . , 2014 ; durmann and kley , 2015 ) . in particular , in the low - viscosity regime migration
is faster than the radial drift of the gas .
additional positive torques exerted on the planets have been searched for in the last years .
benitez - llambay et al .
( 2015 ) found that a few earth - mass planet with a very high temperature because of a vigorous accretion of solids can migrate outwards .
however , this effect wanes in the mass regime of giant planet cores , so it does not help the giant planet formation problem ( bitsch et al . , 2015b ) .
fung et al .
( 2015 ) claimed to have found a new positive torque arising only in 3d simulations of planet - disk interactions .
however , this torque seems to disappear when simulations are conducted with a sufficiently high resolution ( masset , private communication ) .
thus , nothing promising is in view at the current time .
maybe it is the disk structure that is misleading us ( coleman and nelson , 2016 ) . as we saw in sect .
[ disk ] the accretion @xmath4-disks that are usually considered may not be realistic . if the new disk - wind paradigm of transport creates a global positive surface density gradient in the inner few aus of the disk , as advocated by suzuki et al .
( 2010 ) , the migration of planets can be substantially modified ( ogihara et al .
, 2015a , b ) . however , it is not yet sure that this disk structure is real ( see bai , 2016 ) .
more work on the disk structure is needed .
the solar system is characterized by a tri - modal structure . in its inner part
there are the terrestrial planets , rocky and relatively small .
farther out , there is the asteroid belt , which carries cumulatively a negligible mass ( about @xmath25 earth masses ) . beyond 5 au
lies the realm of the giant planets .
the formation of the terrestrial planets requires that , during the lifetime of the protoplanetary disk , a set of planetary embryos formed in that region , with approximately the mass of mars . mars itself could be a surviving embryo ( dauphas and pourmand , 2011 ) . the small mass ratio between mars and the earth suggests a substantial depletion in the total mass of solid beyond 1 au and across the asteroid belt ( hansen , 2009 ; raymond et al 2009 ) , established very early during the history of the solar system .
in contrast , the formation of the giant planets requires the accretion of solid cores of 1020 earth masses before the removal of gas .
how this tri - modal structure was set is still an issue of open debate .
walsh et al . ( 2011 ) proposed that the depletion in solid mass beyond 1 au was caused by a specific pattern of jupiter s migration . keeping in mind the caveats
discussed in the previous section , it is legitimate to postulate that jupiter initially migrated inward , when it was basically alone in the disk , and then reversed migration direction when saturn formed ( masset and snellgrove , 2001 ; morbidelli and crida , 2007 ; pierens and nelson , 2008 ; pierens and raymond , 2011 ; discussed at length in raymond & morbidelli 2014 ) .
if this inward - then - outward migration history brought jupiter down to @xmath26 au before bringing it into the outer disk again , the region beyond 1 au would have been strongly depleted in mass ( fig .
4 ) . indeed ,
the simulations of walsh et al . show that the terrestrial planets would have formed with a mass distribution very similar to the one observed ( see also jacobson and morbidelli , 2014 ) , and the asteroid belt would have acquired a structure consistent with the current one ( see deienno et al . , 2016 )
. if this scenario , dubbed `` grand tack '' is correct , then the solar system had initially just a bi - modal structure , with mars - mass planetary embryos inside of the snowline and giant planet cores outside ; the tri - modal structure ( i.e. the depletion between 1 au and the orbit of jupiter ) was generated by jupiter s migration ( fig .
4 ) . however , even this bimodal structure is puzzling . assuming that planetesimals formed everywhere in the disk with comparable masses ( but see sect .
[ planetesimals ] ) , the subsequent process of planet growth by pebble - accretion should favor the bodies closer to the sun ( ida et al . , 2016 ) .
in other words , giant planet cores should have formed in the inner disk and mars - mass embryos in the outer disk ! the only exception to this is if the pebble sizes drastically changed at the snowline .
morbidelli et al .
( 2015 ) postulated that beyond the snowline the pebbles were dm - size objects made of ice and silicate grains .
when these pebbles drifted across the snowline the ice sublimated : 50% of the solid mass was lost and small ( mm - size ) silicate grains were released ( see also saito and sirono , 2011 ) .
if there was enough turbulence in the disk to make the particle layer thicker than the hill radii of the planetesimals ( a weak turbulence with @xmath27 would have been sufficient ) the accretion of small grains would have been much less efficient than the accretion of large pebbles .
hence , two initially equal - mass planetesimals on either side of the snowline would have grown at drastically different rates : by the time the outer one reached 20 earth masses , the inner one would have barely reached mars - mass ( morbidelli et al . , 2015 ; ida et al . , 2016 ) .
this explanation of the bimodal structure of the solar system , however , is not unique : if planetesimals inside the snowline were smaller for some reason ( see sect .
[ planetesimals ] ) than planetesimals beyond the snowline , the latter would have grown much faster by pebble - accretion even without a drastic change in pebble size across the snowline . but
let s take a step back and wonder whether the tri - modal structure of the solar system could have been primordial ( i.e. not the consequence of a grand tack - like migration of jupiter ) .
starting from a smooth population of planetesimals levison et al .
( 2015b ) obtained a concentration of mass inside of 1 au through pebble - accretion .
their model involves a number of assumptions , three of which appear particularly critical to us .
first , it is assumed that there is no mass flux in solids coming from beyond the snowline ; second , pebbles form by condensation and coagulation throughout the inner disk ; third , the disk is flared everywhere .
the first and the second assumptions together have the consequence that , the closer a planetesimal is to the snowline ( but inwards of its location ) , the smaller the pebble - flux that crosses its orbit .
the third assumption makes the pebble - accretion process more efficient with decreasing heliocentric distance .
these three assumptions may not be well justified .
preventing a flux of pebbles from the outer disk requires the formation of a strong positive gas surface density gradient ( i.e. a pressure `` bump '' ) at the snowline .
it has been proposed in the context on an analytic 1d ( radial ) disk model ( kretke and lin , 2007 ) that the snowline can create a pressure bump because the condensation of icy grains could strongly reduce the ionization of the gas , quench its turbulent viscosity , and increase the gas surface density beyond the snowline .
however , 3d numerical simulations of disk structure in the presence of radial and vertical viscosity transitions ( bitsch et al . , 2014b ) show that this is true only if @xmath28 , where @xmath0 is the viscosity .
it is very unlikely that the condensation of icy grains could have such a dramatic effect on the viscosity . as for the second assumption
, it has been shown that the direct condensation of gas is not relevant after about half a viscous timescale ( morbidelli et al . , 2016 ) .
in fact , after this time , the gas drifts inward faster than the radial displacement of the condensation lines .
so when the temperature drops at a given location the local gas should not carry any condensable species because that species would have already condensed farther out . concerning the third assumption , a warm disk with a snowline at @xmath29 au , as in the levison et al .
model , must be dominated by viscous heating in its inner parts ( see sect . [ disk ] ) , but viscously dominated disks are not flared ( kenyon and hartmann , 1987 ; chiang and goldreich , 1997 ; bitsch et al . , 2014 ) .
having a flared disk in the inner part would requires an ad - hoc radial - dependence of the dust / gas ratio ( levison et al . , 2015b ) .
nevertheless , the possibility that the solid mass was concentrated within 1 au should be taken seriously .
as discussed in sect .
[ planetesimals ] it is possible that the streaming instability formed a first generation of planetesimals in just two places : near the inner edge of the disk ( although it is unclear if this could extend out to 1 au ) and beyond the snowline .
if this was the case , obviously pebble accretion could only operate in these regions .
asteroids may have only formed later , as a low mass second - generation population , from the collisional debris of the first generation .
although this view is far from being established , it would explain the tri - modal structure of the solar system without the need for the grand tack .
not being able to decide a priori which view is the most realistic for the formation of the solar system , we can turn to observational constraints to discriminate them .
izidoro et al .
( 2015c ) pointed out that , in the absence of jupiter s migration , an initially low - mass asteroid belt should have remained dynamically cold , i.e. most asteroids should have today low eccentricity , low inclination orbits .
this contrasts with the observed distribution .
however , in a more recent work ( izidoro et al . ,
2016 ) it was found that , if the giant planets had a secular chaotic motion ( which might have happened , for instance , if jupiter and saturn spent some time in their mutual 1/2 resonance after the removal of the gas from the protoplanetary disk ) , the orbital distribution in the asteroid belt could have been excited by the stochastic stirring exerted by a multitude of secular resonances .
thus , the asteroid belt does not seem to be diagnostic of what really happened , given that its structure can be explained by both the grand tack model and by chaotic stirring ( fig .
4 ) . therefore
, we probably need to turn to the terrestrial planets for more insight .
the amount of water - rich material incorporated by the terrestrial planets in the two scenarios may be significantly different , although this has not been thoroughly explored .
another issue to investigate is the final orbital excitation of the terrestrial planets .
levison et al . ( 2015b ) investigated the growth of terrestrial planets from a local annulus of planetary embryos and planetesimals .
they find that the final terrestrial planet systems have angular momentum deficits ( amd : a mass - weighted mean of the eccentricities and inclinations of the planets ; laskar , 1997 ) that is 2 to 10 times that of the real terrestrial planets .
the reason for this is that the systems of planetary embryos deplete nearly all planetesimals before having any major orbital instability that allows embryo - embryo merging .
this leaves behind insufficient mass in planetesimals to significantly damp the orbits of the final planets ( obrien et al 2006 ; raymond et al 2006 ) .
if this is confirmed by future studies including also the ejection of debris in embryo - embryo collisions , it could become a strong argument in favor of the grand tack hypothesis ( or another mechanism that injects planetesimals into the inner disk ) .
in fact , during the inward migration of jupiter in the grand tack model , a large number of planetesimals are transported from the asteroid belt into the terrestrial planet region ( walsh et al .
this repopulates the terrestrial planet region of planetesimals , which then damp the eccentricities and inclinations of the forming planets by dynamical friction , allowing the final system to meet the amd constraint ( jacobson and morbidelli , 2014 ) .
so - called `` super - earths '' are the most abundant class of planets known to date .
exoplanet searches have found that roughly half of all sun - like stars have one or more planets with radii of 1 - 4 earth radii and orbital periods shorter than @xmath17100 days ( mayor et al 2011 ; howard et al 2012 ; fressin et al 2013 ; petigura et al 2013 ) .
what can we learn about planet formation from the population of super - earths ?
( for simplicity , in this discussion we refer to all exoplanets smaller than @xmath30 and with orbital periods shorter than 100 days as _ super - earths_. ) super - earths are often found in multiple systems , usually in compact orbital configurations with low eccentricities and low mutual inclinations ( lissauer et al 2011 , fabrycky et al 2014 ; tremaine & dong 2012 ; fang & margot 2012 ) . their orbital separations measured in orbital period ratio or in mutual hill radii are similar to those among the solar system s terrestrial planets ( fang & margot 2013 ) .
it is thus tempting to imagine that these planets formed by the same processes as the terrestrial planets , via in - situ growth of rocky planetesimals or planetary embryos . invoking the ad - hoc existence of a large population of embryos close to their stars can indeed reproduce some properties of the observed super - earth systems ( hansen & murray 2013 ) .
however , the in - situ accretion hypothesis fails for several reasons .
first , if super - earths formed locally , then one can construct a `` minimum - mass extra - solar nebula '' ( mmen ) to understand the distribution of solids close to the star ( raymond et al 2008 ; chiang & laughlin 2013 ; note that kuchner , 2004 built an analogous nebula using close - in giant exoplanets ) .
but the mmen is both extremely dense ( 1 - 2 orders of magnitude more massive than the nebula inferred from the solar system or from observations of disks around young stars ; andrews et al 2009 ; williams & cieza 2011 ) and inconsistent among different systems , with a broad range of surface density profiles ( raymond & cossou 2014 ; schlichting , 2015 ) .
it is difficult to reconcile such a nebula with the structure of an actual protoplanetary disk .
second , whereas migration of planetary embryos had a limited effect for mars - mass planetary embryos in the primordial solar system , migration can not be ignored in the case of super - earths . given the super - earths higher masses and closer proximity to their stars ( which likely translates into higher gas densities ) , the timescales for orbital migration are extremely short ( ogihara et al 2015a ) .
in fact , in the in - situ accretion model the inner disk was so dense that even the timescale for in - spiraling of embryos due to aerodynamic drag was shorter than the disk lifetime ( inamdar & schlichting 2015 ) .
this means that , even if the super - earths formed in - situ , their orbits must have changed .
in other words , super - earths could not have formed in - situ in the strictest sense .
it is nevertheless possible that super - earths grew from solids that were transported inward . yet
the size scale of inward - transported solids remains debated .
moriarty and fischer ( 2015 ) proposed that super - earths formed from in - situ planetesimals which efficiently accreted pebbles flowing from the outer disk .
the model of chatterjee & tan ( 2014 ; see also chatterjee & tan 2015 and hu et al 2016 ) instead does not require the pre - existence of planetesimals in the inner disk , but proposes that pebbles drifted inward and were trapped at the inner boundary of the dead zone ( fig . 5 ) . when enough mass was trapped , gravitational instability was triggered that led to the formation of a planet .
the dead zone receded , pebbles were trapped farther out at this new location , and the process continued .
the pebble drift model is appealing in its simplicity yet has only been explored in simple analytical terms .
while promising , the actual trapping , growth , and subsequent orbital migration of super - earths by this mechanism needs further study .
an alternate model proposes that the inward transport occurred at larger size scales via proto - planet migration ( terquem & papaloizou 2007 ; ogihara and ida , 2009 ; mcneil & nelson 2010 ; ida & lin 2010 ; cossou et al 2014 ) . see fig .
5 . planetary embryos similar to those that built the cores of the giant planets formed at orbital radii of a few to a few tens of au .
embryos migrated inward until they reached the inner edge of the disk .
embryos piled up and their orbits formed a chain of mean motion resonances ( sometimes with a planet pushed interior to the disk s inner edge ) . when the gas disk dissipated ( and its dissipative effects disappeared ) , some of the resonant chain destabilized , particularly those where the number of planets of the resonant chain was larger than a threshold value ( matsumoto et al .
these destabilized systems suffered a late stage phase of giant collisions between embryos .
with very simple assumptions the surviving planets quantitatively match the observed distributions ( cossou et al .
, 2014 ; ogihara et al .
this also predicts that some systems of super - earths those that avoided late instabilities should survive in resonant chains such as the recently - identified 4-planet resonant chain in the kepler-223 system ( mills et al .
2016 ) .
while effective and simple , the migration model is built on two other models that have their own uncertainties .
first , it relies on the rapid formation of a population of planetary embryos ( see lambrechts & johansen 2014 ; morbidelli et al 2015 ) .
the formation of super - earths in the context of realistic pebble accretion of embryos has not been simulated self - consistently to date .
second , the migration model is built on the current generation of migration maps ( e.g. , bitsch et al 2015 ) , which are themselves at the mercy of the structure and evolution of the disk ( see sections 2 and 3 ) .
this illustrates the interdependence of the different problems in planet formation .
we now turn our attention to the question of similarities and differences between super - earths and the solar system s terrestrial planets .
we have access to a plethora of data regarding the terrestrial planets including exquisite orbital constraints on the inner and outer planetary system , geochemical measurements ( for earth and mars ) , seismological data ( for earth ) , and a host of other information .
while we can imagine extrapolating our understanding of our solar system , we can not hope to obtain similar data for planets around other stars in a forseable future .
the key question remains : in making an analogy between our terrestrial planets and super - earths , would we be comparing apples with apples ?
regardless of how they form , super - earths grow much faster than earth .
earth s accretion concluded ( i.e. , it reached @xmath1799.5% of its final mass ) with a final giant impact that is thought to have created the moon ( benz et al .
1986 ; canup & asphaug 2001 ) roughly 50 - 150 myr after the start of planet formation ( kleine et al 2009 ; jacobson et al 2014 ) .
accretion times close to the star are extremely short , and simulations show that earth - sized or larger planets grow within 0.1 - 1 myr ( lissauer 2007 ; raymond et al 2007 ; bolmont et al 2014 ; ogihara et al 2015a ) . and any migration must of course have taken place during the gaseous disk phase , which lasted a few myr ( haisch et al 2001 ; hillenbrand 2008 ; but see pfalzner et al 2014 for a different view ) .
the orbital spacing of super - earths is suggestive of a final phase of instability ( cossou et al 2014 ; pu & wu 2015 ) , which may have been triggered by the dissipation of the gaseous disk ( ogihara and ida , 2009 ) and would have concluded within a few million years .
raymond et al ( 2014 ) proposed a distinction between super - earths and `` giant embryos '' , the difference being that giant embryos would have been fully - formed while the gas was still present in the disk .
perhaps a clearer division would be between planets that become massive enough to migrate while the disk is still present and those that do not .
the vast majority of observed super - earths are massive enough that the inferred migration timescales are much shorter than the typical disk lifetime ( rogers et al 2011 ; swift et al 2013 ) .
however , a handful of systems host planets so small that their migration timescales are very long ( e.g. , the kepler-444 system ; campante et al 2015 ) . in principle , very small planets must have been built from solids that did not migrate ; rather , they either accreted in - situ or drifted inward as pebbles . the solar system s terrestrial planets were built from embryos that presumably did not migrate significantly , whereas super - earths grew from migrating embryos .
one might imagine that variations in the protoplanetary disk mass could put super - earths and the terrestrial planets in the same context .
low - mass disks should form small embryos that do not migrate much whereas high - mass disks should form larger embryos that migrate significantly ( e.g. , kokubo & ida 2002 ) .
perhaps the solar system simply formed from a lower - mass disk than most of the super - earth systems .
we can qualitatively test this idea using planets observed around different types of stars , because the disk mass is thought to scale ( to some degree ) with the stellar mass ( e.g. , williams & cieza 2011 ; see discussion in raymond et al 2007 ) .
indeed , mulders et al ( 2015 ) found that the population of super - earths orbiting m stars is statistically different than that orbiting fgk stars . as expected from our simple arguments , m stars host more small super - earths are fewer large ones compared with fgk stars ( with the division between `` small '' and large set at @xmath31 ) .
however , the m stars super - earths are still easily massive enough to have migrated .
if even m stars low - mass disks produce super - earths , then some other factor is needed to explain the inner solar system s very small embryos .
jupiter may be the key .
terrestrial embryos grew from pebbles that drifted inward through the disk , and the source of pebbles moved outward in time ( lambrechts & johansen 2014 ; morbidelli et al 2015 ) .
when jupiter s core reached @xmath32 it became a barrier to the inward flux of pebbles .
this starved the terrestrial embryos , leaving them too small to migrate .
if the disk had been more massive then the terrestrial embryos could have reached the migration mass faster . and in a low - mass disk without a jupiter , the inward - sweeping of the snowline
could have delivered more material to the terrestrial embryos as the disk evolved ( morbidelli et al 2016 ) . by these arguments ,
a system would need two conditions to be met to avoid forming migrating embryos .
first , the disk mass must be below a certain threshold to avoid forming migrating terrestrial embryos very quickly .
second , a gas giant must form exterior to the terrestrial zone early enough to block the pebble flux and starve the inner embryos .
of course , these conditions are linked because giant planet cores are also migrating embryos , and cores that end up at @xmath175 au may plausibly originate across the entire disk ( bitsch et al 2015b ; raymond et al 2016 ) .
observations remain difficult to interpret . by combining radial velocity and transit data ,
bulk density measurements have been made for roughly 100 super - earths ( see marcy et al 2014 ) .
most planets larger than 1.5 - 2 @xmath33 have low measured bulk densities and may thus be `` mini - neptunes '' rather than `` super - earths '' ( weiss & marcy 2014 ; lopez & fortney 2014 ; rogers 2015 ; wolfgang & lopez 2015 ) .
most planets smaller than @xmath34 have bulk densities expected of rocky planets ( dressing et al 2015 ; but exceptions exist e.g. , jontof - hutter et al 2014 ) . of course ,
given the significant error bars and the uncertainty in planetary building materials , the detailed compositions of even the best - measured super - earths remain poorly - constrained ( e.g. , adams et al 2008 ) .
the meaning of the transition between super - earths and mini - neptunes at @xmath35 ( @xmath36 ) remains unclear .
this is unlikely to mark the transition between migrating and non - migrating embryos because there is no clear correlation between orbital radius and planet size .
perhaps @xmath37 is simply the critical mass above which cores can accrete gas from the disk ( see , e.g. , ikoma et al 2001 ; hubickyj et al 2005 ) .
it is interesting to note that izidoro et al ( 2015b ) s simulations to reproduce the ice giants were by far the most successful if the icy building blocks were @xmath38 in mass . or perhaps this transition is related to the mass above which atmospheres are more easily retained during the late stage of giant impacts ( e.g. , lee et al 2014 ; inamdar & schlichting 2015 , 2016 ) .
this remains an active area of research .
we still have not addressed the question of why there are no super - earths in the solar system .
indeed , the absence of terrestrial planets interior to mercury remains a mystery in itself .
this absence as well as mars small mass can be explained if the initial distribution of embryos in the solar system only extended from roughly 0.7 to 1 au ( hansen , 2009 ) . while significant work has been devoted to understanding the origin of the mass depletion beyond 1 au
, relatively few studies have addressed the very inner solar system .
the absence of super - earths can be explained relatively naturally by the migration model .
imagine that the first large planetary embryo that formed migrated to a zero - torque location and grew into jupiter ( cossou et al 2014 ) , carving a radial gap in the disk .
the inward migration of more distant embryos was then blocked by jupiter ( izidoro et al 2015a ) .
uranus and neptune , and even saturn s core , may thus have been trapped in the outer solar system by jupiter , whereas in absence of jupiter these planets or their precursors might have conitnued to migrate into the inner solar system becoming close - in super - earths .
simulations of this process are indeed the first to successfully reproduce the relative masses of the ice giants ( izidoro et al 2015b ) .
what remains unclear are the conditions required for the first embryo to grow into jupiter before other embryos can migrate past .
indeed , simulations of the migration model over - produce systems of super - earths ( e.g. , cossou et al 2014 ) .
raymond et al ( 2016 ) proposed a hybrid pebble+migration model to explain the absence of terrestrial planets closer - in than mercury . in their model ,
jupiter s core grew quickly in the very inner solar system , presumably by accreting inward - drifting pebbles .
following the migration maps of bitsch et al ( 2015 ) , the core subsequently migrated back outward to several au . during its outward migration ,
the very inner solar system was swept clean by resonant shepherding ( tanaka and ida 1999 ; mandell et al 2007 ) .
simulations show that the sweeping mechanism breaks down at 0.5 - 1 au , thus clearing the region interior to mercury but not the venus - earth zone ( raymond et al 2016 ) . while provocative , this model is admittedly built on several assumptions ( pebble drift and outward migration , notably ) with associated uncertainties .
two recent models have proposed that the primordial solar system contained systems of super - earths that were later destroyed ( volk & gladman 2015 ; batygin & laughlin 2015 ) . however ,
each of these studies has significant limitations .
for instance , volk & gladman ( 2015 ) proposed that the solar system s super - earths were destabilized and underwent a series of catastrophic collisions that ground the planets to dust .
however , they did not model the actual collisions or collisional grinding , and a cursory examination of the relevant collisional parameters suggests that they were not in the supercatastrophic regime ( see genda et al 2012 ; leinhardt & stewart 2012 ) .
batygin & laughlin ( 2015 ) proposed that the inward - then - outward migration proposed by the grand tack model ( walsh et al 2011 ; pierens & raymond 2011 ) produced a giant pulse of collisional debris that drifted inward and pushed a set of super - earths onto the sun .
however , batygin & laughlin a ) did not consistently treat the accretion of collisional debris in the earth - venus zone , b ) did not take collisional evolution / growth into account as the debris drifted inward , and c ) ignored the fact that disks have inner edges which are inferred from observations ( e.g. , eisner et al 2005 ) and naturally produced in models of gas accretion onto stars ( e.g. , romanova et al 2013 ) . taking these limitations into account , we do not find either of these studies convincing .
( for a more detailed critique of these studies we refer the reader to section 3.1 of raymond et al ( 2016 ) . )
yet while the proposed mechanisms are not compelling , it remains a possibility that the solar system once contained a system of super - earths .
what is missing are a viable destruction mechanism and , ideally , an observational signature .
just like any individual person , the solar system has its own history .
the probability of any other planetary system following an identical blueprint is zero . but
how typical was the solar system s evolutionary path ? _ ( did it go to college , get married and get a normal job ? ) _ or was the solar system s path unusual in some way ? _ ( did it quit high school to teach scuba diving in belize ? ) _ jupiter is the only solar system planet within the reach of current observations .
based on statistically - sound exoplanet observational surveys , the sun - jupiter system is special at roughly the level of one in a thousand .
first , the sun is an unusually massive star ; the most common type of star are m dwarfs , with masses of 10 - 50% of the sun s .
the sun is unusual at the @xmath1710% level , depending on the definition of `` sun - like '' and the assumed stellar initial mass function .
second , only @xmath1710% of sun - like stars have gas giant planets with orbits shorter than a few to 10 au ( cumming et al 2008 ; mayor et al 2011 ) .
third , only about 10% of giant exoplanets have orbits wider than 1 au and eccentricities smaller than 0.1 . the median eccentricity is @xmath170.25 ( butler et al 2006 ; udry & santos 2007 ) . taken together ,
these constraints suggest the sun - jupiter system is a 0.1% case .
of course , the situation is more complex than described here , as there exist correlations between a number of relevant parameters such as between the stellar mass and the presence of gas giants ( johnson et al 2007 ; lovis & mayor 2007 ) and between the orbital radius and eccentricity ( e.g. , butler et al 2006 ; ford & rasio 2008 ) . the numbers quoted here are a simple order of magnitude , but they clearly illustrate that the solar system is not a typical case in at least one regard : the presence and orbit of jupiter . currently - favored theoretical modeling likewise favors jupiter as the key player in the solar system s evolution .
the grand tack model invokes a two phase migration history for jupiter ( walsh et al 2011 ) .
izidoro et al ( 2015a ) proposed that jupiter held back the inward migration of uranus and neptune and stopped them from become hot super - earths .
assorted other models invoke jupiter s growth ( turrini et al 2012 ) , dynamics ( izidoro et al 2016 ) or migration ( batygin & laughlin 2015 ) to explain the properties of the solar system .
there remain considerable uncertainties at each step of planet formation .
that is the theme of this review : even our most successful models are built on a shaky foundation .
looking to the future , the most important steps forward may be those that show how planet do _ not _ form .
we acknowledge support by the french anr , project number anr-1313-bs05 - 0003 - 01 projet mojo ( modeling the origin of jovian planets ) . this review paper does not report any new data .
all data reported in the figures or in the text are derived from the cited sources .
* adams , e. r. , seager , s. , elkins - tanton , l. 2008 .
ocean planet or thick atmosphere : on the mass - radius relationship for solid exoplanets with massive atmospheres .
the astrophysical journal 673 , 1160 - 1164 . * alexander , r. , pascucci , i. , andrews , s. , armitage , p. , cieza , l. 2014 .
the dispersal of protoplanetary disks .
protostars and planets vi 475 - 496 . *
alibert , y. , mordasini , c. , benz , w. , winisdoerffer , c. 2005 .
models of giant planet formation with migration and disc evolution .
astronomy and astrophysics 434 , 343 - 353 . * andrews , s. m. , wilner , d. j. , hughes , a. m. , qi , c. , dullemond , c. p. 2009 .
protoplanetary disk structures in ophiuchus .
the astrophysical journal 700 , 1502 - 1523 . * armitage , p. j. 2011 .
dynamics of protoplanetary disks
. annual review of astronomy and astrophysics 49 , 195 - 236 .
* asphaug , e. , jutzi , m. , movshovitz , n. 2011 .
chondrule formation during planetesimal accretion .
earth and planetary science letters 308 , 369 - 379 .
* bai , x .-
wind - driven accretion in protoplanetary disks .
ii . radial dependence and global picture . the astrophysical journal 772 , 96 .
* bai , x .-
n . , stone , j. m. 2013 .
wind - driven accretion in protoplanetary disks .
i. suppression of the magnetorotational instability and launching of the magnetocentrifugal wind . the astrophysical journal 769 , 76 . * bai , x .-
. 2014 .
hall - effect - controlled gas dynamics in protoplanetary disks . i. wind solutions at the inner disk .
the astrophysical journal 791 , 137 . * bai , x .-
. 2015 . hall effect controlled gas dynamics in protoplanetary disks .
ii . full 3d simulations toward the outer disk .
the astrophysical journal 798 , 84 . * bai , x .-
. 2016 . towards a global evolutionary model of protoplanetary disks . the astrophysical journal 821 , 80 . * balbus , s. a. , hawley , j. f. 1991
. a powerful local shear instability in weakly magnetized disks .
i - linear analysis .
ii - nonlinear evolution . the astrophysical journal 376 , 214 - 233 . * balbus , s. a. , papaloizou , j. c. b. 1999 . on the dynamical foundations of @xmath4 disks
the astrophysical journal 521 , 650 - 658 .
* batygin , k. , laughlin , g. 2015 .
jupiter s decisive role in the inner solar system s early evolution .
proceedings of the national academy of science 112 , 4214 - 4217 .
* bentez - llambay , p. , masset , f. , koenigsberger , g. , szulgyi , j. 2015 .
planet heating prevents inward migration of planetary cores .
nature 520 , 63 - 65 .
* benz , w. , slattery , w. l. , cameron , a. g. w. 1986 .
the origin of the moon and the single - impact hypothesis .
i. icarus 66 , 515 - 535 . *
birnstiel , t. , andrews , s. m. 2014 . on the outer edges of protoplanetary dust disks .
the astrophysical journal 780 , 153 .
* birnstiel , t. , fang , m. , johansen , a. 2016 .
dust evolution and the formation of planetesimals .
arxiv e - prints arxiv:1604.02952 .
* bitsch , b. , kley , w. 2010 .
orbital evolution of eccentric planets in radiative discs .
astronomy and astrophysics 523 , a30 . * bitsch , b. , morbidelli , a. , lega , e. , crida , a. 2014 .
stellar irradiated discs and implications on migration of embedded planets .
accreting - discs . astronomy and astrophysics 564 , a135 . * bitsch , b. , morbidelli , a. , lega , e. , kretke , k. , crida , a. 2014b .
stellar irradiated discs and implications on migration of embedded planets .
viscosity transitions .
astronomy and astrophysics 570 , a75 . * bitsch , b. , johansen , a. , lambrechts , m. , morbidelli , a. 2015 . the structure of protoplanetary discs around evolving young stars .
astronomy and astrophysics 575 , a28 . * bitsch , b. , lambrechts , m. , johansen , a. 2015b . the growth of planets by pebble accretion in evolving protoplanetary discs .
astronomy and astrophysics 582 , a112 . * blum , j. , wurm , g. 2008 . the growth mechanisms of macroscopic bodies in protoplanetary disks
. annual review of astronomy and astrophysics 46 , 21 - 56 .
* bolmont , e. , raymond , s. n. , von paris , p. , selsis , f. , hersant , f. , quintana , e. v. , barclay , t. 2014 .
formation , tidal evolution , and habitability of the kepler-186 system . the astrophysical journal 793 , 3 . * bottke , w. f. , nesvorn , d. , grimm , r. e. , morbidelli , a. , obrien , d. p. 2006 .
iron meteorites as remnants of planetesimals formed in the terrestrial planet region .
nature 439 , 821 - 824 .
* bouvier , a. , blichert - toft , j. , moynier , f. , vervoort , j.d . ,
albarede , f. 2007 .
pb - pb dating constraints on the accretion and cooling history of chondrites .
geochimica et cosmochimica acta 71 , 1583 - 1604 . * bouvier , j. , and 10 colleagues 2007 .
magnetospheric accretion - ejection processes in the classical t tauri star aa tauri .
astronomy and astrophysics 463 , 1017 - 1028 . * burkhardt , c. , kleine , t. , bourdon , b. , palme , h. , zipfel , j. , friedrich , j.m . ,
ebel , d.s .
hf - w mineral isochron for ca , al - rich inclusions : age of the solar system and the timing of core formation in planetesimals . geochimica et cosmochimica acta 72 , 6177 - 6197 .
* butler , r. p. , johnson , j. a. , marcy , g. w. , wright , j. t. , vogt , s. s. , fischer , d. a. 2006 . a long - period jupiter - mass planet orbiting the nearby m dwarf gj 849 .
publications of the astronomical society of the pacific 118 , 1685 - 1689 .
* campante , t. l. , and 40 colleagues 2015 .
an ancient extrasolar system with five sub - earth - size planets . the astrophysical journal 799 , 170 . * canup , r. m. , asphaug , e. 2001 . origin of the moon in a giant impact near the end of the earth s formation .
nature 412 , 708 - 712 .
* carrera , d. , johansen , a. , davies , m. b. 2015 . how to form planetesimals from mm - sized chondrules and chondrule aggregates .
astronomy and astrophysics 579 , a43 . * chatterjee , s. , tan , j. c. 2014 . inside - out planet formation .
the astrophysical journal 780 , 53 . * chatterjee , s. , tan , j. c. 2015 .
vulcan planets : inside - out formation of the innermost super - earths .
the astrophysical journal 798 , l32 . * chiang , e. i. , goldreich , p. 1997 .
spectral energy distributions of t tauri stars with passive circumstellar disks . the astrophysical journal 490 , 368 - 376 . * chiang , e. , youdin , a. n. 2010 .
forming planetesimals in solar and extrasolar nebulae .
annual review of earth and planetary sciences 38 , 493 - 522 . * chiang , e. , laughlin , g. 2013 .
the minimum - mass extrasolar nebula : in situ formation of close - in super - earths .
monthly notices of the royal astronomical society 431 , 3444 - 3455 .
* coleman , g. a. l. , nelson , r. p. 2014 . on the formation of planetary systems via oligarchic growth in thermally evolving viscous discs .
monthly notices of the royal astronomical society 445 , 479 - 499 .
* coleman , g. a. l. , nelson , r. p. 2016 .
giant planet formation in radially structured protoplanetary discs .
monthly notices of the royal astronomical society 460 , 2779 - 2795 .
* cossou , c. , raymond , s. n. , hersant , f. , pierens , a. 2014 .
hot super - earths and giant planet cores from different migration histories .
astronomy and astrophysics 569 , a56 .
* crida , a. , morbidelli , a. , masset , f. 2006 . on the width and shape of gaps in protoplanetary disks .
icarus 181 , 587 - 604 . * cumming , a. , butler , r. p. , marcy , g. w. , vogt , s. s. , wright , j. t. , fischer , d. a. 2008 .
the keck planet search : detectability and the minimum mass and orbital period distribution of extrasolar planets .
publications of the astronomical society of the pacific 120 , 531 - 554 . *
cuzzi , j. n. , hogan , r. c. , paque , j. m. , dobrovolskis , a. r. 2001 . size - selective concentration of chondrules and other small particles in protoplanetary nebula turbulence .
the astrophysical journal 546 , 496 - 508 .
* cuzzi , j. n. , hogan , r. c. , bottke , w. f. 2010 . towards initial mass functions for asteroids and kuiper belt objects .
icarus 208 , 518 - 538 . * cuzzi , j. n. , hartlep , t. , estrada , p. r. 2016 .
planetesimal initial mass functions and creation rates under turbulent concentration using scale - dependent cascades .
lunar and planetary science conference 47 , 2661 .
* dauphas , n. , pourmand , a. 2011 .
hf - w - th evidence for rapid growth of mars and its status as a planetary embryo .
nature 473 , 489 - 492 .
* deienno , r. , gomes , r. s. , walsh , k. j. , morbidelli , a. , nesvorn , d. 2016 . is the grand tack model compatible with the orbital distribution of main belt asteroids ? .
icarus 272 , 114 - 124 . * dominik , c. , tielens , a. g. g. m. 1997 .
the physics of dust coagulation and the structure of dust aggregates in space .
the astrophysical journal 480 , 647 - 673 . * drakowska , j. , dullemond , c. p. 2014 . can dust coagulation trigger streaming instability ? .
astronomy and astrophysics 572 , a78 . * drakowska , j. , alibert , y. and moore , b. 2016 .
close - in planetesimal formation by pile - up of drifting pebbles . astron . and
astrophys , in press . * dressing , c. d. , charbonneau , d. , newton , e. r. 2015
. the occurrence rate and composition of small planets orbiting small stars .
aas / division for extreme solar systems abstracts 3 , 501.03 . *
duffell , p. c. , haiman , z. , macfadyen , a. i. , dorazio , d. j. , farris , b. d. 2014 .
the migration of gap - opening planets is not locked to viscous disk evolution .
the astrophysical journal 792 , l10 .
* dullemond , c. p. , dominik , c. , natta , a. 2001 . passive irradiated circumstellar disks with an inner hole .
the astrophysical journal 560 , 957 - 969 . * dullemond , c. p. , van zadelhoff , g. j. , natta , a. 2002 . vertical structure models of t tauri and herbig ae / be disks .
astronomy and astrophysics 389 , 464 - 474 .
* drmann , c. , kley , w. 2015 .
migration of massive planets in accreting disks .
astronomy and astrophysics 574 , a52 . * eisner , j. a. , hillenbrand , l. a. , white , r. j. , akeson , r. l. , sargent , a. i. 2005 .
observations of t tauri disks at sub - au radii : implications for magnetospheric accretion and planet formation .
the astrophysical journal 623 , 952 - 966 .
* fabrycky , d. c. , and 21 colleagues 2014 .
architecture of kepler s multi - transiting systems .
ii . new investigations with twice as many candidates .
the astrophysical journal 790 , 146 .
* fang , j. , margot , j .-
architecture of planetary systems based on kepler data : number of planets and coplanarity .
the astrophysical journal 761 , 92 .
* fang , j. , margot , j .-
are planetary systems filled to capacity ?
a study based on kepler results .
the astrophysical journal 767 , 115 .
* ford , e. b. , rasio , f. a. 2008 .
origins of eccentric extrasolar planets : testing the planet - planet scattering model .
the astrophysical journal 686 , 621 - 636 .
* fressin , f. , torres , g. , charbonneau , d. , bryson , s. t. , christiansen , j. , dressing , c. d. , jenkins , j. m. , walkowicz , l. m. , batalha , n. m. 2013 . the false positive rate of kepler and the occurrence of planets . the astrophysical journal 766 , 81 . * fung , j. , artymowicz , p. , wu , y. 2015 . the 3d flow field around an embedded planet . the astrophysical journal 811 , 101 . * genda , h. , kokubo , e. , ida , s. 2012 .
merging criteria for giant impacts of protoplanets .
the astrophysical journal 744 , 137 .
* goldreich , p. , tremaine , s. 1980 .
disk - satellite interactions .
the astrophysical journal 241 , 425 - 441 .
* gressel , o. , turner , n. j. , nelson , r. p. , mcnally , c. p. 2015 .
global simulations of protoplanetary disks with ohmic resistivity and ambipolar diffusion .
the astrophysical journal 801 , 84 .
* guillot , t. , ida , s. , ormel , c. w. 2014 . on the filtering and processing of dust by planetesimals .
i. derivation of collision probabilities for non - drifting planetesimals .
astronomy and astrophysics 572 , a72 . * gttler , c. , blum , j. , zsom , a. , ormel , c. w. , dullemond , c. p. 2009 . the first phase of protoplanetary dust growth : the bouncing barrier .
geochimica et cosmochimica acta supplement 73 , 482 . * hansen , b. m. s. 2009 .
formation of the terrestrial planets from a narrow annulus . the astrophysical journal 703 , 1131 - 1140 . * hansen , b. m. s. , murray , n. 2013 .
testing in situ assembly with the kepler planet candidate sample .
the astrophysical journal 775 , 53 .
* haisch , k. e. , jr . , lada , e. a. , lada , c. j. 2001 .
disk frequencies and lifetimes in young clusters .
the astrophysical journal 553 , l153-l156 .
* hasegawa , y. , pudritz , r. e. 2011 . the origin of planetary system architectures - i. multiple planet traps in gaseous discs .
monthly notices of the royal astronomical society 417 , 1236 - 1259 .
* hellary , p. , nelson , r. p. 2012 .
global models of planetary system formation in radiatively - inefficient protoplanetary discs .
monthly notices of the royal astronomical society 419 , 2737 - 2757 . * helled , r. , bodenheimer , p. , podolak , m. , boley , a. , meru , f. , nayakshin , s. , fortney , j. j. , mayer , l. , alibert , y. , boss , a. p. 2014 .
giant planet formation , evolution , and internal structure .
protostars and planets vi 643 - 665 . *
hillenbrand , l. a. 2008 .
disk - dispersal and planet - formation timescales .
physica scripta volume t 130 , 014024 . * howard , a. w. , and 66 colleagues 2012 .
planet occurrence within 0.25 au of solar - type stars from kepler .
the astrophysical journal supplement series 201 , 15 . *
hu , x. , zhu , z. , tan , j. c. , chatterjee , s. 2016 . inside - out planet formation .
planet - disk interaction at the dead zone inner boundary .
the astrophysical journal 816 , 19 . * hubickyj , o. , bodenheimer , p. , lissauer , j. j. 2005 .
accretion of the gaseous envelope of jupiter around a 5 10 earth - mass core .
icarus 179 , 415 - 431 . * ida , s. , lin , d. n. c. 2004a . toward a deterministic model of planetary formation . i. a desert in the mass and semimajor axis distributions of extrasolar planets . the astrophysical journal 604 , 388 - 413 . * ida , s. , lin , d. n. c. 2004b .
toward a deterministic model of planetary formation .
ii . the formation and retention of gas giant planets around stars with a range of metallicities . the astrophysical journal 616 , 567 - 572 . *
ida , s. , lin , d. n. c. 2010 . toward a deterministic model of planetary formation .
dynamical interaction and coagulation of multiple rocky embryos and super - earth systems around solar - type stars .
the astrophysical journal 719 , 810 - 830 . * ida , s. , guillot , t. , morbidelli , a. 2016 . the radial dependence of pebble accretion rates : a source of diversity in planetary systems i. analytical formulation .
astronomy and astrophysics 591 , a72 . * ikoma , m. , emori , h. , nakazawa , k. 2001 .
formation of giant planets in dense nebulae : critical core mass revisited .
the astrophysical journal 553 , 999 - 1005 .
* inamdar , n. k. , schlichting , h. e. 2015 .
the formation of super - earths and mini - neptunes with giant impacts .
monthly notices of the royal astronomical society 448 , 1751 - 1760 .
* inamdar , n. k. , schlichting , h. e. 2016 . stealing the gas : giant impacts and the large diversity in exoplanet densities . the astrophysical journal 817 , l13 . * izidoro , a. , raymond , s. n. , morbidelli , a. , hersant , f. , pierens , a. 2015a .
gas giant planets as dynamical barriers to inward - migrating super - earths .
the astrophysical journal 800 , l22 .
* izidoro , a. , morbidelli , a. , raymond , s. n. , hersant , f. , pierens , a. 2015b .
accretion of uranus and neptune from inward - migrating planetary embryos blocked by jupiter and saturn .
astronomy and astrophysics 582 , a99 . * izidoro , a. , raymond , s. n. , morbidelli , a. , winter , o. c. 2015c .
terrestrial planet formation constrained by mars and the structure of the asteroid belt .
monthly notices of the royal astronomical society 453 , 3619 - 3634 .
* izidoro , a. , raymond , s.n . ,
pierens , a. , morbidelli , a. , winter , o.c . ,
nesvorny , d. , 2016 .
structure of the asteroid belt from the gas giants 1 growth and chaotic dynamics .
science advances , in press .
arxiv e - prints arxiv:1609.04970 .
* jacobson , s. a. , morbidelli , a. 2014 .
lunar and terrestrial planet formation in the grand tack scenario .
philosophical transactions of the royal society of london series a 372 , 0174 . * jacobson , s. a. , morbidelli , a. , raymond , s. n. , obrien , d. p. , walsh , k. j. , rubie , d. c. 2014 .
highly siderophile elements in earth s mantle as a clock for the moon - forming impact .
nature 508 , 84 - 87 . * johansen , a. , henning , t. , klahr , h. 2006 .
dust sedimentation and self - sustained kelvin - helmholtz turbulence in protoplanetary disk midplanes .
the astrophysical journal 643 , 1219 - 1232 . * johansen , a. , oishi , j. s. , mac low , m .-
m . , klahr , h. , henning , t. , youdin , a. 2007 .
rapid planetesimal formation in turbulent circumstellar disks .
nature 448 , 1022 - 1025 . * johansen , a. , youdin , a. , mac low , m .-
particle clumping and planetesimal formation depend strongly on metallicity . the astrophysical journal 704 , l75-l79 . * johansen , a. , lacerda , p. 2010 .
prograde rotation of protoplanets by accretion of pebbles in a gaseous environment .
monthly notices of the royal astronomical society 404 , 475 - 485 . * johansen , a. , blum , j. , tanaka , h. , ormel , c. , bizzarro , m. , rickman , h. 2014 . the multifaceted planetesimal formation process .
protostars and planets vi 547 - 570 . * johansen , a. , mac low , m .-
m . , lacerda , p. , bizzarro , m. 2015 .
growth of asteroids , planetary embryos , and kuiper belt objects by chondrule accretion .
science advances 1 , 1500109 . *
johnson , j. a. , butler , r. p. , marcy , g. w. , fischer , d. a. , vogt , s. s. , wright , j. t. , peek , k. m. g. 2007 . a new planet around an m dwarf : revealing a correlation between exoplanets and stellar mass .
the astrophysical journal 670 , 833 - 840 . * johnson , b. c. , minton , d. a. , melosh , h. j. , zuber , m. t. 2015 .
impact jetting as the origin of chondrules .
nature 517 , 339 - 341 .
* jontof - hutter , d. , lissauer , j. j. , rowe , j. f. , fabrycky , d. c. 2014 .
kepler-79 s low density planets . the astrophysical journal 785 , 15 .
* kataoka , a. , tanaka , h. , okuzumi , s. , wada , k. 2013 .
fluffy dust forms icy planetesimals by static compression .
astronomy and astrophysics 557 , l4 . * kenyon , s. j. , hartmann , l. 1987 .
spectral energy distributions of t tauri stars - disk flaring and limits on accretion .
the astrophysical journal 323 , 714 - 733 . * klahr , h. h. , bodenheimer , p. 2003 .
turbulence in accretion disks : vorticity generation and angular momentum transport via the global baroclinic instability . the astrophysical journal 582 , 869 - 892 . * kleine , t. , touboul , m. , bourdon , b. , nimmo , f. , mezger , k. , palme , h. , jacobsen , s. b. , yin , q .- z . ,
halliday , a. n. 2009 .
hf - w chronology of the accretion and early evolution of asteroids and terrestrial planets .
geochimica et cosmochimica acta 73 , 5150 - 5188 . * kokubo , e. , ida , s. 2002 .
formation of protoplanet systems and diversity of planetary systems .
the astrophysical journal 581 , 666 - 680 .
* kretke , k. a. , lin , d. n. c. 2007 . grain retention and formation of planetesimals near the snow line in mri - driven turbulent protoplanetary disks . the astrophysical journal 664 , l55-l58 .
* kruijer , t. s. , sprung , p. , kleine , t. , leya , i. , burkhardt , c. , wieler , r. 2012 .
hf - w chronometry of core formation in planetesimals inferred from weakly irradiated iron meteorites .
geochimica et cosmochimica acta 99 , 287 - 304 .
* kuchner , m. j. 2004 . a minimum - mass extrasolar nebula . the astrophysical journal 612 , 1147 - 1151 .
* lambrechts , m. , johansen , a. 2012 . rapid growth of gas - giant cores by pebble accretion .
astronomy and astrophysics 544 , a32 . * lambrechts , m. , johansen , a. 2014 . forming the cores of giant planets from the radial pebble flux in protoplanetary discs .
astronomy and astrophysics 572 , a107 . * laskar , j. 1997 .
large scale chaos and the spacing of the inner planets .. astronomy and astrophysics 317 , l75-l78 . * laughlin , g. , adams , f. c. 1997 .
possible stellar metallicity enhancements from the accretion of planets .
the astrophysical journal 491 , l51-l54 .
* lee , e. j. , chiang , e. , ormel , c. w. 2014 .
make super - earths , not jupiters : accreting nebular gas onto solid cores at 0.1 au and beyond . the astrophysical journal 797 , 95 . * leinhardt , z. m. , stewart , s. t. 2012 .
collisions between gravity - dominated bodies . i. outcome regimes and scaling laws . the astrophysical journal 745 , 79 .
* lesur , g. , kunz , m. w. , fromang , s. 2014 .
thanatology in protoplanetary discs
. the combined influence of ohmic , hall , and ambipolar diffusion on dead zones .
astronomy and astrophysics 566 , a56 . * levison , h. f. , bottke , w. f. , gounelle , m. , morbidelli , a. , nesvorn , d. , tsiganis , k. 2009 .
contamination of the asteroid belt by primordial trans - neptunian objects .
nature 460 , 364 - 366 . * levison , h. f. , kretke , k. a. , duncan , m. j. 2015 . growing the gas - giant planets by the gradual accumulation of pebbles .
nature 524 , 322 - 324 . * levison , h. f. , kretke , k. a. , walsh , k. j. , bottke , w. f. 2015b . growing the terrestrial planets from the gradual accumulation of sub - meter sized objects .
proceedings of the national academy of science 112 , 14180 - 14185 . * libourel , g. , krot , a. n. , chaussidon , m. 2006 .
olivines in magnesian porphyritic chondrules : mantle material of earlier generations of differentiated planetesimals ? .
meteoritics and planetary science supplement 41 , 5295 . *
lin , d. n. c. , papaloizou , j. 1986 . on the tidal interaction between protoplanets and the protoplanetary disk .
iii - orbital migration of protoplanets . the astrophysical journal 309 , 846 - 857 . *
lin , d. n. c. , bodenheimer , p. , richardson , d. c. 1996 .
orbital migration of the planetary companion of 51 pegasi to its present location .
nature 380 , 606 - 607 .
* lin , d. n. c. 1997 .
planetary formation in protostellar disks .
iau colloq . 163 : accretion phenomena and related outflows 121 , 321 .
* lissauer , j. j. 2007 .
planets formed in habitable zones of m dwarf stars probably are deficient in volatiles . the astrophysical journal 660 , l149-l152 . * lissauer , j. j. , and 24 colleagues 2011 .
architecture and dynamics of kepler s candidate multiple transiting planet systems .
the astrophysical journal supplement series 197 , 8 .
* lodders , k. 2003 . solar system abundances and condensation temperatures of the elements .
the astrophysical journal 591 , 1220 - 1247 . * lopez , e. d. , fortney , j. j. 2014 . understanding the mass - radius relation for sub - neptunes : radius as a proxy for composition . the astrophysical journal 792 , 1 . * lovis , c. , mayor , m. 2007 . planets around evolved intermediate - mass stars .
i. two substellar companions in the open clusters ngc 2423 and ngc 4349 .
astronomy and astrophysics 472 , 657 - 664 . *
lynden - bell , d. , pringle , j. e. 1974 . the evolution of viscous discs and the origin of the nebular variables .. monthly notices of the royal astronomical society 168 , 603 - 637 . * mandell , a. m. , raymond , s. n. , sigurdsson , s. 2007 .
formation of earth - like planets during and after giant planet migration .
the astrophysical journal 660 , 823 - 844 . * marcus , p. s. , pei , s. , jiang , c .- h . ,
barranco , j. a. , hassanzadeh , p. , lecoanet , d. 2015 .
zombie vortex instability .
i. a purely hydrodynamic instability to resurrect the dead zones of protoplanetary disks . the astrophysical journal 808 , 87 .
* marcy , g. w. , and 102 colleagues 2014 .
masses , radii , and orbits of small kepler planets : the transition from gaseous to rocky planets . the astrophysical journal supplement series 210 , 20 .
* martin , r. g. , lubow , s. h. , livio , m. , pringle , j. e. 2012 .
dead zones around young stellar objects : fu orionis outbursts and transition discs .
monthly notices of the royal astronomical society 423 , 2718 - 2725 .
* masset , f. , snellgrove , m. 2001 .
reversing type ii migration : resonance trapping of a lighter giant protoplanet .
monthly notices of the royal astronomical society 320 , l55-l59 .
* masset , f. s. , morbidelli , a. , crida , a. , ferreira , j. 2006 .
disk surface density transitions as protoplanet traps .
the astrophysical journal 642 , 478 - 487 . * matsumoto , y. , nagasawa , m. , ida , s. 2012 .
the orbital stability of planets trapped in the first - order mean - motion resonances .
icarus 221 , 624 - 631 . * mayor , m. , queloz , d. 1995 . a jupiter - mass companion to a solar - type star .
nature 378 , 355 - 359 . * mayor , m. , and 13 colleagues 2011 . the harps search for southern extra - solar planets xxxiv .
occurrence , mass distribution and orbital properties of super - earths and neptune - mass planets .
arxiv e - prints arxiv:1109.2497 .
* mcneil , d. s. , nelson , r. p. 2010 . on the formation of hot neptunes and super - earths .
monthly notices of the royal astronomical society 401 , 1691 - 1708 .
* mills , s. , fabrycky , d. c. , migaszewski , c. , ford , e. b. , petigura , e. , isaacson , h. t. 2016 .
kepler-223 : a resonant chain of four transiting , sub - neptune planets .
aas / division of dynamical astronomy meeting 47 , # 101.03 .
* morbidelli , a. , crida , a. 2007 .
the dynamics of jupiter and saturn in the gaseous protoplanetary disk .
icarus 191 , 158 - 171 . * morbidelli , a. , lunine , j. i. , obrien , d. p. , raymond , s. n. , walsh , k. j. 2012 . building terrestrial planets
. annual review of earth and planetary sciences 40 , 251 - 275 .
* morbidelli , a. , szulgyi , j. , crida , a. , lega , e. , bitsch , b. , tanigawa , t. , kanagawa , k. 2014 .
meridional circulation of gas into gaps opened by giant planets in three - dimensional low - viscosity disks .
icarus 232 , 266 - 270 . * morbidelli , a. , lambrechts , m. , jacobson , s. , bitsch , b. 2015 . the great dichotomy of the solar system : small terrestrial embryos and massive giant planet cores .
icarus 258 , 418 - 429 . * morbidelli , a. , bitsch , b. , crida , a. , gounelle , m. , guillot , t. , jacobson , s. , johansen , a. , lambrechts , m. , lega , e. 2016 .
fossilized condensation lines in the solar system protoplanetary disk .
icarus 267 , 368 - 376 . * moriarty , j. , fischer , d. 2015 .
building massive compact planetesimal disks from the accretion of pebbles .
the astrophysical journal 809 , 94 .
* mulders , g. d. , pascucci , i. , apai , d. 2015 .
an increase in the mass of planetary systems around lower - mass stars . the astrophysical journal 814 , 130 . * nelson , r. p. , gressel , o. , umurhan , o. m. 2013 .
linear and non - linear evolution of the vertical shear instability in accretion discs .
monthly notices of the royal astronomical society 435 , 2610 - 2632 .
* obrien , d. p. , morbidelli , a. , levison , h. f. 2006 .
terrestrial planet formation with strong dynamical friction .
icarus 184 , 39 - 58 . * ogihara , m. , ida , s. 2009 .
n - body simulations of planetary accretion around m dwarf stars .
the astrophysical journal 699 , 824 - 838 .
* ogihara , m. , morbidelli , a. , guillot , t. 2015a . a reassessment of the in situ formation of close - in super - earths .
astronomy and astrophysics 578 , a36 . * ogihara , m. , morbidelli , a. , guillot , t. 2015b . suppression of type i migration by disk winds .
astronomy and astrophysics 584 , l1 . * okuzumi , s. , tanaka , h. , kobayashi , h. , wada , k. 2012 .
rapid coagulation of porous dust aggregates outside the snow line : a pathway to successful icy planetesimal formation . the astrophysical journal 752 , 106 . * ogilvie , g. i. , lubow , s. h. 2002 .
on the wake generated by a planet in a disc .
monthly notices of the royal astronomical society 330 , 950 - 954 .
* ormel , c. w. , klahr , h. h. 2010 . the effect of gas drag on the growth of protoplanets .
analytical expressions for the accretion of small bodies in laminar disks .
astronomy and astrophysics 520 , a43 . * paardekooper , s .- j . ,
mellema , g. 2006 .
halting type i planet migration in non - isothermal disks .
astronomy and astrophysics 459 , l17-l20 . * paardekooper , s .- j . ,
baruteau , c. , crida , a. , kley , w. 2010 . a torque formula for non - isothermal type i planetary migration - i. unsaturated horseshoe drag . monthly notices of the royal astronomical society 401 , 1950 - 1964 .
* paardekooper , s .- j . ,
baruteau , c. , kley , w. 2011 . a torque formula for non - isothermal type i planetary migration - ii .
effects of diffusion .
monthly notices of the royal astronomical society 410 , 293 - 303 .
* petigura , e. a. , marcy , g. w. , howard , a. w. 2013 . a plateau in the planet population below twice the size of earth . the astrophysical journal 770 , 69 . * pfalzner , s. , steinhausen , m. , menten , k. 2014 .
short dissipation times of proto - planetary disks : an artifact of selection effects ? .
the astrophysical journal 793 , l34 .
* pierens , a. , nelson , r. p. 2008 .
constraints on resonant - trapping for two planets embedded in a protoplanetary disc .
astronomy and astrophysics 482 , 333 - 340 . * pierens , a. , raymond , s. n. 2011 . two phase , inward - then - outward migration of jupiter and saturn in the gaseous solar nebula . astronomy and astrophysics 533 , a131 .
* pollack , j. b. , hubickyj , o. , bodenheimer , p. , lissauer , j. j. , podolak , m. , greenzweig , y. 1996 .
formation of the giant planets by concurrent accretion of solids and gas .
icarus 124 , 62 - 85 . *
pringle , j. e. 1981 .
accretion discs in astrophysics . annual review of astronomy and astrophysics 19 , 137 - 162 .
* pu , b. , wu , y. 2015 .
spacing of kepler planets : sculpting by dynamical instability . the astrophysical journal 807 , 44 . *
raymond , s. n. , quinn , t. , lunine , j. i. 2006 .
high - resolution simulations of the final assembly of earth - like planets i. terrestrial accretion and dynamics .
icarus 183 , 265 - 282 .
* raymond , s. 2007 . formation and stability of hot earth planets .
bulletin of the american astronomical society 39 , 110.03 . * raymond , s. n. , scalo , j. , meadows , v. s. 2007 . a decreased probability of habitable planet formation around low - mass stars . the astrophysical journal 669 , 606 - 614 . * raymond , s. n. , barnes , r. , mandell , a. m. 2008 .
observable consequences of planet formation models in systems with close - in terrestrial planets .
monthly notices of the royal astronomical society 384 , 663 - 674 . *
raymond , s. n. , obrien , d. p. , morbidelli , a. , kaib , n. a. 2009 .
building the terrestrial planets : constrained accretion in the inner solar system .
icarus 203 , 644 - 662 .
* raymond , s. n. , kokubo , e. , morbidelli , a. , morishima , r. , walsh , k. j. 2014 . terrestrial planet formation at home and abroad .
protostars and planets vi 595 - 618 .
* raymond , s. n. , morbidelli , a. 2014 . the grand tack model : a critical review .
iau symposium 310 , 194 - 203 . * raymond , s. n. , cossou , c. 2014 .
no universal minimum - mass extrasolar nebula : evidence against in situ accretion of systems of hot super - earths .
monthly notices of the royal astronomical society 440 , l11-l15 . *
raymond , s. n. , izidoro , a. , bitsch , b. , jacobson , s. a. 2016 . did jupiter s core form in the innermost parts of the sun s protoplanetary disc ? .
monthly notices of the royal astronomical society 458 , 2962 - 2972 . * rogers , l. a. , bodenheimer , p. , lissauer , j. j. , seager , s. 2011 . formation and structure of low - density exo - neptunes .
the astrophysical journal 738 , 59 . * rogers , l. a. 2015 .
most 1.6 earth - radius planets are not rocky . the astrophysical journal 801 , 41 . * romanova , m. m. , ustyugova , g. v. , koldoba , a. v. , lovelace , r. v. e. 2013 .
warps , bending and density waves excited by rotating magnetized stars : results of global 3d mhd simulations .
monthly notices of the royal astronomical society 430 , 699 - 724 . * saito , e. , sirono , s .-
planetesimal formation by sublimation .
the astrophysical journal 728 , 20 . * sato , t. , okuzumi , s. , ida , s. 2016 . on the water delivery to terrestrial embryos by ice pebble accretion .
astronomy and astrophysics 589 , a15 . * schlichting , h. 2015 .
orbital architecture and mean motion resonances of multiple planet systems .
iau general assembly 22 , 2255741 . * shakura , n. i. , sunyaev , r. a. 1973 .
black holes in binary systems .
observational appearance .. astronomy and astrophysics 24 , 337 - 355 .
* simon , j. b. , lesur , g. , kunz , m. w. , armitage , p. j. 2015 . magnetically driven accretion in protoplanetary discs .
monthly notices of the royal astronomical society 454 , 1117 - 1131 .
* stoll , m. h. r. , kley , w. 2014 . vertical shear instability in accretion disc models with radiation transport . astronomy and astrophysics 572 , a77 . * stoll , m. h. r. , kley , w. 2016 . particle dynamics in discs with turbulence generated by the vertical shear instability .
arxiv e - prints arxiv:1607.02322 .
* stone , j. m. , ostriker , e. c. , gammie , c. f. 1998 .
dissipation in compressible magnetohydrodynamic turbulence . the astrophysical journal 508 , l99-l102 . * suzuki , t. k. , muto , t. , inutsuka , s .-
protoplanetary disk winds via magnetorotational instability : formation of an inner hole and a crucial assist for planet formation .
the astrophysical journal 718 , 1289 - 1304 . * suzuki , t. k. , ogihara , m. , morbidelli , a. , crida , a. and guillot , t. , 2016 .
evolution of protoplanetary discs with magnetically driven disc winds , astron .
, in press * swift , j. j. , johnson , j. a. , morton , t. d. , crepp , j. r. , montet , b. t. , fabrycky , d. c. , muirhead , p. s. 2013 . characterizing the cool kois .
iv . kepler-32 as a prototype for the formation of compact planetary systems throughout the galaxy .
the astrophysical journal 764 , 105 . * tanaka , h. , ida , s. 1999 .
growth of a migrating protoplanet .
icarus 139 , 350 - 366 . * tanaka , h. , takeuchi , t. , ward , w. r. 2002 .
three - dimensional interaction between a planet and an isothermal gaseous disk .
i. corotation and lindblad torques and planet migration . the astrophysical journal 565 , 1257 - 1274 .
* terquem , c. , papaloizou , j. c. b. 2007 . migration and the formation of systems of hot super - earths and neptunes . the astrophysical journal 654 , 1110 - 1120 .
* throop , h. b. , bally , j. 2005 . can photoevaporation trigger planetesimal formation ? . the astrophysical journal 623 , l149-l152 . * tremaine , s. , dong , s. 2012 .
the statistics of multi - planet systems . the astronomical journal 143 , 94 . * turner , n. j. , fromang , s. , gammie , c. , klahr , h. , lesur , g. , wardle , m. , bai , x .-
. 2014 . transport and accretion in planet - forming disks .
protostars and planets vi 411 - 432 .
* turrini , d. , coradini , a. , magni , g. 2012 .
jovian early bombardment : planetesimal erosion in the inner asteroid belt .
the astrophysical journal 750 , 8 .
* udry , s. , santos , n. c. 2007 .
statistical properties of exoplanets . annual review of astronomy and astrophysics 45 , 397 - 439 .
* villeneuve , j. , chaussidon , m. , libourel , g. 2009 .
homogeneous distribution of @xmath39al in the solar system from the mg isotopic composition of chondrules .
science 325 , 985 . * visser , r. g. , ormel , c. w. 2016 . on the growth of pebble - accreting planetesimals .
astronomy and astrophysics 586 , a66 .
* volk , k. , gladman , b. 2015 .
consolidating and crushing exoplanets : did it happen here ? . the astrophysical journal 806 , l26 . * youdin , a. n. , goodman , j. 2005 .
streaming instabilities in protoplanetary disks . the astrophysical journal 620 , 459 - 469 . * walsh , k. j. , morbidelli , a. , raymond , s. n. , obrien , d. p. , mandell , a. m. 2011 . a low mass for mars from jupiter s early gas - driven migration .
nature 475 , 206 - 209 . *
ward , w. r. 1997 .
protoplanet migration by nebula tides .
icarus 126 , 261 - 281 . * weidenschilling , s. j. 1977 .
aerodynamics of solid bodies in the solar nebula .
monthly notices of the royal astronomical society 180 , 57 - 70 .
* weidenschilling , s. j. , cuzzi , j. n. 1993 .
formation of planetesimals in the solar nebula .
protostars and planets iii 1031 - 1060 .
* weidenschilling , s. j. 1995 .
can gravitational instability form planetesimals ? .
icarus 116 , 433 - 435 . * weiss , l. m. , marcy , g. w. 2014 .
the mass - radius relation for 65 exoplanets smaller than 4 earth radii .
the astrophysical journal 783 , l6 .
* wetherill , g. w. 1990 .
formation of the earth
. annual review of earth and planetary sciences 18 , 205 - 256 . * williams , j. p. , cieza , l. a. 2011 .
protoplanetary disks and their evolution . annual review of astronomy and astrophysics 49 , 67 - 117 .
* wolfgang , a. , lopez , e. 2015 .
how rocky are they ? the composition distribution of kepler s sub - neptune planet candidates within 0.15 au . the astrophysical journal 806 , 183 . * zsom , a. , dullemond , c. p. 2008 . a representative particle approach to coagulation and fragmentation of dust aggregates and fluid droplets
astronomy and astrophysics 489 , 931 - 941 . | over the past two decades , large strides have been made in the field of planet formation .
yet fundamental questions remain . here
we review our state of understanding of five fundamental bottlenecks in planet formation .
these are : 1 ) the structure and evolution of protoplanetary disks ; 2 ) the growth of the first planetesimals ; 3 ) orbital migration driven by interactions between proto - planets and gaseous disk ; 4 ) the origin of the solar system s orbital architecture ; and 5 ) the relationship between observed super - earths and our own terrestrial planets . given our lack of understanding of these issues , even the most successful formation models remain on shaky ground . |
flows of most liquid substances are usually studied by modeling the liquid as a continuum , but there are some substances that allow the study of flows at the kinetic level , i.e. , at the level of the individual constituent particles . as examples , we can mention chute flows in granular materials @xcite and capillary flows in colloids @xcite .
the solid particles in these soft materials are large enough that their motion can be tracked by video microscopy , allowing experimenters to record their positions and velocities . like granular materials and colloids ,
dusty plasmas also allow direct observation of individual particle motion .
dusty plasma @xcite is a four - component mixture consisting of micron - size particles of solid matter , neutral gas atoms , free electrons , and free positive ions .
these particles of solid matter , which are referred to as `` dust particles , '' gain a large negative charge @xmath2 , which is about @xmath3 elementary charges under typical laboratory conditions .
the motion of the dust particles is dominated by electric forces , corresponding to the local electric field @xmath4 , where @xmath5 is due to confining potentials and @xmath6 is due to coulomb collisions with other dust particles .
due to their high charges , coulomb collisions amongst dust particles have a dominant effect . the interaction force @xmath7 amongst dust particles is so strong that the dust particles do not move easily past one another , but instead self - organize and form a structure that is analogous to that of atoms in a solid or liquid @xcite .
in other words , the collection of dust particles is said to be a strongly - coupled plasma @xcite . in a strongly - coupled plasma ,
the pressure @xmath8 is due mainly to interparticle electric forces , with only a small contribution from thermal motion @xcite .
even when it has a solid structure , a collection of dust particles is still very soft , as characterized by a sound speed on the order of 1 cm / s @xcite . as a result , a dusty plasma in a solid phase is very easily deformed by small disturbances , and it can be made to flow .
flows can be generated , for example , by applying shear using a laser beam that exerts a spatially - localized radiation force @xcite .
in such an experiment , the reynolds number is usually very low , typically @xmath9 , indicating that the flow is laminar @xcite .
this paper provides further analysis and details of the experiment that was reported in @xcite .
we now list some of the major points of these two papers , to indicate how they are related and how they differ . in this paper
, we present : ( 1 ) a detailed treatment of the continuity equations for both momentum and energy , ( 2 ) our method of simultaneously determining two transport coefficients ( viscosity and thermal conductivity ) , ( 3 ) values of these two coefficients , and ( 4 ) spatially - resolved profiles of the terms of the energy equation , including the terms for viscous heating and thermal conduction , as determined by experimental measurements . in @xcite
, we reported : ( 1 ) a discovery of peaks in a spatially - resolved measurement of kinetic temperature , ( 2 ) a demonstration that these peaks are due to viscous heating in a region of a highly sheared flow velocity , and ( 3 ) a quantification of the role of viscous heating , in competition with thermal conduction , by reporting a dimensionless number of fluid mechanics called the brinkman number @xcite which we found to have an unusually large value due to the extreme properties of dusty plasma as compared to other substances .
the values of viscosity and thermal conduction found in this paper are used as inputs for the calculations of dimensionless numbers in @xcite .
the identification of viscous heating as the cause of the temperature peaks reported in @xcite is supported by the spatially - resolved measurements reported here . in the experiment ,
the dust particles are electrically levitated and confined by the electric field in a sheath above a horizontal lower electrode in a radio - frequency ( rf ) plasma , forming a single layer of dust particles , fig .
the dust particles can move easily within their layer , but very little in the perpendicular direction , so that their motion is mainly two dimensional ( 2d ) .
they interact with each other through a shielded coulomb ( yukawa ) potential , due to the screening provided by the surrounding free electrons and ions @xcite .
as the dust particles move , they also experience frictional drag since individual dust particles in general move at velocities different from those of the molecules of the neutral gas .
this friction can be modeled as epstein drag @xcite , and characterized by the gas damping rate @xmath10 . using experimental measurements of their positions and velocities , the dust particles can of course
be described in a _ particle _ paradigm .
they can also be described by a _ continuum _ paradigm by averaging the particle data on a spatial grid . in transport theory ,
momentum and energy transport equations are expressed in a continuum paradigm , while transport coefficients such as viscosity and thermal conductivity are derived using the particle paradigm because these transport coefficients are due to collisions amongst individual particles . in our experiment
, we average the data for particles , such as velocities , to obtain the spatial profiles for the continuum quantities , such as flow velocity . in the continuum paradigm
, a substance obeys continuity equations that express the conservation of mass , momentum , and energy .
these continuity equations , which are also known as navier - stokes equations , characterize the transport of mass , momentum , and energy . in a multi - phase or multi - component substance , these equations can be written separately for each component .
the component of interest in this paper is dust . in sec .
ii , we review the continuity equations for dusty plasmas . in secs .
iii and iv , we provide details of our experiment and data analysis method .
we designed our experiment to have significant flow velocity and significant gradients in the flow velocity , i.e. , velocity shear . in sec .
v , we simplify the continuity equations using the spatial symmetries and steady conditions of the experiment , and including the effects of external forces .
we will use our experimental data as inputs in these simplified continuity equations in secs .
vi and vii .
we now review the continuity equations for mass , momentum , and energy for dusty plasmas . we will then discuss the significance of some of the terms for our experiments .
the equation of mass continuity , i.e. , conservation of mass , is @xmath11 in this paper , the mass density @xmath12 and the fluid velocity @xmath13 describe the dust continuum .
the momentum equation @xcite is @xmath14\nabla(\nabla \cdot { \bf v } ) & + & { \bf f}_{\rm ext},\end{aligned}\ ] ] here , @xmath15 , @xmath0 , and @xmath16 are the charge density , shear viscosity , and bulk viscosity , respectively ; and @xmath17 is called the kinematic viscosity . in this paper , these parameters describe the dust continuum .
equation ( [ momentum ] ) describes the force per unit mass , i.e. , acceleration , for the continuum .
the confining field @xmath18 and pressure @xmath8 have been discussed in sec .
i. the last term in eq .
( [ momentum ] ) is due to the momentum contribution from forces such as gas friction , laser manipulation , ion drag , and any other forces that are external to the layer of dust particles , as discussed in sec .
the other terms on the right - hand - side of eq .
( [ momentum ] ) correspond to viscous dissipation , which arises from coulomb collisions amongst the charged dust particles .
the viscous term @xmath19 was studied in a previous dusty plasma experiment @xcite . for our experiment
, we can consider eq .
( [ momentum ] ) for two conditions , with and without the application of the external force @xmath20 . without the external force ,
the fluid velocity @xmath13 is zero so that eq .
( [ momentum ] ) is reduced to @xmath21 , meaning that the confining electric force is in balance with the pressure ( which mainly arises from interparticle electric forces ) . with the external force ,
the external confining force remains as it was without @xmath20 ; moreover , the pressure @xmath8 will be affected only weakly because the density is unchanged ( as we will show later ) so that the interparticle electric forces that dominate the pressure will also be unchanged .
therefore , in using eq .
( [ momentum ] ) , we will assume that the first two terms on the right hand side always cancel everywhere . the internal energy equation , as it is expressed commonly in fluid dynamics @xcite , is @xmath22 here
, @xmath23 is the entropy per unit mass , @xmath1 is the thermal conducitivity , and @xmath24 is the thermodynamic temperature of the dust continuum .
the last term @xmath25 is due to the energy contribution from any external forces @xmath20 .
we assume that the continuity equations , eqs .
( [ mass]-[energy ] ) , are valid for the dust particles separately from other components of dusty plasmas that occupy the same volume , such as neutral gas atoms .
the coupling between the dust particles with other components is indicated in @xmath20 and @xmath25 , so that the momentum and energy of the dust particle motion is treated for the dust particles separately from other components .
other external forces , such as those due to laser manipulation , are also indicated in @xmath20 and @xmath25 . the first term on the right - hand - side of eq .
( [ energy ] ) is due to viscous heating .
the viscous heating term @xmath26 depends on the square of the shear , i.e. , the square of the gradient of flow velocity .
a general expression for @xmath26 has many terms ( cf .
( 3.4.5 ) of ref .
@xcite or eq .
( 49.5 ) of ref .
@xcite ) , but it can be simplified for our experiment by taking advantage of symmetries , as explained in sec .
iv . the second term on the right - hand - side of eq .
( [ energy ] ) is due to thermal conduction .
it arises from a temperature gradient .
previous experiments with 2d dusty plasmas include a study of the thermal conduction term in eq .
( [ energy ] ) @xcite . in this paper ,
most of our attention will be devoted to the first two terms on the right - hand - side of eq .
( [ energy ] ) . using our experimental data
, we will compare the magnitudes of these terms . in @xcite
, we demonstrated that viscous heating is measurable and significant , when evaluated using only _ global _ measures like the brinkman number .
here we further evaluate viscous heating by characterizing it _ locally _ using spatially - resolved profiles for the terms in eq .
( [ energy ] ) .
we also develop a method to simultaneously obtain values of two transport coefficients of @xmath0 and @xmath1 .
here we provide a more detailed explanation of the experiment than in @xcite .
an argon plasma was generated in a vacuum chamber at @xmath27 ( or 2.07 pa ) , powered by rf voltages at @xmath28 and @xmath29 peak - to - peak .
we used the same chamber and electrodes as in @xcite .
the dust particles were @xmath30 diameter melamine - formaldehyde microspheres of mass @xmath31 .
the dust particles settled in a single layer above the powered lower electrode .
the layer of dust particles had a circular boundary with a diameter of @xmath32 and contained @xmath33 dust particles .
as individual dust particles moved about within their plane , they experienced a frictional damping @xcite with a rate @xmath34 due to the surrounding argon gas .
the particles were illuminated by a @xmath35-nm argon laser beam that was dispersed to provide a thin horizontal sheet of light , fig .
1(a ) . using a cooled 14-bit digital camera ( pco model 1600 ) viewing from above , we recorded the motion of individual dust particles .
this top - view camera imaged a central portion of the dust layer , as sketched in fig .
the movie is available for viewing in the supplemental material of @xcite .
the portion of the camera s field of view that we will analyze was @xmath36 , and it contained @xmath37 particles .
we recorded @xmath38 frames at a rate of 55 frame / s , with a resolution of @xmath39 .
our choice of 55 frame / s was sufficient for accurate measurement of various dynamical quantities , including the kinetic temperature , although a slightly higher frame rate would have been optimal @xcite .
in addition to the top - view camera , we also operated a side - view camera to verify that there was no significant out - of - plane motion ; this was due to a strong vertical confining electric field .
thus , we will analyze the particle motion data taking into consideration only the motion within a horizontal plane . at first , the dust particles self - organized in their plane to form a 2d crystalline lattice .
the particle spacing , as characterized by a wigner - seitz radius @xcite , was @xmath40 , corresponding to a lattice constant @xmath41 , an areal number density @xmath42 , and a mass density @xmath43 . using the wave - spectra method for thermal motion of particles in the undisturbed lattice
, we found the following parameters for the dust layer : @xmath44 , @xmath45 , and @xmath46 , where @xmath47 is the nominal 2d dusty plasma frequency @xcite , @xmath2 is the particle charge , @xmath48 is the elementary charge , and @xmath49 is the screening length of the yukawa potential .
we used laser manipulation @xcite to generate stable flows in the same 2d dusty plasma layer .
a pair of continuous - wave @xmath50-nm laser beams struck the layer at a @xmath51 downward angle . in this manner , we applied radiation forces that pushed particles in the @xmath52 directions , as shown in fig . 1 .
the power of each beam was @xmath53 as measured inside the vacuum chamber . to generate a wider flow than in @xcite
, the laser beams were rastered in the both @xmath54 and @xmath55 directions in a lissajous pattern as in @xcite , with frequencies of @xmath56 and @xmath57 .
we chose these frequencies to be @xmath58 to avoid exciting coherent waves .
the rastered laser beams filled a rectangular region , which crossed the entire dust layer and beyond , as sketched in fig .
this laser manipulation scheme resulted in a circulating flow pattern with three stable vortices , as sketched in fig .
we designed the experiment so that in the analyzed region the flow is straight in the @xmath52 directions , without any significant curvature .
curvature in the flow , which is necessary for the flow pattern to close on itself as sketched in fig .
1(b ) , was limited in our experiment to the extremities of the dust layer , where it would not affect our observations .
( color online ) .
( a ) side - view sketch of the apparatus , not to scale . a single layer of dust particles of charge @xmath2 and mass @xmath59
are levitated against gravity by a vertical dc electric field .
there is also a weaker radial dc electric field @xmath60 which prevents the dust particles from escaping in the horizontal direction .
further details of the chamber are shown in @xcite .
the two laser beams are rastered in the @xmath55 direction so that they have a finite expanse , and they are offset in the @xmath55 direction as shown in ( b ) .
( b ) top - view sketch of laser - driven flows in the 2d dusty plasma . in the region of interest , the flow is straight , with curvature of the flow limited to the extremities of the dust layer .
a video image of the dust particles within the region of interest is also shown .
the region of interest is divided into 89 bins of width @xmath61 so that particle data can be converted to continuum data . ]
we start our data analysis by analyzing data for individual particles , i.e. , by working in the _ particle _ paradigm . using image analysis software @xcite , with a method optimized as in @xcite to minimize measurement error , we identify individual particles in each video image and calculate their @xmath62 coordinates .
we then track a dust particle between two consecutive frames and calculate its velocity @xmath63 as the difference in its position divided by the time interval between frames @xcite . now having the position and velocity of all the particles in the analyzed region , we can study motion at the particle level .
for example , in sec .
vi we will use data for the individual particles to calculate the rate of their energy dissipation due to their frictional drag with the neutral gas , @xmath25 .
we next convert our data for individual particles to continuum data , i.e. , we change from the _ particle _ paradigm to the _ continuum _ paradigm .
this is done by averaging particle data within spatial regions of finite area , which we call bins .
there are 89 bins , which are all long narrow rectangles aligned in the @xmath55 direction , as shown in fig .
each bin contains @xmath64 dust particles .
we choose the shape of these bins to exploit the symmetry of the experiment , which in the analyzed region has an ignorable coordinate , @xmath54 .
the width of each bin is the same as the wigner - seitz radius , @xmath65 . to reduce the effect of particle discreteness as a particle crosses the boundaries between bins , we use the cloud - in - cell weighting method @xcite which has the effect of smoothing data so that a particle contributes its mass , momentum and energy mostly to the bin where it is currently located and to a lesser extent to the next nearest bin .
the data are binned in this way regardless of their @xmath54 positions , since @xmath54 is treated as an ignorable coordinate .
we also time - average these binned data , exploiting the steady conditions of the experiment .
this procedure ( binning , cloud - in - cell - weighting , and time averaging ) yields our continuum quantities , such as the flow velocity @xmath66 .
it also yields a kinetic temperature @xmath67 which is calculated from the individual particle velocities ; this kinetic temperature is not necessarily identical to the thermodynamic temperature @xmath24 . here , @xmath68 is the boltzmann constant .
we assume that it is valid to use a continuum model when gradients are concentrated in a region as small as a few particle spacings .
in fact , it has been shown experimentally that the momentum equation @xcite and the energy equation @xcite remain useful in 2d dusty plasma experiments with gradients that are as strong as in our experiment .
the notation we use in this paper distinguishes velocities and other quantities according to whether they correspond to individual particles or continuum quantities .
parameters for individual dust particles are denoted by a subscript @xmath69 , for example @xmath63 for the velocity of an individual dust particle .
continuum quantities in a theoretical expression are indicated without any special notation , for example @xmath13 for the hydrodynamic velocities in eqs .
( [ mass ] ) and ( [ momentum ] ) .
finally , continuum quantities that we compute with an input of experimental data , as described above , are indicated by a bar over the symbol , for example @xmath70 .
here , we present our simplification of the continuity equations , eqs .
( [ mass]-[energy ] ) , to describe our 2d dust layer .
these simplifications involve three approximations suitable for the conditions in our 2d layer , and a treatment of two external forces , laser manipulation and gas friction , that are responsible for @xmath20 and @xmath25 .
we describe these simplifications and the resulting continuity equations , next .
the first of our three approximations is @xmath71 .
this approximation is suitable for the steady overall conditions of our experiment . aside from the particle - level fluctuations that one desires to average away ,
when adopting a continuum model , the only time - dependent processes in the experiment were the rastering of the laser beam at @xmath72 and the @xmath28 rf electric fields that powered the plasma .
these frequencies are too high for the dust particles to respond , and the rastering of the beams is not a factor anyway because we will only use the continuum equations outside the laser beams .
the second approximation is that @xmath73 is negligibly small . due to the symmetry in our experiment design , as mentioned in sec .
iii , @xmath54 is treated as an ignorable coordinate , i.e. , @xmath74 . as a verification of this assumption , we observe that the ratio @xmath75 is of order @xmath76 , as a measure of the slightly imperfect symmetry of our experiment .
thus , for our experiment , we consider @xmath77 to be of zeroth order and @xmath73 to be two orders of magnitude smaller , when we approximate eqs .
( [ mass]-[energy ] ) .
the third approximation is that @xmath78 is negligibly small , based on our observation of the flow velocity of our 2d layer .
our results for the calculated flow velocity of the dust layer are shown in fig .
2 . from the velocity @xmath79 in fig 2(a ) , the flow can be easily identified from the two peaks with broad edges .
however , the flow velocity in the @xmath55 direction is two orders of magnitude smaller than in the @xmath54 direction , as shown in fig .
2(b ) .
( color online ) .
profiles of continuum parameters during laser manipulation , including ( a - b ) flow velocity , ( c ) areal number density , and ( d ) a measure @xmath80 of local structural order that would be a value of 1 for a perfect crystal @xcite
. disorder , as indicated by a small value in ( d ) , is found to be greatest where the shear is largest , not where the flow is fastest . ] using the second and third approximations , and omitting terms that are small by at least two orders of magnitude , we can easily see that we can approximate @xmath81 and @xmath82 .
the latter indicates that the dust layer can be treated as an incompressible fluid in our experiment .
using these three approximations in eq .
( [ mass ] ) , we find that @xmath83 . in other words , for our approximations , the density @xmath12 is uniform , which is confirmed by our experimental observation of fig .
because the density is so uniform , we can also assume that plasma parameters , such as the dust charge @xmath2 , are also spatially uniform within the analyzed region .
in addition to these approximations , we also assume that @xmath0 and @xmath1 are valid transport coefficients for our system .
there are theoretical reasons to question whether 2d systems ever have valid transport coefficients , and this is typically tested _ theoretically _ using long - time tails in correlation functions @xcite
. it could be tested _ experimentally _ by repeating the determination of @xmath0 and @xmath1 for vastly different length scales for the gradients of velocity and temperature and verifying that they do not depend on the length scale .
however , such a test is not practical for experiments like ours , which tend to have a limited range of diameters of dust layers that can be prepared .
now we consider external forces that contribute momentum and energy to the two equations , eqs .
( [ momentum ] ) and ( [ energy ] ) . for our experiment
, we can mention six _ external _ forces acting on the dust layer : gas friction , laser manipulation , electric confining force , gravity , electric levitating force , and ion drag . since we only study the 2d motion of dust particles within their plane , the last three forces are of no interest because they are in the perpendicular direction , and will not affect the horizontal motion of particles that is of interest here .
the electric confining force is balanced by the pressure inside the 2d dusty plasma lattice , @xmath84 , as described in sec .
thus , only two of the six forces need to be considered : gas friction and laser manipulation .
gas friction is the main dissipation mechanism in our experiment .
we can consider the effect of this friction first at the level of a single dust particle , and then at the level of a continuum .
for the _ momentum _ equation , we note first that at the particle level , a single dust particle moving at a speed of @xmath63 experiences a drag acceleration of @xmath85 . at the continuum level , the contribution of this drag to eq .
( [ momentum ] ) is simply the average acceleration experienced by all dust particles in a given spatial region , @xmath86 .
for the _ energy _ equation , at the particle level the rate of energy dissipation for one dust particle is @xmath87 , which is the product of a drag force and velocity , where @xmath88 is the kinetic energy of one dust particle . at the continuum level ,
averaging over all the dust particles in the given spatial region , the rate of energy loss per unit mass in eq .
( [ energy ] ) is @xmath89 @xcite .
using the three approximations listed above and taking into account gas friction and laser manipulation forces , the mass , momentum , and energy continuity equations become @xmath90 @xmath91 @xmath92 where @xmath93 is the viscous heating term , after simplifications based on the assumptions of @xmath74 , @xmath94 , and @xmath82 . in this paper , we will restrict our analysis to a spatial region where the laser force and power are zero , i.e. , @xmath95 and @xmath96 . in this region ,
the momentum and energy continuity equations will be further simplified : @xmath97 @xmath98 to simplify the problem , we will assume that @xmath0 and @xmath1 are independent of temperature , as discussed in @xcite .
we can comment on the meaning of these two equations .
equation ( [ momentum_dpg ] ) indicates a balance of the sideways transfer of momentum due to two mechanisms : viscosity arising from interparticle electric forces and frictional loss of momentum due to collisions with gas atoms .
equation ( [ energy_dpg ] ) describes the energy transferred from the viscous heating @xmath26 and thermal conduction ( the second term ) as being balanced by the energy dissipated due to friction as expressed in the last term of eq .
( [ energy_dpg ] ) .
we will use eqs .
( [ momentum_dpg]-[energy_dpg ] ) only in spatial regions where @xmath95 and @xmath96 , i.e. , outside the laser beam . in the next section
, we will present our calculation of terms appearing in eqs .
( [ momentum_dp]-[energy_dpg ] ) using the dust particles position and velocity data from our experiment .
the terms of interest in these equations are @xmath99 , @xmath100 , @xmath101 , and @xmath102 .
( color online ) . profiles of the first ( a ) and second ( b ) derivatives of flow velocity . to make a comparison , the flow velocity profile @xmath103
is also provided in ( c ) . ] our results for the first and second derivatives of the flow velocity are presented in fig .
these results are calculated using the flow velocity profile in fig .
2(a ) , which we reproduce in fig .
3(c ) . the first and second derivatives in fig
. 3(a - b ) will be used in eqs .
( [ viscous_heat ] ) and ( [ momentum_dpg ] ) , respectively . from fig .
3(a ) , we can identify four points of maximum shear , i.e. , maximum @xmath99 ; these are at @xmath104 , @xmath105 , @xmath106 , and @xmath107 .
these points of maximum shear coincide with other features of interest : the minimum in the structural order , fig .
2(d ) , and peaks in the mean - square velocity fluctuation profile , which we will present below .
we find that disorder , as indicated by a small value in fig .
2(d ) , is greatest where the shear is largest , not where the flow is fastest . comparing panels ( b ) and ( c ) of fig . 3
, we find that the profiles for the flow velocity @xmath108 and its second derivative are similar , in regions without laser manipulation , for example in the central region @xmath109 .
this similarity is expected from the momentum equation , eq .
( [ momentum_dpg ] ) , provided that the viscosity @xmath0 is spatially uniform .
we do not expect that @xmath0 would be spatially uniform since viscosity in general depends on temperature and the temperature is highly non - uniform , as we will show below .
nevertheless , we find that the two curves are nearly similar , with a small discrepancy that we will quantify below when we present data for the residual of eq .
( [ momentum_dpg ] ) .
( color online ) .
profiles of the mean squared particle velocity shown separately for particle motion in the @xmath54 ( a ) and @xmath55 ( b ) directions .
these quantities correspond to the mean particle kinetic energy @xmath101 , i.e. , @xmath110 .
the thin curve in each panel is the flow velocity profile @xmath103 ; its scale is shown in fig .
3(c ) . ]
profiles of the mean squared particle velocity , corresponding to the kinetic energy in the energy equation , eq .
( [ energy_dpg ] ) , are shown in fig . 4 .
this kinetic energy includes energy associated with both the macroscopic flow and the fluctuations at the particle level .
we will use these profiles in determining @xmath101 in the next section , where we will find the residual of eq .
( [ energy_dpg ] ) .
( color online ) .
profiles of the mean - square particle velocity fluctuation shown separately for particle motion in the @xmath54 and @xmath55 directions , ( a ) and ( b ) , respectively .
these quantities combined correspond to the kinetic temperature @xmath111 , eq .
( [ kt ] ) .
the profile of @xmath111 were reported in @xcite , where we discovered peaks in the kinetic temperature where the shear is largest .
the second derivative of the averaged mean - square particle velocity fluctuation for the motion in the @xmath54 and @xmath55 directions ( c ) will be used to determine the second derivative of the thermodynamic temperature in eq .
( [ energy_dpg ] ) .
the thin curve in each panel is the flow velocity profile @xmath103 ; its scale is shown in fig .
] in fig . 5
, we present our results for the mean - square velocity fluctuation , which corresponds to the kinetic temperature as in eq .
( [ kt ] ) .
we will use this kinetic temperature in place of the thermodynamic temperature @xmath24 in the energy equation , eq .
( [ energy_dpg ] ) . unlike the kinetic energy @xmath101 ,
the kinetic temperature only includes the energy associated with the fluctuations of particle velocity about the flow velocity @xmath112 @xcite . as reported in @xcite , there are peaks in the kinetic temperature profile that coincide with the position of maximum shear .
these peaks can also be seen in fig .
5(a - b ) . in @xcite , we attributed these peaks to viscous heating .
as an intuitive explanation of viscous heating , consider that higher shear conditions lead to collisions of particles flowing at different speeds , causing scattering of momentum and energy that leads to higher random velocity fluctuations , and therefore higher kinetic temperature . in the next section , we provide further verification that the temperature peaks are due to viscous heating ; we do this by confirming that three terms in the energy equation , including viscous heating , are in balance as indicated by their summing to zero .
we now examine the momentum and energy equations , eqs .
( [ momentum_dpg ] ) and ( [ energy_dpg ] ) , which are written so that the right - hand - side is zero .
when we use these equations with an input of experimental data , however , the terms will not sum exactly to zero , but will instead sum to a finite residual .
we calculate these residuals , and we vary two free parameters , the viscosity @xmath0 and the thermal conductivity @xmath1 , to minimize the residuals .
( specifically , we minimize the square residual summed over all bins in the central region of @xmath113 . )
this minimization procedure yields the best estimation for the values for @xmath0 and @xmath1 , which will be our first chief result .
we will then , as our second chief result , be able to make a spatially - resolved comparison of the magnitude of different terms in the energy equation , eq .
( [ energy_dpg ] ) .
( color online ) .
( a ) a profile of the residual of the momentum equation , eq .
( [ momentum_dpg ] ) , assuming a kinematic viscosity of @xmath114 . for the central region ,
magnified in ( b ) , the summation of the squared residual reaches its minimum when @xmath114 ; this minimization process is how we determine @xmath0 , which is one of our main results .
the thin curve in ( a ) is the flow velocity profile @xmath103 ; its scale is shown in fig .
3(c ) . ]
figure 6 shows the residual of the momentum equation , eq .
( [ momentum_dpg ] ) .
this result is shown for @xmath115 , which is the best estimation of the kinematic viscosity .
the data are shown as a spatial profile because we calculated eq .
( [ momentum_dpg ] ) separately for each bin , i.e. , each value of @xmath55 .
two peaks in fig .
6(a ) , located within the laser manipulation region , are due to the momentum contribution from the laser .
we do not use eq .
( [ momentum_dpg ] ) with these peaks because of a finite laser force @xmath116 there .
instead we will use the flatter region between these peaks , where @xmath95 , as magnified in fig .
the small residuals in this flatter region indicate that the momentum equation , eq .
( [ momentum_dpg ] ) , is able to accurately account for the momentum of our 2d dust layer , and that the minimization process in this region yields a value for the viscosity .
( color online ) .
( a ) a profile of the residual of the energy equation , eq .
( [ energy_dpg ] ) , assuming a thermal diffusivity of @xmath117 . for the central region ,
magnified in ( b ) , the summation of the squared residual reaches its minimum when @xmath118 .
this minimization process is how we determine @xmath1 , which is another of our main results .
the thin curve in ( a ) is the flow velocity profile @xmath103 ; its scale is shown in fig .
3(c ) . ]
figure 7 shows the residual of the energy equation , eq .
( [ energy_dpg ] ) , for @xmath119 , which is the best estimation of the thermal diffusivity . in this calculation
, we used the value of @xmath120 from the momentum equation above , and we varied the value of @xmath1 to minimize the residuals as described above .
two large negative peaks in fig .
7(a ) are due to the energy contribution from the laser manipulation @xmath121 , which is not included in eq .
( [ energy_dpg ] ) .
the small values of residuals in the flatter region between these peaks , as magnified in fig .
7(b ) , show that the energy equation , eq .
( [ energy_dpg ] ) , accurately describe energy transport in our 2d dust layer , and that the minimization process in this region yields a value for the thermal diffusivity . by achieving a small value of the residual ,
we have verified that the three terms in the energy equation , eq .
( [ energy_dpg ] ) , are in balance .
since one of these terms is viscous heating and another is computed from the temperature profile which has peaks , the balance we observe here is consistent with the conclusion of @xcite that the temperature peaks are due to viscous heating . as the first chief result of this paper , we obtain the kinematic viscosity value of @xmath122 and thermal diffusivity value of @xmath119 .
these values are obtained simultaneously in a single experiment . a source of uncertainty in these values
is systematic error in @xmath10 , for example due to particle size dispersion or uncertainty in the epstein drag coefficients @xcite .
this is so because our method actually yields results for @xmath123 and @xmath124 .
another source of uncertainty is our simplification that we neglect heating sources other than laser manipulation @xcite ; in a test we determined that @xmath125 is in a range from 7.5 to @xmath126 , depending on whether these small heating effects are accounted for .
we note that our results of @xmath17 and @xmath127 agree with the results of previous experiments @xcite using the same size of dust particles and a similar value of @xmath128 .
( color online ) . ( a )
profiles of three terms in eq .
( [ energy_dpg ] ) , assuming the transport coefficients , @xmath0 and @xmath1 , obtained above . for the central region without laser manipulation ( b ) , thermal conduction is one order magnitude larger than the viscous heating .
this is our second chief result : a spatially - resolved comparison of the different mechanisms for energy transfer in our 2d dust layer .
the thin curve in ( a ) is the flow velocity profile @xmath103 ; its scale is shown in fig .
3(c ) . ]
our experiment allows us to obtain spatially - resolved quantitative measurements of three heat transfer effects : viscous heating , thermal conduction , and dissipation due to gas friction ( i.e. , cooling ) .
these three effects appear as the three terms on the left - hand - side of eq .
( [ energy_dpg ] ) . as the second chief result of this paper ,
we plot these three terms , presented as spatial profiles , in fig .
8 . examining the spatial profiles for these terms , in fig . 8
, we see the most prominent features are two large peaks for the gas dissipation term where the flow velocity @xmath103 is fastest ; in this region the energy dissipation due to gas friction reaches its maximum . despite the prominence of these features , however , they are not what interest us here . instead , we are more interested in the regions of high shear , near the edge of the laser manipulation . recall that in these high shear regions , the flow velocity gradient @xmath99 is largest , and there are peaks in the profiles of the kinetic temperature and the second derivative of the flow velocity , figs . 3 and 5 . in fig . 8
, our spatially - resolved profiles reveal that viscous heating and thermal conduction terms are peaked in regions of high shear .
the viscous heating term , eq .
( [ viscous_heat ] ) , is always positive , meaning that viscous dissipation is always a source of heat wherever it occurs .
the thermal conduction term partial @xmath129 , on the other hand , can be either positive or negative , indicating that heat is conducted toward or away from the point of interest , respectively . in locations where the shear is strongest , for example at @xmath104 ,
the thermal conduction term is negative indicating that heat is conducted away from that point .
although viscous heating has great importance in all kinds of fluids , and it has been understood theoretically for a very long time @xcite , a spatially - resolved measurement of it is uncommon . in most physical systems
viscous heating is usually hard to measure either because the temperature increase is overwhelmingly suppressed by rapid thermal conduction , as we discussed in @xcite , or because the thinness of the shear layer does not allow convenient _ in - situ _ temperature measurements .
most experimental observations of temperature increases due to viscous heating are either external or global measurements , and not spatially - resolved measurements like those that we report here . indeed , in our literature search , we found no previous spatially - resolved experimental measurements of the viscous heating term , not only for dusty plasma , but also for any other physical system . as we explained in @xcite , our ability to detect strong effects of viscous heating is due to the extreme properties of dusty plasma , as compared to other substances .
our ability to make spatially - resolved measurements is due to our use of video imaging of particle motion .
( color online ) . profiles of ( a ) the ratio of the viscous heating and thermal conduction terms in eq .
( [ energy_dpg ] ) , and ( b ) the absolute value of this ratio .
the sign of @xmath129 determines the sign of the ratio in ( a ) .
a positive ratio indicates that heat is conducted toward the position of interest .
the large values of the ratio in ( b ) at high shear regions indicate significant viscous heating at those locations .
for comparison , in @xcite we found a brinkman number br = 0.5 ( indicated by the arrow ) , which is a global measure of the flow that provides less detailed information than the spatially resolved ratio shown here .
the thin curve in ( b ) is the flow velocity profile @xmath103 ; its scale is shown in fig .
3(c ) . ]
to further analyze the second chief result of this paper , the spatial profiles of the viscous heating and thermal conduction terms in fig .
8(a ) , we plot the ratio of these two terms in fig
. 9(a ) .
this ratio has its largest positive and negative values in the regions of high shear . in fig .
9(a ) , negative values of this ratio are observed to occur in the high shear regions , which indicates that heat is conducted away from these regions .
this result is consistent with observation of kinetic temperature peaks here . to characterize the magnitude of these two terms , we plot in fig .
9(b ) the absolute value of this ratio .
we can see that , within regions of high shear , this ratio can be as large as unity , or even larger .
a typical value of this ratio in the shear region is of order 0.5 for our experiment .
this matches the value of the brinkman number , br = 0.5 that we found in @xcite for the same experiment .
the brinkman number is a global measure of the viscous heating , in competition with thermal conduction .
in summary , we reported further details of the laser - driven flow experiment in a dusty plasma that was first reported in @xcite .
we simplified the momentum and energy continuity equations , exploiting the symmetry and steady conditions of the experiment .
we developed a method to obtain transport coefficients by minimizing the residuals of continuity equations using the input of experimental data .
as our first chief result , we use this method to simultaneously determine , from the same experiment , two transport coefficients : kinematic viscosity and thermal diffusivity , which are based on viscosity and thermal conductivity , respectively . as our second chief result , we obtained spatially - resolved measurements of various terms in the energy equation .
we found that , in a laser - driven dusty plasma flow , viscous heating is significant in regions with high shear , which is consistent with the interpretation of @xcite that the peaks in the temperature profile are due to viscous heating .
a. melzer and j. goree , in _ low temperature plasmas : fundamentals , technologies and techniques _ , 2nd ed . , edited by r. hippler , h. kersten , m. schmidt , and k. h. schoenbach ( wiley - vch , weinheim , 2008 ) , p. 129 .
the external forces can add or remove energy from the collection dust particles , as expressed by @xmath25 , which is the final term of the energy equation , eq .
( [ energy ] ) . besides heating from laser manipulation and cooling from the gas friction as discussed in the text , we can mention two other contributions to @xmath25 for our experiment : heating from random kicks from collisions with the gas atoms and heating by fluctuating electric fields in the plasma . in the absence of any laser manipulation , the other three terms have their background levels which are in balance , so that their total @xmath25 is zero . in the presence of laser manipulation , even outside the region where the laser beams strike the dust particles , there are flows that result in a cooling by gas friction that is enhanced above its background level . when using the energy equation with data from our experiment with laser manipulation , we ignore two heating effects , gas atom collisions and fluctuating electric fields . using data from an experimental run without laser manipulation , we estimated those effects , so that we could report the range of values for @xmath125 in the text .
our laser manipulation method introduces anisotropy in the particle velocities . since our laser manipulation drives a flow only in the @xmath52 directions , @xmath130 .
similarly , the kinetic energy is mostly due to the flow motion in the @xmath54 direction , i.e. , @xmath131 , as shown in fig .
4(a - b ) .
however , the kinetic temperature due to the velocity fluctuations in the @xmath54 direction is almost the same as for fluctuations in the @xmath55 direction . | a shear flow of particles in a laser - driven two - dimensional ( 2d ) dusty plasma are observed in a further study of viscous heating and thermal conduction .
video imaging and particle tracking yields particle velocity data , which we convert into continuum data , presented as three spatial profiles : mean particle velocity ( i.e. , flow velocity ) , mean - square particle velocity , and mean - square fluctuations of particle velocity . these profiles and
their derivatives allow a spatially - resolved determination of each term in the energy and momentum continuity equations , which we use for two purposes .
first , by balancing these terms so that their sum ( i.e. , residual ) is minimized while varying viscosity @xmath0 and thermal conductivity @xmath1 as free parameters , we simultaneously obtain values for @xmath0 and @xmath1 in the same experiment .
second , by comparing the viscous heating and thermal conduction terms , we obtain a spatially - resolved characterization of the viscous heating .
= 1 |
the fundamental principles of physics , as we understand them today , emphasize the role of three constants : @xmath2 ( newton s gravitational constant ) , @xmath3 ( the speed of light ) and @xmath4 ( the planck constant ) . by a suitable choice of units , we can set the numerical value of all these three to unity and the broad structure of physical theories can be represented using a 3-dimensional space in which each of the cartesian coordinates is taken to be one of the above mentioned fundamental constants ( see fig .
[ fig : physicscube ] ) .
it turns out to be convenient to use @xmath5 rather than @xmath3 in such a description , and the entire space of physical theories will be confined within the unit cube so formed .
the examination of this diagram reveals several interesting features .
the origin @xmath6 represents an idealized non - relativistic ( point ) mechanics ( nrm ) with which many physics courses begin . moving along the speed of light axis to @xmath7 , ( keeping @xmath8 ) , will get us special relativistic ( sr ) mechanics .
similarly , moving along the @xmath9axis will lead to non - relativistic , classical , newtonian gravity ( ng ) and travelling along the @xmath10axis will lead to non - relativistic quantum mechanics ( qm ) .
more exact theoretical structures emerge when a _
pair _ of constants are non - zero .
the vertex @xmath11 , @xmath12 represents classical general relativity ( gr ) which combines the principles of special relativity and gravity .
similarly , @xmath13 , @xmath14 leads to flat spacetime quantum field theory ( qft ) which combines the principles of special relativity and quantum theory .
the vertex at which all the three constants are unity , @xmath15 , should represent the domain of quantum gravity but , more importantly for our purpose , _ it also represents the study of quantum field theory in curved space - time _
( qft in cst ) , like , for example , the study of radiation from black holes .
( a description of the thermal features of black holes requires all these three constants to be non - zero . )
while quantum gravity still remains a distant dream , we do have a fair amount of understanding of quantum field theory in curved spacetime and , in this sense , this vertex ( qft in cst ) can be considered as within our grasp . while most of these limiting forms of physical theories have attracted a reasonable amount of attention and made it into textbooks , the above diagram brings out one limiting case which probably has not been explored in comparable detail .
this is the vertex with @xmath16 , which corresponds to exploring the nature of gravity in a quantum _ mechanical _ context ( gqm ) .
much of the discussion in this paper is devoted to the exploration of this vertex and , more generally , to projecting the theories to the @xmath17 plane by taking the @xmath18 limit in different contexts .
this vertex is of interest for several reasons .
first , there are interesting questions related , for example , to the principle of equivalence and the free - fall of atomic systems in newtonian gravitational fields which are conceptually and experimentally important @xcite .
this topic has received a fair amount of attention in the context of neutron interferometry @xcite as well as in the context of neutrino physics @xcite .
there are some controversies ( and possible lack of clarity ) in the literature on this topic , which makes it worth studying .
second , this vertex serves as an interesting test - bed for several limiting processes . for example , note that one can arrive at the gqm vertex from the quantum field theory in curved spacetime ( qft in cst ) vertex by taking the limit @xmath18 .
it will be interesting to ask what limiting forms the known results in quantum field theory in curved spacetime take when we project ( move ) along this direction and do a taylor series expansion in inverse powers of @xmath3 .
( for example , it would be interesting to ask what is the lowest non - trivial order in @xmath19 one should work at to reproduce , say , the thermal nature of the inertial vacuum . ) third , this limit also has interesting implications even in the context of @xmath14 , viz .
, when we move from quantum field theory to quantum mechanics .
how does quantum _ field _ theory reduce to _ single particle _
quantum mechanics when we take the @xmath18 limit , and how does this limiting process get modified when we `` switch on '' @xmath2 ?
we will see that there are some subtleties even in the simplest context of transition from special relativity to non - relativistic mechanics because the symmetry groups the lorentz group and the galilean group have very different structures @xcite . finally , it is interesting to investigate the role of the principle of equivalence in quantum mechanical and field theoretic phenomena .
( this issue has attracted attention in many of the papers quoted in @xcite ) .
we know that one can mimic the local effects of gravity by a coordinate transformation to a suitable non - inertial frame .
therefore , a corresponding transformation to a non - inertial frame in quantum mechanics must allow us to incorporate the effects of gravity and take us from the qm to the gqm vertex .
it is interesting to verify this explicitly and understand the details .
the key new results in our paper are the following : \(i ) we show that the reason for the non - invariance of the galilean action under galilean transformations is intimately connected to the existence of _ rest - mass energy of the particle which leaves a residue even in the non - relativistic limit_. the details of the calculation might help to clarify some controversial conclusions in the literature @xcite .
\(ii ) we point out that the special relativistic action is invariant under a lorentz transformation _ even if we ignore terms of order @xmath20 and beyond _ , and discuss the consequences of this result , which does not seem to have been noticed previously in the literature .
\(iii ) we show that the principle of equivalence does hold good in quantum mechanics , _ if _ it is interpreted as embodying the equivalence of the _ dynamical equations _ in an accelerated frame and in a gravitational field .
we explicitly prove this equivalence for the case of the schrodinger equation in an inertial and an accelerated frame of reference .
\(iv ) we show that the relativistic feynman propagator in quantum field theory can be obtained from a path integral over @xmath21 , just as in the case of non - relativistic quantum mechanics , but now with the standard special relativistic action @xmath22 _ which is not quadratic_. we also show explicitly _ how _ the feynman propagator reduces to the feynman path - integral kernel in the non - relativistic limit , not only in the inertial frame , but also in the rindler frame , and in the more general context of a weak gravitational field .
this procedure also allows us to obtain the leading relativistic _
correction _ to the feynman path - integral .
\(v ) we show that the reason for the _ non - transitivity _ property of the feynman propagator is the same as that for the non - transitivity of the corresponding _ energy kernel _ in the non - relativistic theory , and we elaborate and discuss this connection . throughout the paper , we have tried to be as self - contained as possible by giving the necessary pedagogical details .
we begin by discussing the apparently elementary situation of the transition from special relativity to non - relativistic mechanics ( nrm ) by taking the limit @xmath23 which involves moving along sr to nrm in fig .
[ fig : physicscube ] .
while text books consider this limiting procedure as straightforward , we will see that some curious features arise when we consider the limiting form of the action functional in this context .
( not surprisingly , this issue has led to some controversial conclusions in published literature @xcite which we also hope to clarify . ) as is well known , special relativity is invariant under a lorentz transformation of the coordinates , while non - relativistic mechanics is invariant under a galilean transformation of the coordinates .
given the fact that one recovers the galilean coordinate transformation by setting @xmath24 in the lorentz transformation equations , one would have thought that any theory which is lorentz invariant will lead to a theory which is invariant under galilean transformations in the limit of @xmath18 .
as we shall see , there are several subtleties in the manner by which galilean invariance of the theory arises in this limit from lorentz invariance .
this topic is usually discussed in the literature in relation to the structure of the galilean group and superselection rules in quantum mechanics ( see e.g. , @xcite ) .
we shall explore the issues in a more straightforward and transparent manner .
we begin by enquiring how the action functional for a free particle changes under lorentz transformations in special relativity and galilean transformations in non - relativistic mechanics .
we note that the action @xmath25 is given by @xmath26 where the lagrangian @xmath27 can be expressed as a function @xmath28 of the square of the magnitude of the particle s velocity , by taking into account the homogeneity of space and time and the isotropy of space .
this holds both in the case of relativistic and non - relativistic mechanics .
it is impossible to proceed further and determine the explicit form of the lagrangian , without making some additional assumptions .
we now have to make a distinction between non - relativistic and relativistic mechanics by postulating the invariance of physical laws under different sets of coordinate transformations . in non - relativistic theory
, we postulate that the equations of motion should retain the same form when we make a galilean transformation : @xmath29 from the co - ordinates @xmath30 of an inertial frame @xmath31 to the co - ordinates @xmath32 of another frame @xmath33 moving with a uniform velocity @xmath34 along the positive @xmath35direction with respect to @xmath31 .
( in this and what follows , we suppress the two spatial dimensions and work in ( 1 + 1)-dimensions for simplicity . )
the corresponding velocity transformation is @xmath36 where @xmath37 and @xmath38 are the velocities measured in frames @xmath31 and @xmath33 respectively .
the sufficient ( though not necessary ) condition for the equations of motion to retain the same form in both @xmath31 and @xmath33 is that the action should be invariant under the transformations in .
it is straight forward to see that _ no _ non - trivial function @xmath39 has this property and hence _ we fail to construct _ an action functional which remains invariant under the galilean transformation .
the usual procedure at this stage is to note that the equations of motion will remain invariant even if the lagrangian is not , as long as the lagrangian changes only by the addition of a total time derivative of a function of coordinates and time .
it is again easy to show that this requires the lagrangian for the free particle to be _ proportional _ to the square of the velocity i.e. @xmath40 or @xmath41 where @xmath42 is defined to be the mass of the particle . in this case , the lagrangians @xmath27 and @xmath43 in the frames of reference @xmath31 and @xmath33 respectively are related by : @xmath44 hence , in the non - relativistic theory , the lagrangians @xmath27 and @xmath43 differ by a total time derivative of a function of co - ordinates and time .
the corresponding actions differ by contributions at the end points : @xmath45 note that the canonical momentum ( @xmath46 ) and the energy @xmath47 are _ not _ invariant when we transform from @xmath27 to @xmath43 . in special relativity ,
we replace by the lorentz transformations between @xmath31 and @xmath33 of the form : @xmath48 the transformation of velocities is now given by : @xmath49 in the limit of @xmath50 , the above equations reduce to the corresponding ones for the galilean transformation .
curiously enough , it is now possible to construct an action functional which _ is actually invariant _ ( instead of picking up an extra boundary term ) under the transformations in .
this is given by @xmath51 where @xmath52 is a constant .
the above form of the action remains invariant under lorentz transformations .
it is _ not _ possible to choose @xmath52 such that reduces to the action for non - relativistic mechanics when @xmath18 .
the best we can do is to choose @xmath52 in such a way that in the non - relativistic limit , we get back the non - relativistic form of the action , _ apart from a constant term _ in the lagrangian
. this amounts to setting @xmath53 .
hence , in special relativity , the action for a free particle is taken to be : @xmath54 the above discussion raises a couple of questions which require investigation .
first , we expect the non - relativistic theory to arise as a limiting case of the fully relativistic theory , in the limit of @xmath50 .
it is true that the lorentz transformation equations reduce to the galilean transformation equations in the limit of @xmath50 .
however , the special relativistic action does _ not _ reduce to the non - relativistic action in this limit , but instead picks up an extra term , @xmath55 evaluated at the end points . as we shall see , this term which is usually ignored in textbooks as being due to `` just an addition of a constant to a lagrangian '' has some interesting implications for the structure of special relativity and non - relativistic mechanics .
this is already hinted at by the fact that the relativistic action in blows up in the limit of @xmath18 and does not have a valid limit in the strict mathematical sense .
if one can `` renormalize '' this action by adding a term @xmath56 to , then @xmath57 will have a proper limit .
but then , @xmath58 is _ not _ lorentz invariant , and hence this `` renormalization '' is not a valid procedure .
we will see repeatedly that the term @xmath59 plays a crucial role in our discussions .
second , we find that in the special relativistic theory , which is the more exact theory , the action remains invariant under special relativistic ( lorentz ) transformations . however , in the non - relativistic limit , the corresponding non - relativistic action does _ not _ remain invariant under the analogous non - relativistic ( galilean ) transformations , but again picks up some extra terms .
this is surprising , when seen in the light of the galilean transformation being a limiting case of the lorentz transformation . on @xmath60 or @xmath61 but incorporated galilean invariance by allowing @xmath27 to pick up a total derivative .
it is possible to do the converse .
one can write down free particle lagrangians which are _ strictly invariant _ under galilean transformations but depend explicitly on @xmath61 and @xmath60 through a total time derivative .
a simple example is @xmath62 which is _ invariant _ under a galilean transformation but depends on @xmath61 and @xmath60 through a total time derivative ; @xmath27 differs from @xmath63 by the total time derivative @xmath64 . ] both the above issues are usually ignored by noting that the equations of motion do not change when a total time derivative of a function of coordinates and time is added to the lagrangian , and hence such action functionals are equivalent as far as physical phenomena are concerned .
this is true in _ classical _ physics , but in _ quantum _ theory , the form of the action is closely related to the phase of the wave function .
our analysis suggests that the phase of the wave function of a free particle remains invariant under lorentz transformations , but in the @xmath18 limit , this invariance gets broken .
we will now explore this situation by studying the phase of the free particle wave function in relativistic quantum theory and its limiting form in non - relativistic quantum mechanics .
the key issue can be introduced by examining the following argument which seems to allow one to find the wave function of a _ moving _ free particle from that of a _ stationary _ free particle .
the argument runs as follows : consider a particle of mass @xmath42 which is at rest in a frame @xmath33 that is moving with a velocity @xmath34 along the @xmath35axis of another inertial frame @xmath31 .
we postulate that ( a ) the phase of the wave function in @xmath33 should depend on the rest energy @xmath65 in the usual manner and hence the wave function should be @xmath66 . if we also assume that ( b ) @xmath67 is a scalar under lorentz transformations , then the wave function in @xmath31 is given by @xmath68 \propto \exp[-iet+ipx ] \label{relpsi}\ ] ] where the symbols have their usual meanings and @xmath69 and @xmath70 are the relativistic energy and momentum of the particle in the frame @xmath31 in which it is moving with velocity @xmath34 .
thus , given the postulates ( a ) and ( b ) mentioned above , we can determine the wave function of a moving particle from that of a particle at rest using the lorentz transformation .
, but the key points mentioned here continue to hold even when we take this fact into account .
hence we will not worry about it in this section . ]
if we now take the limit of @xmath18 , we find that @xmath71 and @xmath72
. hence the wave function of a non - relativistic particle seems to be given by @xmath73 \exp[-i(1/2)mv^2t+imvx]\ ] ] thus we seem to have obtained the correct solution to free particle schrodinger equation `` except for '' the phase term @xmath74 which has no interpretation in non - relativistic mechanics .
again , strictly speaking , @xmath67 does not have a limit when @xmath18 .
one needs to consider , instead , the quantity @xmath75 $ ] to get the correct limit , but , just as in the case of the action functional , this is not a valid prescription .
the correct @xmath67 in a relativistic theory is indeed the one in and not the one with an @xmath76 term subtracted out from the phase .
the situation becomes more curious when we realize that we _
could not have done any of these _ using galilean transformations alone .
that is , there is no way of obtaining the wave function of a moving particle from that of a stationary particle by the application of a galilean transformation . to begin with
, the postulate ( a ) above has no meaning in non - relativistic quantum mechanics and so we can not write down a wave function with a time dependent phase in its rest frame . second , as we shall see , wave functions do not transform as `` scalars '' under galilean transformations .
to get everything right , one has to add an extra phase , the origin of which we will now discuss .
let us again start with the phase of a relativistic wave function @xmath77 .
we know that under a lorentz transformation , this goes over to @xmath78 where @xmath79 ; @xmath80 . in this transformation
, we treat @xmath81 as just two real numbers parameterizing the wave function , and are _ not _ assuming that they are components of a four - vector , etc .
but once we have applied the lorentz transformation _ to the coordinates , _ we can interpret the coefficients of @xmath82 ( which are @xmath83 ) as the energy and momentum of the particle in @xmath33 , and this interpretation , as we know , is correct .
none of this works with galilean transformations and non - relativistic quantum mechanics . under a galilean transformation ,
the phase of the free particle wave function in non - relativistic quantum mechanics transforms as @xmath84 which does not allow us to read off the coefficients of @xmath85 and @xmath86 as the physical energy and momentum of the non - relativistic particle in @xmath33 .
so , unlike in the relativistic case , we can not identify the correct energy and momentum from the phase of the wave function after a galilean transformation .
we can , of course , introduce by hand the two parameters @xmath87 and re - express the right hand side of in terms of @xmath88 .
the result is best presented in terms of the identity : @xmath89 so , if we express the phase in @xmath33 in terms of @xmath90 and @xmath91 , then the phase is not invariant and we need to add an extra phase in the right hand side of .
this extra phase is essentially the difference between the two actions in .
since the phases of wave functions are classical actions for a free particle , it is clear that the non - invariance of the action under galilean transformations reflects itself in the additional phase that arises in the wave function .
these facts are also related to the contrasting behaviour of the klein - gordon equation and the schrodinger equation under lorentz transformations and galilean transformations , respectively . with future applications in mind
, we will consider the transformation of the schrodinger equation under a more general co - ordinate transformation from a frame of reference @xmath92 to a frame @xmath93 where @xmath94 is some arbitrary function of time .
( the galilean transformation is a special case when @xmath95 ; when @xmath94 is not a linear function of @xmath61 this represents a transformation to a non - inertial frame . )
using the fact that @xmath96 , @xmath97 we can transform the free particle schrodinger equation @xmath98 into the @xmath99 coordinates and obtain @xmath100 ( we have set @xmath101 for convenience . ) in the case of the galilean transformation , we have @xmath95 , leading to @xmath102 obviously the free particle schrodinger equation is not form invariant under the galilean transformation if we consider @xmath103 to be a scalar unlike the klein - gordon equation , which is form invariant , with the wave function remaining a scalar . it is the second term on the right hand side of which requires an extra phase to be added to the wave function for everything to be consistent . once again
, all this is closely related to the fact that the action in special relativity is invariant under lorentz transformations while the action in non - relativistic mechanics picks up an additional term under galilean transformations . to see this explicitly
, we will again consider the behaviour of the _ action _ for a free particle under the transformation from a frame of reference @xmath92 to a frame @xmath93 as before , and relate it to the extra phase in the wave function explicitly .
the lagrangian of the free particle in @xmath33 is given by @xmath104 which can be written in the form @xmath105 where @xmath106 and @xmath27 is a new lagrangian : @xmath107 which is equivalent to @xmath43 as far as the equations of motion are concerned , since the total time derivative @xmath108 does not contribute to the equations of motion .
this @xmath27 represents the lagrangian for a particle acted upon by a force @xmath109 or , equivalently , a particle located in a spatially homogeneous but time dependent gravitational field @xmath110 .
( in fact , the entire phase @xmath111 acquires a simple physical meaning when we take gravitational time dilation into account ; we will say more about this in section [ sec : qmequivalence ] . )
we recall that when we add @xmath112 , to the lagrangian @xmath27 , thus transforming it to @xmath113 , both the canonical momentum and the hamiltonian change , becoming @xmath114 and @xmath115 respectively . in quantum mechanics , the time evolution of the wave function
is determined by the hamiltonian operator and hence , the form of the wave function must change when we make a co - ordinate transformation from the frame @xmath31 to the frame @xmath33 .
let @xmath116 be the quantum - mechanical wave function for the free particle in the frame @xmath33 .
then , it can be shown that the corresponding wave function @xmath117 for the same particle in the frame @xmath31 is given by : @xmath118 and it satisfies the equation @xmath119 in the frame of reference @xmath31 , which is just the schrodinger equation with a `` pseudo - potential '' energy term , @xmath120 .
this term arises because the frame @xmath31 is an accelerated frame of reference and has a close link with the principle of equivalence , which we shall take up later on in section [ sec : qmequivalence ] .
( this result has been derived numerous times in the literature ; for completeness we reproduce the derivation in appendix [ app2 ] ) . coming back to the case of galilean transformation , we have @xmath95 and @xmath121 therefore , takes the form @xmath122\ ] ] that is
, we need to transform the wave function , treating it as a scalar , _ and then add an extra phase _ which is consistent with what we found in . as we have said before , all this is perfectly consistent as regards the application of the galilean transformation in quantum mechanics .
how is it that the klein - gordon equation is invariant under the lorentz transformation , but the schrodinger equation which is presumably obtained in the @xmath18 limit of the klein - gordon equation is _ not _ invariant under the galilean transformation , given the fact that the lorentz transformation reduces to the galilean transformation in the appropriate limit ?
this has to do with the manner in which one obtains the schrodinger equation from the klein - gordon equation and brings to the center stage the role of the @xmath123 term in the phase .
we will outline how the extra phase in can be obtained from a fully invariant klein - gordon equation . consider the wave function @xmath124 which is the solution to a free particle klein - gordon equation . we know that under a lorentz transformation , @xmath125 , thus transforming as a scalar . to obtain the schrodinger equation for a wave function
@xmath126 we first have to separate the @xmath59 term from the phase of the @xmath127 by writing @xmath128\ ] ] it is then straightforward to show that in the limit of @xmath18 , @xmath126 will satisfy a free particle schrodinger equation .
[ we will demonstrate a more general result in the presence of a gravitational field later on , and hence we skip the algebraic details here ; see . ] to obtain the schrodinger equation in @xmath33 , we have to similarly write @xmath129 .
the fact that @xmath127 transforms as a scalar can now be used to relate @xmath67 and @xmath130 , and we find that @xmath131\ ] ] we see that , in addition to the scalar transformation , the wave function picks up a phase which is just @xmath132 .
_ incredibly enough , this expression has a finite , nonzero limit when @xmath18 ! _ evaluating this quantity in the limit of @xmath18 , we get @xmath133 - mc^2 t ' \nonumber\\ & = & mvx ' + \frac{mv^2 t'}{2 } + \mathcal{o}\left ( \frac{1}{c^2}\right ) \label{wavephase}\end{aligned}\ ] ] this is precisely the mysterious phase which occurs in the schrodinger equation under a galilean transformation .
it has a simple interpretation as being equal to @xmath132 , thus emphasizing the role of rest energy _ even in the non - relativistic limit_. we will see later that this interpretation holds even in the presence of gravity , if we take the gravitational time dilation effect into account .
there is another peculiar result closely related to the one obtained above which deserves mention .
we have already seen that the action for a relativistic particle is invariant under the lorentz transformation while the action for the non - relativistic particle is not invariant under the galilean transformation .
but it turns out somewhat surprisingly that the action for the relativistic particle remains invariant even if we only retain the terms in the action upto and including @xmath0 ( and ignore terms of @xmath134 and higher ) and also treat the lorentz transformation to the same order of approximation .
this may be seen as follows : in the expression for the relativistic action for a free particle , we expand the integrand in a power series retaining terms upto and including order @xmath135 , and ignoring all higher order terms .
we then find : @xmath136 however , to the same order of approximation , we also know that : @xmath137 hence we find that @xmath25 and @xmath138 are identical in form , even to this order in @xmath5 !
( this is not at all an obvious result and it occurs due to the cancellation of the @xmath139 term coming from the velocity transformation , with that coming from the @xmath85 to @xmath61 transformation . ) clearly , when both these transformations are carried out consistently , the action remains form - invariant , even if we disregard @xmath134 and higher terms ! note that the action in is what a text book would have considered as being the `` same as the action in non - relativistic mechanics except for a constant added to the lagrangian '' .
but we have proved that the action for a non - relativistic particle is _ not _ invariant under a galilean transformation , but picks up an extra term .
if we express the change in the non - relativistic action under the galilean transformation as @xmath140 , we can write : @xmath141 = mc^2 \int dt \left(1 - \frac{v'^2}{2 c^2}\right ) - mc^2 \int dt \left(1 - \frac{v^2}{2 c^2}\right)\ ] ] we see that this difference @xmath140 bears a close resemblance to the difference between the corresponding relativistic action functionals , with the crucial difference being that the _ time has been kept invariant _ in while the expressions in and have integrals over @xmath85 and @xmath61 .
this shows clearly that the reason for the non - invariance of the galilean transformation lies in equating @xmath61 to @xmath85 .
if we take into account the difference between @xmath61 and @xmath85 , even to the lowest order in @xmath142 , it will lead to invariance of the non - relativistic action functional .
the invariance of the approximate action in allows us to obtain the result @xmath143\ ] ] the left hand side can be interpreted as the difference in the phase of a wave function arising due to the rest mass energy in two coordinate systems .
we see that this quantity is equal to the expression on the right hand side which is completely independent of @xmath3 ! in other words , this phase difference which occurs in a wave function through the factor @xmath144 $ ] survives to a _ completely non - relativistic expression _ in the @xmath18 limit as already seen in .
this result , which is surprising at first sight , arises from the fact that the difference between the coordinate time and the proper time carried by a moving clock is given by @xmath145 where the last equality holds to @xmath134 .
it follows that the phase difference term @xmath146 survives in the non - relativistic limit even though it has no simple interpretation in the non - relativistic limit .
this algebraic fact ( but not its origin or connection with the limiting process , etc . ) has been noted earlier and discussed in a different context in ref .
@xcite where it was concluded that there are some difficulties in reconciling the galilean transformation with quantum mechanics .
our analysis shows that this is not the case and that such a result is indeed _ expected _ when the galilean transformation is used correctly in quantum mechanics . the real surprise for which we have no simple explanation is the result that the relativistic action remains form invariant even when we work to the accuracy of @xmath134 .
the above result , as well as the invariance of the truncated action in , rely crucially upon our retaining the @xmath123 term in all the expressions .
in fact , if we define a non - relativistic energy as @xmath147 , then one can again show that the phase of the wave function given by ( @xmath148 ) transforms to ( @xmath149 ) under a lorentz transformation , if we again retain terms only to the accuracy of @xmath134 .
we next take up issues which arise when we switch on @xmath2 and move from quantum mechanics towards gqm in fig .
[ fig : physicscube ] .
one of the key questions which has been debated extensively in the literature is the validity of the principle of equivalence in quantum mechanics @xcite . in newtonian gravity
, the principle of equivalence can be introduced in many ways and it also leads to several different conclusions .
we can not , _ a priori _ , expect all these descriptions of the principle of equivalence ( or the resulting conclusions ) to be equally applicable when we proceed from the vertex qm to gqm in fig .
[ fig : physicscube ] .
we provide a brief critique of this issue in this section , essentially arguing that much of the debate can be settled by using an appropriate definition of the principle of equivalence in the quantum mechanical domain .
consider the following two versions of the principle of equivalence : ( a ) the dynamics of a system in a spatially homogeneous ( but possibly time dependent ) gravitational field @xmath150 is the same as the dynamics of the same system viewed in an accelerated frame with coordinates @xmath151 and @xmath152 .
( b ) the accelerations of different masses @xmath153 in a gravitational field are independent of the masses @xmath154 .
therefore , any measurement based on the trajectories of a particle moving in a given gravitational field can not be used to determine its mass .
it is obvious that the statement ( a ) implies statement ( b ) in non - relativistic mechanics ; we know that the acceleration @xmath155 produced in the non - inertial frame is just @xmath156 which is indeed independent of the mass of the particle . if the dynamics in the non - inertial frame is the same as that in a gravitational field , then we must have @xmath157 in the gravitational field implying statement ( b ) .
further , newton s law tells us that @xmath158 implying equality of inertial and gravitational masses .
( in this paper , the term ` gravitational mass ' refers to the _ passive gravitational mass _ of the object . ) obviously , a measurement of the trajectory of the particle can not be used to determine its mass .
a moment of thought shows that the last statement requires qualification in quantum theory .
we do not have trajectories in quantum theory but one assumes that operators like @xmath159 , etc . are observables in quantum mechanics .
if so , one should be able to devise a suitable experiment which measures , say , the dispersion in the position : @xmath160 .
this quantity certainly depends on the mass of the particle when the particle moves in a gravitational field or even if the particle is free .
in fact , the case of the free particle drives home the point without clouding the issue by involving acceleration , gravity etc . the trajectory of a free particle in non - relativistic mechanics is certainly independent of its mass , and any observation related to the trajectory can not determine its mass . but consider two free particles having masses @xmath161 and @xmath162 ( @xmath163 ) with identical initially prepared states in which the wave functions are gaussian wave packets of mean zero and width @xmath164 .
if we allow these wave packets to propagate in time , then the widths @xmath165 and @xmath166 of these wave packets at any later time @xmath167 will be given by : @xmath168 hence , the width of the wave packet at any later time depends on @xmath42 , the mass of the free particle . by measuring the width at a later time and knowing the initial width
, we can deduce the mass of the particle .
the same result holds for particles moving under a gravitational field ( see e.g. , @xcite ) .
clearly , this can not be used to prove any violation of the principle of equivalence .
note that the _ heisenberg equations of motion _ for the quantum mechanical particle are identical in structure to the classical equations of motion and read @xmath169 , and hence , are independent of mass .
nevertheless , the dispersion in particle position will depend on the mass , in general .
this shows that all the consequences which we derive from a particular version of the principle of equivalence in the classical domain may not hold in the quantum domain .
as another example illustrating the same result , consider the context of the gravitational hydrogen atom , with the coulomb force between the proton and the electron being replaced by the gravitational force between two particles of masses @xmath42 and @xmath170 respectively , with @xmath42 orbiting around @xmath170 and @xmath171 . for this system ,
the energy levels are given by : @xmath172 we see that the frequency of transition from one energy state to another will also be proportional to @xmath173 , and hence will depend on the mass of the orbiting particle .
thus , in principle , this frequency can be used to determine @xmath42 .
more generally , we can use the frequency of transitions between successive energy levels of a system composed of a massive particle orbiting around a `` nucleus '' , to determine the mass of the orbiting particle .
this is clearly impossible in the classical theory .
again , this experimental determination of mass should not be taken as a violation of the principle of equivalence . to see what is involved ,
let us consider another example consisting of a particle of mass @xmath42 , moving under the action of a potential : @xmath174 this represents a `` non - hookean '' spring when @xmath175 .
we will now choose the spring constant @xmath52 such that it is proportional to the mass of the particle : @xmath176 .
classically , the mass @xmath42 cancels out of the equations of motion , @xmath177 , showing that the acceleration is independent of the mass of the particle .
( more elaborately , one can consider several masses @xmath178 , each moving under a separate potential with the same @xmath179 and @xmath180 , so that each potential scales with the corresponding mass . )
once again , classically , we can not determine the mass by observations of the trajectory .
the situation is different in quantum mechanics .
we can evaluate the energy levels of this `` oscillator '' in the wkb approximation using the bohr - sommerfeld quantization rule : @xmath181 where @xmath182 is the quantum number corresponding to the energy level @xmath183 , and is assumed to be large compared to unity . using this integral
we can determine the scaling of @xmath183 with @xmath182 and @xmath42 for different values of the exponent @xmath179 .
we find ( appendix [ app3 ] ) that : @xmath184 when @xmath185 , @xmath186 which is the case of the harmonic oscillator ; in this case , @xmath183 is also independent of the mass @xmath42 .
however , when @xmath175 the energy levels @xmath183 and hence the spacing between energy levels depend on @xmath42 .
therefore , we can use quantum mechanical transitions between successive energy levels of any of these oscillators to determine their mass @xmath42 !
clearly , quantum mechanics allows such a determination which is not possible in classical theory based on measurements related to the trajectory of a particle . as the above example shows , this fact has nothing to do with gravity _ per se _ and is a general phenomenon .
( in fact , even in classical mechanics , if one measures dynamical variables like momentum , energy etc . , one can certainly determine the mass of a free particle . )
we believe that issues related to the determination of the inertial mass of a particle by quantum mechanical observations are red herrings to the proper formulation of the principle of equivalence in quantum mechanics . such a formulation can be based on the statement ( a ) in the beginning of the last section which embodies the mathematical equivalence of dynamical equations in an accelerated frame and in a gravitational field . given the schrodinger equation for a free particle in one frame of reference , if we transform it to any other arbitrarily accelerated frame of reference , we should obtain an equation identical in form to the schrodinger equation for a particle in the presence of an appropriate , in general time - dependent , _
gravitational field_. for example , if we transform from a frame @xmath187 to an arbitrarily accelerated frame @xmath188 , where @xmath94 is a function of @xmath61 , the schrodinger equation should pick up a `` potential energy term '' @xmath189 where @xmath190 is the instantaneous acceleration of @xmath33 with respect to @xmath31 .
we have already demonstrated this result earlier in .
the only new feature is that the wave functions in the two frames are not related by a scalar transformation , but require an additional phase factor to be added .
this phase factor , as we demonstrated , was required for consistency under galilean transformations , but the situation is less clear for the case of an accelerated transformation . in what follows , we shall investigate this result in greater detail from different perspectives .
in section [ subsec : lorgal ] , we showed that while the phase of the free particle klein - gordon equation is lorentz invariant , the corresponding phase of the schrodinger wave function is not invariant under a galilean transformation .
we also demonstrated that the transformation to an accelerated frame in the non - relativistic context leads to the schrodinger equation with a gravitational potential .
however , to obtain this result , we needed to explicitly include a phase in the wave function in .
the question arises as to how we can understand this phase in a general context when we are no longer dealing with the lorentz transformation , and the galilean transformation as its limiting form , but have to deal with non - inertial frames .
it is important to obtain the result in directly from a fully generally covariant equation , just as we obtained the results for the galilean transformation by taking the appropriate limit of the klein - gordon equation .
we shall now show how this can be done for a time dependent acceleration , which does not seem to have been discussed in the literature before . in the relativistic case
, we have a quantum scalar field satisfying the free - particle klein - gordon equation in one frame of reference , which we call @xmath30 ; the frame @xmath30 being arbitrarily accelerated with time - dependent acceleration @xmath150 with respect to an inertial coordinate system @xmath191 . in this frame ( known as the _ generalized rindler frame _ )
the metric is given by : @xmath192 the explicit co - ordinate transformation @xcite between @xmath31 and the inertial frame @xmath33 is given by : @xmath193 where @xmath194 is related to @xmath150 by : @xmath195 the generally covariant klein - gordon equation for a scalar field @xmath196 in this frame is given by : @xmath197 hereafter , unless mentioned otherwise , we use units in which @xmath198 . using the form of the metric as given in , we find : @xmath199 we now substitute @xmath200 and note the following : @xmath201 and @xmath202 @xmath203 substituting the above relations into , we find ( putting back the @xmath3- and @xmath4- factors ) : @xmath204 \left[-\frac{m^2 c^2 \psi}{\hbar^2 } - \frac{2i m}{\hbar } \frac{\partial \psi}{\partial t } + \frac{1}{c^2}\frac{\partial^2 \psi}{\partial t^2}\right ] & - & \nonumber\\ \left(\frac{imc^2\psi}{\hbar } - \frac{\partial \psi}{\partial t}\right)\frac{x}{c^4}\frac{dg}{dt } + \left(1 + \frac{g(t ) x}{c^2}\right)^3 \frac{\partial^2 \psi}{\partial x^2 } & + & \nonumber \\ \left(1 + \frac{g(t ) x}{c^2}\right)^2 \frac{\partial \psi}{\partial x}\frac{g}{c^2 } = \frac{m^2 c^2}{\hbar^2 } \left(1 + \frac{g(t ) x}{c^2}\right)^3\psi \end{aligned}\ ] ] retaining terms to the lowest order ( upto but excluding order @xmath205 ) , we get on simplification : @xmath206 which is identical to the schrodinger equation for a particle of mass @xmath42 in an accelerated frame of reference moving with acceleration @xmath207 or equivalently , in a time - dependent gravitational field of strength @xmath150 . hence , we see that the klein - gordon equation reduces in the appropriate limit to the schrodinger equation , with the term @xmath208 indicating the accelerated nature of the frame .
notice that we have excluded the terms of @xmath135 order .
this is consistent with our earlier comments on the `` accuracy '' of the galilean transformation . for the galilean transformation and the non - relativistic limit to be valid
, one must keep only those terms upto order @xmath5 and not beyond .
in fact , the phase @xmath111 acquired by the non - relativistic wave function acquires a direct physical meaning when we consider the schrodinger equation as a limit of the klein - gordon equation .
we can transform the free particle solution to the klein - gordon equation in the inertial frame , @xmath209 , as a scalar to the non - inertial frame , thus obtaining @xmath210 .
the non - relativistic limits of @xmath209 and @xmath210 will differ by a phase term @xmath211 , which , in the appropriate limit , will give the correct phase dependence arrived at in when we consider the effect of gravitational time dilation . in the presence of a gravitational potential @xmath212 , the proper time lapse @xmath213 of a co - moving clock is related to the coordinate time lapse @xmath214 by @xmath215\ ] ] so that , when @xmath216 , we get @xmath217\nonumber\\ & \approx&-m \int dt \left ( - \frac{\dot\xi^2}{2 } + x\ddot\xi \right ) = -mx\dot\xi+\frac{1}{2}m\int dt \dot\xi^2 \end{aligned}\ ] ] which is precisely the phase @xmath111 found in !
once again we see that a result in non - relativistic quantum mechanics acquires a simple interpretation when we treat it as a limit of relativistic theory , thanks to the factor @xmath211 in the phase .
the result also shows that in the instantaneous rest frame of the particle , the phase of the wave function evolves as @xmath218 , where @xmath219 is the proper time shown by the co - moving clock , thereby again validating the principle of equivalence in quantum mechanics .
we conclude this section with a comment related to the structure of the schrodinger equation in and its relationship to quantum field theory in the rindler frame . for a general @xmath94 , the potential in depends explicitly on time , and hence it does not allow stationary state solutions .
the exception occurs in the case of a uniformly accelerated frame for which @xmath220 is a constant . in that case , possesses stationary state solutions with the energy eigenfunctions being airy functions .
a similar situation arises in the fully relativistic context . for a time
dependent @xmath150 , the metric in is not stationary and the klein - gordon equation does not possess mode functions with an @xmath221 dependence . the exception is the case of constant @xmath222 , leading to the rindler metric , when such stationary solutions exist .
( in fact , these solutions , given in terms of hankel functions , reduce to the corresponding airy functions in the @xmath1 limit , as they should . )
. we know , however , that the positive frequency mode functions in the rindler frame are a superposition of positive and negative frequency mode functions in the inertial frame , leading to @xcite the well - known davies - unruh temperature @xmath223 .
when one takes the @xmath1 limit , this effect vanishes at the lowest order ( which is independent of @xmath3 ) but leads to a nontrivial result at the next order .
we hope to describe this and related features in a separate publication @xcite .
in the previous sections we dealt with the klein - gordon equation and the schrodinger equation without worrying about the physical interpretation of their solutions in the relativistic and non - relativistic contexts .
it is , however , well known that while the non - relativistic wave function has a standard interpretation , as the probability amplitude in a single particle theory , the solution to the klein - gordon equation can not be interpreted as a probability amplitude for the relativistic particle . in fact , consistent interpretation of quantum field theory requires us to recognize the existence of anti - particles and treating the solutions of relativistic wave equations as _ operator - valued _ entities .
this requires a closer scrutiny of the limits which we studied in the previous sections , which we shall now turn our attention to .
the solution @xmath224 of a klein - gordon equation treated as an operator , can be expanded using creation ( @xmath225 ) and annihilation ( @xmath226 ) operators in the second quantized formalism : @xmath227 where @xmath228 is a c - number solution to the klein - gordon equation labelled by the momentum @xmath229 .
the transition element @xmath230 between the standard vacuum state @xmath231 and the one - particle state @xmath232 with momentum @xmath233 is what we have taken to be the relativistic analog of a wave function in the previous sections . in the @xmath18 limit
, the combination @xmath234 will reduce to the schrodinger wave function for the free particle . in a similar way
, one can think of the state @xmath235 as representing a state with a particle at position @xmath236 and the quantity @xmath237|0\rangle\ ] ] as the amplitude for a particle to propagate from the event @xmath238 to event @xmath239 with the time ordering operator @xmath240 to take care of the distinction between particles and anti - particles .
relativistic quantum field theory makes use of the ( feynman ) propagator @xmath241 extensively and , in fact , all the standard results of interacting field theory can be obtained using this propagator .
it is , of course , possible to introduce a similar propagator in non - relativistic quantum mechanics , which can also be interpreted as the amplitude for a particle to propagate from one event to another , even though it is not as extensively used as the propagator in field theory .
given the fact that the solutions to the klein - gordon equation have no direct interpretation as probability amplitudes in quantum field theory , it is better to study the non - relativistic limit of quantum field theory in terms of the two propagators . in the present section
, we shall consider different aspects of this issue . in non - relativistic quantum mechanics , one can determine the wave function at @xmath242 from the wave function at @xmath243 , where @xmath244 , by a suitable integral kernel @xmath245 via the equation : @xmath246 this propagation kernel can be expressed formally as a sum over paths @xmath247 of an amplitude @xmath248)$ ] where @xmath249 $ ] is the action for the path satisfying the appropriate boundary conditions at @xmath250 .
for a free particle of mass @xmath251 , the kernel is given by : @xmath252 the sum over paths can be rigorously defined by a time - slicing procedure which assumes that at any intermediate time @xmath61 , with @xmath253 , the particle has a unique position . if the action is a quadratic functional , it can be shown that ( see e.g.,@xcite ) the propagation kernel has the form : @xmath254 where @xmath255 is the action for the classical path . in particular , the kernel for a free particle is given by : @xmath256 \,\theta(t_2-t_1)\nonumber\\ & \equiv & \theta(t_2-t_1 ) f(\mathbf{x}_2 , t_2 ; \mathbf{x}_1 , t_1)\end{aligned}\ ] ] where @xmath257 is the number of spatial dimensions .
( the @xmath258 function is just conventional ; purely algebraically , holds even for @xmath259 . )
it is also often convenient to define the propagator in _ the energy space _ by the definition : @xmath260 where @xmath261 .
this holds for any time - independent action functional ; for a free particle , we have the result : @xmath262}\ ] ] from , it also follows that : @xmath263 which we will call the _ transitivity property _ of the kernel . the same result , of course , holds for @xmath245 as well but we explicitly express it in the above form ( using @xmath264 ) to stress that purely algebraically this result is valid even without any time ordering .
we do _ not _ expect something analogous to to hold in the relativistic case .
a mathematical reason for this fact is simple : the relativistic mode function satisfies a second order differential equation in time , unlike its nonrelativistic counterpart which satisfies the schrodinger equation that is first order in time . hence the mode function at time @xmath265 can not be determined given _ only _ the mode function at an earlier time , unlike in the case of .
in fact , if the mode function @xmath212 and its first time derivative are known at some initial time @xmath266 , then , at some other time @xmath265 , we have : @xmath267\ ] ] where @xmath268 is the retarded green s function .
it is obvious that both @xmath212 and its first derivative are required to be known at the initial time @xmath266 in order to determine @xmath212 at the later time @xmath265 .
nevertheless , given the fact that we do have a valid action functional for a relativistic particle , one can formally define a kernel using the path integral as @xmath269/\hbar ] \label{eleven}\ ] ] where @xmath270 the second form of action expresses the trajectory @xmath271 in terms of a parameter @xmath272 chosen such that @xmath273 are the boundary values .
unfortunately , this action is not quadratic in the dynamical variables and hence the standard procedure for evaluating the path integral fails .
it is , however , possible to evaluate the path integral in _ exactly _ by at least two other methods .
the first involves utilizing the fact that the integral in is just the proper length of a given path connecting the end points .
we can implement the sum over paths in a euclidean lattice and take an appropriate limit to the continuum at the end , thereby evaluating .
this is worked out in detail in reference @xcite .
a second procedure is more heuristic , but physically transparent , and is based on the fact that even in non - relativistic quantum mechanics , there exists an action functional , called the jacobi action , which has a structure similar to the one in involving a square root .
we shall briefly indicate how this procedure can be used to explicitly evaluate the path integral in .
we shall essentially prove that the following relation holds : @xmath274\right\ } \label{nineteen}\end{aligned}\ ] ] ( on either side of the equation , the overdot denotes differentiation with respect to the integration variable used . )
the expression on the left hand side is a path integral defined using the jacobi action for a particle propagating from one spatial point to another with energy @xmath275 in an external potential @xmath34 .
the right hand side involves the fourier transform of the usual path integral ( given in the curly brackets ) which we know how to evaluate .
this relation allows us to give meaning to path integrals involving re - parameterization invariant action functionals which have square - roots in them .
the proof of is given in appendix [ app4 ] .
we can evaluate the path integral for the relativistic particle involving a square root in the action using the result in by making the following substitutions in : ( i ) @xmath276 , ( ii ) @xmath277 , ( iii ) @xmath278 , and a change of sign on the integration variable .
thus , becomes : @xmath279 on transforming the integration variable on the left - hand side of the above equation from @xmath280 to @xmath281 , we find that it becomes : @xmath282 which is exactly the relativistic propagator with the square - root action that we were interested in evaluating ( ) .
the sum over paths within the curly brackets on the right - hand side of can be evaluated by the standard procedure for a free particle which allows the relativistic kernel to be expressed as the integral @xmath283 where we have done a rescaling from @xmath219 to @xmath284 to express it in the standard form .
the @xmath241 is the standard feynman propagator in schwinger s proper time representation , which differs from the kernel by an unimportant constant .
( dimensionally , this constant is to be expected , because the feynman propagator has dimensions of square of inverse length , while the kernel which we have evaluated has the dimensions of cube of inverse length ) .
the way we have obtained the relativistic kernel shows that it is actually the result of evaluating a path integral , with the action for a relativistic particle as to be expected .
hereafter , we shall work with the feynman propagator @xmath241 , as it is the usual quantity of interest in quantum field theory .
as usual , the factor @xmath285 is not strongly convergent in the upper limit of the integration .
this difficulty can be tackled by interpreting @xmath286 as @xmath287 or by rotating the integration contour to the imaginary axis by going over to the euclidean sector .
the euclidean propagator is given by : @xmath288 where @xmath289 is the modified bessel function of the second kind .
the above expression can indeed be interpreted as the amplitude for a relativistic particle to propagate from one event @xmath290 to another event @xmath236 , and hence is equal to @xmath291|0\rangle$ ] .
but it does not propagate any `` wave function '' in the relativistic theory for reasons explained earlier .
we , however , would expect this propagator to reduce to the standard non - relativistic kernel in the @xmath18 limit .
we shall now describe how this comes about .
let us first consider the non - relativistic ( @xmath1 ) limit of .
it is easy to see that , in this limit , the integral in can be evaluated in the saddle point approximation .
( in this section , we set @xmath292 . ) to avoid mathematical ambiguities , it is convenient to work in the euclidean space and start with : @xmath293 where we have rescaled @xmath294 in to @xmath295 so that @xmath294 has the dimensions of length ( or time ) .
the stationary phase of this integral occurs at @xmath296 , where @xmath297 , so that the integral evaluates to @xmath298 where we have put back the @xmath299factors in the phase .
the prefactor is given by : @xmath300 where @xmath301 is the saddle point of the integrand and @xmath302 is the argument of the exponential .
we now take the @xmath303 limit of the above expression . noting that , in this limit , @xmath304 where @xmath305 , we get ( relabeling @xmath241 to @xmath245 , the kernel ) : @xmath306 finally , going back to lorentzian time with @xmath307 , we obtain : @xmath308 which matches exactly with the non - relativistic path - integral kernel for a free particle in three dimensions , apart from two standard factors : ( i ) the factor @xmath309 is the standard extra term in the action functional which arises on taking the non - relativistic limit of the fully relativistic action functional .
this is exactly the phase term which we factored out in the wave function , while going from the klein - gordon equation to the schrodinger equation for an accelerated particle , previously in section [ sec : nringrav ] .
( ii ) the factor @xmath310 arises due to the difference in normalization conventions for the relativistic and non - relativistic mode functions . in non - relativistic quantum mechanics ,
the free - particle wave functions are normalized with the factor @xmath311 , while in quantum field theory , the mode functions are normalized with the factor @xmath312 .
since the feynman propagator is bilinear in the mode functions , this leads to an extra factor @xmath313 in the feynman propagator as compared to the path - integral kernel .
the factor @xmath313 in the non - relativistic limit reduces to just @xmath314 . by the same procedure
, we can also compute the first - order relativistic correction to the path integral . to do this ,
we retain one more order in @xmath5 in the expansion of @xmath315 in to obtain : @xmath316 this gives the correction to the non - relativistic propagator to one higher order in @xmath317 .
we hope to discuss the physics from these corrections in a future work .
the above analysis shows that the feynman propagator of the field theory does reduce to the correct free particle propagator in non - relativistic quantum mechanics in spite of all the interpretational differences between the two theories .
this is related to the following facts .
\(i ) the time component of the conserved probability current for the klein - gordon equation reduces to @xmath318 in the appropriate limit , allowing for the standard probabilistic interpretation .
since , in this limit , the klein - gordon equation becomes effectively first order in time , we find that also holds to the correct order of approximation . more formally ,
the second order propagation equation for the mode functions of the klein - gordon equation does reduce to ( with @xmath241 reducing to @xmath245 ) in the appropriate limit , making everything consistent .
\(ii ) in the non - relativistic path integral defining the propagator @xmath245 , we sum over paths which always go forward in time .
however , in the expression in the curly bracket in which gives the corresponding path integral in four dimensions , particles go forward _ in proper time _ but their paths may go forward or backward in coordinate time . on the left hand side of ,
this will correspond to paths with @xmath319 for which the amplitude is exponentially suppressed .
all these effects vanish in the @xmath320 limit , thus allowing a single particle interpretation .
\(iii ) the situation is very similar to that of paraxial propagation in optics , which allows us to approximate the second - order wave equation into one which involves only first derivatives with respect to a given coordinate , viz . , the coordinate along which light is propagating .
the mathematics is identical in both the cases @xcite . in the case involving a beam of light ( of frequency @xmath321 ) traveling along the @xmath322axis from a slit located at @xmath323 , say , to a point on a screen with coordinates @xmath324 , with @xmath325 ( ` paraxial propagation ' ) , the relevant physics is contained in the phase difference acquired by the beam corresponding to a path difference @xmath326 : @xmath327^{1/2}-(z_2-z_1 ) \right\ } \approx \frac{\omega}{c}\frac{(\mathbf{x}_2^\perp-\mathbf{x}_1^\perp)^2}{2(z_2-z_1 ) } \label{optics}\ ] ] for a relativistic particle of mass @xmath42 , the corresponding phase is related to the proper time lapse @xmath328 and is given by : @xmath329^{1/2}-c(t_2-t_1 ) \right\ } \label{relpart}\ ] ] which reduces , in the @xmath1 limit ( which implies the paraxial propagation condition : @xmath330 along the time axis ) , to : @xmath331 which is precisely the phase of the quantum mechanical kernel for the nonrelativistic particle , arising again from the rest mass energy term in the appropriate limit .
just as in paraxial optics we retain the propagation of light along the forward @xmath332 direction ( from slit to screen , say ) , ignoring backward propagation , in arriving at nonrelativistic quantum mechanics from field theory , we retain the propagation of the particle forward in the time axis , ignoring backward propagation . in both the contexts ,
the approximation effectively reduces the degree of the relevant differential equation from two to one in the appropriate independent variable ( which is @xmath332 in optics and @xmath61 in quantum theory ) .
that is , in arriving at non - relativistic quantum mechanics from quantum field theory , one considers paraxial propagation of the klein - gordon modes with the axis now being along the time direction .
we have already seen in section [ sec : nringrav ] that , at the level of differential equations , the relevant limits work out correctly : the @xmath18 limit of the klein - gordon equation in a homogeneous , time dependent , gravitational field gives us the correct schrodinger equation .
further , except for a phase whose origin we can understand , the same schrodinger equation is obtained in a non - inertial frame mimicking the same gravitational field .
this establishes the principle of equivalence in the context of quantum mechanics at the level of the field equations .
but , as mentioned earlier , there is a significant difference in the interpretation of the solutions of the klein - gordon and schrodinger equations in the two contexts .
therefore , it is not clear whether the formal equivalence of the two differential equations in the appropriate limit ensures the validity of the principle of equivalence . a more rigorous way of establishing this result would be to work at the level of propagators , and show that the feynman green s function in a non - inertial frame reduces in the @xmath1 limit to the non - relativistic kernel for the appropriate gravitational field .
we shall now describe how this comes about in the case of a constant acceleration @xmath333 in the negative @xmath35direction , or equivalently , a constant gravitational field @xmath222 in the @xmath334 direction .
the non - relativistic kernel for this case is given by ( @xcite ) : @xmath335 the transformation from an inertial frame @xmath336 to an accelerated frame @xmath337 is given by @xmath338 the feynman propagator in this uniformly accelerated frame can be obtained from the propagator in the inertial frame , given by @xmath339 by expressing the line interval @xmath340 in terms of the rindler coordinates .
( this result holds because of the bi - scalar nature of the feynman propagator . ) elementary algebra using gives the line interval between the two events @xmath341 and @xmath342 to be @xmath343\ ] ] so that the propagator becomes @xmath344 \end{aligned}\ ] ] the above integral can be again approximated by the method of stationary phase .
the phase of the integrand becomes stationary for the value of @xmath280 given by : @xmath345^{1/2 } \end{aligned}\ ] ] substituting for @xmath280 in , we get : @xmath346^{1/2 } \end{aligned}\ ] ] where @xmath182 is the standard normalization factor arising from the saddle - point approximation .
we will now obtain the non - relativistic @xmath347 limit of the above expression by expanding the exponent in a power series retaining terms upto order @xmath348 inside the radical .
this gives : @xmath349\right]^{1/2 } \end{aligned}\ ] ] on simplification , ignoring all terms of order @xmath350 and higher inside the bracket , we obtain : @xmath351 \end{aligned}\ ] ] which simplifies to : ( with @xmath352 ) @xmath353 which , apart from the term @xmath354 , is identical to the phase of the standard , non - relativistic feynman path integral for a particle of mass @xmath42 , moving with a constant acceleration @xmath222 in the negative @xmath35direction , or equivalently , situated in a uniform gravitational field of strength @xmath222 in the positive @xmath35direction , as given in ; see also in appendix [ app5 ] with @xmath355 replaced by @xmath356 .
the normalization factor @xmath182 also works out correctly just as in the case of the free particle .
hence , we have shown that the relativistic feynman propagator in the rindler frame reduces , in the appropriate limit , to the corresponding non - relativistic feynman path integral for the case of a particle moving with uniform acceleration .
we now generalize the above result and formally show how the limiting form of the non - relativistic propagator can be obtained for _ any _ weak newtonian gravitational field , starting from a fully relativistic description .
such a weak field is described by a metric of the form @xmath357 where @xmath212 is the newtonian potential .
the feynman propagator again has the integral representation @xmath358 where @xmath340 now represents the proper length of a path between two events evaluated using the metric in . assuming that we can formally interchange the sum over paths and the integration over @xmath280
, we can express the above result as @xmath359\right\}\nonumber\\ & = & \sum_{\rm paths } \int_0^\infty ds\ e^{-im^2c^2s/\hbar^2 } \exp\left\ { -\frac{ic^2}{4 } \int_0^s d\tau \ \dot t^2 \right\ } f(s)\end{aligned}\ ] ] where @xmath360 is treated as a function of the parameter @xmath280 .
thus , we can write the expression for the propagator as : @xmath361 where we have identified the sum over paths as functional integrals over spatial and time coordinates : @xmath362 in order to estimate the above integral in the @xmath303 limit , we note that this limit is equivalent to taking a saddle point on the integrand treated as a function of the parameter @xmath280 .
the saddle point of the integrand is given by : @xmath363 we now evaluate the integral in in the saddle point limit given by .
the functional integral over all paths @xmath364 reduces , in saddle point , to the condition that @xmath365 = constant . from our preceding argument for the saddle point of @xmath280
, we see that the value of this constant is @xmath366 .
if we use these results in the expression for @xmath367 , we find : @xmath368 \ , \exp\left [ -\frac{ic^2}{4 } \frac{\hbar t}{2
m } \left(\frac{4m^2}{\hbar^2}\right)\right ] \nonumber\\ & & \qquad\qquad \qquad \times \qquad \exp\left [ - \frac{i}{4}\int_0^s d\tau \left ( 2\phi \ \dot t^2 - \dot{\mathbf { x}}^2\right)\right]\nonumber\\ & = & \int \mathcal{d}\mathbf{x } \ , e^{-imc^2t/\hbar } \exp\left\
{ -\frac{i}{4 } \int_0^s d\tau\ , \dot t^2 \left [ 2\phi - \left(\frac{d\mathbf{x}}{dt}\right)^2\right]\right\ } \end{aligned}\ ] ] with these , admittedly formal , manipulations we have succeeded in converting a four - dimensional functional integral ( over @xmath61 and @xmath369 ) and an ordinary integral over @xmath280 into the standard three - dimensional functional integral over @xmath369 which is appropriate for the non - relativistic limit . according to our previous arguments , the validity condition for the saddle - point approximation is that @xmath365 is a constant , and hence , it can be taken out of the integral in the phase of the exponential , and substituted by @xmath370 .
we also change the variable of integration form @xmath219 to @xmath61 ( noting that the entire operation is performed under the condition that , at @xmath371 , @xmath372 ) .
we get : @xmath373\right\ } \nonumber\\ & = & \int \mathcal{d}\mathbf{x } \ , e^{-imc^2t/\hbar}\exp \left\ { -\frac{im}{\hbar } \int_0^{t } \phi \ , dt+ \int_0^{t } \frac{im}{2\hbar } \left(\frac{d\mathbf{x}}{dt}\right)^2 dt \right\ } \nonumber\\ & = & \int \mathcal{d}\mathbf{x } \ , e^{-imc^2t/\hbar}\exp \left\ { \frac{i}{\hbar}\left [ \int_0^{t}\frac{m}{2 } \left(\frac{d\mathbf{x}}{dt}\right)^2 dt - \int_0^{t } m\,\phi\ , dt \right]\right\}\end{aligned}\ ] ] the above expression is identical in form ( except , of course , for the standard @xmath374 phase ) to the non - relativistic feynman path integral for a particle of mass @xmath42 in an external potential @xmath212 .
this proves that in the @xmath303 limit , the feynman propagator does indeed reduce to the non - relativistic feynman path integral for the case of an external weak gravitational potential .
in non - relativistic quantum mechanics , the kernel satisfies the transitivity condition given by .
this represents the fact that a particle which propagates from @xmath375 to @xmath376 must exist at some unique spatial location @xmath377 at any intermediate time @xmath265 .
we can not have a similar transitivity condition obeyed in terms of spatial coordinates , which is apparent from the fact that time and space coordinates enter differently in the kernel . in the path integral
, one evaluates paths which go forward in time but can go forward or backward in spatial coordinates . in the case of the feynman propagator , we instead work with the path integral in in which the particles go forward in the _ proper time _
@xmath280 , but can go forward or backward both in coordinate time and spatial coordinates . after evaluating the path integral ,
we do a fourier transform with respect to the proper time to obtain the feynman green s function .
it is obvious that , just as we can not have transitivity vis - a - vis spatial coordinates in the non - relativistic kernel , we can not have transitivity in spatial _ or _ time coordinates for the feynman propagator .
one can , in fact , prove that the transitivity integral for the feynman propagator leads to the result @xmath378 \label{twentynine } \end{aligned}\ ] ] where the factor in square brackets expresses the lack of transitivity .
we shall provide a simple demonstration of this result since we have not seen it discussed in the literature
. the result can be obtained from the schwinger representation of the propagator by straightforward integration ( as shown in appendix [ app6 ] ) .
however , it is somewhat easier to prove the result using the well known momentum space representation of the feynman propagator given by @xmath379 explicit substitution of in gives : @xmath380\left [ \frac{d^4k}{(2\pi)^4 } \
, \frac{i e^{-ik(x_2-x_1)}}{k^2 - m^2 + i\epsilon}\right]\nonumber\\ & & \qquad\qquad= -\frac{1}{(2\pi)^4}\int \frac{d^4p \ , e^{-ip(x_3-x_1)}}{(p^2 - m^2 + i\epsilon)^2}\end{aligned}\ ] ] since @xmath381 we immediately get the result in . once again , in the non - relativistic limit , the extra factor in reduces to unity which can be demonstrated by straightforward manipulation .
it is also interesting to notice that the kernel in , which contains the first - order relativistic correction to the non - relativistic path integral , does not possess the transitivity property due to the presence of additional terms in the integrand .
the origin of this non - transitivity can be understood from the fact that , _ even in the non - relativistic case _ , the _ energy kernel _ is not transitive .
this is to be expected purely from dimensional analysis and can be seen as follows .
the energy kernel in non - relativistic quantum mechanics is given by : @xmath382 so that @xmath2 has the dimensions of @xmath383 $ ] .
if @xmath2 is to be transitive , we must have @xmath384 which is clearly impossible since the left - hand side has dimensions of @xmath385 $ ] and the right - hand side , that of @xmath383 $ ] .
clearly , one needs a factor having dimensions of time ( or equivalently , of inverse energy ) on the right - hand side of to make it dimensionally consistent .
the correct result can be easily obtained when the action has no explicit time dependence .
the non - relativistic energy kernel can be expressed as : @xmath386 where @xmath387 is the energy eigenfunction corresponding to the eigenvalue @xmath388 .
hence , we get , on using the orthonormality of energy eigenfunctions : @xmath389\end{aligned}\ ] ] which is similar in form to the result we obtained above in for the feynman propagator .
in fact , the equivalence arises from the fact that the feynman propagator is just the energy fourier transform ( with @xmath275 replaced by @xmath123 ) of the path - integral with the relativistic quadratic action , as is clear from .
( the sign flip between the right - hand sides of and is unimportant , being the result of the difference in the fourier transform conventions used in their derivations . ) hence , the non - transitivity of the feynman propagator can be understood directly from this fact , and the result does not require any deeper quantum field theoretic concepts ( like that of antiparticles ) for its explanation .
we have discussed several aspects of the non - relativistic limit of quantum field theory in both inertial and noninertial frames and in weak gravitational fields . starting from curious subtleties in the transition from special relativity to classical mechanics ,
we have explained the physical origin of the extra term in the non - relativistic action under galilean transformations , thereby clarifying some points discussed previously in the literature .
we have taken the non - relativistic limit of the generally covariant klein - gordon equation , and shown that it reduces to the schrodinger equation in the presence of the corresponding gravitational field , thus providing an interpretation of the principle of equivalence in quantum mechanics .
we also showed how such limiting processes can be described , fairly rigorously , in the language of propagators , which is essential because relativistic field theory does not allow an interpretation in terms of a single particle wave function . along the way
, we have indicated how the relativistic feynman propagator can be explicitly obtained from the standard special relativistic square - root action functional and connected it up with the schwinger proper time description .
we have also obtained the non - relativistic limit of the feynman propagator in the inertial and rindler frames , and in a weak gravitational field , and demonstrated its reduction to the corresponding path integral kernels .
thus , we have effectively explored the gqm vertex of the physics cube ( fig.[fig : physicscube ] ) , and studied the @xmath390 plane as a projection from other parts of the cube . while some of these results are obvious with hindsight , we were persuaded to present the details because we have not seen them discussed properly in textbooks or in published literature . as fig.[fig
: physicscube ] shows , the exact theories of physics allow useful , but approximate , limiting forms . sometimes , these approximate versions though incorrect can be internally consistent with all their features being understandable within the approximation .
but in some other contexts , approximate theories can host features which are residues of the more exact description .
this paper highlights some features of the nonrelativistic limit which belong to the latter class .
we saw that the behaviour of the action functional and the phase of the wave function in the nonrelativistic limit have no simple explanation within the nonrelativistic theory .
but they emerge very naturally from the nonrelativistic limit of the term @xmath391 , thereby acquiring a direct physical meaning in the relativistic domain .
this interpretation , as we saw , holds even in a noninertial frame and in the presence of a weak gravitational field . what is curious is the fact that this phase has a nonzero , finite , limit even when @xmath1
this is essential to make quantum mechanics and galilean relativity consistent ; essentially , one needs to start with relativistic quantum theory and lorentz transformations , and take appropriate limits to understand what happens when one combines quantum mechanics with galilean relativity !
this is an example of a more exact theory leading to curious residues in its approximate version .
this paper was written when one of the authors ( t.p . )
was visiting the institute of astronomy , cambridge and the hospitality of the ioa is gratefully acknowledged .
the research of t.p is partially supported by the j.c .
bose fellowship of dst , india .
the research of h.p is supported by the kvpy fellowship of dst , india .
we thank d.v .
ahluwalia and c.s .
unnikrishnan for detailed comments on the manuscript .
in this appendix we prove that the wave function satisfies the schrodinger equation , .
we set @xmath392 for convenience , so that becomes : @xmath393 the co - ordinate transformation is given by @xmath394 .
we now substitute @xmath395 into the above equation , where @xmath396 .
we have the following relations : and @xmath398 hence , @xmath399 where we have used the facts that @xmath400 and @xmath401 using these relations , becomes : @xmath402 we also know that @xmath403 and @xmath404 using the above relations in , it readily transforms to : @xmath405 which is satisfied identically , since we know that @xmath406 is a solution to the free particle schrodinger equation in the @xmath407 frame of reference .
hence , we see that the wave function in satisfies . in this appendix we determine the scaling of the energy levels @xmath183 of the ( in general ) non - hookean oscillator with the quantum number @xmath182 . consider which , for the case of the power - law potential ( [ powerlawpot ] ) , becomes : @xmath408 this integral is to be evaluated between the two turning points .
as the potential is symmetric in @xmath239 , we can write : @xmath409 substituting @xmath410 , we get : @xmath411 since we are only interested in the scaling of @xmath183 with @xmath182 and @xmath42 , we can see that ( using the fact that @xmath412 ) : @xmath184 hence , we see that , only in the case @xmath185 , we have @xmath186 , with the mass dependence scaling out . in all other cases , the energy levels @xmath183 , and consequently , the transition frequency between successive energy levels , will depend on @xmath42 , the mass of the oscillator . [ the full expression for @xmath183 in the wkb approximation has been worked out in , for e.g. , @xcite . ]
the result in can be proved by using the hamiltonian form of the feynman path integral and converting the path integral sums over the coordinates and momenta in an appropriate form .
we will need to use the following three standard results in functional integration : @xmath413 the is just the definition of delta functional ; and can be easily obtained in the euclidean sector by usual time - slicing techniques and can be analytically continued .
they represent direct generalizations of the corresponding results for ordinary integrals .
to prove , we begin by writing the standard feynman path integral in the hamiltonian form : @xmath416 where the sum is over all functions @xmath417 but only those functions @xmath247 which satisfy the boundary conditions .
( we will assume @xmath418 . )
we now introduce inside the functional integral , the factor unity in the form : @xmath419 leading to : @xmath423 \sum_{\mathbf{p } } \exp i\int dt\,\left [ \mathbf{p \cdot \dot x } + \frac{\lambda(t)}{2 m } p^2\right]\nonumber\\ & = & \sum_{\mathbf{x}}\sum_{\lambda(t ) } \exp i \int dt\ , \lambda(t)\left [ v(\mathbf{x } ) - e\right ] \exp i\int dt\,\frac{1}{2 } \frac{m}{\lambda(t)}\mathbf{\dot{x}}^2\nonumber\\ & = & \sum_{\mathbf{x}}\sum_{\lambda(t ) } \exp i \int dt\ , \left[\lambda(t ) \left ( v(\mathbf{x } ) - e\right ) + \frac{m}{2\lambda(t ) } \mathbf{\dot{x}}^2\right]\nonumber\\ & = & \sum_{\mathbf{x } } \exp
i \int dt\ , \sqrt { 2 m ( e - v)\ , |\mathbf{\dot{x}}|^2}\end{aligned}\ ] ] the left hand side is just the propagator for a particle with energy @xmath275 , which now has an explicit path integral representation in terms of a square - root action .
using this result in , we obtain . in this appendix
, we describe how the non - relativistic _ kernel _ transforms when one goes over from an inertial frame to an accelerating frame , and use our general result to verify explicitly the form of the kernel for the case of a particle moving with uniform acceleration .
we consider the two frames of reference @xmath31 and @xmath33 , where the particle is free in @xmath424 and accelerated in the frame @xmath425 ; @xmath31 and @xmath33 being related by the transformation @xmath426 ; with @xmath427 .
let @xmath428 be the non - relativistic propagator in @xmath33 , where @xmath429 and @xmath430 are the co - ordinates of the final and initial points respectively .
then , it turns out that in the frame @xmath31 , in which the particle has an acceleration @xmath355 , the propagator @xmath431 takes the form : @xmath432 where @xmath433 here , the overdot denotes differentiation with respect to time , and an expression like @xmath434 denotes the time derivative of @xmath435 evaluated at @xmath436 .
the above relationship is easy to understand along the following lines : recall that the lagrangian picks up a total time derivative term as we go from frame @xmath33 to frame @xmath31 ( or vice - versa ) .
since we are dealing with a free particle , the lagrangian is a quadratic function of the co - ordinates and velocities , and the transformation preserves the quadratic nature of the lagrangian . we know
that , for quadratic lagrangians , the propagator contains a phase term that is proportional to the action functional for the classical trajectory between the initial and the final points . from
, we see that @xmath25 and @xmath138 , the classical action functionals in the two frames , are related by @xmath437 .
hence , we would expect that the propagators in the same two frames are related by @xmath438 $ ] where @xmath439 and @xmath440 . from we
can see that @xmath441 from which the result ( [ ktrans ] ) follows .
we now specialize to the case of uniform acceleration .
we have a free particle in the frame of reference @xmath442 and we transform from this frame to frame @xmath443 where @xmath444 , @xmath445 , and @xmath110 is a constant .
consider first the free particle kernel @xmath446 in the frame @xmath407 which is given by : @xmath447 where @xmath182 is a normalization constant , which depends on the initial and final time co - ordinates @xmath448 and @xmath436 respectively , and we have used the fact that @xmath445 . since we have translational invariance as far as the time co - ordinate is concerned , we can replace @xmath448 and @xmath436 by @xmath449 and @xmath240 respectively .
we now substitute for the co - ordinate @xmath86 in terms of @xmath239 and also add the phase factor @xmath450 . here , @xmath451 is given by : @xmath452 substituting and into , we find : @xmath453 \end{aligned}\ ] ] on simplification , we get : @xmath454 which matches identically with the standard expression for the propagator for a uniformly accelerated particle moving with acceleration @xmath110 @xcite .
hence , we have verified the validity of our general formula for the special case of the frame @xmath30 being a uniformly accelerated frame of reference .
we provide a direct proof of working in the euclidean sector and evaluating the integral on the left hand side explicitly . from the expression for the euclidean feynman propagator : @xmath455 in the left hand side of
, we get : @xmath456 now , using the standard result : @xmath457 we find that @xmath458\ ] ] transforming the variables to @xmath459 and @xmath460 gives : @xmath461 this demonstrates the result in .
d. v. ahluwalia , _ gen .
* 29 * , 1491 ( 1997 ) ; c. s. unnikrishnan , _ mod . phys . lett .
a _ * 17 * , 1081 ( 2002 ) ; c. s. unnikrishnan , g. t. gillies , [ arxiv:1106.4151 ] ; g. z. adunas , e. rodriguez - milla , d. v. ahluwalia , _ gen .
_ * 33 * , 183 , ( 2001 ) [ arxiv : gr - qc/0006022 ] ; e. kajari et al . , _ app
b , lasers and optics _ * 100 * , 43(2010 ) ; e. okon and c. callender , _ eur . jour .
phil . of sci .
_ * 1 * , 133 ( 2011 ) ; md . manirul ali , a. s. majumdar , d. home , a. k. pan , _ class .
* 23 * , 6493 ( 2006 ) ; r. aldrovandi , j. g. pereira and k. h. vu,_class .
quan . grav _ ,
* 21 * , 51 ( 2004 ) ; a. herdegen and j. wawrzycki , _ phys .
* 66 * , 044007 ( 2002 ) ; m. p. silverman , _ gen .
* 20 * , 247 ( 1988 ) ; p. r. holland , _ found .
* 2 * , 471 ( 1989 ) ; k. k. nandi and yuan - zhong zhang _ phys .
d _ * 66 * , 063005 ( 2002 ) ; p. chingangbam and p. sharan , _ phys .
rev . a _ * 64 * , 042107 ( 2001 ) ; y. q. cai and g. papini , _ gen
* 22 * , 259 ( 1990 ) ; p. d. mannheim , _ phys .
a _ * 57 * , 1260 ( 1998 ) ; j. -l .
staudenmann et al .
* 21 * , 1419 ( 1980 ) ; t. goldman , m. v. hynes , and m. m. nieto , _ gen .
grav . _ * 18 * , 67 ( 1986 ) ; c. lmmerzahl , _ gen . rel
_ * 28 * , 1043 ( 1996 ) .
p. yu zhu ,
z. j. shen and y. s. wu , _ phys .
d _ * 32 * , 368 ( 1985 ) ; a. peters , k. y. chung and s. chu , _ metrologia _ * 38 * , 25 ( 2001 ) ; f. v. kowalski , _ phys . rev .
a _ * 51 * , 120 ( 1995 ) ; k. c. littrell , b. e. allman and s. a. werner , _ phys . rev .
a _ * 56 * , 1767 ( 1997 ) ; h. rauch , _ pramana _ * 25 * , 383 ( 1985 ) ; u. bonse and t. wroblewski , _ phys .
_ * 51 * , 1401 ( 1983 ) ; u. bonse and t. wroblewski , _ phys . rev .
d _ * 30 * , 1214 ( 1984 ) ; a. camacho , and a. camacho - galvn _ rep .
phys . _ * 70 * , 1937 ( 2007 ) ; d. m. greenberger , _ rev .
phys . _ * 55 * , 875 ( 1983 ) ; d. m. greenberger , a. w. overhauser , _ rev .
phys . _ * 51 * , 4378 ( 1979 ) .
d. v. ahluwalia , _
phys.lett.a _ * 275 * , 31 ( 2000 ) ; d. v. ahluwalia , _ mod.phys.lett.a _ * 13 * , 1393 ( 1998 ) ; d.v .
ahluwalia , c. burgard , _ gen .
_ * 28 * , 1161 ( 1996 ) ; ( errata ) * 29 * , 681 ( 1997 ) ; d. v. ahluwalia , c. burgard , _ phys.rev.d _ * 57 * , 4724 ( 1998 ) ; n. fornengo , c. giunti , c. w. kim and j. song , _ phys . rev
. d _ * 56 * , 1895 ( 1997 ) ; m. m. nieto , _ hyperfine interactions _ * 100 * , 193 ( 1996 ) ; r. m. crocker , c. giunti and d. j. mortlock , _ phys .
* 69 * , 063008 ( 2004 ) .
kaempffer , _ concepts in quantum mechanics _ ,
appendix 7 ( if galileo had known quantum mechanics ) ( academic press , new york - london , 1965 ) ; d. giulini , _ ann . of phys .
_ * 249 * , 222 ( 1996 ) ; v. bargmann , _ ann . of math . _
* 59 * , 1 ( 1954 ) ; j - m .
levy - leblond , _ jour .
* 4 * , 776 ( 1963 ) ; d. v. ahluwalia , _ class.quant.grav _ * 22 * , 1433 ( 2005 ) . | we discuss the non - relativistic limit of quantum field theory in an inertial frame , in the rindler frame and in the presence of a weak gravitational field , and attempt to highlight and clarify several subtleties . in particular , we study the following issues : ( a ) while the action for a _ relativistic _
free particle is invariant under the lorentz transformation , the corresponding action for a _ non - relativistic _
free particle is not invariant under the galilean transformation , but picks up extra contributions at the end points .
this leads to an extra phase in the non - relativistic wave function under a galilean transformation , which can be related to the rest energy of the particle _ even in the non - relativistic limit_. we show that this is closely related to the peculiar fact that the relativistic action for a free particle remains invariant even if we restrict ourselves to @xmath0 in implementing the lorentz transformation .
( b ) we provide a brief critique of the principle of equivalence in the quantum mechanical context .
in particular , we show how solutions to the generally covariant klein - gordon equation in a noninertial frame , which has a time - dependent acceleration , reduce to the non - relativistic wave function in the presence of an appropriate ( time - dependent ) gravitational field in the @xmath1 limit , and relate this fact to the validity of the principle of equivalence in a quantum mechanical context .
we also show that the extra phase acquired by the _ non - relativistic _ wave function in an accelerated frame , actually arises from the gravitational time dilation and survives in the non - relativistic limit .
( c ) while the solution of the schrodinger equation can be given an interpretation as being the probability amplitude for a single particle , such an interpretation fails in quantum field theory .
we show how , in spite of this , one can explicitly evaluate the path integral using the ( non - quadratic ) action for a relativistic particle ( involving a square root ) and obtain the feynman propagator .
further , we describe how this propagator reduces to the standard path integral kernel in the non - relativistic limit .
( d ) we show that the limiting procedures for the propagators mentioned above work correctly even in the presence of a weak gravitational field , or in the rindler frame , and discuss the implications for the principle of equivalence . |
the thomas fermi ( tf ) equation has proved useful for the treatment of many physical phenomena that include atoms@xcite , molecules@xcite , atoms in strong magnetic fields@xcite , crystals@xcite and dense plasmas@xcite among others .
for that reason there has been great interest in the accurate solution of that equation , and , in particular , in the accurate calculation of the slope at origin@xcite . besides
, the mathematical aspects of the tf equation have been studied in detail@xcite .
some time ago liao@xcite proposed the application of a technique called homotopy analysis method ( ham ) to the solution of the tf equation and stated that `` it is the first time such an elegant and explicit analytic solution of the thomas fermi equation is given '' .
this claim is surprising because at first sight earlier analytical approaches are apparently simpler and seem to have produced much more accurate results@xcite .
recently , khan and xu@xcite improved liao s ham by the addition of adjustable parameters that improve the convergence of the perturbation series .
the purpose of this paper is to compare the improved ham with a straightforward analytical procedure based on pad approximants@xcite supplemented with a method developed some time ago@xcite . in section [ sec : ham ]
we outline the main ideas of the ham , in section [ sec : hpm ] apply the hankel pad method ( hpm ) to the tf equation , and in section [ sec : conclusions ] we compare the ham with the hpm and with other approaches .
in order to facilitate later discussion we outline the main ideas behind the application of the ham to the tf equation .
the tf equation @xmath0 is an example of two point nonlinear boundary value problem . when solving this ordinary differential equation one faces problem of the accurate calculation of the slope at origin @xmath1 that is consistent with the physical boundary conditions indicated in equation ( [ eq : tf ] ) . in what follows
we choose the notation of khan and xu@xcite whose approach is more general than the one proposed earlier by liao@xcite .
they define the new solution @xmath2 , where @xmath3 and rewrite the tf equation as @xmath4 where @xmath5 is the inverse of the slope at origin ( @xmath6 ) and @xmath7 is an adjustable parameter .
khan and xu@xcite state that the solution to eq .
( [ eq : tf2 ] ) can be written in the form @xmath8 that reduces to liao s expansion@xcite when @xmath9 . in principle
there is no reason to assume that the series ( [ eq : g_series ] ) converges and no proof is given in that sense@xcite . besides , the partial sums of the series ( [ eq : g_series ] ) will not give the correct asymptotic behaviour at infinity@xcite as other expansions do@xcite .
liao@xcite and kahn and xu@xcite do not use the ansatz ( [ eq : g_series ] ) directly to solve the problem but resort to perturbation theory .
for example ,
kahn and xu@xcite base their approach on the modified equation @xmath10 = q\hbar \mathcal{n}% \left [ \phi ( \xi ; q),\gamma ( q)\right ] \label{eq : ham}\ ] ] where @xmath11 and @xmath12 are linear and nonlinear operators , respectively , @xmath13 is a perturbation parameter and @xmath14 is another adjustable parameter . besides , @xmath15 is a conveniently chosen initial function and @xmath16 becomes the solution to equation ( [ eq : tf2 ] ) when @xmath17@xcite .
both @xmath16 and @xmath18 are expanded in a taylor series about @xmath19 as in standard perturbation theory , and @xmath20 is another adjustable parameter@xcite .
the authors state that ham is a very flexible approach that enables one to choose the linear operator and the initial solution freely@xcite and also to introduce several adjustable parameters@xcite .
however , one is surprised that with so many adjustable parameters the results are far from impressive , even at remarkable great perturbation orders@xcite .
for example the @xmath21 $ ] pad approximant of the ham series yields @xmath22 with three exact digits@xcite , while the @xmath23 $ ] pad approximant of the @xmath24 expansion@xcite provides slightly better results@xcite . a more convenient expansion of the solution of the tf equation leads to many more accurate digits@xcite with less terms .
in what follows we outline a simple , straightforward analytical method for the accurate calculation of @xmath1 . in order to facilitate the application of the hpm we define the variables @xmath25 and @xmath26 ,
so that the tf equation becomes @xmath27 -f(t)f^{\prime } ( t)-2t^{2}f(t)^{3}=0 \label{eq : tf3}\ ] ] we expand the solution @xmath28 to this differential equation in a taylor series about @xmath29 : @xmath30 where the coefficients @xmath31 depend on @xmath32 . on substitution of the series ( [ eq : f_series ] ) into equation ( [ eq : tf3 ] ) we easily calculate as many coefficients @xmath31 as desired ; for example , the first of them are @xmath33 the hpm is based on the transformation of the power series ( [ eq : f_series ] ) into a rational function or pad approximant @xmath34(t)=\frac{\sum_{j=0}^{m}a_{j}t^{j}}{\sum_{j=0}^{n}b_{j}t^{j } } \label{eq:[m / n]}\ ] ] one would expect that @xmath35 in order to have the correct limit at infinity ; however , in order to obtain an accurate value of @xmath36 it is more convenient to choose @xmath37 , @xmath38 as in previous applications of the approach to the schrdinger equation ( in this case it was called riccati pad method ( rpm))@xcite .
the rational function ( [ eq:[m / n ] ] ) has @xmath39 coefficients that we may choose so that @xmath40,t)=\mathcal{o}(t^{2n+d+1})$ ] and the coefficient @xmath41 remains undetermined .
if we require that @xmath40,t)=\mathcal{o}% ( t^{2n+d+2})$ ] we have another equation from which we obtain @xmath36 . however , it is convenient to proceed in a different ( and entirely equivalent ) way and require that @xmath34(t)-\sum_{j=0}^{2n+d+1}f_{j}t^{j}=\mathcal{o}(t^{2n+d+2 } ) \label{eq:[m / n]2}\ ] ] in order to satisfy this condition it is necessary that the hankel determinant vanishes @xmath42 where @xmath43 is the dimension of the hankel matrix .
each hankel determinant is a polynomial function of @xmath36 and we expect that there is a sequence of roots @xmath44}$ ] , @xmath45 that converges towards the actual value of @xmath46 for a given value of @xmath47 .
we compare sequences with different values of @xmath47 for inner consistency ( all of them should give the same limit ) .
notice that a somewhat similar idea was also proposed by tu@xcite , although he did not develop it consistently .
present approach is simple and straightforward : we just obtain the taylor coefficients @xmath31 from the differential equation ( [ eq : tf3 ] ) in terms of @xmath36 , derive the hankel determinant , and calculate its roots . since @xmath48 is the first nonzero coefficient that depends on @xmath36 we choose hankel sequences with @xmath49 .
the hankel determinant @xmath50 exhibits many roots and their number increases with @xmath51 .
if we compare the roots of @xmath50 with those of @xmath52 we easily identify the sequence @xmath44}$ ] that converges towards the actual value of @xmath36 .
[ fig : logconv ] shows @xmath53}-2f_{2}^{[d-1,d]}\right| $ ] for @xmath54 that provides a reasonable indication of the convergence of the sequence of roots .
we clearly appreciate the great convergence rate of the sequences with @xmath55 and @xmath56 . for example , for @xmath55 and @xmath57 it is approximately given by @xmath58}-2f_{2}^{[d-1,3]}\right|=14.2\times 10^ { -0.705d}$ ] . from the sequences for @xmath57 we estimate @xmath59 which we believe is accurate to the last digit .
we are not aware of a result of such accuracy in the literature with which we can compare our estimate .
it is certainly far more accurate than the result obtained by kobayashi et al@xcite by numerical integration that is commonly chosen as a benchmark@xcite .
present rational approximation to the tf function is completely different from previous application of the pad approximants , where the slope at origin was determined by the asymptotic behaviour of at infinity@xcite .
our approach applies to @xmath60 and the slope at origin is determined by a local condition at that point ( [ eq:[m / n]2 ] ) which results in the hankel determinant ( [ eq : hankel ] ) . in this sense
our approach is similar to ( although more systematic and consistent than ) tu s one@xcite as mentioned above .
once we have the slope at origin we easily obtain an analytical expression for @xmath61 in terms of the rational approximation ( [ eq:[m / n ] ] ) to @xmath28 . in order to have the correct behaviour at infinity we choose @xmath62@xcite .
table [ tab : u(x ) ] shows values of @xmath61 and its first derivative for @xmath63 ( the approximation is obviously much better for @xmath64 ) given by the approximant @xmath65 $ ] .
our results are in remarkably agreement with the numerical calculation of kobayashi et al@xcite and are by far much more accurate than those provided by the ham@xcite .
notice that we are comparing a @xmath65 $ ] pad approximant on the straightforward series expansion ( [ eq : f_series ] ) with @xmath66 $ ] and @xmath67 $ ] approximants on an elaborated perturbation series@xcite .
any accurate analytical expression of the solution @xmath61 to the tf equation requires an accurate value of the unknown slope at origin @xmath1 , and the hpm provides it in a simple and straightforward way . in this sense
the hpm appears to be preferable to other accurate approaches@xcite and is far superior to the ham@xcite .
notice for example that our estimate @xmath68}=-1.588 $ ] , based on a rational approximation @xmath69 $ ] , is better than the result provided by a @xmath21 $ ] pad approximant on the improved ham perturbation series@xcite . besides , by comparing table 2 of khan and xu@xcite with our fig .
[ fig : logconv ] one realizes the different convergence rate of both approaches .
one should also take into account that the hpm does not have any adjustable parameter for tuning up its convergence properties , while , on the other hand , the `` flexible '' ham with some such parameters plus a pad summation results in a much smaller convergence rate@xcite . we also constructed a pad approximant
@xmath65 $ ] from the series ( [ eq : f_series ] ) and obtained the tf function and its derivative with an accuracy that outperforms the @xmath66 $ ] and @xmath21 $ ] pad approximants on the ham perturbation series@xcite .
it is clear that the hpm is by far simpler , more straightforward , and much more accurate than the ham .
in addition to the physical utility of the hpm we think that its mathematical features are most interesting .
although we can not provide a rigorous proof of the existence of a convergent sequence of roots for each nonlinear problem , or that the sequences will converge towards the correct physical value of the unknown , a great number of successful applications to the schrdinger equation@xcite suggest that the hpm is worth further investigation .
notice that we obtain a global property of the tf equation @xmath1 from a local approach : the series expansion about the origin ( [ eq : f_series ] ) .
the fact that our original rational approximation ( [ eq:[m / n ] ] ) does not have the correct behaviour at infinity is not at all a problem because we may resort to a more conventient expansion@xcite once we have an accurate value of the unknown slope at origin .
d .. 2d .. 12d .. 12 & & + 1 & 0.424008 & 0.273989 + 5 & 0.078808 & 0.023560 + 10 & 0.024315 & 0.0046028 + 20 & 0.005786 & 0.00064727 + 30 & 0.002257 & 0.00018069 + 40 & 0.001114 & 0.00006969 + 50 & 0.000633 & 0.00003251 + 60 & 0.000394 & 0.0000172 + 70 & 0.0002626&0.000009964 + 80 & 0.0001838 & 0.000006172 + 90 & 0.0001338 & 0.000004029 + 100&0.0001005 & 0.000002743 + 1000&0.000000137&0.00000000040 + | we discuss a recently proposed analytic solution to the thomas fermi ( tf ) equation and show that earlier approaches provide more accurate results .
in particular , we show that a simple and straightforward rational approximation to the tf equation yields the slope at origin with unprecedented accuracy , as well as remarkable values of the tf function and its first derivative for other coordinate values . |
recent decades have seen the application of mathematical control theory to quantum systems in both physics and chemistry , as technological advances have allowed for greater precision in manipulation of these systems @xcite@xcite@xcite@xcite@xcite .
one particular area of interest is the possible construction of quantum computers , which have the power to perform algorithms not accessible to conventional computers .
a major experimental obstacle to any implementation of such a computer , however , is the decoherence of the system under influence of the environment .
while much progress has been made on the control of closed quantum systems @xcite@xcite@xcite , work on open quantum systems has proved more challenging @xcite@xcite@xcite@xcite@xcite .
one important issue is that controls are nearly always in the form of hamiltonian operators .
this type of control is unable to directly affect the purity of a state @xcite or transfer the state between unitary orbits . to control purity
, one must use the dissipative dynamics to move between orbits . to this end
, we wish to derive a differential equation that captures only the inter - orbit dynamics , and collects the remaining dynamics ( along the orbits ) into a new control variable .
this can be done if we assume arbitrary control over the hamiltonian dynamics .
the resulting differential equation can tell us how the location along the orbit affects the motion between orbits . in this paper , we show that this can be done for two - dimensional systems subject to lindblad dissipation , and the formalism can be extended to consider the problem of purification ( for related work in two dimensions , see @xcite@xcite ) .
the preliminaries of lindblad dissipation are outlined in section 2 , and the projection of the lindblad differential equation onto the set of unitary orbits is discussed in section 3 . in section 4 , we analyze the controllability of this equation for various choices of system parameters @xmath0 in particular , the case where the anti - symmetric part of the dissipation is non - zero . in section 5
, we present a theorem that specifies necessary and sufficient conditions for purifiability .
the state of a closed quantum system is described by a norm - one vector in a complex hilbert space that evolves according to the schrdinger equation : @xmath1 in order for the norm to be preserved , the hamiltonian operator @xmath2 must be hermitian .
an open quantum system , on the other hand , is described by a trace - one , positive - semidefinite operator @xmath3 on the hilbert space , known as the density operator ( or density matrix when working in finite dimensions , as we shall ) .
the interpretation of this matrix is the following : an eigenvalue of @xmath3 is the probability that the system is in the corresponding eigenstate . since the matrix is trace - one and positive - semidefinite , these eigenvalues are non - negative real numbers that sum to one .
a state @xmath4 in the closed system becomes a rank - one projection operator @xmath5 and its dual as @xmath6 .
inner products are written @xmath7 and outer products ( or rank - one matrices ) @xmath8 . ] .
the schrdinger equation , when extended to the density matrix , becomes the von neumann equation : @xmath9\ ] ] certain relevant quantities are invariant under the von neumann equation .
the density matrix at any time can be written @xmath10 , where @xmath11 is unitary .
since matrices at different times are similar , the eigenvalues are constant .
the purity of the system , which is defined to be @xmath12 , is also invariant since it is the 2-norm of the vector of eigenvalues .
this has implications for quantum control .
since control variables typically appear in the hamiltonian only , the control dynamics can not directly alter the probabilities , or purify the state ( _ i.e. _ achieve a purity of one ) .
however , a system that interacts with the environment will have non - hamiltonian dynamics . in general
, this will be an integro - differential equation , but if one assumes the dynamics depends only the present state and not its history ( _ i.e. _ the markovian condition ) and there is not explicit time dependence , the resulting differential equation is the lindblad equation@xcite@xcite : @xmath13 + \sum_{j=1}^{m } \left(l_j\rho l_j^\dagger -\frac{1}{2 } \{l_j^\dagger l_j,\rho\}\right ) \label{lindblad}\ ] ] the lindblad operators @xmath14 can be taken to be traceless , as adding a multiple of the identity @xmath15 to @xmath16 is equivalent to adding an operator @xmath17 to the hamiltonian .
an alternate equation , known as the lindblad - kossakowski equation , chooses a basis @xmath18 of the set of traceless @xmath19-dimensional matrices that is orthonormal relative to the inner product @xmath20 : @xmath13 + \sum_{j , k=1}^{n^2 - 1 } a_{jk}\left(l_j\rho l_k^\dagger -\frac{1}{2 } \{l_k^\dagger l_j,\rho\}\right ) \label{lindkoss}\ ] ] where the coefficients @xmath21 s form a positive - semidefinite matrix , known as the gorini - kossakowski - sudarshan matrix .
a lindblad operator can be thought as a stochastic jump with recoil . under the influence of one lindblad operator , a pure state @xmath22 in time
@xmath23 becomes a mixture of two states , @xmath24 .
here , @xmath25 and @xmath26 .
in other words , @xmath4 jumps to the state @xmath27 with probability @xmath28 .
this is a jump because the @xmath23 appears in the probability only , meaning the destination state does not approach the original state as @xmath29 .
conversely , the second state in the mixture is @xmath30 , which is infinitesimally close to the original state . in other words ,
depending on @xmath31 , there may be an infinitesimal recoil needed to compensate for the jump process .
when @xmath32 is a multiple of the identity ( for example , when the lindblad operator is a multiple of a pauli matrix ) , the second state reduces to the original state @xmath4 , so that the jump is recoil - less .
as mentioned in the introduction , control of open quantum systems typically involves control variables in the hamiltonian .
hamiltonian operators , however , can only move states _ along _ unitary orbits , and not _ between _ orbits .
the goal of this paper is to isolate the between - orbit dynamics for a generic two - dimensional system under lindblad dissipation .
our starting point is the following control system : @xmath33 + \frac{1}{2 } \sum_{j , k = x , y , z } a_{jk}\left(\sigma_j\rho \sigma_k -\frac{1}{2 } \{\sigma_k\sigma_j,\rho\}\right ) \label{bigde}\ ] ] where @xmath34 are the pauli matrices .
the controls @xmath35 are unbounded and may take any value in @xmath36 .
note that we have chosen our set of control hamiltonians to span @xmath37 .
in other words , we can make any unitary operator up to a non - physical phase difference , and therefore we can move between any two states on a given unitary orbit in arbitrary time .
we are neglecting any drift hamiltonian @xmath38 , since the component along the identity matrix does not contribute to the dynamics , and the components along the pauli matrices can be treated by re - calibrating the control variables : @xmath39 .
the density operator can be written in terms of the pauli matrices : @xmath40 , where the @xmath41 s are components of the bloch vector , such that _ i.e. _ @xmath42 . substituting this expressions into the equation ( [ bigde ] ) ,
we get : @xmath43 + \frac{1}{4}\sum_{jk } a_{jk}[\sigma_j , \sigma_k ] + \frac{1}{4}\sum_{jkl } a_{jk}n_l ( \sigma_j\sigma_l\sigma_k-\frac{1}{2}\{\sigma_k\sigma_j,\sigma_l\ } ) \\ & = \sum_{j , k } ( \frac{-i}{2}u_j n_k+\frac{1}{4}a_{jk } ) [ \sigma_j,\sigma_k ] + \frac{1}{4}\sum_{jkl } a_{jk}n_l ( \sigma_j\sigma_l\sigma_k-\frac{1}{2}\{\sigma_k\sigma_j,\sigma_l\ } ) \end{aligned}\ ] ] the pauli matrices obey the relations @xmath44
= 2i\sum_l \epsilon_l \sigma_l\ ] ] @xmath45 @xmath46 using these relations , the lindblad - kossakowski equation above becomes : @xmath47 if we define @xmath48 , and @xmath49 , we have @xmath50 in vector notation , we can write : @xmath51 where @xmath52 is the matrix with elements @xmath53 .
now we want to decompose this equation into dynamics along and between unitary orbits .
@xmath3 has eigenvalues @xmath54 , where @xmath55 and eigenvectors @xmath56 note the spectra correspond one - to - one with the values of @xmath57 , the bloch radius .
it follows that the unitary orbits are concentric spheres , except for the completely mixed state , which corresponds to the point @xmath58 .
so we can parametrize the orbits by @xmath57 , which lives on the closed interval @xmath59 $ ] , and characterize the motion along orbits with the unit vector @xmath60
. we must be careful with respect to the innermost orbit however .
@xmath61 is not defined there , which means that the differential equations which we will derive for @xmath57 and @xmath61 will have solutions that exist for finite times ( those solutions correspond to trajectories of @xmath3 that pass through the completely mixed state ) .
since @xmath62 , @xmath63 and therefore @xmath64 .
so : @xmath65 the middle term vanishes , the first term is constant in @xmath57 and the third is linear in @xmath57 .
we can write : @xmath66 to find the ode for @xmath61 , we use @xmath67 , which gives @xmath68 .
so we get : @xmath69 our goal here is to view equation ( [ proj ] ) as a control ode where @xmath61 is the control .
this view requires that we have full control over @xmath61 , and we claim that we do , in terms specified by the following lemma .
let @xmath70 be the sphere centered at the origin with radius one , let @xmath71 be the associated closed ball , and let @xmath72 be the closed ball with the origin removed .
let @xmath73 be a piecewise differentiable function from a time interval @xmath74 $ ] onto @xmath70 such that the corresponding solution @xmath75 of equation ( [ proj ] ) is contained in the interval @xmath76 $ ] .
then there are piecewise continuous control functions @xmath77 , @xmath78 and @xmath79 such that equation ( [ unproj ] ) has the piecewise differentiable solution @xmath80 on @xmath72 . first re - write equation ( [ hat ] ) : @xmath81 any equation of the form @xmath82 , where @xmath83 , has solution @xmath84 .
it follows that we can choose the controls to be : @xmath85 since @xmath73 , @xmath86 and @xmath75 are piecewise continuous , so is @xmath87 .
note that the prescription for @xmath87 is unbounded as @xmath88 because of the middle term .
this is because the system can not approach the completely mixed state from any direction : when @xmath89 , @xmath90 is fixed to be @xmath91 regardless of the controls @xmath87 .
we finish this section by writing down an alternate version of ( [ proj ] ) in terms of the eigenvalues of @xmath52 , which allows us to specify a given system in terms of six real parameters .
let @xmath92 be the eigenvalues of @xmath52 .
let @xmath93 and @xmath94 be the components of @xmath91 and @xmath95 relative to the intrinsic axes of @xmath52 ( whereas the subscripts @xmath96 , @xmath97 and @xmath98 denote the components relative to the eigenvectors of the pauli matrices ) .
this gives : @xmath99 the six parameters obey the following inequality , which arises from the positive semi - definiteness of @xmath100 : @xmath101 the positive semi - definiteness of @xmath100 also ensures the positive semi - definiteness of @xmath52 , so we also have @xmath102 .
for a fixed @xmath57 , the right - hand side of equation can be seen as a map from @xmath103 , the set of available controls , to the set of possible values of @xmath104 .
since this is a smooth map from a compact set to @xmath36 , the image should be a closed finite interval . to analyze the controllability of ( [ paraproj ] )
, we define functions @xmath105 and @xmath106 to be the right and left endpoints , respectively , of this interval .
that is , @xmath105 is the maximum possible rate at which @xmath57 can increase , and @xmath107 the minimum , for a given value of @xmath57 .
it is clear that ( [ paraproj ] ) is controllable on a closed subinterval of @xmath108 if @xmath109 and @xmath110 everywhere on the subinterval . to steer between two points @xmath111 and @xmath112
, we choose our controls so that @xmath113 if @xmath114 , or @xmath115 if @xmath116 .
some properties of @xmath117 and @xmath118 can be gleaned from inspection of the differential equation , which we collect into a proposition : if @xmath119 and @xmath120 , 1 .
@xmath117 and @xmath118 are non - increasing .
2 . @xmath121 and @xmath122 .
3 . @xmath123 .
@xmath124 for all @xmath57 and system parameters .
@xmath125 for @xmath126 only for the trivial where @xmath127 ( which requires that all @xmath128 s and @xmath129 s are zero . 5
if @xmath91 has non - zero magnitude , @xmath105 has an isolated intercept @xmath130 $ ] . 1 .
if a control vector @xmath131 achieves the maximum @xmath104 at @xmath132 , then choosing that control for all @xmath133 can only achieve a larger or equal @xmath104 , since the coefficient of @xmath57 in the differential equation , @xmath134 , must be non - negative .
similarly , if a control @xmath135 achieves the minimum at @xmath132 , then choosing that control for all @xmath136 can only achieve a smaller or equal @xmath104 .
furthermore , if @xmath137 and @xmath138 are positive , the coefficient of @xmath57 can not be made zero , so in this case , we can strengthen non - increasing " to decreasing " .
2 . as @xmath139 , the linear term in ( [ paraproj ] ) can be neglected , and we must extremize @xmath140 .
the range of this is clearly @xmath141 $ ] 3 . since @xmath57 can not exceed one , @xmath142 .
4 . non - positivity follows from 1 ) and 2 ) . if @xmath143 , @xmath104 can be always made negative by choosing @xmath144 .
non - zero @xmath91 implies that at @xmath137 and @xmath138 are positive , which means that @xmath117 is strictly decreasing on @xmath108 .
this , together with 2 ) and 3 ) imply the existence of @xmath145 .
if @xmath91 is nonzero , there is an interval @xmath146 , which we call a _ trap _ , inside of which the system is controllable . outside of the trap , on @xmath147 $ ]
, the system is one - way controllable ; that is , @xmath111 can be steered to @xmath112 in finite time if and only if @xmath148 .
the statements in the proposition imply that @xmath149 on @xmath146 , in which case we can steer @xmath111 to @xmath150 by choosing the control that satisfies @xmath151 provided @xmath152 .
conversely , to steer @xmath111 to @xmath153 , we can choose the control that satisfies @xmath154 .
on the interval @xmath147 $ ] , @xmath155 , so @xmath111 can not be steered to @xmath156 , but can be steered to @xmath157 by choosing the control that satisfies @xmath158 , which must be negative . in the case
that @xmath159 , there is no trap : @xmath160 for all @xmath57 , and in fact we can say that @xmath161 where we can achieve the upper and lower bounds by choosing @xmath95 to be @xmath162 and @xmath163 , respectively . in the case
that @xmath164 , the decay of @xmath57 may be halted , but otherwise @xmath57 will decay exponentially to zero at a rate above or equal to @xmath165 .
it is evident , then , that the presence of an asymmetric part in the dissipative term ( represented by @xmath91 ) significantly enhances the possibility of control . in order to calculate @xmath117 and @xmath118 for given @xmath57
, we can use the method of lagrange multipliers . in some cases
, we can solve the resulting equations analytically , but in general one must find the roots of a sixth - order polynomial , so we must resort to numerics . before considering the general case , we will look at a particular case that can be treated analytically .
we consider the possibility that a two - level system can undergo one of two processes represented by the raising and lowering operators @xmath166 and @xmath167 at rates @xmath168 and @xmath169 , respectively . if one constructs the lindblad equation using this scenario , and expresses it in the basis of the pauli matrices , one finds that @xmath170 , @xmath171 , @xmath172 and @xmath173 .
the fact that @xmath91 has only one non - zero component simplifies the equations so that we can treat the system analytically .
if we apply the method of lagrange multipliers to the right - hand side of ( [ paraproj ] ) and set @xmath172 and @xmath174 , we get : @xmath175 where @xmath176 is the lagrange multiplier .
this has the following solutions : @xmath177 where @xmath178 and @xmath179 in can be any pair that satsifies the normalization condition .
solutions do not exist for all @xmath57 , since the magnitude of @xmath180 must not exceed one .
they exist only on @xmath181 $ ] . to determine which solutions correspond to @xmath117 and @xmath118 , we substitute back into .
solutions give @xmath182 and solutions give @xmath183 we can easily conclude that @xmath184 . furthermore , the right - hand side of is greater than or equal to those of , but since it has a limited interval of definition , we have : @xmath185 it happens that @xmath145 in this case coincides with the point at which @xmath117 switches between and , _
i.e. _ @xmath186 .
this is not a general phenomenon , however : if @xmath187 , the switching point and the trap radius would not coincide .
[ analyt ] depicts these solutions for @xmath188 and @xmath189 . vs. @xmath57 for a case that can be solved analytically .
system parameters : @xmath188 , @xmath171 , @xmath172 , @xmath189 .
solid lines represent @xmath117 and @xmath118 .
blue and purple indicate solutions and , respectively .
dotted lines indicate where these solutions do not coincide with @xmath117 . ] vs. @xmath57 for a case that must be solved numerically .
the trap radius is where the maximum achievable @xmath190 passes from positive to negative .
system parameters : @xmath191 , @xmath192 , @xmath193 , @xmath194 , @xmath195 , @xmath196 more generally , one can perform this analytical treatment in the following cases : ( 1 ) if @xmath91 has one non - zero component , ( 2 ) if @xmath91 has two non - zero components , and the corresponding @xmath128 s are equal , and ( 3 ) @xmath91 has three non - zero components , and @xmath197 . if the system does not fall into any of those three categories , lagrange multipliers lead to either a fourth - degree polynomial in @xmath176 ( technically solvable , but inordinately messy ) or a sixth - degree polynomial ( generally not solvable ) .
the fourth - degree polynomial arises in the cases ( 1 ) @xmath91 has two non - zero components but corresponding @xmath128 s are not equal and ( 2 ) @xmath91 has three non - zero components and @xmath198 or @xmath199 .
the sixth - degree polynomial arises if @xmath91 has three non - zero components and @xmath200 . in those cases
, we can find the real roots of the polynomial numerically .
then we can compute the corresponding values of @xmath104 , choose the maximum and minimum values , and assign the values to @xmath117 and @xmath118 . in fig .
[ numer ] , @xmath117 and @xmath118 are shown for a particular system that required solving a sixth - order polynomial .
we have computed the curves for 10,000 points apiece . @xmath145 can be found by numerically interpolating @xmath117 .
for the case depicted in fig .
[ numer ] , @xmath145 was computed to be @xmath201 ( to machine precision ) .
an important goal in quantum control is purification : the process of steering a mixed state to a pure state , which can be characterized by a purity @xmath202 .
alternatively , a system is pure if the leading eigenvalue is one , with remaining eigenvalues being zero . in terms of the above analysis
, we say a system is _ purifiable _ if and only if the trap radius @xmath203 .
in other words , the function @xmath105 has an isolated intercept at @xmath204 .
this section is devoted to proving a theorem that characterizes the possible purifiable systems .
first , we will use the following lemma : @xmath205 at @xmath204 if and only the system is in a state that is an eigenvector of all contributing lindblad operators . because the bloch radius can be written @xmath206 , where @xmath207 are the eigenvalues of @xmath3 , we can write @xmath208 . differentiating this , we get an alternative expression for @xmath104 : @xmath209 where in the last step , the normalization of the vectors makes the quantities in parentheses vanish .
now , if the dissipation is characterized by a collection of lindblad operators @xmath210 s , which are not necessarily orthogonal we can use to specify @xmath211 : @xmath212|\psi_+\rangle - \langle\psi_-|[-ih,\rho]|\psi_-\rangle \\ & + \sum_j \left ( \langle\psi_+|l_j\rho l_j^\dagger|\psi_+\rangle -\frac{1}{2}\langle\psi_+|l_j^\dagger l_j\rho |\psi_+\rangle-\frac{1}{2}\langle\psi_+|\rho l_j^\dagger l_j|\psi_+\rangle \right.\\ & \left .
- \langle\psi_-|l_j\rho l_j^\dagger|\psi_-\rangle + \frac{1}{2}\langle\psi_-|l_j^\dagger l_j\rho |\psi_-\rangle+\frac{1}{2}\langle\psi_-|\rho l_j^\dagger l_j|\psi_-\rangle \right ) \end{aligned}\ ] ] the hamiltonian terms vanish since they are diagonal elements of a skew - symmetric matrix .
we are interested in @xmath104 when @xmath204 , so insert @xmath213 .
we get : @xmath214 if we insert the identity operator between @xmath215 and @xmath16 in the middle term , we get the expression : @xmath216 for @xmath104 to vanish , we need @xmath217 to vanish for each @xmath16 .
this is only possible however if @xmath218 is an eigenvector of each @xmath16 , since otherwise @xmath219 would have some component in the @xmath220 direction .
this proves the lemma .
this leads to the following theorem : a two - level lindblad system is purifiable if and only if one of the following characterizations hold : * there is one lindblad operator , and it is singular .
* there is one linblad operator and it is non - singular with non - orthogonal eigenvectors .
* there is no more than one singular lindblad operator and any number of non - singular operators .
all share a common eigenvector .
* there are any number of non - singular lindblad operators that share a common eigenvector .
we are required to show two things to prove a system is purifiable : @xmath221 , and @xmath222 .
the latter ensures that @xmath117 is strictly decreasing rather than constant in @xmath57 .
when combined with the former condition , this implies that @xmath117 is positive for all @xmath224 , and therefore controllable .
it follows from the lemma that a system is purifiable only if all contributing lindblad operators share a common eigenvector , or else @xmath225 will be strictly negative .
this is only a necessary condition however and not a sufficient one , since the condition implies only that @xmath221 .
we also require that @xmath222 .
so consider the case @xmath226 .
this implies that @xmath227 and @xmath91 are also zero ( due to ) , so that @xmath100 has only one non - zero entry in its natural basis .
we claim that @xmath100 in this form corresponds to a non - singular operator with orthogonal eigenvectors .
it is a rank - one real positive matrix , and therefore can be written @xmath228 for some real 3-vector @xmath229 .
when one diagonalizes the lindblad equation however , this results in a single lindblad operator @xmath230 .
this operator is hermitian and traceless , however , so neglecting the trivial zero operator , it is non - singular with orthogonal eigenvectors .
in other words , as long as the system obeys the terms of the lemma , and does not consist of a single hermitian operator , the system is purifiable .
the first two cases in the theorem cover the remaining possible single - operator cases .
the remaining two cases can be seen by noting that two singular operators can not share eigenvectors , since they have only one ( we consider two operators that are multiples of each other to be essentially one process ) .
the third case covers the possibility of one singular operator : it has only one eigenvector , and that eigenvector must be shared with the other non - singular operator .
the fourth case in the theorem covers the possibility of no singular operators but more than one non - singular operator .
note that the non - singular operators in the third and fourth cases need not have non - orthogonal eigenvectors .
we have shown that the inter - orbit dynamics of a controlled quantum system can be isolated from the intra - orbit dynamics by projecting onto the set of spectra of the density matrix .
if one makes certain assumptions about the controllability of the system along the orbits , the position of the system along the orbit can be viewed as a new control variable , since the intra - orbit dynamics can be made arbitrarily faster than the inter - orbit dynamics . in two dimensions ,
we have derived a equation describing this inter - orbit dynamics , where the new control is the normalized bloch vector , and the most general lindblad system can be described by six real parameters : three describing the symmetric part of the dissipation , and three describing the anti - symmetric part .
we have analyzed the controllability of a general two - dimensional system under lindblad dissipation , particularly for dissipation with non - zero anti - symmetric part . for systems of this type ,
there exists a trap , or a subinterval of the state space where each state is reachable from any other , but from which states may not escape .
the size of this trap can be calculated analytically for certain simple cases , but in general must be calculated numerically .
we have shown how this can be done using the method of lagrange multipliers , and shown results for a particular generic system .
furthermore , we have applied this formalism to categorize the set of purifiable systems .
a necessary condition for purifiability is that all lindblad operators share a common eigenvector . to strengthen this to a sufficient condition
, one must eliminate the case of a single hermitian lindblad operator .
the immediate direction of future work is to apply this formalism to three and higher dimensional systems .
it is well - known that the structure of density matrices is richer and less well - understood than the case for two dimensions @xcite .
for one , the set of pure states no longer constitutes the boundary of the set , but a ( measure - zero ) subset of the boundary .
currently we are studying how our formalism translates to higher dimensions and what obstructions are posed by the richer geometry .
furthermore , we would like to know whether the set of purifiable higher - dimensional systems can be categorized as it has been done in this paper .
c. altafini , controllability properties for finite dimensional quantum markovian master equations , _ j. math .
44 , no . 6 , 2003 , pp . 2357 - 72 . c. altafini , coherent control of open quantum dynamical systems , _ phys .
a _ , vol .
70 , 2004 , 062321 .
g. dirr , u. helmke and i. kurniawan , lie - semigroup structures for reachabilitiy and control of open quantum systems : kossakowski - lindblad generators form lie wedge to markovian channels , _ rep .
_ , vol . 64 , 2009 , pp . | control of quantum dissipative systems can be challenging because control variables are typically part of the system hamiltonian , which can only generate motion along unitary orbits of the system . to transit between orbits
, one must harness the dissipation super - operator . to separate the inter - orbit dynamics from the hamiltonian dynamics for a generic two - dimensional system , we project the lindblad master equation onto the set of spectra of the density matrix , and we interpret the location along the orbit to be a new control variable .
the resulting differential equation allows us to analyze the controllability of a general two - dimensional lindblad system , particularly systems where the dissipative term has an anti - symmetric part .
we extend this to categorize the possible purifiable systems in two dimensions . |
over the last two decades , the study of driven diffusive systems kls , sz , vp , schutz has formed a significant branch in the general pursuit to understand non - equilibrium statistical mechanics .
the underlying dynamics violates detailed balance , so that the long - time behavior is governed by a true _ non - equilibrium _ steady state ( ness ) .
highly unexpected and non - trivial properties are manifested , even in one - dimensional systems with purely local dynamics .
a paradigmatic model in the last class is the totally asymmetric simple exclusion process ( tasep ) krug , derrida92,dehp , s1993,derrida , schutz .
characterized by open boundaries and particle transport , it displays the key physical signatures of a system driven far from equilibrium .
in particular , its ness develops a complex phase diagram controlled by the interactions of the system with its environment . while an understanding of these properties is of fundamental theoretical interest , the tasep and its generalizations
have also acquired fame as models for practical problems in traffic flow @xcite and biological transport @xcite . in this paper
, we study a simple variant of the basic tasep , by introducing a global constraint on the available number of particles . deferring the motivations for such a model to the next paragraphs ,
let us briefly summarize the essentials .
each site of a one - dimensional lattice is either empty or occupied by a single particle .
following random sequential dynamics , the particles enter the lattice at one end ( e.g. , the `` left '' edge ) with a given rate @xmath0 , hop to the right with rate @xmath1 ( scaled to unity ) , subject to an excluded volume constraint , and exit at the far end with a rate @xmath2 . in the standard version of the model ,
both @xmath3 and @xmath2 are constant rates , independent of the number of particles on the lattice .
thus , we may regard the lattice as being coupled to a reservoir or a pool with an arbitrarily large number of particles . here , we consider a tasep with a _
finite _ supply of particles and report our numerical and analytical findings .
a fixed number of particles , @xmath4 , is shared between the lattice and the reservoir , in such a way that the entrance rate , @xmath0 , depends on the number in the pool : @xmath5 . as a result ,
if more particles are found on the lattice , the number of available particles in the pool is lowered , leading to a corresponding decrease of @xmath0 . since the number of particles on the tasep lattice , @xmath6 , feeds back into @xmath0 through the constraint @xmath7 constant @xmath8 , we will refer to our model as `` a constrained tasep . '' for simplicity , we assume the reservoir to be so large that particles exiting the lattice are unaffected by @xmath5 , leaving @xmath2 unchanged . our focus here is to explore the consequences of limited resources : how will the phase diagram and the properties of the phases be affected as the total number of available particles is reduced ? this behavior mimics the limited availability of resources required for a given physical or biological process . for example , in protein synthesis , ribosomes bind near the start sequence of a messenger rna , in order to translate the genetic information encoded on the rna into the associated protein .
ribosomes are large molecular motors which are assembled out of several basic units , and numerous ribosomes can be bound to the same , or other , mrnas , so that multiple proteins are translated in parallel .
when a protein is completed , the ribosome is disassembled and recycled into the cytoplasm . under conditions of rapid cell growth , ribosomes or their constituents
can find themselves in short supply , so that a self - limitation of translation , mimicked via a modification of @xmath0 , can occur .
in a previous study @xcite , a different aspect of ribosome recyling was considered .
this work focused on the enhancement of the ribosome concentration at the initiation site , as a result of diffusion of the ribosome subunits from the termination site ( of the same mrna ) . due to the spatial proximity of initiation and termination sites , @xmath0
is effectively increased . by contrast
, our investigation of constrained resources leads to an effective _ reduction _ of @xmath0 . in the context of traffic models , our problem corresponds to a generalization of the parking garage problem @xcite . here
, one special site ( the `` parking garage '' , or reservoir ) is introduced into a tasep on a ring ( lattice with periodic boundary conditions ) and , for this site only , the occupancy is unlimited .
particles ( `` cars '' ) jump into the garage with unit rate ( corresponding to @xmath9 ) , irrespective of its occupancy .
particles exit the garage with rate @xmath0 , provided the site following the garage is empty . as we will see presently , this is a special case of our more general model .
we begin with a description of our model and its observables of interest .
next , we outline results from simulations and compare them to a simple theory which builds on exact analytic results for the standard tasep .
we find excellent agreement for nearly all ( @xmath0,@xmath2 ) and conclude with some comments and open questions .
the standard tasep is defined on a one - dimensional lattice of length @xmath10 , with sites labeled by @xmath11 , @xmath12 .
each site can be empty or occupied by a single particle , reflected through a set of occupation numbers @xmath13 which take the values @xmath14 or @xmath15 .
the boundaries of the lattice are open .
particles hop from a reservoir onto the lattice with a rate @xmath3 . once on the lattice
, particles will hop to the nearest - neighbor site on the right , provided it is empty . once a particle reaches the end of the lattice , it hops back into the reservoir with rate @xmath2 .
the dynamics is random sequential , leading to fluctuations in the local occupations as well as in the total number of particles , defined via @xmath16 the overall density is given by @xmath17 .
ensemble averages are denoted by @xmath18 .
three distinct phases can be identified and are displayed in an @xmath19 phase diagram ( see fig .
[ fig : phase_diagram ] ) : a low density ( ld ) , a high density ( hd ) , and a maximum current ( mc ) phase . the phases are distinguished , respectively , by their average densities : @xmath20 , @xmath21 , and @xmath22 .
the corresponding stationary currents are given by @xmath23 , @xmath24 , and @xmath25 .
in all of these expressions , finite - size corrections have been neglected .
because of particle - hole symmetry , all aspects of the hd and ld phases are related .
the phase boundaries between the hd and mc phases and between the ld and mc phases mark continuous transitions .
the line separating the hd and ld phases , @xmath26 , is a coexistence line . here , the system consists of a region with low density ( @xmath27 ) followed by one with high density ( @xmath28 ) , connected by a microscopically sharp `` shock . ''
this shock diffuses freely between the ends of the system , so that the _ average _ density profile is linear .
often this is referred to as the `` shock phase '' ( sp ) .
the system of interest here differs from the standard tasep in one important way : the number of particles in the reservoir , @xmath5 , is finite , and we choose the on - rate @xmath0 to depend on @xmath5 as follows .
so as to distinguish this varying rate from the @xmath0 of the ordinary tasep , we denote the former as the _ effective _ on - rate @xmath29 and write @xmath30 where @xmath31 is a function that satisfies three conditions : ( i ) @xmath32 , ( ii ) @xmath33 , and ( iii ) @xmath31 is monotonically increasing .
the first of these conditions is self - evident ; the second simply connects our model to the standard one with unlimited resources ; the last is just common sense . in this notation , the `` parking garage '' problem @xcite
is characterized by @xmath34 where @xmath35 is the heaviside step function . here , we wish to model , say , a cell with finite number of ribosomes , so that it is natural to assume that a smaller number of particles in the pool results in a lower on - rate , etc . , and that @xmath36 would be a smoother function .
specifically , for the simulations shown below we choose @xmath37 where @xmath38 provides the scale for crossover to saturation .
the detailed choice of @xmath38 is unimportant , but it is convenient to make it extensive in @xmath10 . for easier comparison ,
we choose @xmath38 to be the average ( stationary ) density of the unconstrained tasep defined by the parameters @xmath39 : for example , if @xmath40 ( hd phase , ordinarily ) , we set @xmath38 to @xmath41 . for the sp , we arbitrarily choose @xmath42 . to summarize , the control parameters of our model are the same as the ordinary tasep ( @xmath43 ) plus the total number of particles , @xmath4 ( i.e. , @xmath44 ) .
to characterize our system in steady state , we measure the average density @xmath45 , the average ( local ) density profile @xmath46 , and the average current @xmath47 . for the last quantity
, we record the total number of particles entering and leaving the chain over the run ( thereby improving statistics ) and divide the average of these by the length of the run . in our monte carlo simulations
, we pick randomly from the @xmath6 particles on the lattice and one additional _ virtual _ particle .
if a lattice particle is selected , we attempt to update its position as in ordinary tasep ; if the virtual particle is selected , we attempt to place a new particle on the first site of the chain .
@xmath48 attempts constitute one monte carlo step ( mcs ) .
initially , the chain is empty , i.e. , @xmath49 , @xmath50 .
typically , the first @xmath51 ( @xmath52 for sp ) mcs are discarded , to ensure that the system has reached steady state , before data are taken .
then , configurational observables are recorded every @xmath53 mcs and averaged until the run terminates .
the typical length of each run is @xmath52 ( @xmath54 for sp ) mcs .
longer runs are required to obtain low noise data for density profiles and currents .
statistical errors were estimated through visual inspections of time series and through multiple runs , to ensure the reproducibility of the data .
we simulated system sizes in the range of @xmath55 to @xmath56 , with most data taken for @xmath57 . to obtain the phase diagram
, we simulated more than a dozen @xmath39 pairs . here ,
we report results for four points : @xmath58 , @xmath59 , @xmath60 , and @xmath61 .
since these correspond , respectively , to ld , hd , mc , and sp in a standard tasep , we will use these letters to refer to the four cases studied .
in this section , we summarize our simulation results . since only the on - rate depends on the feedback , the behavior of our system is no longer governed by particle - hole symmetry . as a result ,
the simplest case is the first in the above , i.e. , `` ld '' . both @xmath62 and @xmath47 increase with @xmath4 in an expected fashion .
the next case , mc , is already slightly more interesting : both @xmath62 and @xmath47 increase until the effective on - rate reaches 1/2 , after which both become constant .
the end result is a pronounced kink in @xmath62 .
the hd case provides an even more interesting scenario : the system properties range through _ three _ regions as @xmath4 is increased .
finally , the sp provides the most unexpected behavior . in the next section ,
we turn to a simple theoretical description of our findings . _ the ld case . _ here , we set @xmath63 , @xmath64 , and vary the total number of particles , @xmath4 , in the system . with @xmath36 given by eqn ( [ alpha - mcs ] ) , our simulation results for @xmath57 and @xmath65 are shown in fig .
fig : ld . as expected , we see that , for large @xmath4 , the system density @xmath66 approaches the value for the standard tasep for @xmath67 ( here , @xmath68 ) , namely , @xmath69 ( which is @xmath70 in our case ) .
more interestingly , for @xmath71 , we find a reduced density , @xmath72 , a signal that the system is responding to the limited availability of particles from the reservoir . in the limit @xmath73 , @xmath62 naturally vanishes , with a predictable slope ( see below ) .
the current @xmath47 in this `` phase '' also decreases monotonically with decreasing @xmath4 , from its asymptotic limit @xmath74 ( given by @xmath75 ) . and current @xmath47 vs @xmath76 for an ld case ( @xmath57 , @xmath77 , and @xmath78 ) .
open circles and diamonds are from the analysis in section 4.,scaledwidth=35.0% ] _ the mc case .
_ this domain of the phase diagram is characterized by @xmath79 and @xmath80 in the standard tasep .
[ fig : mc ] shows the averages @xmath62 and @xmath47 , as @xmath4 varies , for @xmath81 . the former displays a single kink , just below @xmath82 ,
accompanied by a rather smooth crossover in the current . as we will see below ,
this kink is associated with a crossover from an mc - like behavior to an ld - like behavior , as @xmath4 drops below the value required to sustain a density @xmath83 on the chain . and current @xmath47 vs @xmath76 for an mc case ( @xmath57 , @xmath84 , and @xmath78 ) .
open circles and diamonds are from the analysis in section 4 . ] and current @xmath47 vs @xmath76 for an hd case ( @xmath57 , @xmath84 , and @xmath85 ) .
open circles and diamonds are from the analysis in section 4 .
the predictions from domain wall theory ( not shown ) are similiar . ] _ the hd case .
_ here , we set @xmath86 and @xmath87 .
fig : hd shows our results for @xmath88 . as expected , for sufficiently high @xmath4 , the system settles into the density and current associated with the hd phase of the standard tasep : @xmath89 , and @xmath90 .
as @xmath4 is reduced , to the extent that the chain is prevented from sustaining a density of @xmath28 , the naive expectation is that a crossover to an ld - like behavior appears .
remarkably , however , fig .
[ fig : hd ] shows not just two , but _
three _ distinct regimes , separated by two pronounced `` kinks '' at @xmath91 and @xmath92 . as fig .
[ fig : hd_zoom ] illustrates , these kinks become sharper with increasing @xmath10 . looking for potentially critical behavior
, we measured the variance of @xmath93 near the kinks and compared it to its values deep within each regime .
to our surprise , at least for the @xmath57 system , the variance was not noticeably larger at the kinks than at other locations .
simulations at larger values of @xmath10 would be required to settle this issue with more certainty .
the current @xmath47 is also displayed in fig .
[ fig : hd ] . for low values of @xmath94
, @xmath47 increases in an expected way . around @xmath95
, it shows a distinct kink and then becomes essentially constant , with a value of @xmath96 .
this is close to @xmath75 , the value of @xmath47 for infinite @xmath10 . since the current appears to have already reached the `` asymptotic '' value here , it is not too surprising that there is no sign of a `` second kink '' ( near @xmath97 , where @xmath62 has another one ) .
below , we offer a theoretical description for these findings .
, near @xmath98 , for the hd case in fig .
4:@xmath99 , @xmath100 , and @xmath101 .
] _ the sp case .
_ defined in the standard tasep by @xmath26 , this line of first - order phase transitions separates the hd and ld phases . though the _ average
_ density is 1/2 , its fluctuations are large . in particular , since a shock ( which separates a region with density @xmath28 from one with @xmath27 ) diffuses freely between the ends of the system , the probability @xmath102 of finding the system with a density @xmath62 approaches a flat distribution : @xmath103 , in the limit @xmath104 . of course
, finite - size effects will smooth out the steps at @xmath28 and @xmath27 . imposing finite resources
, we may expect that @xmath62 increases simply with @xmath4 as in the ld case , perhaps rising more slowly to the asymptotic value of 1/2 . instead ,
its properties are much more subtle . as illustrated in fig .
[ fig : sp ] ( for @xmath105 ) , @xmath62 increases with @xmath76 * * , * * slows , speeds up , and slows down again .
the behavior in the low @xmath4 regime is intuitively understandable , since there are too few particles to sustain any phase other than ld .
the large @xmath4 regime is also expected . in the crossover regime ( @xmath106 ) ,
an intriguing `` shoulder '' develops .
though this behavior is accounted for by the domain wall theory @xcite ( see below ) , we have found no simple and intuitive way to understand it .
perhaps the presence of large fluctuations , as the system gets closer to the sp , smooths out the two kinks observed in the hd case . to gain more insight
, we investigated @xmath102 .
[ fig : sp - hist ] shows particle density histograms ( gathered from 50k data points and plotted with arbitrary normalization here ) , for several choices of @xmath76 . for low @xmath76 ( e.g. , 0.5 in the figure ) ,
our system is dominated by the lack of resources , so that this distribution is essentially gaussian , quite similar to those measured deep within the ld phase .
as @xmath76 increases , the system enters the crossover regime and @xmath102 becomes quite asymmetric , as a detailed fitting of the @xmath107 case shows . at the other extreme ,
this distribution is indeed relatively flat .
however , the true asymptotic regime is reached only for very large @xmath76 .
thus , the slope for even the @xmath108 case ( cf . fig .
[ fig : sp - hist ] ) is discernably non - zero . throughout the intermediate regime ,
the structure of @xmath109 is more interesting .
each histogram has the cross - section of a `` lean - to '' ( a shack with a slanted roof ) .
a more detailed inspection shows that the `` roof '' is predominantly a ( slowly ) decaying exponential .
nearly identical to @xmath2 .
] this picture is consistent with the one from domain wall theory @xcite , as will be discussed below .
it is this slow crossover - from well - defined gaussians to a completely flat distribution - which accounts for the interesting structure in the rise of @xmath62 with @xmath94 . finally , the behavior of the current is also sensitive to the slow crossover , although it displays no interesting structures like those in @xmath110 . comparing fig . [
fig : sp ] with the earlier ones , we see that it saturates at values of @xmath76 which are 2 to 3 times higher than in the other cases . and current @xmath47 vs @xmath76 for an sp case ( @xmath57 , @xmath77 , and @xmath85 ) .
open circles are from the domain wall theory in section 4 .
open diamonds are the current computed from eqn ( [ ld - th ] ) . ] for @xmath57 , @xmath111 , and @xmath85 .
( for ease of comparisons , the histograms with solid lines only are scaled down by 2 . )
legend shows various values of @xmath76 . ]
in the steady state of a standard tasep , the overall density of particles is controlled only by the entrance and exit rates .
it is natural to expect that these relations would also hold in our case , except that the stationary density needs to be determined from the ( variable ) entrance rate in a self - consistent way . in the following ,
we pursue this approach and compare it to the simulation results reported in the preceding section . _ the ld case . _
again , we begin with a choice of @xmath112 such that the system is in an ld phase , with @xmath113 . invoking the exact expression for the density of the standard tasep , namely , @xmath69 , we write the density of particles in the resource - limited case as @xmath114 . with @xmath115 ,
this becomes an implicit equation for @xmath62 : @xmath116 which can be solved for given @xmath0 and @xmath4 .
specifically , since @xmath117 , the solution will lie between @xmath14 and @xmath3 .
the upper bound will be reached if @xmath118 so that the right hand side of eqn ( [ ld - th ] ) remains essentially constant as a function of @xmath62 . for any finite value of @xmath76 ,
the right hand side is a monotonically decreasing function of @xmath62 so that ( i ) eqn ( [ ld - th ] ) always has a solution , and ( ii ) the solution decreases monotonically with @xmath76 .
the resulting density , and the associated current , @xmath119 , are shown as open circles and diamonds , respectively , in fig .
[ fig : ld ] , for the choice of @xmath120 given in eqn ( [ alpha - mcs ] ) .
we see , in particular , that for small values of @xmath76 , we may use @xmath121 whence @xmath122 .
clearly , the agreement with the simulation data is impressive . _ the mc case .
_ after these considerations , the analysis of the mc phase is quite straightforward . for large @xmath4 , the density saturates at @xmath123 , with @xmath124 . for small @xmath4 ,
the system is ld - like and follows eqn ( [ ld - th ] ) .
the transition occurs when @xmath29 is forced to drop below @xmath125 , i.e. , at the intersection of eqn ( [ ld - th ] ) and @xmath126 . here
, @xmath4 assumes the critical value of @xmath127 , given by @xmath128 the current follows the behavior of the density . _ the hd case .
_ next , we turn to the high - density phase , with @xmath129 and @xmath87 .
if the number of available particles is very large , we may again invoke the exact relation for the density of the standard tasep , namely @xmath130 this reflects the constant density observed in the simulations .
clearly , this can not be sustained as the number of available particles is reduced , since the effective @xmath0 of the system is also reduced .
a transition should be expected when it reaches @xmath2 .
@xmath131 . translating @xmath2 @xmath132 into an equation for the system density @xmath62 , we arrive at @xmath133 solving for @xmath62 , and setting it to the one from eqn ( [ hd - th1 ] ) , we find the critical value of the total particle number for this transition : @xmath134 .
( the subscript , @xmath135 , helps to distinguish it from the lower transition point . ) for the specific @xmath36 used in simulations , this results in @xmath136 .
\label{hd - c2}\ ] ] as we see , @xmath137 \simeq 1.01 $ ] is in excellent agreement with the `` second kink '' shown in fig .
[ fig : hd ] . at the opposite extreme ,
the resources are very low , so that @xmath29 is significantly less than @xmath2 , and the system is pushed into ld - like behavior . for such low @xmath4 , the density follows eqn ( [ ld - th ] ) . in this regime , @xmath36 is well approximated by a linear function , so that @xmath138 , i.e. , @xmath139 of course , eqn ( [ ld - th ] ) is also not `` sustainable , '' in the sense that @xmath140 for large @xmath4 instead of the correct limit : @xmath141 .
the critical value of @xmath4 again occurs when @xmath29 reaches @xmath2 , i.e. , when the density given by eqn ( [ ld - th ] ) reaches @xmath2 .
the result is @xmath142 again , for the case shown in fig .
[ fig : hd ] , there is excellent agreement between the data and this prediction : @xmath143 . between @xmath144 and @xmath145 ,
eqn ( [ hd - th1a ] ) provides a _ linear _ dependence of @xmath62 on @xmath76 @xmath146 with _ unit _ slope .
again , with _ no _ fitting parameters , this theoretical prediction agrees remarkably well with the observations . in this regime ,
the reservoir occupation ( and so @xmath29 ) remains constant , while the change in @xmath4 is balanced by the change in total lattice population : @xmath147 .
since @xmath148 ( on the average ) , the current remains at the _ constant _ value @xmath149 throughout .
while this `` three - piece '' approach is both intuitively appealing and surprisingly successful in reproducing simulation data , it begs the question : is there a unified theory that is just as successful with predictions ?
the answer is affirmative , namely , the domain wall theory dw , sa . since this theory is absolutely indispensable for providing good predictions for the next case , we will discuss its details in the next subsection .
_ the sp case .
_ here , monte carlo simulations presented us with unexpected phenomena : the `` shoulder '' in fig .
[ fig : sp ] for @xmath62 .
the simple approach taken above succeeds only in predicting the behavior for the extreme values of @xmath76 .
dominated by the limited resources , the solution of @xmath150 fits the data very well for @xmath151 . at the other extreme ( @xmath152 )
, the system is necessarily in the sp , where @xmath126 on the average . in the crossover regime ,
the understanding of the structure in @xmath153 requires a more sophisticated theory @xcite .
based on a biased random walk of the shock ( i.e. , domain wall ) , santen and appert @xcite derived the average occupation for a _
finite _ tasep : @xmath154 where @xmath155 , and @xmath156 is the ratio of the diffusion constants to the right / left . to apply to our sp case , we simply set @xmath0 to be @xmath157 in this equation .
that sets up a relationship between @xmath62 and @xmath94 which can be computed numerically .
the agreement between this theory and data is quite respectable ( fig .
[ fig : sp ] ) , though not as spectacular as in the other cases . not surprisingly , most of the disagreement occurs in the crossover region .
we believe that the main difficulty lies with the large fluctuations of the domain wall , feeding back in non - trivial ways to the on - rate . for another perspective on the successes and limitations of this theory , we can compare @xmath102 above with the predicted @xmath158 , the probability of finding the domain wall to be at site @xmath11 .
the latter is , for the steady state , a pure exponential @xcite : @xmath159 .
since our focus here is the dependence on @xmath62 , @xmath158 can be translated into @xmath102 , via the relation @xmath160 .
we find that the _ predicted _ @xmath102 is proportional to @xmath161 , and nonzero only in the interval @xmath162 $ ] .
our simulations show several non - trivial deviations from this theoretical prediction , however .
( i ) the slopes in the exponent are systematically steeper than predicted .
( ii ) quadratic terms in @xmath163 are far from negligible .
( iii ) the tails of @xmath102 extend substantially outside the range @xmath164 $ ] . in other words , there is room for improvements to the domain wall theory .
another intriguing phenomenon , not shown here explicitly , is the complex dependence of this crossover on @xmath10 .
in particular , both simulations and theory show that the `` shoulder '' completely disappears for small @xmath10 .
for example , in an @xmath165 run , @xmath153 rises linearly for @xmath166 and simply `` slows down '' into the asymptotic value of @xmath125 for @xmath167 . for large @xmath10 ,
in contrast , there exists a subtle competition between the factors @xmath168 and @xmath169 in eqn (
[ < n > ] ) .
the result is that the crossover region gets longer with larger @xmath10 , producing challenges at the simulations front as well .
of course , these difficulties are hardly surprising , since the complexities here are undoubtedly due to large fluctuations associated with the infrared scales .
further studies into this issue , especially more detailed investigations of the lean - to s in @xmath102 , should provide insight for a better understanding of such fluctuations . to conclude this section
, we briefly note that we also simulated an alternate form of the on - rate , namely , @xmath170 $ ] .
we will not present any details here , but the agreement of simulation data and analytic results for the three regimes ( hd , ld , and mc ) is equally convincing .
however , the details of the crossover regime in sp are likely to be quite sensitive to functional form of @xmath36 .
in this paper , we studied a tasep with open boundary conditions .
in contrast to the standard model , where the particles are supplied by an infinite reservoir , the total number of available particles in our model , @xmath4 , is _ fixed_. mimicking limited resources , the available particles are shared between the reservoir and the system itself : if more particles bind to the chain , the reservoir is depleted , and vice versa .
we account for this situation through an effective on - rate , @xmath29 , which depends on @xmath5 , the number of particles in the pool , and hence , on @xmath6 , the number on the chain .
specifically , we mainly used eqn ( [ alpha - mcs ] ) , @xmath171 , here .
as the number of available particles is reduced from infinity , the system crosses over from the asymptotic behavior of the standard tasep to different behaviors dominated by limited resources . if the corresponding standard model ( i.e. , @xmath172 limit ) lies in the ld phase , there are no surprises in either @xmath62 , the average particle density on the chain , or @xmath47 , the average current : both are proportional to @xmath4 for @xmath173 , crossing gently over to the asymptotic values much like the @xmath174 function . for the mc phase
, there is already a sublety : both @xmath62 and @xmath47 increase monotonically and then , at a characteristic @xmath29 ( @xmath175 ) become constant .
@xmath62 displays a notable kink there . for the other two cases ( hd and sp )
, @xmath62 exhibits an additional , intermediate regime ( though the behavior of @xmath47 shows little hint of this regime ) . in the hd case , @xmath62 is linear in @xmath76 with _ unit _ slope ; only the intercept depends on @xmath176 . here
, @xmath5 remains constant , as any increase in @xmath76 is `` absorbed '' by the chain ( through a shift in the domain wall ) .
this behavior may be regarded in the same light as an equilibrium system with phase co - existence .
considering , e.g. , the pressure _ vs. _ volume isotherm for a liquid - gas system below criticality , the pressure remains constant over some range , as any increase in volume is absorbed by a shift in the liquid - to - gas ratio .
the most interesting and challenging case is the sp . in the intermediate regime ,
a `` shoulder '' appears in @xmath153 .
in addition , this structure displays a non - trivial dependence on @xmath10 .
although the predictions of domain wall theory agree reasonably well with mc data , there are small ( @xmath177 ) systematic discrepancies .
we also investigated histograms of particle densities , which cross over from sharply peaked gaussians for @xmath173 to a flat distribution ( between @xmath178 and @xmath179 ) for @xmath180 . in the intermediate regime , this distribution displays a much richer behavior than a truncated exponential the result of domain wall theory .
several questions remain open for further study .
first , are the transitions between the three regimes in hd associated with true thermodynamic singularities ? we find that the changes in slope accompanying these `` crossovers '' become sharper with increasing chain length @xmath10 .
while this speaks for a true transition , the absence of large fluctuations is somewhat surprising .
clearly , a more careful analysis is required to explore these questions more fully .
second , though domain wall theory is reasonably successful for @xmath181 in the sp case , we would like to find improvements of this approach , so as to account for the crossover regime better . arriving at a good understanding of @xmath102 and the large fluctuations would be desirable .
progress along these lines may also help us to develop an intuitive picture for the complex @xmath10 dependence .
finally , we are mindful of one of the motivations of this study , namely , finite resources in biological systems , e.g. , ribosomes ( and aa - trna s ) for modeling protein synthesis in a cell @xcite . for that application , the above tasep needs to be generalized , to include exclusion at a distance ( particles covering more than one site ) and inhomogeneous hopping rates .
how these systems are affected by finite resources will pose many interesting new challenges , especially on the theoretical frontier . _ acknowledgements . _
we thank travis merritt for preliminary simulation results and meesoon ha and marcel den nijs for helpful discussions .
this work was supported in part by the national science foundation through dmr-0414122 and dmr-0705152 . | the totally asymmetric simple exclusion process ( tasep ) is a well studied example of far - from - equilibrium dynamics . here
, we consider a tasep with open boundaries but impose a global constraint on the total number of particles . in other words , the boundary reservoirs and the system must share a finite supply of particles . using simulations and analytic arguments , we obtain the average particle density and current of the system , as a function of the boundary rates and the total number of particles .
our findings are relevant to biological transport problems if the availability of molecular motors becomes a rate - limiting factor .
_ keywords _ : non - equilibrium statistical physics , totally asymmetric exclusion process , biological transport |
visual localization / tracking plays a central role for many applications like intelligent video surveillance , smart transportation monitoring systems . localization and tracking algorithms aim to find the most similar region to the target in an image .
recently , kernel - based tracking algorithms @xcite have attracted much attention as an alternative to particle filtering trackers @xcite .
one of the most crucial difficulties in robust tracking is the construction of representation models ( likelihood models in bayesian filtering trackers ) that can accommodate illumination variations , deformable appearance changes , partial occlusions , .
most current tracking algorithms use a single static template image to construct a target representation based on density models . for both kernel - based trackers and particle filtering trackers ,
a popular method is to exploit color distributions in simple regions ( region - wise density models ) .
generally semi - parametric kernel density estimation techniques are adopted .
however , it is difficult to update this target model @xcite , and the target representation s fragility usually breaks these trackers over a long image sequence . considerable effort has been expended to ease these difficulties .
we believe that the key to finding a solution is to find the right representation . in order to accommodate appearance changes ,
the representation model should be learned from as many training examples as possible .
fundamentally two methods , namely on - line and off - line learning , can be used for the training procedure .
on - line learning means constantly updating the representation model during the course of tracking .
@xcite proposes an incremental eigenvector update strategy to adapt the target representation model .
a linear probabilistic principal component analysis model is used .
the main disadvantage of the eigen - model is that it is not generic and is usually only suitable for characterizing texture - rich objects . in @xcite
a wavelet model is updated using the expectation maximization ( em ) algorithm .
a classification function is progressively learned using adaboost for visual detection and tracking in @xcite and @xcite respectively .
@xcite adopts pixel - wise gaussian mixture models ( gmms ) to represent the target model and sequentially update them . to date , however , less work has been reported on how to elegantly update _ region - wise density _ models in tracking .
in contrast , classification is a powerful bottom - up procedure : it is trained off - line and works on - line . due to the training
being typically built on very large amounts of training data , its performance is fairly promising even without on - line updating of the classifier / detector .
inspired by image classification tasks with color density features and real - time detection , we learn off - line a density representation model from multiple training data . by considering tracking as a binary classification problem , a discriminative classification rule is learned to distinguish between the tracked object and background patterns . in this way
a robust object representation model is obtained .
this proposal provides a basis for considering the design of enhanced kernel - based trackers using robust kernel object representations .
a by - product of the training is the classification function , with which the tracking problem is cast into a binary classification problem .
an object detector directly using the classification function is then available .
combining a detector into the tracker makes the tracker more robust and provides the capabilities of automatic initialization and recovery from momentary tracking failures . in theory ,
many classifiers can be used to achieve our goal . in this paper
we show that the popular kernel based non - linear support vector machine ( svm ) well fits the kernel - based tracking framework . within this framework
the traditional kernel object trackers proposed in @xcite and @xcite can be expressed as special cases . because we use probabilistic density features , the learning process is closely related to probabilistic kernels based svms @xcite .
it is imperative to minimize computational costs for real - time applications such as tracking .
a desirable property of the proposed algorithm is that the computational complexity is independent of the number of support vectors .
furthermore we empirically demonstrate that our algorithm requires fewer iterations to achieve convergence .
our approach differs from @xcite although both use the svm classification score as the cost function . in @xcite
, avidan builds a tracker along the line of standard optical flow tracking . only the homogeneous quadratic polynomial kernel ( or kernels with a similar quadratic structure ) can be used in order to derive a closed - form solution .
this restriction prevents one using a more appropriate kernel obtained by model selection .
an advantage of @xcite is that it can be used consistently with the optical flow tracking , albeit only gray pixel information can be used .
moreover , the optimization procedure of our approach is inspired by the kernel - based object tracking paradigm @xcite .
hence extended work such as @xcite is also applicable here , which enables us to find the global optimum .
if joint spatial - feature density is used to train an svm , a fixed - point optimization method may also be derived that is similar to @xcite .
the classification function of the svm trained for vehicle recognition is not smooth spatial mis - registration ( see fig . 1 in @xcite ) .
we employ a spatial kernel to smooth the cost function when computing the histogram feature . in this way ,
gradient based optimization methods can be used . using statistical learning theory
, we devise an object tracker that is consistent with ms tracking .
the ms tracker is initially derived from kernel density estimation ( kde ) .
our work sheds some light on the connection between svm and kde .
another important part of our tracker is its on - line re - training in parallel with tracking .
continuous updating of the representation model can capture changes of the target appearance / backgrounds .
previous work such as @xcite has demonstrated the importance of this on - line update during the course of tracking .
the incremental svm technique meets this end @xcite , which efficiently updates a trained svm function whenever a sample is added to or removed from the training set . for our proposed tracking framework , the target model can be learned in either batch svm training or on - line svm learning .
we adopt a sophisticated on - line svm learning proposed in @xcite for its efficiency and simplicity .
we address the crucial problem of adaptation , , the on - line learning of discriminant appearance model while avoiding drift .
the main contributions of our work are to solve ms trackers two drawbacks : the template model can only be built from a single image ; and it is difficult to update the model . the solution is to extend the use of statistical learning algorithms for object localization and tracking .
svm has been used for tracking by means of spatial perturbation of the svm @xcite .
we exploit svm for tracking in a novel way ( along the line of ms tracking ) .
the key ingredients of our approach are : * probabilistic kernel based svms are trained and incorporated into the framework of ms tracking . by carefully selecting the kernel ,
we show that no extra computation is required compared with the conventional single - view ms tracking . * an on - line
svm can be used to adaptively update the target model .
we demonstrate the benefit of on - line target model update . *
we show that the annealed ms algorithm proposed in @xcite can be viewed as a special case of the continuation method under an appropriate interpretation . with the new interpretation , annealed ms
can be extended to more general cases .
extension and new discovers are discussed .
an efficient localizer is built with global mode seeking techniques .
* again , by exploiting the svm binary classifier , it is able to determine the scale of the target .
an improved annealed ms - like algorithm with a cascade architecture is developed .
it enables a more systematic and easier design of the annealing schedule , in contrast with _ ad hoc _ methods in previous work @xcite .
the remainder of the paper is organized as follows . in [ sec : pre ] , the general theory of ms tracking and svm is reviewed for completeness .
our proposed tracker is presented in [ sec : gkt ] .
finally experimental results are reported in [ sec : exp ] .
we conclude this work in [ sec : conclusion ] .
for self - completeness , we review mean shift tracking , support vector machine and its on - line learning version in this section . mean shift ( ms ) tracking was firstly presented in @xcite . in ms tracking ,
the object is represented by a square region which is cropped and normalized into a unit circle . by denoting @xmath0 as the color histogram of the target model , and @xmath1 as the target candidate color histogram with the center at @xmath2 ,
the similarity function between @xmath0 and @xmath3 is ( when bhattacharyya divergence @xcite is used ) , @xmath4 here @xmath5 is the dissimilarity measurement . let @xmath6 be a region s pixel positions in image @xmath7 with the center at @xmath2 . in order to make the cost function smooth
otherwise gradient based ms optimization can not be applied a kernel with profile @xmath8 is employed to assign smaller weights to those pixels farther from the center , considering the fact that the peripheral pixels are less reliable .
an @xmath9-bin color histogram is built for an image patch located at @xmath2 , @xmath10 , where @xmath11 here @xmath8 is the homogeneous spatial weighting kernel profile and @xmath12 is its bandwidth .
@xmath13 is the delta function and @xmath14 normalizes @xmath0 .
the function @xmath15 maps a feature of @xmath16 into a histogram bin @xmath17 .
@xmath2 is the kernel center ; and for the target model usually @xmath18 .
the representation of candidate @xmath19 takes the same form .
given an initial position @xmath20 , the problem of localization / tracking is to estimate a best displacement @xmath21 such that the measurement @xmath22 at the new location best matches the target @xmath0 , , @xmath23 by taylor expanding @xmath24 at the start position @xmath20 and keeping only the linear item ( first - order taylor approximation ) , the above optimization problem can be resolved by an iterative procedure : @xmath25 } = \frac { \sum_{\ell=1}^n { \bi_\ell { \widetilde w}_\ell g ( \vert \frac{\bc ^ { [ \tau ] } - \bi_\ell}{h } \vert^2 ) } } { \sum_{\ell=1}^n { { \widetilde w}_\ell g ( \vert \frac{\bc ^ { [ \tau ] } - \bi_\ell}{h } \vert^2 ) } } , \ ] ] where @xmath26 and the superscript @xmath27 , indexes the iteration step .
the weights @xmath28 are calculated as : @xmath29 see @xcite for details .
we limit our explanation of the support vector machine classifiers algorithm to an overview .
large margin classifiers have demonstrated their advantages in many vision tasks .
svm is one of the popular large margin classifiers @xcite which has a very promising generalization capacity .
the linear svm is the best understood and simplest to apply .
however , linear separability is a rather strict condition .
kernels are combined into margins for relaxing this restriction .
svm is extended to deal with linearly non - separable problems by mapping the training data from the input space into a high - dimensional , possibly infinite - dimensional , feature space , , @xmath30 . using the kernel trick ,
the map @xmath31 is not necessarily known explicitly . like other kernel methods
, svm constructs a symmetric and positive definite kernel matrix ( gram matrix ) which represents the similarities between all training datum points .
given @xmath32 training data @xmath33 , the kernel matrix is written as : @xmath34 .
when @xmath35 is large , the labels of @xmath36 and @xmath37 , @xmath38 and @xmath39 , are expected to be the same . here , @xmath40 .
the decision rule is given by @xmath41 with @xmath42 where @xmath43 , @xmath44 , are support vectors , @xmath45 is the number of support vectors , @xmath46 is the weight associated with @xmath47 , and @xmath48 is the bias .
the training process of svm then determines the parameters @xmath49 by solving the optimization problem @xmath50 where @xmath51 is the slack variable set and the regularization parameter @xmath52 determines the trade - off between svm s generalization capability and training error .
@xmath53 corresponds to @xmath54-norm and @xmath55-norm svm respectively .
the solution takes the form @xmath56 . here
, @xmath57 and most of them are @xmath58 , yielding sparseness .
the optimization can be efficiently solved by linear programming ( @xmath59-norm svm ) or quadratic programming ( @xmath55-norm svm ) in its dual .
refer to @xcite for details .
a simple on - line kernel - based algorithm , termed , has been proposed for a variety of standard machine learning tasks in @xcite .
the algorithm is computationally cheap at each update step .
we have implemented here for on - line svm learning .
see fig .
@xmath59 in @xcite for the backbone of the algorithm .
we omit the details due to space constraint .
as mentioned , visual tracking is naturally a time - varying problem .
an on - line learning method allows updating the model during the course of tracking .
the standard kernel - based ms tracker is generalized by maximizing a sophisticated cost function defined by svm . measuring
the similarity between images and image patches is of central importance in computer vision . in svms
, the kernel @xmath60 plays this role .
most commonly used kernels such as gaussian and polynomial kernels are not defined on the space of probability distributions . recently various probabilistic kernels have been introduced , including the fisher kernel @xcite , top @xcite , kullback - leibler kernel @xcite and probability product kernels ( ppk ) @xcite , to combine generative models into discriminative classifiers .
a probabilistic kernel is defined by first fitting a probabilistic model @xmath61 to each training vector @xmath36 .
the kernel is then a measure of similarity between probability distributions .
ppk is an example @xcite , with kernel given by @xmath62 where @xmath63 is a constant .
when @xmath64 , ppk reduces to a special case , termed the bhattacharyya kernel : @xmath65 in the case of discrete histograms , , @xmath66^\t $ ] and @xmath67^\t $ ] , becomes @xmath68 when @xmath69 , @xmath70 computes the expectation of one distribution over the other , and hence is termed the expected likelihood kernel @xcite . in @xcite
its corresponding statistical affinity is used as similarity measurement for tracking .
the bhattacharyya kernel is adopted in this work due to : * the standard ms tracker @xcite uses the bhattacharyya distance .
it is clearer to show the connection between the proposed tracker and the standard ms tracker by using bhattacharyya kernel . *
it has been empirically shown , at least for image classification , that the generalization capability of expected likelihood kernel @xmath71 is weaker than the bhattacharyya kernel . meanwhile , non - linear probabilistic kernels including bhattacharyya kernel , kullback - leibler kernel , rnyi kernel perform similarly @xcite .
moreover , bhattacharyya kernel is simple and has no kernel parameter to tune .
the ppk has an interesting characteristic that the mapping function @xmath72 is explicitly known : @xmath73 .
this is equivalent to directly setting @xmath74 and the kernel @xmath75 .
consequently for discrete ppk based svms , in the test phase the computational complexity is independent of the number of support vectors .
this is easily verified .
the decision function is @xmath76^\t \bp ( \bx ) ^\rho + b } \\ & = \left [ \sum_{i=1}^{n_s } { \hbeta_i \bq ( \bx_i ) ^\rho } \right]^\t \bp ( \bx ) ^\rho + b. \end{aligned}\ ] ] the first term in the bracket can be calculated beforehand .
for example , for histogram based image classification like @xcite , given a test image @xmath77 , the histogram vector @xmath78 is immediately available .
in fact we can interpret discrete ppk based svms as _ linear _ svms in which the input vectors are @xmath79the features _ non - linearly _ , it is linear .
the non - linear probabilistic kernels induce a transformed feature space ( as the bhattacharyya kernel does ) to smooth density such that they significantly improve classification over the linear kernel @xcite . ]
extracted from image densities .
again , one might argue that , since the bhattacharyya kernel is very similar to the linear svm , it might not have the same power in modelling complex classification boundaries as the traditional non - linear kernels like the gaussian or polynomial kernel .
the experiments in @xcite indicate that the classification performance of a probabilistic kernel which consists an exponential calculation is not clearly better : exponential kernels like the kullback - leibler kernel and rnyi kernel performs similarly as bhattacharyya kernel on various datasets for image classification .
moreover our main purpose is to learn a representation model for visual tracking .
unlike other image classification tasks in which high generalization accuracy is demanded for visual tracking achieving very high accuracy might not be necessary and may not translate to a significant increase in tracking performance .
note that ppks are less compelling when the input data are vectors with no further structure .
however , even the gaussian kernel is a special case of ppk ( @xmath80 in equation and @xmath81 is a single gaussian fit to @xmath82 by maximum likelihood ) @xcite .
by contrast , the reduced set method is applied in @xcite to reduce the number of support vectors for speeding up the classification phase .
applications which favour fast computation in the testing phase , such as large scale image retrieval , might also benefit from this discrete ppk s property .
it is well known that the magnitude of the svm score @xmath83 measures the _ confidence _ in the prediction .
the proposed tracking is based on the assumption that the local maximum of the svm score corresponds to the target location we seek , starting from an initial guess close to the target . if the local maximum is positive , the tracker accepts the candidate .
otherwise an exhaustive search or localization process will start . the tracked position at time @xmath84 is the initial guess of the next frame @xmath85 and so forth .
we now show how the local maximum of the decision score is determined . as in @xcite ,
a histogram representation of the image region can be computed as equation . with equations , and , we have to represent the image region .
we also use the image center @xmath2 to represent the image region @xmath86 . for clarity
we define notation @xmath87 . ]
@xmath88 we assume the search for the new target location starts from a near position @xmath89 , then a taylor expansion of the kernel around @xmath90 is applied , similar to @xcite . after some manipulations and putting those terms independent of @xmath2 together , denoted by @xmath91 , becomes @xmath92 where @xmath93 and @xmath94 } { \sqrt{p_u ( \bc_0 ) } } \delta ( { \vartheta ( \bi_\ell -u ) } ) } .
\label{eq : svtweight2}\ ] ] here is obtained by swapping the order of summation .
the first term of @xmath95 is the weighted kernel density estimate with kernel profile @xmath8 at @xmath2 .
it is clear now that our cost function @xmath96 has an identical format as the standard ms tracker .
can we simply set @xmath97 which leads to a fixed - point iteration procedure to _ maximize _ @xmath95 as the standard ms does ?
if it works , the optimization would be similar to .
unfortunately , @xmath97 can not guarantee a local maximum convergence .
that means , the fixed point iteration can converge to a local minimum .
we know that only when all the weights @xmath98 are positive , converges to a local maximum as the standard ms does .
see appendix for the theoretical analysis .
however , in our case , a negative support vector s weight @xmath46 is negative , which means some of the weights computed by could be negative .
the traditional ms algorithm requires that the sample weights must be non - negative .
@xcite has discussed the issue on ms with negative weights and a _ heuristic _
modification is given to make ms able to deal with samples with negative weights . according to @xcite ,
the modified ms is @xmath99 } = \frac { \sum_{\ell=1}^n { \bi_\ell \hw_\ell g ( \vert \frac{\bc ^ { [ \tau ] } - \bi_\ell}{h } \vert^2 ) } } { \sum_{\ell=1}^n { \vert \hw_\ell g ( \vert \frac{\bc ^ { [ \tau ] } - \bi_\ell}{h } \vert^2 ) \vert } } .\ ] ] here @xmath100 is the absolute value operation . alas this heuristic solution is problematic .
note that no theoretical analysis is given in @xcite .
we show that the methods in @xcite can not guarantee converging to a local maximum mode .
see appendix for details .
the above problem may be avoided by using @xmath59-class svms @xcite in which @xmath98 is strictly positive .
however the discriminative power of svm is also eliminated due to its unsupervised nature . in this work , we use a quasi - newton gradient descent algorithm for maximizing @xmath101 in .
in particular , the l - bfgs algorithm @xcite is adopted for implementing the quasi - newton algorithm .
we provide callbacks for calculating the value of the svm classification function @xmath101 and its gradient .
typically , only few iterations of the optimization procedure are performed at each frame .
it has been shown that quasi - newton can be a better alternative to ms optimization for visual tracking @xcite in terms of accuracy .
quasi - newton was also used in @xcite for kernel - based template alignment . besides
, in @xcite the authors have shown that quasi - newton converges around twice faster than the standard ms does for data clustering .
the essence behind the proposed svm score maximization strategy is intuitive .
the cost function favors both the dissimilarity to negative training data ( , background ) and the similarity to positive training data . compared to the standard ms tracking
, our strategy provides the capability to utilize a large amount of training data .
the terms with positive @xmath102 in the cost function play the role to attract the target candidate while the negative terms repel the candidate . in @xcite zhao
have extended ms tracking by introducing a background term to the cost function , , @xmath103 .
@xmath104 is the background color histogram in the corresponding region .
it also linearly combines both positive and negative terms into tracking and better performance has been observed .
it is simple and no training procedure is needed .
nevertheless it lacks an elegant means to exploit available training data and the weighting parameters @xmath105 and @xmath106 need to be tuned manually .
the original ms tracker s analysis relies on kernel properties @xcite .
we argue that the main purpose of the kernel weighting scheme is to smooth the cost function such that iterative methods are applicable .
kernel properties then derive an efficient ms optimization .
as observed by many other authors @xcite , the kernels used as weighting kernel density estimation @xcite .
we can simply treat the feature distribution as a weighted histogram to smooth the cost function and , at the same time , to account for the non - rigidity of tracked targets .
note that ( 1 ) the optimization reduces to the standard ms tracking if @xmath107 ; ( 2 ) other probability kernels like @xmath108 are also applicable here .
the only difference is that @xmath109 in will be in other forms . in previous contents
we have shown that in the testing phase discrete ppk s support vectors do not introduce extra computation .
again , for our tracking strategy , no computation overhead is introduced compared with the traditional ms tracking in @xcite .
this can be seen from equation .
the summation in ( the bracketed term ) can be computed off - line .
the only extra computation resides in the training phase : the proposed tracking algorithm has the _ same _ computation complexity as the standard ms tracker .
it is also straightforward to extend this tracking framework to spatial - feature space @xcite which has proved more robust .
+ a technique is proposed in @xcite , dubbed annealed mean shift ( ) , to reliably find the global _ density _ mode .
is motivated by the observation that the number of modes of a kernel density estimator with a gaussian kernel is monotonically non - increasing the bandwidth of the kernel .
here we re - interpret this global optimization and show that it is essentially a special case of the _ continuation _ approach @xcite . with the new interpretation ,
it is clear now that this technique is applicable to a broader types of cost functions , not necessary to a density function .
the continuation method is one of the unconstrained global optimization techniques which shares similarities with deterministic annealing . a series of gradually deformed but smoothed cost functions are successively optimized , where the solution obtained in the previous step serves as an initial point in the current step .
this way the convergence information is conveyed . with sufficient smoothing , the first cost function will be concave / convex such that the global optimum can be found .
the algorithm iterates until it traces the solution back to the original cost function .
we now recall some basic concepts of the continuation method . given a non - linear function @xmath110 , the transformation @xmath111 for @xmath110 is defined such that @xmath112 , @xmath113 where @xmath114 is a smoothing function ; usually the gaussian is used .
@xmath12 is a positive scalar which controls the degree of smoothing .
@xmath115 is a normalization constant such that @xmath116 note the similarity between the smoothing function @xmath114 and the definition of the kernel in kde . from
, the defined transformation is actually the convolution of the cost function with @xmath117 . in the frequency domain ,
the frequency response of @xmath118 equals the product of the frequency responses of @xmath119 and @xmath120 . being a smoothing filter ,
the effect of @xmath117 is to remove high frequency components of the original function .
therefore one of the requirements for @xmath114 is its frequency response must be a low - pass frequency filter .
we know that popular kernels like gaussian or epanechnikov kernel are low - pass frequency filters .
this is one of the principle justifications for using gaussian or epanechnikov to smooth a function .
when @xmath12 is increased , @xmath111 becomes smoother and for @xmath121 , the function is the original function .
the annealed version of mean shift introduced in @xcite for global mode seeking is a special case of the general continuation method defined in equation .
let the original function @xmath122 take the form of a dirac delta comb ( impulse train in signal processing ) , , @xmath123 , where @xmath47 is known . with the fundamental property that @xmath124 for any function @xmath125 , we have @xmath126 this is exactly same as a kde .
this discovers that is a special case of the continuation method .
when @xmath127 with @xmath128 ( @xmath129 can be negative ) , the above analysis still holds and this case corresponds to the svm score maximization in [ sec : dsm ] .
it is not a trivial problem to determine the optimal scale of the spatial kernel bandwidth , , the size of the target , for kernel - based tracking .
a line search method is introduced in @xcite . for
, an important open issue is how to design the annealing schedule .
armed with an svm classifier , it is possible to determine the object s scale . if only the color feature is used , due to its lack of spatial information and insensitive to scale change , it is difficult to estimate a fine scale of the target . by combining other features , better
estimates are expected . as we will see in the experiments ,
reasonable results can be obtained with only color .
it is natural to combine into a cascade structure , like the cascade detector of @xcite .
we start ms search from a large bandwidth @xmath130 .
after convergence , an extra verification is applied to decide whether to terminate the search .
if @xmath131 , it means @xmath130 is too large . then we need to reduce the bandwidth to @xmath132 and start ms with the initial location @xmath133 .
this procedure is repeated until @xmath134 , @xmath135 .
@xmath136 and @xmath137 are the final scale and position .
little extra computation is needed because only a decision verification is introduced at each stage .
in this section we implement a localizer and tracker and discuss related issues .
experimental results on various data sets are shown .
for the first experiment , we have trained a face representation model .
@xmath138 faces cropped from caltech-@xmath139 are used as positive raw images , and @xmath140 negative images are randomly cropped from images which do not contain faces .
the image size is reduced to @xmath141 pixels .
kernel - weighted rgb colour histograms , consisting of @xmath142 bins , are extracted for classification . by default
we use a soft svm trained with libsvm ( slightly modified to use customized kernels ) .
test accuracy on the training data is @xmath143 ; and @xmath144 on a test data set which contains totally @xmath145 negative data .
note that our main purpose is not to train a powerful face detector ; rather , we want to obtain an appearance model that is more robust than the single - view appearance model .
we now test how well the algorithm maximizes the svm score .
first , we feed the algorithm a rough initial guess and run ms .
see fig .
[ fig : trackingexp1 ] for details .
the first example in fig .
[ fig : trackingexp1 ] comes from the training data set .
the initial svm score is negative . in this case
, a single step is required to switch to a positive score it moves closely to the target after one iteration .
we plot the corresponding cost function in fig .
[ fig:1 ] . by comparison ,
the cost function of the standard ms is also plotted ( the target template is cropped from the same image ) .
we can clearly see the difference .
the other two test images are from outside of the training data set . despite the significant face color difference and variation in illumination ,
our svm localizer works well in both tests . to compare the robustness
, we use the first face as a template to track the second face in fig .
[ fig : trackingexp1 ] , the standard ms tracker fails to converge to the true position .
we now apply the global maximum seeking algorithm to object localization . in @xcite
, it has been shown that it is possible to locate a target no matter from which initial position the ms tracker starts . here
we use the learned classification rule to determine when to stop searching .
we start the annealed continuation procedure with the initial bandwidth @xmath146
. then the bandwidth pyramid works with the rule @xmath147 , @xmath148 .
@xmath149 is the maximum number of iterations .
we stop the search when for some @xmath9 the svm score is positive upon convergence .
the image center is set to be the initial position of the search for these @xmath150 tests .
we present the results in fig .
[ fig : localisationexp1 ] . in the first test ,
our proposed algorithm works well : it successfully finds the face location , and also the final bandwidth well fits the target .
[ fig : localisationexp1 ] ( right ) shows how the svm score evolves .
it can be seen that every bandwidth change significantly increases the score .
if the target size is large and there is a significant overlap between the target and a search region at a coarse bandwidth , @xmath136 , the overlap can make the cascade search stop _ prematurely _ ( see the second test in fig .
[ fig : localisationexp1 ] ) .
again this problem is mainly caused by the color feature s weak discriminative power .
a remedy is to include more features .
however , for certain applications where the scale - size is not critically important , our localization results have been usable .
furthermore , better results could be achieved when we train a model for a specific object ( , train an appearance model for a specific person ) with a single color feature .
effectiveness of the proposed generalized kernel - based tracker is tested on a number of video sequences .
we have compared with two popular color histogram based methods : the standard ms tracker @xcite and particle filters @xcite .
unlike the first experiment , we do not train an _ off - line _ svm model for tracking .
it is not easy to have a large amount of training data for a general object , therefore in the tracking experiment , an on - line svm described in [ sec : onlinesvm ] is used for training .
the user crops several negative data and positive data for initial training . during the course of tracking the on - line svm updates its model by regarding the tracked region as a positive example and randomly selecting a few sub - regions ( background area ) around the target as negative examples .
a @xmath142-binned color histogram is used for both the generalized kernel tracker and standard ms tracker . for the particle filter , with @xmath151 or @xmath152 particles
, the tracker fails at the first a few frames .
so we have used @xmath153 particles .
and the frame rate is @xmath154 frames per second ( fps ) .
, title="fig:",scaledwidth=11.7% ] and the frame rate is @xmath154 frames per second ( fps ) .
, title="fig:",scaledwidth=11.7% ] and the frame rate is @xmath154 frames per second ( fps ) .
, title="fig:",scaledwidth=11.7% ] and the frame rate is @xmath154 frames per second ( fps ) . , title="fig:",scaledwidth=11.7% ] + and the frame rate is @xmath154 frames per second ( fps ) . , title="fig:",scaledwidth=11.7% ] and the frame rate is @xmath154 frames per second ( fps ) .
, title="fig:",scaledwidth=11.7% ] and the frame rate is @xmath154 frames per second ( fps ) .
, title="fig:",scaledwidth=11.7% ] and the frame rate is @xmath154 frames per second ( fps ) .
, title="fig:",scaledwidth=11.7% ] + and the frame rate is @xmath154 frames per second ( fps ) . , title="fig:",scaledwidth=11.7% ] and the frame rate is @xmath154 frames per second ( fps ) .
, title="fig:",scaledwidth=11.7% ] and the frame rate is @xmath154 frames per second ( fps ) . , title="fig:",scaledwidth=11.7% ] and the frame rate is @xmath154 frames per second ( fps ) . , title="fig:",scaledwidth=11.7% ] and the frame rate is @xmath154 fps .
, title="fig:",scaledwidth=11.7% ] and the frame rate is @xmath154 fps . , title="fig:",scaledwidth=11.7% ] and the frame rate is @xmath154 fps . , title="fig:",scaledwidth=11.7% ] and the frame rate is @xmath154 fps . ,
title="fig:",scaledwidth=11.7% ] + and the frame rate is @xmath154 fps .
, title="fig:",scaledwidth=11.7% ] and the frame rate is @xmath154 fps .
, title="fig:",scaledwidth=11.7% ] and the frame rate is @xmath154 fps . ,
title="fig:",scaledwidth=11.7% ] and the frame rate is @xmath154 fps . , title="fig:",scaledwidth=11.7% ] + and the frame rate is @xmath154 fps . , title="fig:",scaledwidth=11.7% ] and the frame rate is @xmath154 fps .
, title="fig:",scaledwidth=11.7% ] and the frame rate is @xmath154 fps .
, title="fig:",scaledwidth=11.7% ] and the frame rate is @xmath154 fps .
, title="fig:",scaledwidth=11.7% ] in the first experiment , the tracked person moves quickly .
hence the displacement between neighboring frames is large .
the illumination also changes .
the background scene is cluttered and contains materials with similar color as the target .
the proposed algorithm tracks the whole sequence successfully .
[ fig : tracking1 ] summarizes the tracking results .
the standard ms tracker fails at frame # 57 ; recovers at frame # 74 and then fails again .
the particle filter also loses the target due to motion blur and fast movement .
our on - line adaptive tracker achieves the most accurate results .
[ fig : tracking2 ] shows that the results on a more challenging video .
the target turns around and at some frames it even moves out of the view . at frame # 194
, the target disappears .
generalized kernel tracker and particle filter recovers at the following frames while the ms tracker fails .
again we can see the proposed tracker performs best due to its learned template model and on - line adaptivity .
when the head turns around , all trackers can lock the target because compared with the background , the hair color is more similar to the face color .
these two experiments show the proposed tracker s robustness to motion blur , large pose change and target s fast movement over the standard ms tracker and particle filter based tracker . in the experiments , to initialize the proposed tracker , we randomly pick up a few negative samples from the background .
we have found this simple treatment works well .
we present more samples from three more sequences in figs .
[ fig : tracking3 ] , [ fig : tracking4 ] and [ fig : tracking5 ] .
we mark only our tracker in these frames . from figs .
[ fig : tracking3 ] and [ fig : tracking4 ] we see that despite the target moving into shadow at some frames , our tracker successfully tracks the target through the whole sequences .
we have shown promising tracking results of the proposed tracker on several video clips .
we now present some quantitative comparisons of our algorithm with other trackers .
first , we run the proposed tracker , ms , and particle filter trackers on the cubicle sequence 1 . in fig .
[ fig : cubicle1 ] , we show some tracking frames of our method and particle filtering . compared with particle filtering , ours are much better in terms of accuracy and much faster in terms of the tracking speed .
our results are also _ slightly _ better than the standard ms tracker .
but visually there is no significant difference , so we have not included ms results in fig .
[ fig : cubicle1 ] .
again , the particle filter tracker uses @xmath153 particles .
we have run the particle filter @xmath155 times and the best result is reported .
[ fig : qant1 ] shows the absolute deviation of the tracked object s center at each frame .
clearly the generalized kernel tracker demonstrates the best result .
we have reported the average tracking error ( the euclidean distance of the object s center against the ground truth ) in table [ tab : quat1 ] , which shows the proposed tracker outperforms ms and particle filter . in table
[ tab : quat1 ] , the error variance estimates are calculated from the tracking results of all frames regardless the target is lost or not .
we have also proved the importance of on - line svm update .
as mentioned , when we switch off the on - line update , our proposed tracker would behave similarly to the standard ms tracker .
we see from table [ tab : quat1 ] that even without updating , the generalized kernel tracker is slightly better than the standard ms tracker .
this might be because the initialization schemes are different : the generalized kernel tracker can take multiple positive as well as negative training examples to _ learn _ an appearance model , while ms can only take a single image for initialization .
although we only use very few training examples ( less than @xmath156 ) , it is already better than the standard ms tracker . in this sequence ,
when the target object is occluded , the particle filter tracker only tracks the visible region such that the deviation becomes large .
our approach updates the learned appearance model using on - line svm .
the region that partially contains the occlusion is added to the object class database gradually based on the on - line update procedure .
this way our tracker tracks the object position close to the ground truth .
we also report the tracking failure rate ( fr ) for this video , which is the percentage of the number of failure frames in the total number of frames .
if the distance between the tracked center and the ground truth s center is larger than a threshold , we mark it a failure . we have defined the threshold as @xmath157 or @xmath158 of the diagonal length of the ground truth s bounding box , which results in two criteria : and respectively .
the former is more strict than the latter .
as shown in table [ tab : quat1 ] , our tracker with on - line update produces lowest tracking failures under either criterion . and @xmath159 fps . , title="fig:",scaledwidth=11.7% ] and @xmath159 fps . ,
title="fig:",scaledwidth=11.7% ] and @xmath159 fps .
, title="fig:",scaledwidth=11.7% ] and @xmath159 fps . ,
title="fig:",scaledwidth=11.7% ] + and @xmath159 fps .
, title="fig:",scaledwidth=11.7% ] and @xmath159 fps .
, title="fig:",scaledwidth=11.7% ] and @xmath159 fps . ,
title="fig:",scaledwidth=11.7% ] and @xmath159 fps . ,
title="fig:",scaledwidth=11.7% ] [ tab : quat1 ] we also compare the running time of trackers , which is an important issue for real - time tracking applications .
table [ tab : time1 ] reports the results on two sequences .. a desktop with intel core@xmath160 duo 2.4-ghz cpu and @xmath150-g ram is used for running all the experiments . ]
the generalized kernel tracker ( around @xmath161 fps ) is comparable to the standard ms tracker , and much faster than the particle filter .
this coincides with the theoretical analysis : our generalized kernel tracker s computational complexity is independent of the number of support vectors , so in the test phrase , the complexity is almost same as the standard ms .
one may argue that the on - line update procedure introduces some overhead .
but the generalized kernel tracker employs the l - bfgs optimization algorithm which is about twice faster than ms , as shown in @xcite .
therefore , overall , the generalized kernel tracker runs as fast as the ms tracker .
because the particle filter is stochastic , we have run it @xmath155 times and the average and standard deviation are reported . for our tracker and ms , they are deterministic and the standard deviation is negligible .
note that the computational complexity if the particle filter tracker is linearly proportional to the number of particles .
-norm absolute error ( pixels ) of the object s center against the ground truth on the cubicle sequence 1 .
the two figures correspond to @xmath162- , and @xmath163-axis , respectively .
the proposed tracker with on - line updating gives the best result .
as expected , the proposed tracker without updating shows a similar performance with the standard ms tracker .
, title="fig:",scaledwidth=40.0% ] -norm absolute error ( pixels ) of the object s center against the ground truth on the cubicle sequence 1 .
the two figures correspond to @xmath162- , and @xmath163-axis , respectively .
the proposed tracker with on - line updating gives the best result .
as expected , the proposed tracker without updating shows a similar performance with the standard ms tracker . , title="fig:",scaledwidth=40.0% ] .running time per frame ( seconds ) .
the stochastic particle filter tracker has run @xmath155 times and the standard deviation is also reported . [
cols="^,^,^,^",options="header " , ] [ tab : time1 ] we have run another test on cubicle sequence 2 .
we show some results of our method and particle filtering in fig .
[ fig : cubicle2 ] .
although all the methods can track this sequence successfully , the proposed method achieves most accurate results .
we see that when the tracked object turns around , our algorithm is still able to track it accurately .
table [ tab : quat2 ] summarizes the quantitative performance .
our method is also slightly better ms .
again we see that on - line update does indeed improve the accuracy .
we have also reported the tracking failure rates on this video .
our tracker with on - line update has the lowest tracking failures and the one without on - line update is the second best .
these results are consistent with the previous experiments .
[ tab : quat2 ] and frame rate @xmath159 fps .
, title="fig:",scaledwidth=11.7% ] and frame rate @xmath159 fps . ,
title="fig:",scaledwidth=11.7% ] and frame rate @xmath159 fps .
, title="fig:",scaledwidth=11.7% ] and frame rate @xmath159 fps .
, title="fig:",scaledwidth=11.7% ] + and frame rate @xmath159 fps . ,
title="fig:",scaledwidth=11.7% ] and frame rate @xmath159 fps .
, title="fig:",scaledwidth=11.7% ] and frame rate @xmath159 fps . , title="fig:",scaledwidth=11.7% ] and frame rate @xmath159 fps . ,
title="fig:",scaledwidth=11.7% ] -norm absolute error ( pixels ) of the object s center against the ground truth on the walker sequence 3 .
the two figures correspond to @xmath162- , and @xmath163-axis , respectively .
it clearly shows that on - line update of the generalized kernel tracker is beneficial : without on - line update , the error is larger .
, title="fig:",scaledwidth=40.0% ] -norm absolute error ( pixels ) of the object s center against the ground truth on the walker sequence 3 .
the two figures correspond to @xmath162- , and @xmath163-axis , respectively .
it clearly shows that on - line update of the generalized kernel tracker is beneficial : without on - line update , the error is larger .
, title="fig:",scaledwidth=40.0% ] to demonstrate the effectiveness of the on - line svm learning , we switch off the on - line update and run the tracker on the walker sequence 3 .
we plot the @xmath164-norm absolute deviation of the tracked object s center in pixels at each frame in fig .
[ fig : qant3 ] . apparently , at most frames , on - line update produces more accurate tracking results .
the average euclidean tracking error is @xmath165 pixels with on - line update and @xmath166 pixels without on - line update .
conclusions that we can draw from these experiments are : ( 1 ) the proposed generalized kernel - based tracker performs better than the standard ms tracker on all the sequences that we have used ; ( 2 ) on - line learning often improves tracking accuracy .
to summarize , we have proposed a novel approach to kernel based visual tracking , which performs better than conventional single - view kernel trackers @xcite . instead of minimizing the density distance between the candidate region and the template , the generalized ms tracker works by maximizing the svm classification score .
experiments on localization and tracking show its efficiency and robustness . in this way
, we show the connection between standard ms tracking and svm based tracking .
the proposed method provides a generalized framework to the previous methods .
future work will focus on the following possible avenues : * other machine learning approaches such as relevance vector machines ( rvm ) @xcite , might be employed to learn the representation model . since in the test phrase , rvm and svm
take the same form , rvm can be directly used here .
rvm achieves comparable recognition accuracy to the svm , but requires substantially fewer kernel functions .
it would be interesting to compare different approaches performances ; * the strategy in this paper can be easily plugged into a particle filter as an observation model . improved tracking results are anticipated than for the simple color histogram particle filter tracker developed in @xcite .
generally collins modified mean shift @xcite ( equation ) can not guarantee to converge to a local maximum .
it is obvious that a fixed point @xmath167 obtained by iteration using equation will not satisfy @xmath168 @xmath169 is the original cost function .
therefore , generally , @xmath167 will not even be an extreme point of the original cost function . in the following example
, @xmath167 obtained by collins modified mean shift converges to a point which is close to a local _ minimum _ , but not the exact minimum . in fig .
[ fig : negmsexp ] we give an example on a mixture of gaussian kernel which contains some negative weights . in this case both the standard ms and collins modified ms fail to converge to a maximum .
p. prez , c. hue , j. vermaak , and m. gangnet , `` color - based probabilistic tracking , '' in _ proc .
conf . comp .
_ , copenhagen , denmark , 2002 , vol .
2350 of _ lecture notes in computer science _ , pp . 661675 . c. shen , a. van den hengel , and a. dick , `` probabilistic multiple cue integration for particle filter based tracking , '' in _ proc .
conf . digital image computing techniques & applications _ , sydney , australia , 2003 , pp .
309408 . o. javed , s. ali , and m. shah , `` online detection and classification of moving objects using progressively improving detectors , '' in _ proc .
ieee conf . comp
_ , san diego , ca , 2005 , vol . 1 ,
696701 .
p. j. moreno , p. ho , and n. vasconcelos , `` a kullback - leibler divergence based kernel for svm classification in multimedia applications , '' in _ proc .
neural inf . process .
_ , vancouver , canada , 2003 .
r. jenssen , d. erdogmus , j. c. principe , and t. eltoft , `` towards a unification of information theoretic learning and kernel methods , '' in _ proc .
ieee workshop on machine learning for signal proce .
_ , sao luis , brazil , 2004 , pp . 93102 .
r. jenssen , d. erdogmus , j. c. principe , and t. eltoft , `` the laplacian pdf distance : a cost function for clustering in a kernel feature space , '' in _ proc . adv .
neural inf . process .
_ , 2004 , vol .
17 , pp . 625632 .
c. yang , r. duraiswami , and l. davis , `` efficient spatial - feature tracking via the mean - shift and a new similarity measure , '' in _ proc .
ieee conf . comp .
_ , san diego , ca , 2005 , vol . 1 ,
176183 .
[ ] chunhua shen received the b.sc . and m.sc .
degrees from nanjing university , china , and the ph.d .
degree from university of adelaide , australia .
he has been working as a research scientist in nicta , canberra research laboratory , australia since october 2005 .
he is also an adjunct research follow at australian national university and an adjunct lecturer at university of adelaide .
his research interests include statistical machine learning , convex optimization and their application in computer vision .
junae kim is a phd student at the research school of information sciences and engineering , australian national university .
she is also attached to nicta , canberra research laboratory .
she received the b.sc .
degree from ewha womans university , korea in 2000 , m.sc . from pohang university of science and technology , korea in 2002 , and m.sc . from australian national university in 2007 .
she was a researcher in electronics and telecommunications research institute ( etri ) , korea for 5 years before she moved to australia .
her research interests include computer vision and machine learning .
[ ] hanzi wang received his b.sc .
degree in physics and m.sc .
degree in optics from sichuan university , china , in 1996 and 1999 , respectively .
he received his ph.d .
degree in computer vision from monash university , australia , in 2004 .
he is a senior research fellow at the school of computer science , university of adelaide , australia .
his current research interest are mainly concentrated on computer vision and pattern recognition including robust statistics , model fitting , optical flow calculation , visual tracking , image segmentation , fundamental matrix estimation and related fields .
he has published more than 30 papers in major international journals and conferences .
he is a member of the ieee society . | kernel - based mean shift ( ms ) trackers have proven to be a promising alternative to stochastic particle filtering trackers . despite its popularity , ms trackers have two fundamental drawbacks : ( 1 ) the template model can only be built from a single image ; ( 2 ) it is difficult to adaptively update the template model . in this work
we generalize the plain ms trackers and attempt to overcome these two limitations .
it is well known that modeling and maintaining a representation of a target object is an important component of a successful visual tracker .
however , little work has been done on building a robust template model for kernel - based ms tracking .
in contrast to building a template from a single frame , we train a robust object representation model from a large amount of data .
tracking is viewed as a binary classification problem , and a discriminative classification rule is learned to distinguish between the object and background .
we adopt a support vector machine ( svm ) for training . the tracker is then implemented by maximizing the classification score .
an iterative optimization scheme very similar to ms is derived for this purpose .
compared with the plain ms tracker , it is now much easier to incorporate on - line template adaptation to cope with inherent changes during the course of tracking . to this end , a sophisticated on - line support vector machine is used .
we demonstrate successful localization and tracking on various data sets .
= 1 kernel - based tracking , mean shift , particle filter , support vector machine , global mode seeking . |
transmembrane voltage is often recorded during physiological study of biological neurons . however , voltage - gated ion channel activity and neurotransmitter levels are quite difficult to measure directly and are usually unobserved in such studies . in addition , there is a great diversity of neuron morphology , protein expression , and plasticity which may affect voltage dynamics and synaptic transmission @xcite . early development and senescence
may also be major determinants of voltage response profiles @xcite .
synaptic tuning in particular is thought to be an essential mediator of learning , stimulus response integration , and memory .
there is evidence that memory and learning may depend critically on several distinct types of dynamic behavior in the voltage of neurons .
the ml model reproduces the voltage of a single neuron and , depending on parameterization and initial conditions , can exhibit many of the experimentally observed behaviors of biological neurons @xcite . in this paper , we explore a simple neural network consisting of two biologically identical , reciprocally coupled ml neurons . @xcite have shown that this modest model can exhibit a wide range of oscillating or non - oscillating voltage depending on the values of just a few parameters , specifically in this study , @xmath1and @xmath0 . in the absence of noise , the model can predict synchronous or asynchronous firing , as well as either equal or unequal action potential amplitudes .
additionally , in the presence of even small noise in the applied current and weak synaptic coupling , the system can exhibit mixed - mode oscillations ( mmo ) characterized by periods of small amplitude oscillation interrupted by large amplitude excursions .
in further work with the two ml neuron model , @xcite explored two synaptically decoupled neurons driven by both common and independent intrinsic noise terms .
they found that shared common noise promotes synchronous firing of the two neurons , while separate intrinsic noise terms promoted asynchronous firing .
the relative scaling of the two noise sources was observed to be key in predicting the degree of synchrony .
in addition , while they did not specifically look at mmo , they hypothesized that such synchrony in a synaptically coupled network would increase the probability of mmo , by facilitating longer residence times within the unstable periodic orbits adjacent to the system s stable periodic orbits .
indeed , in this paper we will detail the relative positions of these parameter regions as they are of key importance to our conditioned likelihood approach . specifically , we will provide a quick look - up table for the region in parameter space where stable periodic orbits are possible .
@xcite develop a expectation - maximization ( em ) stochastic particle filter method to estimate the parameters in a single ml neuron based on observation of voltage only .
a key aspect of their approach is that they assume both the voltage and the channel gating variables are in an oscillatory regime , but stochastically perturbed . these perturbations are considered nuisance parameters which their method marginalizes away . specifically , they treat the unobserved channel gating variable from the model as a completely latent variable . starting from estimates of the initial conditions for the voltage and channel gating variables , they iteratively predict the gating variable and voltage and
then update the predicted voltage to the next time step using a modification of the well - known euler differential equation solver .
they discuss that an assumption of stationarity in their method limits applicability to only short time windows over which current input can be considered constant ( e.g. 600ms ) .
they also note that certain parameters , conductances and reversal potentials in particular , are sensitive to choice of tuning parameters required by the method .
these studies demonstrate the active progress as well as the challenges of model parameter estimation for biological neuronal models and , more generally , for relaxation oscillator models .
each of these studies derives asymptotic approximations or general forms for model likelihood , but use fundamentally different techniques and assumptions in doing so . in each
study the approach is specifically crafted to the model . in this paper
we attempt to develop a convenient bayesian estimation scheme with only a few tuning parameters and relatively few mild assumptions .
we focus our attention on deterministic synaptically coupled ml neurons .
application of our method to stochastically coupled ml neurons is on - going work in our group . in the case of ml , estimation of @xmath1and @xmath0is non - trivial due to the diversity of possible dynamic behavior and the abrupt transitions among these seen with just small changes in these parameters values .
however , we can better understand the critical values of these parameters by studying the system s bifurcation structure .
we are able to locate parameter regimes where dramatic changes in the system appear .
the neurons analyzed in this study are classified as type ii neurons , characterized by discontinuous drastic shifting between behavioral states . because there is a distinct switch in behavior , bifurcation analyses determine a closed region of parameter space over which the relevant dynamics may occur .
sampling over such a feasibility region amounts to conditioning the inference on an _ a priori _ assumed class of dynamics ( e.g. stable node , limit cycle , steady state etc . ) . while facilitating conditioning the likelihood on feature statistics of the voltage
, this may translate into increased confidence and reduced bias in the parameter estimates .
our goal is parameter inference based on the temporal voltage response of two synaptically coupled neurons which are deterministically coupled to voltage - gated ionic conductance dynamics @xcite .
a single ml model has a two - dimensional phase space and is known to reproduce many of the behaviors experimentally observed in biological neurons @xcite .
therefore , systems of coupled ml neurons may offer a reasonable starting point for developing statistical inference methods for models of neuronal networks .
the ml network we study is , @xmath2 where @xmath3 note that in the stochastic version of this model @xmath4 and @xmath5 and @xmath6 are standard independent wiener process variables . in this paper , however we will be concerned with the deterministic version of this model where @xmath7 . .
* * [ cols="^,^,^,^",options="header " , ] let @xmath9 have fourier series representation , + @xmath10 + then , it it to be shown that @xmath11 proof is by induction and is adapted from @xcite . in the base case ( @xmath12 ) , @xmath13 t & \overbrace{+2\pi^3 a_1 ^ 2 \phi_1 ^ 3 \sin(4\pi\phi_1 t)}^{g(t)~\in{~{\ensuremath{\mathcal{o}\!\left(1\right)}\xspace}}}\\ & - 2\pi^3 b_1 ^ 2 \phi_1 ^ 3 \sin(4\pi\phi_1t)\\ & + 16\pi^4a_1b_1\phi_1 ^ 4 \cos^2(2\pi\phi_1t)\\ & -16\pi a_1b_1\phi_1 ^ 4\end{aligned}\ ] ]
then it is supposed that for @xmath14 we have , @xmath15t+g_{n-1}(t)\end{aligned}\ ] ] @xmath16t+g_n(t)\end{aligned}\ ] ] next , collecting the @xmath17 terms from the summation yields , @xmath18 expanding the square in the previous result gives , @xmath19 the first term is recognized as the induction hypothesis and so has the form @xmath20 .
the integrals of the remaining terms are evaluated with the extensive use of trigonometric identities .
@xmath21t+g_{n-1}(t)\\ & + \sum_{k=1}^{n-1}8\pi^3a_ka_n\phi_k^2\phi_n^2\left[\frac{\sin(2\pi(\phi_k-\phi_n)t)}{\phi_k-\phi_n}+\frac{\sin(2\pi(\phi_k+\phi_n)t)}{\phi_k+\phi_n}\right]\\ & -\sum_{k=1}^{n-1}8\pi^3b_ka_n\phi_k^2\phi_n^2\left[\frac{\cos(2\pi(\phi_k-\phi_n)t)}{\phi_k-\phi_n}+\frac{\cos(2\pi(\phi_k+\phi_n)t)}{\phi_k+\phi_n}\right]\\ & + \sum_{k=1}^{n-1}8\pi^3b_ka_n\phi_k^2\phi_n^2\left[\frac{1}{\phi_k-\phi_n}+\frac{1}{\phi_k+\phi_n}\right]\\ & -\sum_{k=1}^{n-1}8\pi^3a_kb_n\phi_k^2\phi_n^2\left[\frac{\cos(2\pi(\phi_k-\phi_n)t)}{\phi_k-\phi_n}+\frac{\cos(2\pi(\phi_k+\phi_n)t)}{\phi_k+\phi_n}\right]\\ & + \sum_{k=1}^{n-1}8\pi^3a_kb_n\phi_k^2\phi_n^2\left[\frac{1}{\phi_k-\phi_n}+\frac{1}{\phi_k+\phi_n}\right]\\ & + \sum_{k=1}^{n-1}8\pi^3b_kb_n\phi_k^2\phi_n^2\left[\frac{\sin(2\pi(\phi_k-\phi_n)t)}{\phi_k-\phi_n}+\frac{\sin(2\pi(\phi_k+\phi_n)t)}{\phi_k+\phi_n}\right]\\ & \mathbf{+8\pi^4\phi_n^4a_n^2t}+2\pi^3a_n^2\phi_n^3\sin(4\pi\phi_nt)\\ & + 16\pi^4a_nb_n\phi_n^4\sin^2(2\pi\phi_nt)\\ & \mathbf{+8\pi^4\phi_n^4b_n^2t}-2\pi^3b_n^2\phi_n^3\sin(4\pi\phi_nt)\end{aligned}\]]the bold terms may be combined with the leading term of the induction hypothesis raising the upper bound of the summation from @xmath14 to @xmath22 .
the remaining terms , only containing @xmath23 as arguments of sines and cosines , can be merged with @xmath24 from the induction hypothesis . calling
this merger @xmath25 completes the induction .
cumulative power has been written in the desired form @xmath26t+g_n(t)\\ & = & c\cdot{t } + { \ensuremath{\mathcal{o}\!\left(1\right)}\xspace}\end{aligned}\ ] ] since clearly @xmath27 @xmath28 it follows that their sum @xmath29 proving the desired result .
this report summarizes work that was done as part of the summer undergraduate research institute of experimental mathematics ( suriem ) held at the lyman briggs college of michigan state university .
we are very grateful to the national security agency and the national science foundation for funding this research .
we would also like to thank our advisor , professor daniel p. dougherty , for his guidance throughout the summer , and our graduate assistant , joseph e. roth , for his assistance . | the morris - lecar ( ml ) model has applications to neuroscience and cognition .
a simple network consisting of a pair of synaptically coupled ml neurons can exhibit a wide variety of deterministic behaviors including asymmetric amplitude state ( aas ) , equal amplitude state ( eas ) , and steady state ( ss ) .
in addition , in the presence of noise this network can exhibit mixed - mode oscillations ( mmo ) , which represent the system being stochastically driven between these behaviors . in this paper , we develop a method to specifically estimate the parameters representing the coupling strength ( @xmath0 ) and the applied current ( @xmath1 ) of two reciprocally coupled and biologically similar neurons .
this method employs conditioning the likelihood on cumulative power and mean voltage .
conditioning has the potential to improve the identifiability of the estimation problem .
conditioning likelihoods are typically much simpler to model than the explicit joint distribution , which several studies have shown to be difficult or impossible to determine analytically .
we adopt a rejection sampling procedure over a closed defined region determined by bifurcation continuation analyses .
this rejection sampling procedure is easily embedded within the proposal distribution of a bayesian markov chain monte carlo ( mcmc ) scheme and we evaluate its performance .
this is the first report of a bayesian parameter estimation for two reciprocally coupled morris - lecar neurons , and we find a proposal utilizing rejection sampling reduces parameter estimate bias relative to naive sampling .
application to stochastically coupled ml neurons is a future goal . |
in the course of the last decade redshift surveys have provided a major advance in our knowledge of the large scale distribution of galaxies and its statistical properties ( see for example the review of giovanelli and haynes 1991 ) .
bright , wide angle surveys ( e.g. cfa2 , geller and huchra 1989 ; ssrs2 , da costa et al .
1994 ; perseus
pisces , giovanelli and haynes 1988 ) cover a large fraction of the sky and provide a clear picture of the nearby universe , up to about 10,000 km / s or @xmath5 ( where @xmath6 ) .
these surveys reveal large structures in the distribution of galaxies : voids of sizes up to @xmath7 ( de lapparent et al .
1986 ) , and large ( @xmath8 ) bidimensional sheets as for instance the great wall within cfa2 ( geller and huchra 1989 ) , the southern wall within ssrs2 ( da costa et al 1994 ) , or the perseus
pisces filament ( giovanelli and haynes 1988 ) .
strategies alternative to those adopted for the above mentioned surveys have been used to study the large scale structure in depth without paying the price of an excessively large increase of the observing time .
these strategies include : a ) sparse sampling ( see kaiser 1986 ) , as in the stromlo apm redshift survey ( loveday et al . 1992 ) ; b ) chessboard surveys consisting of separated fields covering a large solid angle ; c ) pencil beam surveys , as in the broadhurst et al .
( 1990 ; beks ) survey .
these deeper surveys have confirmed the texture detected in shallower surveys up to a depth of 40000 50000 km / s . these strategies , while allowing a faster completion of the surveys , sometimes do not allow an unambiguous interpretation of the data . for example , beks report evidence of periodic structures in their first pencil beam redshift survey .
however , the real nature of this periodicity is not clear .
moreover , the periodicity is not confirmed in other directions of the sky by the same authors ( koo 1993 ) and is not detected in other similarly deep redshift surveys even in directions close to the original beks pencil beam ( bellanger and de lapparent 1995 ) . the present survey ( hereafter esp : eso slice project )
was designed to provide an unbiased spectroscopic sample of galaxies brighter than @xmath9 , with a high level of completeness , over a region of the sky significantly extended in the right ascension direction .
the geometry of the survey , a slice , is the most efficient for mapping three - dimensional structures like those observed in shallower surveys ( de lapparent et al .
1988 ) , provided its thickness is not smaller than the correlation length of the galaxy distribution .
our sample can be used to derive the basic statistical properties of the galaxy distribution , averaged over a volume large enough to smooth out `` local '' inhomogeneities . with a magnitude limit @xmath10
we include @xmath11 galaxies up to @xmath12 and obtain a redshift distribution peaking at @xmath2 .
because of the target selection criteria , the esp is unbiased with respect to a number of statistical descriptors , such as , for example , the luminosity function ( see zucca et al . 1997 , paper ii ) .
the redshift distribution of the esp galaxies is similar to that of the galaxies in the las campanas redshift survey ( lcrs , shectman et al .
1996 ) , which covers over 700 square degrees in six strips , each @xmath13 , and consists of 26418 redshifts of galaxies .
the comparison between the two surveys , however , is not straightforward .
in fact , esp galaxies are selected in the photographic blue band , while lcrs galaxies are selected on red ccd frames .
furthermore , contrary to what was done for the lcrs , we did not apply any `` a priori '' selection on the surface brightness of the target galaxies . in section 2
we briefly describe the survey region , the galaxy catalogue , the observations and the data reduction ; in section 3 we discuss the main spectral properties of the galaxies in our sample ; in section 4 we describe qualitatively the large scale structures detected in our survey .
we summarize the main properties of the esp redshift survey in section 5 .
a detailed description of the survey will be presented in the paper containing the whole data catalogue ( vettolani et al .
1997 , paper iii , in preparation ) . here
we briefly summarize the main characteristics of the esp project .
the main area of the esp survey is a strip @xmath14 long in right ascension and @xmath15 thick in declination ( hereafter strip a ) . in order to make full use of the allotted nights
, we were able to survey also an area of @xmath16 ( hereafter strip b ) , five degrees west of strip a. both strips are located in the region of the south galactic pole .
the position was chosen in order to minimize the galactic absorption effects ( @xmath17 ) .
the right ascension limits are from @xmath18 to @xmath19 for strip b and from @xmath20 to @xmath21 for strip a , at the mean declination @xmath22 ( 1950 ) .
the target galaxies , with a limiting magnitude @xmath9 , were extracted from the edinburgh
durham southern galaxy catalogue ( edsgc , heydon dumbleton et al .
1988 , 1989 ) , which has been obtained from cosmos scans of serc j survey plates .
the edsgc has a @xmath23 completeness at @xmath24 and an estimated stellar contamination @xmath25 ( heydon dumbleton et al . 1989 ) .
preliminary analysis of ccd data , obtained with the 0.9 m dutch / eso telescope for about 80 galaxies in the magnitude range @xmath26 in the region of our survey , shows a linear relation between @xmath27(edsgc ) and @xmath28(ccd ) , with a dispersion ( @xmath29 ) of about 0.2 magnitudes around the fit ( garilli et al . in preparation ) .
since the ccd pointings cover the entire right ascension range of our survey , this @xmath29 includes both statistical errors within single plates and possible plate to plate zero point variations .
we do not have enough information to assess the reliability of magnitudes for the 71 esp galaxies with @xmath30 .
it is known that in this bright range various problems may affect the measures , as saturation , ghosts and spikes on plates , the presence of substructure in bright galaxies , or contamination from overlapping bright stars ( e.g. loveday 1996 ) . with this _ caveat _ in mind , we simply note that presently we do not have evidence for large errors on the magnitudes of bright galaxies in our catalogue , on the basis of a few cases we could check .
lllllll sample & catalogue & observed & galaxies & stars & failed + & objects & objects & with z & & spectra + strip a & 3391 & 3265 & 2749 & 368 & 147 + strip b & 1096 & 779 & 593 & 125 & 61 + total & 4487 & 4044 & 3342 & 493 & 208 + the observations have been obtained during six observing runs ( in the period 19911993 ) with the multifiber spectrograph optopus ( lund 1986 , avila et al .
1989 ) at the cassegrain focus of the eso 3.6 m telescope at la silla .
the geometry of the survey ( see figure 1 ) is a regular grid consisting of two rows of adjacent circular fields .
each field has a diameter of 32 arcmin , corresponding to the field of view of the optopus spectrograph .
the two rows of fields are separated by 15 arcmin and slightly overlap each other .
the total solid angle is @xmath31 steradians ( i.e. 23.3 square degrees ) .
the instrument optopus has 50 fibers , each with a diameter projected on the sky of @xmath32 arcsec .
the fibers are manually plugged into holes drilled in aluminum plates at the galaxy positions . for each field
we reserved at least 5 fibers for the measurement of the spectrum of the blank sky . on average
, there are 35 galaxies brighter than @xmath9 per optopus field and @xmath33 of the fields have fewer than 45 galaxies . when the number of galaxies in a field exceeded the number of available fibers , we selected at random the galaxies to be observed from the total galaxy list . whenever possible , we re - observed overdense fields in order to reach a higher completeness .
another set of observations has been obtained in october 1994 with the multifiber mefos spectrograph ( felenbok et al .
thanks to these observations we have been able to observe some of the galaxy spectra that had no redshift determination after the optopus runs . in particular , we have used these observations to reach a completeness as uniform as possible over all the optopus fields in strip a. the spectra cover the wavelength range from 3730 to 6050 , with an average pixel size of 4.5 .
we measured the redshifts of the galaxies by cross correlating the sky - subtracted spectra with a set of 8 template stars .
the template stars have been observed with the same instrumental set - up used to obtain the galaxy spectra .
we also measured redshifts from the emission lines , whenever present .
the median internal velocity error is of the order of @xmath34 km / s . from a comparison of our 8 templates with three sao radial velocity standard stars we estimate that the zero point error should be smaller than @xmath35 km / s
. we will give a full description of our observations and our reduction procedure in paper iii .
we observed a total of 4044 objects , corresponding to @xmath36 of the parent photometric sample of 4487 objects .
out of the 4044 observed objects , 493 turned out to be stars and 208 have a too low signal to noise ratio to provide a reliable redshift ( failed spectra ) . in the end , our final sample consists of a total of 3342 galaxies with reliable redshifts ( @xmath37 one quasar ) . about half of the galaxies in our sample have detectable emission lines ( see section 3 ) . in table 1
we report a summary of the basic numbers relative to strip a and strip b. the mean completeness of our galaxy sample is @xmath38 .
we derive this value by assuming that all the failed spectra correspond to galaxies and that the percentage of stars among the 443 objects that were not observed is the same as in the spectroscopic sample ( i.e. @xmath39 ) .
the completeness level is significantly different in strip a and in strip b , being @xmath40 for strip a and @xmath41 for strip b. the different completeness level between the two strips is due to our choice of repeating observations only for fields in strip a. figure 2 shows the completeness for each optopus field : panel a ) refers to the northern row ( field numbers @xmath42 ) , panel b ) to the southern row ( field numbers @xmath8 ) .
a detailed understanding of the completeness and the selection effects is extremely important in any analysis of our data .
we therefore performed a number of tests to assess the statistical properties of the galaxies for which we did not measure the spectrum or we did not get a measurable spectrum .
first of all , in the fields where the number of objects is higher than the number of available fibers , the objects we observed are a random subset of the total catalog with respect to both magnitude and surface brightness . the set of observed galaxies departs from randomness only with respect to the selection of the position of the target objects , due to the instrumental constraint that two fibers can not be put closer than about 25 arcsec ; obviously this means that when dealing with pairs of galaxies at small projected separation , we could observe only one galaxy ( except when the other galaxy was selected for a second exposure ) .
as far as failed spectra are concerned , we find that the fraction of failed spectra increases at fainter magnitude .
however , the fractional increase of failed spectra at fainter magnitudes is not strong , reaching @xmath43% at the limit @xmath9 .
this effect is taken into account in our statistical analysis of the survey ( see for example paper ii ) .
the esp data provides a large data - base of galaxy spectra which , when fully analyzed , will constitute a reference for the next generation of larger and deeper redshift surveys that will probe galaxy evolution ( @xmath44 ) .
these data are well suited for the study of the intrinsic properties of galaxies in terms of their stellar populations , and their relation with the environment and redshift ( cosmic time ) .
this will be presented in a future paper .
emission lines are present in a large fraction ( @xmath45 ) of the galaxies in our sample .
we mainly detect [ oii ] @xmath4 , @xmath46 , [ oiii ] @xmath47 and @xmath48 .
our preliminary analysis focused on the 2 doublet , since this line is the most useful star formation tracer ( kennicutt 1992 ) in the wavelength range covered by our spectra .
galaxies showing the 2 line correspond , at increasing equivalent width , to three main categories : spiral galaxies , where the line originates mostly from hii regions in the disks , galaxies undergoing a significant burst of star formation , or agns .
great care is required when dealing with statistical analysis of emission lines properties of galaxies , because the detectability of lines in a spectrum strongly depends on the signal to noise ratio @xmath49 of the adjacent continuum .
the @xmath49 ratio in the blue part of our spectra ranges from @xmath50 2 to @xmath50 10 , with an average value of the order of 4 .
with such a value the 2 line can be detected only if its equivalent width is larger than about 5 . for the spectra of poorer quality , however , the minimum detectable 2 equivalent width is of the order of 20 ( see , for example , figure 7 in vettolani et al .
for this reason , in order to properly analyze the distribution of the equivalent widths below 20 and to study the possible correlations of the equivalent width with other intrinsic properties , one should also take into account the information carried by the upper limits . in this paper , in order to avoid dealing with upper limits , we present a preliminary analysis of the galaxies with an 2 equivalent width greater than 20 . up to now the measurement of the equivalent width of the 2 line , or of an upper limit for the galaxies in which the line is not detected , has been done for a subsample of 2550 galaxies , corresponding to 76% of the total sample . from this subsample
we find that about 13% of the galaxies show an 2 equivalent width greater than 20 and can therefore be classified as actively star forming galaxies .
this fraction , however , is a strong function of absolute magnitude , ranging from a few percent for the brightest galaxies up to about 40% for the galaxies fainter than @xmath1 ( see figure 3 ) .
this result is qualitatively similar to what has been found in the autofib redshift survey ( see figure 8 in ellis et al .
obviously , this correlation has to be taken into account in the determination of the evolution of the star formation rate as estimated by comparing bright and local samples of galaxies with fainter and more distant ones .
one of the main goals of our survey is to study the properties of the large scale structure in the local universe .
our survey is significantly deeper than local surveys ( e.g. cfa2 & ssrs2 ) since @xmath51 galaxies are sampled out to @xmath12 , corresponding to a comoving distance @xmath52 ( @xmath53 ) , which is the effective depth of the sample .
this depth , together with a wide extension in the right ascension direction ( @xmath14 corresponding to @xmath54 at the effective depth ) , could be enough to ensure the characteristic of a `` fair sample '' to esp .
clearly , the discovery of structures larger than those observed in the existing shallower surveys may prevent even esp from being a fair sample . in particular
, esp is deep enough to detect at least the first peaks of the beks survey and it is sufficiently wide and thick ( @xmath55 at @xmath56 ) to clarify the nature of the peaks that could correspond either to isolated groups / clusters or to the intersection of the beam with a larger connected structure like the great wall . detailed statistical results , including an analysis of the groups and clusters in the survey , will be presented in forthcoming papers ; here we describe the main characteristics of the volume sampled by the esp .
figure 4 shows the histogram of the distribution in comoving distance of the 3342 galaxies with measured redshift ( panel a ) and the corresponding wedge diagram ( panel b ) . for comparison
we show in panel c ) the wedge diagram corresponding to the esp right ascension range from the nearest strip of the lcrs centered at @xmath57 ( shectman et al .
1996 ) .
the most outstanding feature in figure 4a is the peak at @xmath58 .
although about 50 redshifts in this peak are due to galaxies within 1 abell radius from the centers of two abell clusters ( aco 2840 and aco 2860 ) both located near the eastern edge of strip a , this peak remains highly significant even if these galaxies are removed from the sample .
in fact , it is clear from figure 4b that the peak in the redshift distribution corresponds to an ensemble of large
scale structures nearly perpendicular to the line of sight .
if this ensemble can be considered as a single , connected structure , it would have a linear dimension of about 120 , before entering into the gap between regions a and b. at least part of this structure is visible in the lcrs data ( panel c ) where it seems to extend towards west and possibly connecting to other structures .
other foreground / background peaks in the redshift histogram correspond to structures less evident to the eye in the wedge diagram .
close examination of the wedge diagram shows the existence of many more dense regions and structures , some of which are elongated along the line of sight such as the one in region b stretching from about 130 to 200 .
the lcrs data in panel c ) allow to follow some of these structures , even if the different sampling , color selection and depth of the samples do not allow a quantitative comparison . as well as all the other redshift surveys ( see for instance panel c ) ,
our data suggest the presence of large underdense regions .
figure 4a clearly shows one such region at @xmath59 and a second one corresponding to a nearby ( @xmath60 ) underdensity . on the basis of our data and the comparison with the lcrs data ,
we conclude that this nearby underdense region is statistically significative ( see figure 4a ) ; however , given the small solid angle covered by esp and lcrs in this region , it is impossible to assess its extension in the transverse direction .
this underdensity has interesting consequences for the interpretation of the bright galaxy counts as we discuss in paper ii .
the vertical lines in figure 4a show the location of the regularly spaced density enhancements found in the beks pencil beam survey , located at @xmath61 , @xmath62 , i.e. @xmath63 north of the eastern part of the esp .
the two main peaks in our redshift distribution ( at comoving distances of @xmath64 and @xmath65 ) are in reasonably good agreement with the beks peaks .
the same two peaks are clearly visible in the lcrs data extracted from their slice immediately north ( @xmath66 ) of our strip . moreover , the peak at @xmath65 appears to be present in the deep pencil beam of bellanger and de lapparent ( 1995 ) , centered at @xmath67 and @xmath68 , i.e. @xmath69 away from the beks direction .
it is also visible in the shallower optopus probes obtained in this region of the sky by ettori et al .
( 1995 , see their figure 2 ) .
the coincidence in depth of the two nearest peaks of the redshift distribution within esp with similar peaks in all other surveys in the same region suggests the presence of large extended structures ( walls ) , approximately orthogonal to the line of sight . under this hypothesis
, the structure at @xmath70 would have minimum linear dimensions of the order of @xmath3 , comparable with the size of the great wall ( geller and huchra 1989 ) .
the existence of these two peaks does not imply , however , a periodicity with a preferred scale as claimed by broadhurst et al .
( 1990 ) , which can be tested only with much deeper samples .
figure 5 shows the density of the _ planar projection _ of the data .
isodensity contour levels are spaced as @xmath71 , with @xmath72 : they represent the ratio @xmath73 of the esp projected number density @xmath74 , gaussianly smoothed over @xmath75 , to the projected number density @xmath76 of a similarly smoothed random field , derived from an average of 100 @xmath77 random samples .
each galaxy was given a weight inversely proportional to the selection function ( as derived from the esp luminosity function , see paper ii ) , and to the incompleteness of its field .
this procedure implicitly assumes that the galaxies we did not observe follow the same distribution in depth as the observed galaxies ; this assumption is justified by our random selection of targets in overdense fields .
the weighting with the selection function allows to detect structures even at large distances , which would be otherwise `` washed out '' ; however , as any magnitude limited sample , such structures are inferred from the few high luminosity galaxies that can still be detected , and are therefore affected by a large uncertainty . the random samples were built using the esp average luminosity function and k corrections described in paper ii , and within a three dimensional space which faithfully takes into account the sample geometry , i.e. not only the unobserved region , but also the presence of a few holes " in the original catalogue , due to the presence of bright stars .
the final result is a representation of the large scale galaxy distribution which is more `` objective '' than the cone diagram . in figure 5
the dense regions corresponding to the peaks in figure 4a are clearly identifiable , in particular the two structures crossing the entire region from one side to the other at comoving distances of of @xmath64 and @xmath65 .
note also that the most prominent structure at large comoving distance in the small strip b appears to be elongated along the line of sight .
we have presented the main characteristics of the eso slice project , a redshift survey limited to @xmath9 in a region near the south galactic pole .
this survey , consisting of 3342 galaxies with reliable redshift determination , provides a large data - base of galaxy spectra which will be useful to study the spectral properties of galaxies and will constitute a reference for the next generation of larger and deeper redshift surveys .
preliminary analysis of a large sub sample of 2550 galaxies shows a strong anti
correlation between the fraction of galaxies with an 2 equivalent width greater than 20 and the absolute magnitude .
this fraction of actively star forming galaxies increases from a few percent for the brightest galaxies up to about 40% for the galaxies fainter than @xmath1 .
finally , this survey allows the study of the large scale structure at intermediate depth ( @xmath78 ) .
the most outstanding feature in the redshift distribution is a very significant peak at @xmath58 .
similar peaks , at the same distance , are detected in other surveys in the same region of the sky .
we conclude that there is evidence for the presence of a large bidimensional structure perpendicular to the line of sight .
the minimum size of this structure would be of the order of @xmath3 , comparable with the size of the great wall .
this work has been partially supported through nato grant crg 920150 , eec contract erb
ct920033 , cnr contract 95.01099.ct02 and by institut national des sciences de lunivers and cosmology gdr . + it is a pleasure to thank the support we had from the eso staff both in la silla and in garching .
in particular , we are grateful to gerardo avila for his advice and his help in solving every instrumental problem we have been facing during this project .
+ finally , we thank the referee for his constructive comments . | the eso slice project ( esp ) is a galaxy redshift survey we have recently completed as an eso key
project .
the esp covers 23.3 square degrees in a region close to the south galactic pole .
the survey is nearly complete ( 85% ) to the limiting magnitude @xmath0 and consists of 3342 galaxies with reliable redshift determination . in this paper ,
the first in a series that will present the results of the esp survey , we describe the main characteristics of the survey and briefly discuss the properties of the galaxy sample . from a preliminary spectral analysis of a large sub sample of 2550 galaxies
we find that the fraction of actively star forming galaxies increases from a few percent for the brightest galaxies up to about 40% for the galaxies fainter than @xmath1 .
the most outstanding feature in the esp redshift distribution is a very significant peak at @xmath2 .
the detection of similar peaks , at the same distance , in other surveys in the same region of the sky , suggests the presence of a large bidimensional structure perpendicular to the line of sight .
the minimum size of this structure would be of the order of @xmath3 , comparable with the size of the great wall .
2[oii ] @xmath4 |
it is widely accepted that gamma - ray burst ( grb ) afterglows originate from the interaction of an ultrarelativistically expanding shell into the circumburst medium ( cbm ) .
differences exists on the detailed kinematics and dynamics of such a shell ( see e.g. refs . and refs . therein ) . due to the ultrarelativistic velocity of the expanding shell ( lorentz gamma factor @xmath0 ) , photons emitted at the same time in the laboratory frame ( i.e. the one in which the center of the expanding shell is at rest ) from the shell surface but at different angles from the line of sight
do not reach the observer at the same arrival time .
therefore , if we were able to resolve spatially the grb afterglows , we would not see the spherical surface of the shell .
we would see instead the projection on the celestial sphere of the equitemporal surface ( eqts ) , defined as the surface locus of points which are source of radiation reaching the observer at the same arrival time ( see e.g. refs .
and refs . therein ) .
the knowledge of the exact shape of the eqtss is crucial , since any theoretical model must perform an integration over the eqtss to compute any prediction for the observed quantities ( see e.g. refs . and refs .
therein ) .
one of the key problems is the determination of the angular size of the visible region of each eqts , as well as the distribution of the luminosity over such a visible region . in the current literature
it has been shown that in the latest afterglow phases the luminosity is maximum at the boundaries of the visible region and that the eqts must then appear as expanding luminous `` rings '' ( see e.g. refs . and refs .
therein ) . such an analysis is applied only in the latest afterglow phases to interpret data from radio observations @xcite or gravitational microlensing @xcite .
the shell dynamics is usually assumed to be fully adiabatic and to be described by a power - law @xmath1 , following the blandford - mckee self similar solution@xcite , where @xmath2 and @xmath3 are respectively the lorentz gamma factor and the radius of the expanding shell .
such a power - law behavior has been extrapolated backward from the latest phases of the afterglow all the way to the prompt emission phase . in refs .
there have been presented the analytic solutions of the equations of motion for grb afterglow , compared with the above mentioned approximate solutions , both in the fully radiative and adiabatic regimes , and the corresponding analytic expressions for the eqtss .
it has been shown that the approximate power - law regime can be asymptotically reached by the lorentz gamma factor only in the latest afterglow phases , when @xmath4 , and only if the initial lorentz gamma factor @xmath5 of the shell satisfies @xmath6 in the adiabatic case or @xmath7 in the radiative case .
therefore , in no way the approximate power - law solution can be used to describe the previous dynamical phases of the shell , which are the relevant ones for the prompt emission and for the early afterglow .
starting from these premises , in this paper we present the distribution of the extended afterglow luminosity over the visible region of a single eqtss within the `` fireshell '' model for grbs .
such a model uses the exact solutions of the fireshell equations of motion and assumes a fully radiative dynamics ( see refs .
therein for details ) .
we recall that within the fireshell model the peak of the extended afterglow encompasses the prompt emission .
we focus our analysis on the prompt emission and the early afterglow phases .
our approach is therefore complementary to the other ones in the current literature , which analyze only the latest afterglow phases , and it clearly leads to new results when applied to the prompt emission phase . for simplicity , we consider only the bolometric luminosity@xcite , since during the prompt phase this is a good approximation of the one observed e.g. by bat or gbm instruments@xcite .
the analysis is separately performed over different selected eqtss . the temporal evolution of the luminosity distribution over the eqtss visible region is presented . as a consequence of these results ,
we show the novel feature that at the beginning of the prompt emission the most luminous regions of the eqtss are the ones closest to the line of sight . on the contrary , in the late prompt emission and in the early afterglow phases the most luminous eqts regions are the ones closest to the boundary of the visible region .
this transition in the emitting region may lead to specific observational signatures , i.e. an anomalous spectral evolution , in the rising part or at the peak of the prompt emission .
we also present an analytic expression for the temporal evolution , measured in arrival time , of the apparent radius of the eqtss in the sky .
we finally discuss analogies and differences with other approaches in the current literature which assumes an adiabatic dynamics instead of a fully radiative one .
for the case of a spherically symmetric fireshell considered in this letter , the eqtss are surfaces of revolution about the line of sight . the general expression for their profile , in the form @xmath8
, corresponding to an arrival time @xmath9 of the photons at the detector , can be obtained from ( see e.g. ref . ):
@xmath10 where @xmath11 is the initial size of the expanding fireshell , @xmath12 is the angle between the radial expansion velocity of a point on its surface and the line of sight , @xmath13 is its equation of motion , expressed in the laboratory frame , and @xmath14 is the speed of light . in the case of a fully radiative regime , the dynamics of the system
is given by the following solution of the equations of motion ( see e.g. refs . and refs .
therein ) : @xmath15}{\gamma_\circ^{-1}+\left(m_\mathrm{cbm}/m_b\right)\left(1+\gamma_\circ^{-1}\right)\left[1+\left(1/2\right)\left(m_\mathrm{cbm}/m_b\right)\right]}\ , , \label{gamma_rad}\ ] ] where @xmath2 is the lorentz gamma factor of the fireshell , @xmath16 is the amount of cbm mass swept up within the radius @xmath3 and @xmath5 and @xmath17 are respectively the values of the lorentz gamma factor and of the mass of the fireshell at the beginning of the extended afterglow phase .
correspondingly , the exact analytic expression for @xmath18 is@xcite : @xmath19 \\[6pt ] & + \tfrac{{r_\circ \sqrt c } } { { 12cm_i^\circ a^2 } } \ln \left\ { { \tfrac{{\left [ { a + \left(r / r_\circ\right ) } \right]^3 \left(a^3 + 1\right)}}{{\left[a^3 + \left ( r / r_\circ \right)^3\right ] \left ( { a + 1 } \right)^3 } } } \right\ } + t_\circ + \tfrac{m_i^\circ r_\circ } { 8c\sqrt c } \left(\tfrac{r^4-r_\circ^4}{r_\circ^4}\right)\ , , \end{split } \label{analsol}\ ] ] where @xmath20{(m_b - m_i^\circ)/m_i^\circ}$ ] , @xmath21 , @xmath22 is the value of the time @xmath23 at the beginning of the extended afterglow phase and @xmath24 . inserting eq.([analsol ] ) into eq.([ta_g ] ) we have the analytic expression for the eqts in the fully radiative regime@xcite : @xmath25 \\[6pt ] & + \tfrac{m_i^\circ r_\circ } { 8r\sqrt{c}}\left(\tfrac{r^4-r_\circ^4}{r_\circ^4}\right)+\tfrac{{r_\circ \sqrt{c } } } { { 12rm_i^\circ a^2 } } \ln \left\ { { \tfrac{{\left [ { a + \left(r / r_\circ\right ) } \right]^3 \left(a^3 + 1\right)}}{{\left[a^3 + \left ( r / r_\circ \right)^3\right ] \left ( { a + 1 } \right)^3 } } } \right\ } + \tfrac{ct_\circ - ct_a + r^\star}{r}. \end{split } \label{eqts_g_dopo}\ ] ] instead , the corresponding equations in the adiabatic regime are@xcite : @xmath26 @xmath27 @xmath28 ) , and in the adiabatic regime ( blue line ) , given by eq.([gamma_ad ] ) . ] a comparison between the lorentz gamma factors in the two regimes is presented in fig .
[ gamma_rad_ad ] . here and
in the following we assume the same initial conditions as in ref . , namely @xmath29 , @xmath30 cm , @xmath31 s , @xmath32 cm , @xmath33 particles/@xmath34 , @xmath35 g. for simplicity , and since we are interested in the overall behavior of the luminosity distribution , we assume a constant cbm density , neglecting the inhomogeneities which are responsible of the temporal variability of the prompt emission @xcite .
+ + + within the fireshell model , the grb extended afterglow bolometric luminosity in an arrival time @xmath36 and per unit solid angle @xmath37 is given by ( details in refs . and refs .
therein ) : @xmath38 where @xmath39 is the energy density released in the interaction of the abm pulse with the cbm measured in the comoving frame , @xmath40 is the doppler factor , @xmath41 is the surface element of the eqts at arrival time @xmath9 on which the integration is performed , and it has been assumed the fully radiative condition .
we are here not considering the cosmological redshift of the source , which is constant during the grb explosion and therefore it can not affect the results of the present analysis .
we recall that in our case such a bolometric luminosity is a good approximation of the one observed in the prompt emission and in the early afterglow by e.g. the bat or gbm instruments@xcite .
we are now going to show how this luminosity is distributed over the eqtss , i.e. we are going to plot over selected eqtss the luminosity density @xmath42 .
the results are represented in fig .
[ eqts_multi ] .
we chose eight different eqtss , corresponding to arrival time values ranging from the prompt emission ( @xmath43 seconds ) to the early ( @xmath44 hour ) afterglow phases .
for each eqts we represent also the boundaries of the visible region due to relativistic collimation , defined by the condition ( see e.g. refs . and refs .
therein ) : @xmath45 we obtain that , at the beginning ( @xmath46 seconds ) , when @xmath2 is approximately constant , the most luminous regions of the eqts are the ones along the line of sight .
however , as @xmath2 starts to drop ( @xmath47 seconds ) , the most luminous regions of the eqtss become the ones closest to the boundary of the visible region .
this transition in the emitting region may lead to specific observational signatures , i.e. an anomalous spectral evolution , in the rising part or at the peak of the prompt emission .
from sec . [ bld ] we obtain that within the fireshell model the eqtss of grb extended afterglows should appear in the sky as point - like sources at the beginning of the prompt emission but evolving after a few seconds into expanding luminous `` rings '' , with an apparent radius evolving in time and always equal to the maximum transverse eqts visible radius @xmath48 which can be obtained from eqs.([ta_g],[vis ] ) : @xmath49 where @xmath13 is given by eq.([analsol ] ) . with a small algebra
we get : @xmath50 where @xmath51 is given by eq.([gamma_rad ] ) and @xmath52 is given by eq.([analsol ] ) , since we assumed the fully radiative condition .
eq.([rbotfin ] ) defines parametrically the evolution of @xmath53 , i.e. the evolution of the maximum transverse eqts visible radius as a function of the arrival time . in fig .
[ eqts_multi ] we saw that such @xmath48 coincides with the actual value of the eqts apparent radius in the sky only for @xmath47 s , since for @xmath54 s the most luminous eqts regions are the ones closest to the line of sight ( see the first three plots in fig .
[ eqts_multi ] ) .
therefore , in fig . [ rmaxvsta_rad ]
we plot @xmath48 given by eq.([rbotfin ] ) in the fully radiative regime together with the actual values of the eqts apparent radius in the sky taken from fig .
[ eqts_multi ] in the three cases in which they are different .
it is clear that , during the early phases of the prompt emission , even the `` exact solution '' given by eq.([rbotfin ] ) can be considered only an upper limit to the actual eqts apparent radius in the sky . in the fully radiative regime .
the red line represents the maximum transverse eqts visible radius @xmath48 following eq.([rbotfin ] ) .
the three blue points represents the actual eqts apparent radius in the sky in the three cases @xmath55 s , @xmath56 s and @xmath57 s in which , as shown in fig .
[ eqts_multi ] , it is smaller than @xmath48 . the blue line is a linear interpolation of such points . ] in the current literature ( see e.g. refs . )
there are no analogous treatments , since it is always assumed an adiabatic dynamics instead of a fully radiative one and only the latest afterglow phases are addressed .
it is usually assumed the blandford - mckee self similar solution@xcite for the adiabatic dynamics @xmath1 .
a critical analysis of the applicability to grbs of this approximate dynamics , instead of the exact solutions in eqs.([gamma_ad],[analsol_ad],[eqts_g_dopo_ad ] ) , has been presented in ref .
, as recalled in the introduction .
the most widely applied formula for the eqts apparent radius in the above mentioned current literature is the one proposed by sari@xcite : @xmath58^{5/8 } \
, \mathrm{cm } , \label{gps99}\ ] ] where @xmath59 is the initial energy of the shell in units of @xmath60 ergs , @xmath61 is the cbm density in units of @xmath44 particle/@xmath34 , @xmath62 is the arrival time at the detector of the radiation measured in days , and @xmath63 is the source cosmological redshift .
waxman et al.@xcite however derived a numerical factor of @xmath64 instead of @xmath65 .
eq.([gps99 ] ) can not be compared directly with eq.([rbotfin ] ) , since they assume two different dynamical regimes .
therefore , we first plot in fig .
[ rmaxvsta_rad_ad ] the exact solution for @xmath48 given by eq.([rbotfin ] ) both in the fully radiative case , using eqs.([gamma_rad],[analsol ] ) , and in the adiabatic one , using eqs.([gamma_ad],[analsol_ad ] ) .
we see that they almost coincide during the prompt emission while they start to diverge in the early afterglow phases , following the behavior of the corresponding lorentz gamma factors ( see fig .
[ gamma_rad_ad ] and details in ref . ) . computed with eq.([rbotfin ] ) in the adiabatic ( blue line ) and in the fully radiative ( red line ) cases respectively . ]
now , in fig .
[ rmaxvsta_ad ] and for the same initial conditions assumed in previous figures , we plot the maximum transverse eqts visible radius @xmath48 given by eq.([rbotfin ] ) in the adiabatic case , using eqs.([gamma_ad],[analsol_ad ] ) , together with eq.([gps99 ] ) , proposed in ref . , and with the corresponding modification in the numerical factor proposed in ref .
from such a comparison , we can see that the approximate regime @xmath66 overestimates the exact solution for @xmath48 during the prompt emission and the early afterglow phases .
it is asymptotically reached only in the latest afterglow phases ( @xmath67 s ) , in the very small region in which the approximate power - law dynamics starts to be applicable ( see details in ref . ) .
however , there is still a small discrepancy in the normalization : the constant numerical factor in front of eq.([gps99 ] ) should be @xmath68 instead of @xmath69 or @xmath70 to reproduce the behavior of the exact solution for large @xmath9 .
moreover , we must emphasize that , in analogy with what we obtained in the fully radiative case ( see fig .
[ rmaxvsta_rad ] ) , in the early phases of the prompt emission , when the fireshell lorentz @xmath2 factor is almost constant , the maximum transverse eqts visible radius @xmath48 given by eq.([rbotfin ] ) can only be considered an upper limit to the actual value of the eqts apparent radius in the sky . in such phases
the approximation implied by eq.([gps99 ] ) can therefore be even worse that what represented in fig .
[ rmaxvsta_ad ] . in the adiabatic regime .
the red line represents the maximum transverse eqts visible radius @xmath48 following eq.([rbotfin ] ) .
the green line represents eq.([gps99 ] ) proposed by sari .
the blue line represents the corresponding modification in the numerical factor proposed by waxman et al .
these last two lines are almost coincident on the scale of this plot . ]
within the fireshell model , using the exact analytic expressions for the fireshell equations of motion and for the corresponding eqtss in the fully radiative condition , we analyzed the temporal evolution of the distribution of the extended afterglow luminosity over the eqts during the prompt emission and the early afterglow phases .
we find that , at the beginning of the prompt emission ( @xmath46 seconds ) , when @xmath2 is approximately constant , the most luminous regions of the eqts are the ones along the line of sight . as @xmath2 starts to drop ( @xmath47 seconds ) , the most luminous regions of the eqtss become the ones closest to the boundary of the visible region .
this transition in the emitting region may lead to specific observational signatures , i.e. an anomalous spectral evolution , in the rising part or at the peak of the prompt emission .
the eqtss of grb extended afterglows should therefore appear in the sky as point - like sources at the beginning of the prompt emission but evolving after a few seconds into expanding luminous `` rings '' , with an apparent radius evolving in time and always equal to the maximum transverse eqts visible radius @xmath48 .
we also derive an exact analytic expression for @xmath48 , both in the fully radiative and in the adiabatic conditions , and we compared it with the approximate formulas commonly used in the current literature in the adiabatic case .
we found that these last ones can not be applied in the prompt emission nor in the early afterglow phases .
even when the asymptotic regime is reached ( @xmath67 s ) , it is necessary a correction to the numerical factor in front of the expression given in eq.([gps99 ] ) which should be @xmath68 instead of @xmath69 or @xmath70 .
r. ruffini , a. g. aksenov , m. g. bernardini , c. l. bianco , l. caito , p. chardonnet , m. g. dainotti , g. de barros , r. guida , l. izzo , b. patricelli , l. j. r. lemos , m. rotondo , j. a. r. hernandez , g. vereshchagin and s. xue , the blackholic energy and the canonical gamma - ray burst iv : the `` long , '' `` genuine short '' and `` fake - disguised short '' grbs , in _
xiii brazilian school on cosmology and gravitation _
, eds . m. novello and s. perez bergliaffa , american institute of physics conference series , vol .
1132 ( 2009 ) , pp . 199266 . | due to the ultrarelativistic velocity of the expanding `` fireshell '' ( lorentz gamma factor @xmath0 ) , photons emitted at the same time from the fireshell surface do not reach the observer at the same arrival time . in interpreting gamma - ray bursts ( grbs ) it is crucial to determine the properties of the equitemporal surfaces ( eqtss ) : the locus of points which are source of radiation reaching the observer at the same arrival time . in the current literature
this analysis is performed only in the latest phases of the afterglow . here
we study the distribution of the grb bolometric luminosity over the eqtss , with special attention to the prompt emission phase .
we analyze as well the temporal evolution of the eqts apparent size in the sky .
we use the analytic solutions of the equations of motion of the fireshell and the corresponding analytic expressions of the eqtss which have been presented in recent works and which are valid for both the fully radiative and the adiabatic dynamics .
we find the novel result that at the beginning of the prompt emission the most luminous regions of the eqtss are the ones closest to the line of sight . on the contrary , in the late prompt emission and in the early afterglow phases the most luminous eqts regions are the ones closest to the boundary of the visible region .
this transition in the emitting region may lead to specific observational signatures , i.e. an anomalous spectral evolution , in the rising part or at the peak of the prompt emission .
we find as well an expression for the apparent radius of the eqts in the sky , valid in both the fully radiative and the adiabatic regimes .
such considerations are essential for the theoretical interpretation of the prompt emission phase of grbs . |
it has been known that nearby elliptical galaxies have color gradients ; colors in an elliptical galaxy gradually become bluer with increasing radius ( e.g. , vader et al .
1988 ; franx , illingworth , & heckman 1989 ; peletier et al . 1990a ; peletier , valentijn , & jameson 1990b , goudfrooij et al . 1994 ; michard 1999 ) .
since many of elliptical galaxies show radial gradients in metal absorption line strengths such as mg@xmath2 , fe@xmath3(5270 ) and fe@xmath2(5335 ) ( e.g. , carollo , danziger , & buson 1993 ; davies , sadler , & peletier 1993 ; gonzalez 1993 ; kobayashi & arimoto 1999 ) , the origin of the color gradients has been naively interpreted to be the metallicity gradients . however , such an interpretation for the origin of the color gradient is premature , because both metallicity gradient and age gradient in stellar population can cause the same color gradient , and we can not distinguish the cause for the gradient .
this is called _ age - metallicity degeneracy _ , which was originally pointed out by worthey , trager , & faber ( 1996 ) in terms of the origin of the color - magnitude relation of nearby elliptical galaxies ( see also arimoto 1996 ) . in order to break this degeneracy and to know the primary origin of the color gradients in elliptical galaxies ,
comparing the observed color gradients in distant ellipticals with predicted model gradients caused by either the metallicity gradient or the age gradient is a very effective approach , as was successful for examining the origin of the color - magnitude ( cm ) relation ( kodama & arimoto , 1997 ) .
tamura et al . ( 2000 ; hereafter called paper i ) constructed the two models both of which reproduce a typical color gradient of elliptical galaxies at @xmath4 using a population synthesis model .
in one model , the mean metallicity of the stellar population decreases with increasing radius at a fixed old mean age . while in the other one , the mean age decreases with a radius at a fixed mean metallicity . these models
were then made evolve back in time .
the evolution of color gradients thus predicted are confronted with the observed ones in distant ( @xmath5 ) ellipticals sampled from the _ hubble deep field - north _
( hdf - n ; williams et al . 1996 ) .
as a result , paper i found that the metallicity gradient is the primary origin of color gradients and the age gradient model can not reproduce the observed gradient at such redshift .
the elliptical galaxies in the hdf - n , however , are only those in field environment .
it has never been obvious that ellipticals in clusters evolve similarly as those in field . in rich clusters
, it has been found that the color - magnitude relation still holds even at around @xmath6 ( e.g. , stanford , eisenhardt , & dickinson 1998 ) and these observational results seem to favor the classical monolithic collapse scenario associated with the galactic wind and high-@xmath7 formation ( e.g. , @xmath8 ) of elliptical galaxies ( e.g. , kodama et al .
however , this kind of evolution has not been established for ellipticals in lower density environment ( but see kodama , bower , & bell 1998 ) .
some predictions either theoretically or observationally show that field ellipticals formed by recent ( at @xmath9 ) merging processes ( e.g. , baugh , cole , & frenk 1996 ; barger et al .
an internal structure of a galaxy such as a metallicity gradient and an age gradient must depend on its formation process .
if cluster ellipticals pass different formation histories from those for field ellipticals , their internal structures , thus the origin of the color gradients , may not be the same . or some environmental effects on color gradients may exist .
thus , the same approach is needed for cluster ellipticals to clarify the origin of their color gradients .
it is noted that dust extinction in elliptical galaxies may also have some effects on the color gradients ( goudfrooij & de jong 1995 ; wise & silva 1996 ; silva & wise 1996 ) .
however , about half of the detection towards ellipticals in far infrared with iras are around @xmath10 threshold and confirmation is needed to be definitive ( bregman et al . 1998 ) .
in addition , spatial distribution of dust in an elliptical galaxy as well as dust mass which could affect a color gradient are not established yet .
these are still open problems and will be examined in detail in our forthcoming papers .
therefore , in this paper , we have chosen to focus on age and metallicity effects only .
this paper is organized as follows .
the sample selection and data analysis of elliptical galaxies are described in 2 .
histograms of color gradients are presented in 3 together with the representative color profiles of the sample ellipticals .
discussion is given in
the cosmological parameters adopted throughout this paper are the same as those in paper i ; @xmath11 km s@xmath12 mpc@xmath12 , @xmath13 and @xmath14 .
in order to examine color gradients in elliptical galaxies in distant clusters , deep imaging data with a high angular resolution are necessary in more than two bands .
thus , we choose to use the archival data taken with the wfpc2 on - board the hst .
smail et al .
( 1997 ) obtained deep imaging data of 11 distant rich clusters at the redshift from 0.37 to 0.56 , most of which have exposure times more than 10,000 sec to examine their morphology in detail , and presented catalogs of the detected objects . in this paper ,
their reduced imaging data which are available at their web site are used . from these data ,
we select six clusters whose images were taken in two bands to obtain galaxy colors .
cluster ids and their basic properties taken from smail et al .
( 1997 ) are given in table 1 . among these clusters ,
cl 0024@xmath1516 , cl 0016@xmath1516 , and cl 0054@xmath1627 are classified as high concentration clusters and cl 0939@xmath1547 and cl 0412@xmath1665 are low concentration clusters ( dressler et al .
1997 ) . for a370 and cl 0939@xmath1547 , the images of their outer fields were taken .
therefore , environmental effects on color gradients may be examined .
early type galaxies ( e , e / s0 , s0/e ) in these clusters are sampled based on the catalog by smail et al .
( 1997 ) . since our main interest is to examine their color gradients , high signal - to - noise ratio ( s / n ) is required and thus galaxies brighter than the apparent magnitude of 21 mag in @xmath17 band are selected in all of the sample clusters .
this apparent magnitude roughly corresponds to the absolute magnitude of @xmath18 mag in @xmath19 band .
our sample galaxies are listed in table 2 with their basic parameters , most of which are taken from smail et al .
( 1997 ) . in each cluster
, the sample galaxies form the tight cm relations as shown in figure 1 , though the relations in a370 , cl 0939@xmath1547 , and cl 0412@xmath1665 are somewhat loose .
since cl 0939@xmath1547 and cl 0412@xmath1665 are classified as low concentration clusters , and the outer regions of the clusters were imaged for a370 and cl 0939@xmath1547 , some environmental effects on the cm relations might be seen . however , a discussion for this subject is beyond our scope in this paper . in the figure ,
a solid square with an open circle shows the object removed from the following analysis , because their color gradients could not be derived accurately enough to use our discussion due to their low signal - to - noise ratios ( s / ns ) or due to their close positions to an edge of the image .
in deriving color profiles of the sample galaxies , we should take care of a positional coincidence between a blue image and a red image , and a difference of the sizes of point spread functions ( psfs ) between the two images .
( one pixel corresponds to @xmath20 for the wide field cameras and @xmath21 for the planetary camera . )
positional displacement between a blue image and a red one causes spurious asymmetry of a color distribution in a galaxy and has a serious effect on the color gradient .
we examined the centroids of the sample galaxies in both blue and red images and corrected a displacement , if a systematic shift between the two images in each cluster was found .
the sizes of psfs should be also estimated and be adjusted between blue and red images .
we simulated the psf at each position of a sample galaxy in each band by using tinytim v4.4 ( krist 1995 ; krist & hook 1997 ) , and corrected the difference of the sizes between the two images , which were estimated by fitting a gaussian to the psfs .
sky value was determined by `` mode '' in an annulus with an inner radius of @xmath22 and a width of 3@xmath23 by using the `` phot '' task in the iraf apphot package and the obtained sky was subtracted .
we next made azimuthally averaged radial surface brightness profiles in both blue and red images with a radial sampling of 0.@xmath231 .
these profiles in each galaxy are made along the ellipses fitted with a fixed position angle to the isophotes of the galaxy images in the red band .
( the position angles are taken from table 2a by smail et al .
( 1997 ) . )
thus the surface brightness profiles in the two bands are derived based on the same fitted elliptical isophotes . in figure 2 ,
representative surface brightness profiles of our sample galaxies are shown .
the profiles of the brightest , intermediate , and the faintest sample galaxies in each cluster are shown in top , middle , and bottom panels , respectively .
the profiles of almost all the galaxies in our sample are well represented by the @xmath24 law .
to derive color profiles , the surface brightness profile in the red band is subtracted from the profile in the blue band .
figure 3 shows the resulting color profiles with the same arrangement as in figure 2 .
an error bar to each data point includes a photometric error , a local sky subtraction error ( 1@xmath25 of a sky value around each object is adopted ) , and a dispersion of colors along each elliptical isophote .
it is important to describe here the two dimensional color distribution in the sample galaxies .
the color maps were constructed by dividing the blue image by the red image after adjusting the positional displacement and difference of the sizes of the psfs .
almost all the color maps show smooth color distribution and do not show any asymmetric or peculiar features .
thus the color profiles well represent the color distribution in each galaxy .
finally , slopes of the color profiles , i.e. , color gradients , in the sample galaxies are derived by the least square fitting within effective radii , which are obtained from the surface brightness profiles in the red band assuming the @xmath24 law .
the fitting for the color gradients are done after removing data points with errors larger than 0.3 mag .
we do not derive the gradients of the profiles with accepted data points fewer than 3 .
the average number of the data points used for the fitting is about 8 and the rejected objects are minor .
resulting color gradients in the sample galaxies are listed in table 2 with the 1@xmath26 fitting errors .
figure 3 indicates resulting slopes of the color gradients as well as the slopes with @xmath27 . in this figure
, abscissa refers to logarithm of a radius normalized by the outermost radius ( @xmath28 ) used for the fitting in each color profile . for most of the sample galaxies , @xmath28 is roughly equal to @xmath29 .
figure 4 shows a histogram for the color gradients of the sample galaxies in each cluster .
each bin of the histograms is set to be 0.2 mag / dex which is comparable to the average value of the fitting error .
it is found that the distributions of the gradients are very narrow except for a few outliers , which are # 535 and # 738 in cl 0024@xmath1516 , # 2005 in cl 0939@xmath1547 , and # 2050 in cl 0016@xmath1516 .
the former three are significantly out of the cm relation towards the blue side , but the last one is almost on the relation . considering that the range of the distribution of the color gradients is comparable with or only slightly larger than the estimated error for the slopes , the intrinsic dispersions of the color gradients must be considerably small .
( the dispersion of the color gradients in nearby elliptical galaxies is about 0.04 mag / dex ( peletier et al . 1990a ; paper i ) . )
it is intriguing that the color gradients of elliptical galaxies are uniform even at intermediate redshift .
furthermore , this encourages the comparison between model gradients and observed gradients in distant clusters , despite rather large errors for the observed slopes .
in order to examine whether the origin of the color gradient is the stellar metallicity or the age , we adopt the same approach as that in paper i and the reader should refer to it in detail .
we briefly summarize it here .
an observed color gradient can be reproduced by either a metallicity gradient or an age gradient of stellar populations in an elliptical galaxy at @xmath4 . however , since the color gradient caused by a metallicity gradient is expected to follow a different evolution from that by an age gradient , the origin of the color gradients can be found by comparing the observed ones at high redshift with those predicted by model .
for this purpose , using the population synthesis model ( kodama & arimoto 1997 ; kobayashi , tsujimoto , & nomoto 2000 ) , we construct the two model galaxies ; one model galaxy has the color gradient made by the pure metallicity gradient ( thereafter called _ metallicity gradient model _ ) without age gradient , and the other made by the pure age gradient ( _ age gradient model _ ) without metallicity gradient . in the metallicity gradient model ,
the metallicity gradient is produced by assuming that a galactic wind blowed later in the inner region in an elliptical galaxy ; star formation continued longer and thus the mean stellar metallicity became higher at the inner region . for the age gradient model ,
star formation started earlier in the inner region and thus the mean age of stellar populations is older than that in the outer region .
the stellar population in each region in an elliptical galaxy is assumed to be made by a single burst and to evolve independently of other regions .
model parameters used here are set to the same as those in paper i , which are chosen so as to reproduce the typical color gradient at @xmath4 .
the mean value of @xmath30 of @xmath31 mag@xmath32dex obtained by peletier et al .
( 1990a ) is adopted as the typical color gradient at @xmath4 .
note that these model galaxies must be old ( 8 @xmath1 15 gyr ) to reproduce colors in nearby elliptical galaxies .
then , we calculate the spectral evolution in each region of the model galaxies and their color gradients at any redshifts using the response functions including each filter on the hst . it should be emphasized that we do not intend to study physical formation process of elliptical galaxies in this paper , but aim at depicting the evolution of the color gradient caused by either metallicity gradient or age gradient to be compared with the observed ones .
actual physical process that made the metallicity / age gradient may be different from our brief ideas presented in the model description .
however it is not a problem here , because once such gradient formed , subsequent evolution of the stellar population is unique and does not depend on the formation process .
the two lines in each panel of figure 5 show the evolutionary track of the model color gradients ; the solid curve indicates the evolution for the metallicity gradient model and the dotted curve for the age gradient model .
the model color gradient by the metallicity gradient is almost constant with a redshift within @xmath6 , while that by the age gradient changes abruptly and shows a quite steep gradient even at @xmath33 .
we will compare the model gradients with the observed ones in the next subsection .
the mean values of the color gradients in each cluster sample are plotted at their redshifts in figure 5 .
an error bar attached to each point indicates a mean error of the gradients in each cluster .
as clearly shown , the metallicity gradient is much more favorable as the origin of the color gradients .
this result does not depend on cosmological parameters or parameters for an evolutionary model of galaxy within a reasonable range , and does not change even if we consider the dispersion of the color gradients in the sample galaxies ( see figure 4 ) and that in nearby ellipticals ( @xmath34 mag / dex ) . although the sample galaxies of which memberships in the clusters are spectroscopically confirmed are minor ( dressler et al . 1999 ) , background or foreground contaminations are not expected to affect the result for the origin of the color gradients , because the result does not change even if we remove the galaxies which significantly deviate from the cm relation in each cluster .
the color gradients in several sample galaxies may be affected by other galaxies close to them , and the color profile of a galaxy which locates close to an edge of the chip or on a joint between the cameras may be somewhat spurious .
however , our result still holds even after removing the galaxies which may suffer from these effects . considering the result in paper i , in both cluster and field , the primary origin of the color gradients in elliptical galaxies
is considered to be the stellar metallicity .
however , it is interesting to point out that the mean values of the color gradients seem to deviate upwards from the line for the metallicity gradient model .
our models are calibrated by the color gradients seen in nearby ellipticals by peletier et al .
( 1990a ) , in which most of the sample ellipticals reside in field or group environment .
therefore , the upward deviation might indicate an environmental effect on the color gradients of elliptical galaxies between in rich clusters and in field .
however , the correlation between the mean value and the degree of the concentration in each cluster is not seen .
in addition , the mean gradients of the clusters of which outer field images were taken do not show larger values than others .
further detailed study on the color gradients in cluster ellipticals and field ones at @xmath4 as well as at high redshift should be done in the future .
we would like to thank c. kobayashi , n. arimoto , and t. kodama for fruitful collaboration in paper i. this work was financially supported in part by grant - in - aid for the scientific research ( no .
11740123 ) by the ministry of education , science , sports and culture of japan .
figure . 1 color - magnitude ( cm ) diagrams for the sample galaxies in the clusters .
a solid square with an open circle indicates the object whose color gradient can not be obtained due to low s / n or the close position to an edge of the image .
`` # 2 '' in the cluster i d refers to the outer field of the cluster . figure .
2 azimuthally averaged radial surface brightness profiles of representative galaxies in the sample are presented .
the profiles of the brightest , intermediate , and the faintest sample galaxies in each cluster are shown in top , middle , and bottom panels , respectively .
solid squares are the profile in a blue band and open circles in a red band .
an object i d is shown at the upper left in each panel .
3 color profiles and fitted slopes of the representative galaxies in the sample are indicated .
the galaxies presented and the arrangement of the panels are the same as those in figure 2 . among the three solid lines in each panel ,
the middle one shows the best fit slope .
other two lines show the slopes with @xmath35 of the best fit .
abscissa refers to logarithm of a radius normalized by the outermost radius ( @xmath28 ) used for the fitting in each color profile . figure .
4 histograms of the color gradients in the sample galaxies in each sample cluster .
a total number of the objects in each histogram is shown at the upper left in each panel and a number in parentheses indicates a total number of the sample galaxies in each cluster .
cluster i d and its redshift are shown on the top of each panel . figure . 5 a mean color gradient in each cluster versus redshift is shown .
a solid curve in each panel represents the evolutionary track of the color gradient caused by the metallicity gradient and a dotted curve shows the track by the age gradient ( see text in detail ) . | color gradients in elliptical galaxies in distant clusters ( @xmath0 ) are examined by using the archival deep imaging data of wide field planetary camera 2 ( wfpc2 ) on - board the hubble space telescope ( hst ) .
obtained color gradients are compared with the two model gradients to examine the origin of the color gradients . in one model , a color gradient is assumed to be caused by a metallicity gradient of stellar populations , while in the other one , it is caused by an age gradient .
both of these model color gradients reproduce the average color gradient seen in nearby ellipticals , but predict significantly different gradients at a redshift larger than @xmath10.3 .
comparison between the observed gradients and the model gradients reveals that the metallicity gradient is much more favorable as the primary origin of color gradients in elliptical galaxies in clusters .
the same conclusion has been obtained for field ellipticals by using those at the redshift from 0.1 to 1.0 in the hubble deep field - north by tamura et al .
( 2000 ) .
thus , it is also suggested that the primary origin of the color gradients in elliptical galaxies does not depend on galaxy environment . |
euso - ta @xcite is a pathfinder detector of the jem - euso project @xcite .
the aim of jem - euso is to detect the rare ultra - high energy cosmic rays ( uhecrs ) from space , observing the uv fluorescence light emitted by cosmic ray extensive air showers ( eass ) through the atmosphere , which is used as an huge calorimeter .
+ the purpose of the euso - ta experiment is to validate some aspects of the observation principle of jem - euso detecting eass from ground .
indeed , it is located at the telescope array ( ta ) @xcite site in utah ( usa ) , in front of the black rock mesa fluorescence detectors ( brm - fds ) station @xcite , both visible in fig .
[ fig : euso - ta - site ] .
euso - ta has the same basic design used also in euso - balloon @xcite and euso - spb @xcite telescopes , other two jem - euso pathfinders dedicated to the validation of the detector design .
+ the optical system is made by two @xmath1 flat lenses , 8 mm thick each .
the design of the optical system is visible in fig .
[ fig : euso - ta - design ] left .
the front side of the front lens and the back side of the rear lens have a plane surface , which is useful for euso - ta in order to clean the dust easily , since it is placed in a desert area .
the back side of the front lens and the front side of the rear lens have a fresnel structure .
they are made of uv grade polymethyl - methacrylate ( pmma ) and details about their manufacture can be found in @xcite .
the optical system allows to euso - ta a field of view ( fov ) of .
+ the euso - ta focal surface has a concave shape .
it consists of one photo detector module ( pdm ) , which is a @xmath2 cm active surface composed by an array of @xmath3 hamamatsu r11265-m64 mapmts .
groups of @xmath4 mapmts form the elementary cells ( ecs ) .
the pdm is shown in fig .
[ fig : euso - ta - design ] right .
+ each mapmt is a matrix of @xmath5 pixels with 2.88 mm side for a total of 2304 pixels over the whole pdm .
an uv transmitting band pass filter ( schott bg3 ) in the range 290 nm-430 nm is glued on top .
each pixel has a gain ( electron multiplication ratio ) of more than @xmath6 which allows single photon counting .
the total photon detection efficiency has a maximum of 35% .
as all the fluorescence cosmic ray detectors , euso - ta works in night time and possibly in clear sky conditions , to reduce effects of atmospheric and cloud attenuation . the elevation of the instrument can be manually changed from to , whereas the azimuth is fixed to from north counterclockwise , being the telescope pointing to the clf direction .
+ one advantage of the euso - ta location is the possibility to use one of the systems developed by the ta collaboration for calibration purposes : the central laser facility ( clf ) @xcite .
the clf is a laser facility 21 km distant from brm - fds which shoots every half an hour 300 laser pulses vertically with a frequency of 10 hz at a known energy , with a typical energy of 4 mj that can be adjusted .
the light scattered from the beam is equivalent to that produced by an equally distant uhecr with energy of @xmath7 ev .
+ the second advantage is to have the possibility to acquire data in coincidence with brm - fds ( _ external trigger _ ) .
indeed , in case of a trigger signal received by brm - fds , euso - ta starts to save the data from the buffer in packets of 128 gate time units ( gtus ) , which otherwise would be overwritten ( one gtu corresponds to @xmath8 , and represents the time resolution of the data acquisition ) .
this is an important point since this acquisition mode allows to know that euso - ta could have detected an event and , from the shower reconstruction made by the ta experiment , the nature of the event ( i.e. direction , distance , energy etc . )
would be then known and used for further analysis . however , since the fov of ta is about 30 times the fov of euso - ta , the amount of data saved is larger than the actual data containing events , that makes the research of events more challenging .
another acquisition mode consists in using an _ internal trigger _ , also called _ first level trigger _ ( l1 ) @xcite , implemented for the jem - euso experiment .
the l1 principle consists in counting an excess of signals over background in groups of @xmath9 pixels lasting more than a preset time
. the internal trigger has been tested , but the current fpga allows the implementation of the trigger logic only for a few ecs .
for this reason , most of the data have been acquired with the external trigger .
the detected events have been analyzed and simulated , but as an example , only the first detected event is discussed in detail in this contribution .
it has been detected on the 13th may 2015 in external trigger mode and , since it was with about 2.5 km relatively close to the detector , it lasted 1 gtu .
a study on the variability of the background has been done to evaluate if the presence of cr event signal influences in an evident manner the mean of the pixels content of a whole frame . for this purpose ,
128 frames corresponding to the packet of gtus containing the detected event have been taken into account .
for each frame the mean value of the pixel contents has been calculated and the 128 mean values are plotted in fig .
[ fig : mean128_13may ] .
it shows the sequence of 128 mean values and , with red mark , the frame with signal from the cr event is highlighted ( frame number 83 in the graph ) .
it is evident that the contribute of the cr event is easily visible over the base line of the background .
unfortunately , from the same study on other detected events it has been found that for events spread on more than one gtu , it is impossible to find the frames with the cr event since the contribution of just a segment of the whole event track has the same magnitude as the background fluctuations .
two simulation frameworks are used in the jem - euso collaboration : euso - offline @xcite and esaf @xcite . in this work ,
results of simulations made with euso - offline are discussed .
improvements regarding the telescope and electronics simulators have been recently made in euso - offline , that made the simulations properly comparable with the data .
+ continuing the discussion on the 13th may 2015 event , we consider the comparison between the data and simulation . tab .
[ tab:13may_sim ] shows the event reconstruction of this event performed by ta experiment , together with the elevation angle that euso - ta had during the data acquisition .
most of these information are required as input in the simulation and are marked with a star in the table : zenith and azimuth ( counterclockwise from east ) angles of the shower , energy of the primary particle , coordinates of the shower core on ground ( in the utm coordinate system , zone `` 12 '' band `` s '' ) and the elevation of euso - ta .
* zenith & + * azimuth & + * energy & @xmath10 ev + @xmath11 ( impact param . )
& 2.5 km + * core coordinate & 349976 m east , 4340200 m north + distance core & 2.43 km + number of gtus & 1 + + * elevation & + + in fig .
[ fig : may13_dat_sum ] the comparison between the detected ( a ) and the simulated ( b ) event is shown , where both the frames show the number of counts on the full pdm .
the simulation itself has been summed to a frame of background extrapolated from the data , from the same packet of gtus where the event is saved .
the white blocks correspond to not working mapmts , where correspondingly also the simulation has been masked .
they look in good agreement from a visual point of view : the number of counts , the inclination and width of the tracks look similar . in ( c ) , the same event as detected by brm - fds within brm - fd s @xmath12 window is shown , with euso - ta s fov overlapped .
it is visible that the spatial resolutions are higher for euso - ta , although the fov is quite smaller .
a quantitative comparison is done considering the counts distribution of the data , the simulation with background and the background itself , as shown in fig .
[ fig : may13_diff ] .
the not working pmts are not considered in the analysis .
the distributions of data and simulation differ slightly .
the difference probably resides , at least in part , in the impossibility to reproduce the exact background frame existing at the moment when the frame with the event was taken .
but it is interesting to see how both the distributions start to show an excess respect to the background at @xmath13 .
in the year 2015 four data tacking campaigns have been done in coincidence with brm - fds , and one in 2016 , for a total of about 140 hours . up to now
four events are considered surely detected .
two of them are related to close showers , lasting 1 gtu , while the other two , at further distances , last 2 and 3 gtus . to give a reference , events with distance @xmath14
km should have an energy @xmath15 ev to be detected . +
a study on all the events detected by brm - fds and passing inside the euso - ta fov is ongoing .
first , each event is simulated and , if visible in the simulation , it is searched in the data , with the help the timestamp of the corresponding ta event .
a direct research of the events in euso - ta data just via the timestamp of the ta events is not efficient , because of a not constant time shift between the timestamps of the events of the two experiments and because of the different time resolutions .
currently , four more events found in the data are considered as candidate events , and further analysis is proceeding in order to ensure their classification .
the collected statistics is compatible with the expectations from preliminary analysis .
euso - ta is currently active and is successfully collecting data .
the offline simulation framework has been improved and it is able to simulate quite well the events .
further analysis on the not detected events are ongoing to search for not yet seen events in euso - ta data . the support received by ta collaboration is deeply acknowledged . + this work was partially supported by basic science interdisciplinary research projects of riken and jsps kakenhi grant ( 22340063 , 23340081 , and 24244042 ) , by the italian ministry of foreign affairs , general direction for the cultural promotion and cooperation , by the helmholtz alliance for astroparticle physics hap funded by the initiative and networking fund of the helmholtz association , germany , by nasa award 11-apra-0058 in the usa , and by slovak academy of sciences mvts jem - euso as well as vega grant agency project 2/0076/13 .
russia is supported by the russian foundation for basic research grant no 13 - 02 - 12175-ofi - m . the spanish consortium involved in the jem - euso space mission
is funded by micinn & mineco under the space program projects : aya2009 - 06037-e / aya , aya - esp2010 - 19082 , aya - esp2011 - 29489-c03 , aya - esp2012 - 39115-c03 , aya - esp2013 - 47816-c4 , mineco / feder - unah13 - 4e-2741 , csd2009 - 00064 ( consolider multidark ) and by comunidad de madrid ( cam ) under projects s2009/esp-1496 & s2013/ice-2822 .
j. h. adams et al .
, `` ground - based tests of jem - euso components at the telescope array site , euso - ta '' exp . astr . ,
40 , pp .
301 - 314 ( 2015 ) j.h
. adams jr .
et al . , `` jem - euso observational technique and exposure '' , astropart .
40 , pp .
117 - 134 ( 2014 ) m. fukushima et al .
, `` telescope array project for extremely high energy cosmic rays '' , prog .
151 , pp .
206 - 210 ( 2003 ) h. tokuno et al .
, `` new air fluorescence detectors employed in the telescope array experiment '' , nucl .
instrum . & meth .
a , vol . 676 , pp . 54 - 65 ( 2012 ) p. von ballmoos et al . ,
`` the euso - balloon mission '' , 34th icrc , pos(icrc2015)322 ( 2015 ) l. wiencke et al .
, `` euso - balloon mission to record extensive air showers from near space '' , 34th icrc , pos(icrc2015)631 ( 2015 ) y. hachisu et al .
, `` manufacturing of the ta - euso and the euso - balloon lenses '' , 32nd icrc , id1040 ( 2013 ) . s. udo et al . , `` the central laser facility at the telescope array '' , 30th icrc , vol . 5 , pp .
1021 - 1024 ( 2008 ) m. bertaina et al .
( jem - euso coll . ) , `` the first level trigger of jem - euso : concept and tests '' nucl .
& meth . , vol . a 824 p. 253 - 255
( 2016 ) t. c. paul et al .
, `` a new design for simulation and reconstruction software for the jem - euso mission '' , 34th icrc , pos(icrc2015)578 ( 2015 ) f. fenu et al .
, `` the esaf simulation framework for the jem - euso mission '' 32nd icrc , vol . 3 , pp .
116 - 119 ( 2011 ) | euso - ta is a pathfinder experiment for the space based jem - euso mission for the detection of ultra - high energy cosmic rays .
euso - ta is an high - resolution fluorescence telescope installed in front of the black rock mesa fluorescence detectors of the telescope array ( ta ) experiment , in utah ( usa ) . at the ta site , a central laser facility
is installed for calibration purposes , since it emits laser beams with known energy and geometry .
euso - ta consists of two 1 @xmath0 fresnel lenses , with a field of view of that focus the light on a photo detector module ( pdm ) .
the pdm currently consists of 36 hamamatsu multi - anode photo - multipliers tubes ( mapmts ) with 64 channels each .
front - end readout is performed by 36 asics , with two fpga boards that send the data to a cpu and a storage system .
the detector was installed in february 2015 .
tests using the mentioned light sources have been performed and observations of cosmic ray events , as well as those of stars with different magnitude and color index have been done .
the data acquisition is triggered by ta fluorescence detectors , although a self - trigger algorithm is currently in the last phases of development and test . with ta , thanks to its large field of view and the surface detectors , the cosmic ray shower events are reconstructed and the parameters are used to perform simulations of the response of euso - ta detector using _
offline_. simulations of the detected events are compared with data and the results are shown in this work . |
the study of transport mechanisms at the mesoscale level is of great importance nowadays .
specifically , the so - called ratchet systems have shown to be proper candidates for explaining unidirectional motion to biological systems @xcite , and have important physical applications for nano- and micro - scale technologies @xcite .
many of these models have been developed in the simple picture of point - like particles @xcite ( see the reviews @xcite for details ) .
such scenario has been subsequently generalized to spatially extended systems @xcite , where much attention has been paid to situations where the net motion arises through time - symmetry breaking @xcite .
this kind of ratchet phenomenon has been recently observed in long josephson junction ( ljj ) devices @xcite .
another possibility that has been considered in the literature is that of nonlinear klein - gordon system where the on - site potential is ratchet - like @xcite .
notwithstanding , to our knowledge the case of spatial - symmetry breaking by inhomogeneities has not been studied in depth .
one such study has been done by carapella et al . who used an inhomogeneous magnetic field to create an effective inhomogeneous junction profile for fluxons to propagate @xcite .
recently an alternative to the generation of motion for extended systems with a disorder in the chain has been proposed @xcite .
the novelty of the procedure is the design of a ratchet device from a lattice of _ point - like inhomogeneities_. for this system , net motion arises from the interplay between disorder and nonlinearity of the nonlinear systems @xcite . in this paper
we elaborate on the preliminary results reported in @xcite .
our aim is to carry out an in - depth analysis of the system , including a careful comparison to related point - like ratchets @xcite and an extension of our results , originally obtained for the sine - gordon ( sg ) model , to other nonlinear klein - gordon models such as the @xmath0 equation .
additional motivation for this work arises from research on models of energy propagation along microtubule filaments inside the cells @xcite .
this application is specially interesting in view of the possible connection with the dynamics of transport in molecular motors in biological systems , with features similar to those of solitons as extended objects . in this context
, the present work sheds light on the role played by the length scale competition between the point - like inhomogeneities ( disorder ) and the size of kinks in the transport dynamics . for this purpose
, we use the framework of collective coordinates ( cc ) in order to gain insight in the cause of the motion and the degrees of freedom that take part in it .
emphasis will be given to the kink width oscillations and their role in the transport properties : indeed , in general , the width of the nonlinear topological excitations is crucial for the movement of these coherent excitations .
the coupling between the translational and kink width degrees is such that motion takes place @xcite .
in particular , in the present work we will see that the oscillation range of the kink width is determined by the interplay with the inhomogeneities .
an additional , relevant issue is the analysis of the motion dynamics under thermal fluctuations . in ratchets , the noise is an important source of energy and , for some biological systems , it is regarded as the main cause of transport . here
we will consider the robustness of our rocking ratchet system under thermal fluctuations . in this case ,
activation of unidirectional motion was observed for a certain range of frequencies in the simulations as well as in the cc . in order to achieve the above mentioned goals ,
our paper is organized as follows : in sec .
ii we formulate the basis for the ratchet device and explain the origin and physical reasons of the rectification process .
a discussion in the cc framework is devoted to the length scale competition between the inhomogeneities and the kink width , and its influence on the motion dynamics . in the same context
, we establish an analogy with a simple model used for describing unidirectional motion in molecular motors @xcite , pointing out the relevant role of the kink width for the ratchet dynamics and its application in biological systems .
subsequently , an analysis of the efficiency in terms of the quantization of the transport is done for the relevant parameters of our ratchet system , including the interference effects among the inhomogeneities .
an example is given for the @xmath0 model , not only for its known rich internal dynamics reinforced by the presence of an internal mode , but also for its potential application to macromolecules ( e.g. , in transfer of energy in microtubules @xcite ) .
next , in sec .
iii , we analyze the kink dynamics subject to noise .
in particular , a new phenomenon of activation of motion induced by noise is described . in order to explain this new behavior ,
different collective coordinate approaches are implemented and thoroughly discussed .
finally , in the last section we summarize the main contributions of our work and make a discussion about the perspectives on this line .
we include appendixes where we detail the cc approaches for one and two collective variables , extending the _ generalized traveling wave ansatz _ ( gtwa ) to the case where inhomogeneities and noise act together with damping and ac forces .
kink dynamics in the presence of inhomogeneities can exhibit different and interesting behaviors , depending on the interplay between those inhomogeneities and the nonlinearity @xcite among other factors .
the generation of net motion using a lattice of point - like inhomogeneities is a good example of such non - trivial phenomenon @xcite .
although in that previous work the problem was discussed for the sg model , it can be generalized in principle for any nonlinear klein gordon system .
therefore , aiming at that general viewpoint , for our analysis we formulate the model in a general way as follows : @xmath1=a\sin(\omega t+\delta_{0 } ) , \label{1}\ ] ] where @xmath2 is the potential for the nonlinear klein - gordon equations and @xmath3 is an external ac force whose parameters @xmath4 , @xmath5 and @xmath6 represent the amplitude , frequency and phase of the periodic force , respectively . in particular , we will focus on the @xmath0 and sg models as specific examples ; the corresponding nonlinear potentials are @xmath7 and @xmath8 $ ] respectively . for our potential to exhibit ratchet - like phenomenon
, we choose @xmath9 to be given by a periodically repeated unit cell , formed by an asymmetric array of delta functions ( inhomogeneities ) .
the unit cell configuration , of length @xmath10 , is defined by three inhomogeneities , the first one located at the beginning of the cell , the second at distance @xmath11 from the first , and the third at distance @xmath12 from the second .
the corresponding mathematical expression is @xmath13\end{aligned}\ ] ] where the parameters satisfy the following constraints @xmath14 ( static kink width in absence of inhomogeneities ) ; @xmath15 with @xmath16 , where @xmath17 , @xmath18 , @xmath19 and @xmath20 , with @xmath21 . for our study we have taken @xmath22 , where in the case of sg , specifically for ljj the point - like inhomogeneities represent microshorts @xcite . however , the same scheme of arrays of point like inhomogeneities can be implemented for @xmath23 , where particularly for the sg model many works have been devoted @xcite .
the choice of three inhomogeneities is inspired by biological polymers like dna where the existence of three bases per codon seems to be the ideal configuration for the occurrence of net transport @xcite . in principle
it is possible to obtain unidirectional motion by using an array whose configuration presents more than three inhomogeneities per unit cell if the distances between the delta functions are in the same length scale as the kink width ( otherwise , different behaviors could arise like obtained e.g. in @xcite ) .
however , the inclusion of more inhomogeneities diminishes the efficiency of the transport as we will see below . ) for sg case : position of kink center vs time : upper panel : different arrays with the same amplitude of the force @xmath24 : @xmath25 , @xmath26 , @xmath27 ( @xmath28 ) ( solid line ) ; @xmath25 , @xmath29 , @xmath27 ( @xmath30 ) ( dashed line ) ; @xmath25 , @xmath31 , @xmath27 ( @xmath32 ) ( dash - dotted line ) .
lower panel : for different amplitudes of the ac force : @xmath24 ( solid line ) ; @xmath33 ( dashed line ) ; @xmath34 ( dash - dotted line ) with the array @xmath25 , @xmath31 , @xmath27 .
the other parameters used are @xmath35 , @xmath36 , @xmath37 and period @xmath38.,width=283 ] ) for sg case : position of kink center vs time : upper panel : different arrays with the same amplitude of the force @xmath24 : @xmath25 , @xmath26 , @xmath27 ( @xmath28 ) ( solid line ) ; @xmath25 , @xmath29 , @xmath27 ( @xmath30 ) ( dashed line ) ; @xmath25 , @xmath31 , @xmath27 ( @xmath32 ) ( dash - dotted line ) .
lower panel : for different amplitudes of the ac force : @xmath24 ( solid line ) ; @xmath33 ( dashed line ) ; @xmath34 ( dash - dotted line ) with the array @xmath25 , @xmath31 , @xmath27 .
the other parameters used are @xmath35 , @xmath36 , @xmath37 and period @xmath38.,width=283 ] contrary to the case of point particles , where motion through point - like inhomogeneities ( delta functions ) is physically meaningless , in our case we deal with kinks ( extended objects ) with a determined width .
this is an intrinsic feature of these nonlinear excitations , and correspondingly the competition between their width and the distances among the inhomogeneities is crucial for the kink motion .
the interference effects among the inhomogeneities @xcite create an effective potential for the motion of the kink center , the location of the inhomogeneities determining the direction of motion .
for the particular configuration of three inhomogeneities per unit cell , directional motion takes place only under the condition @xmath39 . in the upper panel of fig .
[ ratchet0 ] , results of simulations of eq .
( [ 1 ] ) for the sg case with different values of @xmath11 and @xmath12 are depicted .
such a picture shows that our ratchet device is an authentic rectifier .
in addition , as in standard ratchet systems , the directional motion of the kink center takes place only for certain values of the amplitude of the ac force ( see lower panel of fig . [ ratchet0 ] ) , a behavior that is dependent on the ac force frequency . a more detailed picture of the dynamics of the mean velocity for the sg kink center as a function of the ac force amplitude for different frequencies can be found in @xcite .
we have restricted ourselves to the overdamped case by taking @xmath40 , where the inertial effects are small , reducing the generation and propagation of phonons , and for which the kink center moves on a tilted effective potential due to the external ac force .
this regime prevents also the dependence on the initial conditions for the dynamics @xcite . for the integration of eq .
( [ 1 ] ) we have used a strauss - vzquez numerical scheme @xcite with free boundary conditions and spatial and temporal steps @xmath41 and @xmath42 , respectively .
we have checked our results with different spatial steps @xmath43 and @xmath44 . the simulations were done for the spatial interval @xmath45 $ ] with inhomogeneities arranged periodically according to our period in @xmath46 $ ] .
we have used the numerical representation for delta functions given by @xmath47 such representation is not unique but is one of the simplest forms for the numerical implementation @xcite . in order to understand the dynamics of the kink center different cc approaches
have been implemented @xcite . in this previous work
, it was shown that a 1-cc approach is not enough for the correct quantitative description of the results of the simulations even considering possible relativistic effects for high values of ac the force .
however , this approach allows us to understand the underlying physics of this ratchet system in the simplest way . on the other hand , satisfactory agreement with the simulations
was found by an improved , 2-cc approach , which takes into account the kink width dynamics @xcite . at this point
we will proceed directly with the discussion of the improved theory .
the corresponding collective variable equations of two degrees of freedom for the eq .
( [ 1 ] ) ( see appendix b , eqs .
( [ b21])-([b22 ] ) for @xmath48 ) can be expressed as @xmath49 @xmath50 where @xmath51 stands for the position of the kink center , @xmath52 represents the kink width , the internal potential energy of the kink is @xmath53 and @xmath54 is an effective potential depending on the specific equation one is dealing with .
in particular , for the sg case @xmath55 , @xmath56 , @xmath57 , @xmath58 and the effective potential is given by @xmath59}\hspace{2cm}\right .
\left.+\frac{1}{\cosh^2[(x - x_{2}-nl)/l]}+\frac{1}{\cosh^2[(x - x_{3}-nl)/l]}\right].\end{aligned}\ ] ] as we can see from the previous equations the kink width dynamics is coupled to the motion of the center of the kink .
therefore , changes in the kink width directly affect the translational motion .
it is possible to observe , for instance , that decreasing the kink width decreases the effective ac force , making necessary to increase the amplitude of the ac force in order to compensate for such effect .
this is an important factor that explains in part the shift observed in the locations of the windows of motion of the simulations with respect to those obtained from the 1-cc approach ( see fig .
1b in @xcite ) .
another relevant conclusion is the feedback between the effective potential landscape and the kink width , determined in turn by the potential . in this fashion ,
the 2-cc approach reflects the non - trivial interaction of the kink with the inhomogeneities , which is otherwise known to exhibit many counterintuitive phenomena @xcite . to deepen our understanding of the dynamics ,
let us look into the oscillations of the kink width .
as in the case of simulations we restrict ourselves to the overdamped case ( taking @xmath40 ) .
-[5 ] ) ( dashed line ) .
the parameters are @xmath60 , @xmath61 , for the array @xmath25 , @xmath31 , @xmath27 , period @xmath38 and @xmath36 .
see text for a discussion of the loop .
inset : enlargement of the loop indicated by an arrow in the main figure .
the motion of the kink center is indicated by numbered arrows.,width=302 ] ) .
b ) frontal view rotated by a small angle .
the array used is the same as in fig.[ancho].,title="fig:",width=313 ] ) .
b ) frontal view rotated by a small angle .
the array used is the same as in fig.[ancho].,title="fig:",width=321 ] ) ( sg case ) .
the inset shows the part of curve where the oscillations of the kink width take place.,width=302 ] a picture of the kink width oscillations versus kink center position is shown in fig .
[ ancho ] . as we can see , the agreement with the cc approach is excellent , confirming the validity of our predictions .
it is particularly interesting the existence of loops , which arises as a consequence of the kink center motion back and forth into the wells of the effective potential ( see fig . [ rat3pot ] ) , i.e. , describing half an oscillation before overcoming the barrier .
interestingly , another feature that stands out clearly is that the oscillations are around a value different from @xmath56 , the width of the unperturbed kink .
figure [ ancho ] shows that they take place around @xmath62 and , furthermore , that @xmath63 is not even included in the range of oscillations .
this phenomenon is the result of the balance between two opposite forces . on the one hand
, the inclusion of inhomogeneities increases the potential energy of the system .
this fact is reflected in the effective potential energy landscape fig . [ rat3pot ] .
such picture shows that when the kink width decreases , the potential energy decreases as well . taking two points with the same value for @xmath51 but with different kink widths @xmath52 , for example @xmath64 and @xmath65 in fig.[rat3pot]b )
, we see this difference in potential energy , i.e. , @xmath66 where @xmath67 . therefore , as the tendency of the system is to go to the minimum of the potential energy , the kink width would decrease . on the other hand , the kink internal potential energy , eq .
( [ 6 ] ) , has a minimum in @xmath63 , and hence the energy increases when the kink width decreases ( see fig .
( [ curveuint ] ) for @xmath68 ) ; notice that the first term of this equation accounts for a repulsive interaction and the second is given by an attractive one .
as a result of this balance , a new minimum will appear for the oscillations of the kink width .
it is important to note that the difference @xmath69 of the internal potential energy for the kink width @xmath70 with respect to the value @xmath56 ( inset of fig
. [ curveuint ] ) is of the same order as the energy difference @xmath71 between the points mentioned before for the effective potential introduced by the inhomogeneities , in agreement with this discussion .
a problem closely related to our 2-cc approach , given by a point particle ratchet with two degrees of freedom , has been studied in @xcite .
this model was designed for describing molecular motor dynamics consisting of two particles joined by a spring moving in a ratchet potential .
the corresponding equations of motion are given by @xmath72 where @xmath73 , @xmath74 represent the coordinates of the particles , @xmath75 is a sawtooth potential and @xmath76 is the internal potential energy . here @xmath77 with @xmath78 are gaussian white noises . ignoring the noise terms and their influence on the net motion
, we see that the change of variables @xmath79 and @xmath80 casts the system in a similar shape as eqs .
( [ 4]-[5 ] ) in the overdamped case , for which in good approximation the inertial terms could be neglected .
in such a new context the variables @xmath51 , @xmath52 can be interpreted as the mass center and elongation ( distance between the particles ) respectively , and obviously they resemble to the variables mass center and width of the kink in our system .
notice that in both models we have an asymmetric potential . in our case
it is given by eq .
( [ 7 ] ) , which is asymmetric at the cc level if the already mentioned conditions for the distances between the inhomogeneities are satisfied . in both systems
, there are internal potential energies that characterize their elastic properties . in the model in @xcite
, the internal potential is expressed through a harmonic function ( in the original variables ) : @xmath81 ^ 2,\ ] ] which in our collective coordinates can be written as @xmath82 ^ 2.\ ] ] where @xmath83 is the elasticity constant .
the links between the two models can be made more explicit by using a value for @xmath84 close to the minimum around which the kink width oscillates in our simulations ( cf .
the discussion in the preceding subsection ) .
+ it is important to point out that we have defined @xmath52 as a kink width variable as given in the expression ( [ b7 ] ) of the appendix .
however , what the quantity @xmath52 actually means is the distance at which the kink shape approaches its asymptotic values , measured from the center .
this means that @xmath52 in our notation is half the `` real kink width '' .
consequently the ratio ( real kink width)/(period of the effective potential ) becomes @xmath85 for which a very interesting dynamics for point particles dynamics has been reported in related 2-particle model @xcite .
+ this comparison between our model and that in @xcite allows to point out their main differences as well .
it is particularly important that in our framework , the internal energy can describe satisfactorily the repulsive interaction between real molecules where a van - der - waals like - force prevents their overlap .
this is very close to what occurs in molecular motors : if we take again the motion of kinesin as an example , this molecule has two dimer heads that act as ` feet ' , allowing the molecule to ` walk ' along a microtubule @xcite .
the repulsion would then appear when the two dimer heads are too close .
such a repulsive interaction can not be naturally accounted for within the model of two particles . for solving this problem the authors of @xcite resort to fix arbitrary values for @xmath84 which in our case is not necessary .
note , however , that in spite of the technical differences between both models , phenomenologically they are very similar : both of them try to understand how the motion of molecular motors , which proceeds in steps accompanied by deformations ( in the case of kinesin , when one step advances in front of the other ) can arise .
the common conclusion is that a point particle ratchet would not be a good model because the second degree of freedom is needed to capture the whole mechanism of the motion .
the advantage in our approach is that this second degree of freedom arises on its own , without a priori constructions , as an emergent property of the nonlinear excitation .
recent studies @xcite show similar phenomena for the two degrees of freedom point particle ratchet of @xcite when the ratchet is of flashing type .
the close relationship of the model of @xcite to ours suggests that nonlinear klein - gordon models can also exhibit rectification working as flashing ratchets , an issue we will address in future work
. it should be clear , from the results discussed so far , that in order to obtain a ratchet device for extended nonlinear systems with topological nonlinear excitations , the configuration of the inhomogeneities should be designed in such a way that the distance between the inhomogeneities is of the order of the kink width . however , this picture is somewhat too simple , and as we will see below , another important factor to take into account is the existence of interference effects . naively , one may try to design a similar ratchet system for the @xmath0 equation . considering only the kink width factor , it would seem that enlarging the sg array by a factor of @xmath86 ( the ratio between the kink widths in both models ) similar phenomena would be observed . ) -([14 ] ) , for two different arrays with @xmath36 .
( a ) : @xmath25 , @xmath31 , @xmath27 and @xmath38 .
( b ) : @xmath87 , @xmath29 , @xmath88 , @xmath89 . in both panels
sg ( solid line ) ; @xmath0 ( dashed line).,title="fig:",width=283 ] ) -([14 ] ) , for two different arrays with @xmath36 .
( a ) : @xmath25 , @xmath31 , @xmath27 and @xmath38 .
( b ) : @xmath87 , @xmath29 , @xmath88 , @xmath89 . in both panels
sg ( solid line ) ; @xmath0 ( dashed line).,title="fig:",width=283 ] let us make a more specific comparison between both models . to this end , we use the 1-cc framework in the nonrelativistic approach , where the equation of motion can be written as @xmath90 where @xmath91 is the normalized effective potential . for the sg case
we have the following expression @xmath92}\ ] ] with @xmath56 , and @xmath93 , whereas for the case of @xmath0 we have @xmath94}\ ] ] with @xmath95 , and @xmath96 .
the parameter @xmath97 is the topological charge and it is given by @xmath58 and @xmath98 for the systems sg and @xmath0 , respectively .
the normalized effective potentials for two different arrays are depicted in fig .
[ ratchet1 ] .
panel a ) shows standard asymmetric potentials for ratchet systems obtained with an array that satisfies the conditions mentioned above for the location of the inhomogeneities in the sg case .
however , in case b ) the effective potentials obtained for an array approximately given by the multiplication of the factor @xmath99 of the first one , shows a local minimum similar to an array of asymmetric double - well traps .
this last one has been used as a device for generating motion of vortices in superconductor materials @xcite . according to the our previous arguments based on the kink width role ,
a similar picture is expected for the normalized effective potential of @xmath0 and sg if the arrays verify the same length ratio as the full systems .
strikingly , fig .
[ ratchet1 ] shows that the normalized effective potentials are almost the same _ but for the same array length_. this apparent discrepancy can be explained if we take a detailed look at the potential given by eqs .
( [ 13]-[14 ] ) for both cases ( sg and @xmath0 ) .
: mean velocity vs driving amplitude @xmath4 for the frequency @xmath100.(a ) @xmath36 , @xmath25 , @xmath31 , @xmath27 , @xmath38.(b ) @xmath101 , @xmath25 , @xmath31 , @xmath27 , @xmath38 .
( c ) @xmath36 , @xmath87 , @xmath29 , @xmath88 , @xmath89 .
( d ) @xmath101 , @xmath87 , @xmath29 , @xmath88 , @xmath89 .
the lines connecting the points serve as guides for the eye.,title="fig:",width=283,height=188 ] : mean velocity vs driving amplitude @xmath4 for the frequency @xmath100.(a ) @xmath36 , @xmath25 , @xmath31 , @xmath27 , @xmath38.(b ) @xmath101 , @xmath25 , @xmath31 , @xmath27 , @xmath38 .
( c ) @xmath36 , @xmath87 , @xmath29 , @xmath88 , @xmath89 .
( d ) @xmath101 , @xmath87 , @xmath29 , @xmath88 , @xmath89 .
the lines connecting the points serve as guides for the eye.,title="fig:",width=283,height=188 ] + + : mean velocity vs driving amplitude @xmath4 for the frequency @xmath100.(a ) @xmath36 , @xmath25 , @xmath31 , @xmath27 , @xmath38.(b ) @xmath101 , @xmath25 , @xmath31 , @xmath27 , @xmath38 .
( c ) @xmath36 , @xmath87 , @xmath29 , @xmath88 , @xmath89 .
( d ) @xmath101 , @xmath87 , @xmath29 , @xmath88 , @xmath89 .
the lines connecting the points serve as guides for the eye.,title="fig:",width=283,height=188 ] : mean velocity vs driving amplitude @xmath4 for the frequency @xmath100.(a ) @xmath36 , @xmath25 , @xmath31 , @xmath27 , @xmath38.(b ) @xmath101 , @xmath25 , @xmath31 , @xmath27 , @xmath38 .
( c ) @xmath36 , @xmath87 , @xmath29 , @xmath88 , @xmath89 .
( d ) @xmath101 , @xmath87 , @xmath29 , @xmath88 , @xmath89 .
the lines connecting the points serve as guides for the eye.,title="fig:",width=283,height=188 ] it is clear from those expressions that , while in the case of @xmath0 we have a @xmath102 factor in the denominator , sg has a @xmath103 factor . therefore , the peaks and valleys in the effective potential for the @xmath0 system are much narrower than for sg , thus compensating for the increment in length . in addition , as in the sg model we will have dynamical changes of the effective potential due to the kink width variations , making more complicated the dynamics of motion . in any event , these effective potentials obtained in the simple approach highlight the importance of interference effects ( see also @xcite ) and make it clear that the kink width is not the unique quantity to take into account . the consequences of choosing either the first or the second array for the kink dynamics are revealed in fig .
[ ratchet2 ] .
we have taken for the analysis the @xmath0 model with a relative low frequency for the ac force , for which the mean velocity as a function of the ac force amplitude shows a staircase structure .
the range for the amplitude values was taken from the following rescaling expression : @xmath104 .
this rule is deduced from the comparison between the 1-cc approaches for sg and @xmath0 models , considering the similitude of the normalized potentials discussed above .
[ ratchet2 ] shows the dependence of the kink mean velocity as a function of the amplitude for two different arrays and heights of the perturbations introduced by the inhomogeneities .
as we can see , the motion is quantized as in standard ratchet systems @xcite and is characterized by the existence of gaps for which the net motion is absent ( i.e. , pure oscillating states ) .
the mean velocity can be expressed as @xmath105 as usual @xcite , where the indices @xmath106 , @xmath107 @xmath108 quantize the motion . using the expression for @xmath109
we can characterize the motion for each frequency and period of the array . comparing the values obtained from the simulations with the results derived from the expression for @xmath110 with corresponding parameters @xmath10 and @xmath5
, we find that @xmath106 and @xmath107 can take the following values : for panel a ) , @xmath111 and @xmath112 ; for panel b ) , @xmath113 and @xmath112 ; for panel c ) , @xmath114 with @xmath112 and @xmath115 , and for panel d ) , @xmath114 with @xmath112 and @xmath115 . although the mean velocity increases with the spatial period , the maximum value of the index @xmath106 significantly decreases , leading to a global decreasing of the velocity .
these results prove that the inclusion of more inhomogeneities per cycle , which obviously enhances the spatial period , is not a good option if we want to reach high velocities .
furthermore , a very low frequency would be required to obtain windows of motion . in the case of the dependence on the inhomogeneities height ,
the starting point of the stair - steps structure shows a shift towards greater amplitude of the ac force when increases the height , which is natural in order to overcome the barrier .
nevertheless , a higher speed is found , arising from a higher @xmath106 and observable also as a broadening in the windows of motion .
this far , we have analyzed the ratchet - like behavior of our system in the deterministic case .
however , it is clear that for our model to be more realistic , for instance , in the context of ljj , the effect of the temperature has to be taken into account .
the behavior of ratchet systems for nonzero temperature has been extensively studied both for point particles @xcite and for nonlinear extended systems @xcite . as for the problem we are considering here
, the fact that there has not been much effort on soliton ratchet - like phenomena induced by spatial inhomogeneities carries over to the stochastic effects .
therefore , it is important that we address this issue here . for the present work
, we will focus on the robustness of our rocking ratchet under thermal fluctuations .
another relevant issue would be the possible activation , resonances or modifications of the transport features induced by noise , but this topic deserves a detailed analysis and will be the subject of future work . for the sake of definiteness
, we consider the sg model under the influence of a gaussian white noise ; the results for the @xmath0 equation are similar . introducing the effect of the temperature through the fluctuation - dissipation relationship and considering the overdamped case as before ,
taking @xmath40 , we find the following equation : @xmath116=f(t)+ \eta(x , t),\ ] ] with @xmath117 vs driving amplitude @xmath4 for different intensities of the noise .
( a ) @xmath35 : circles @xmath118 , filled squares @xmath119 , squares @xmath120 ; ( b ) @xmath60 : circles @xmath118 , filled squares @xmath119 , squares @xmath120 .
lines serve as guides for the eye .
the inset shows several realizations for the motion of the kink center with @xmath121 , @xmath37 and @xmath120 .
, title="fig:",width=283 ] vs driving amplitude @xmath4 for different intensities of the noise .
( a ) @xmath35 : circles @xmath118 , filled squares @xmath119 , squares @xmath120 ; ( b ) @xmath60 : circles @xmath118 , filled squares @xmath119 , squares @xmath120 .
lines serve as guides for the eye .
the inset shows several realizations for the motion of the kink center with @xmath121 , @xmath37 and @xmath120 .
, title="fig:",width=283 ] where @xmath122 and the intensity of the noise @xmath123 . for the numerical simulations of the full partial differential equation as well as for the numerical solution of the collective variables ( to be discussed in the next subsection ) ,
we have used the heun method with the box - muller - wiener algorithm for generating gaussian random numbers of mean zero and variance one @xcite . in fig .
[ ratchet3 ] we show the behavior of the kink center dynamics under thermal fluctuations .
hereafter , we have set the array parameters to be @xmath25 , @xmath124 , @xmath27 , @xmath38 and @xmath36 for our study .
the mean velocity was calculated using the expression in @xcite , namely @xmath125 where the average is to be understood over many realizations of the noise . from this figure
we see that the steps of the deterministic case are now smoothed , a typical feature for the dynamics under noise .
it is important to realize that this smoothing affects the regions between the windows , which become minima of the mean velocity modulus @xmath126 instead of gaps with null mean velocity ( see fig.[ratchet3]a ) .
this phenomenon is directly related to the strength of the noise , i.e , when the noise increases , the absolute value for the mean velocity decreases but simultaneously the connection between the windows becomes more evident and the windows of motion become less pronounced . as in most other ratchet systems , in our model
the stochastic fluctuations due to temperature assist the jumps of the kink center from one well to the next one , allowing in some cases jumps in the opposite direction of the rectification ( see the inset graphic in fig .
[ ratchet3]b ) which is not possible in the absence of noise .
accordingly , the thermal fluctuations affect the mechanism of rectification whereas , on the other hand , they yield the dynamics of the pure oscillating states of the kink center unstable ( i.e. , destabilizing the regions with locked directional motion for zero temperature ) .
the joint action of both effects leads to the smoothing of the windows and the connection of the deterministic gaps .
vs intensity of noise @xmath127 .
circles : @xmath60 and @xmath128 ; squares : @xmath129 and @xmath130 .
inset shows one realization for the motion of the kink center for @xmath60 , @xmath128 , @xmath131 and @xmath119.,width=302 ] for relative high temperatures the thermal kink energy is sufficient for overcoming the barrier of the effective potential , and the kink motion is in practice diffusive , the influence of the barrier becoming negligible .
for this reason the rectification of motion takes place only for not so large values of the noise intensity ( see discussion in @xcite ) . a remarkable feature we have observed in the simulations
is shown in fig .
[ ratchet3]b for frequency @xmath60 , where new windows ( absent in the deterministic case ) appear .
this scenario is very similar to the one reported in @xcite where a similar surprising and intriguing phenomenon was noted .
there , the authors discussed that these new windows arose due to jumps of the fluxons between stable and unstable pinned fixed points of the deterministic dynamics .
considering the interest for this stochastic phenomenon , we carried out a careful analysis of the corresponding zone . to summarize this work , in fig .
[ activa ] we plot the mean velocity as a function of the noise intensity for different values of the frequency , showing the existence of an effective value for the intensity of the noise for which a maximum absolute value of the mean velocity is obtained .
the inset in fig .
[ activa ] makes it clear that , as expected and suggested in @xcite , the mechanism of activation occurs through jumps between multistable states ( states of the kink center which in absence of noise are purely oscillating ) .
therefore , a higher velocity is obtained when the residence time in these multistates is reduced or , in other words , when the intervals between consecutive jumps decrease . once again , this process of activation becomes more effective when the noise intensity increases , but above a certain value of the noise intensity the kink center starts to jump in the direction opposite to that of the rectification , leading to a global loss in efficiency .
this explains the existence of an effective value for the noise intensity for which the velocity reaches a maximum value .
another interesting characteristic observed in fig .
[ activa ] is the dependence of the maximum mean velocity on the frequency .
specifically , for a frequency value slightly larger than @xmath60 , the maximum velocity decreases , the peak moving towards greater values of the noise and the corresponding window of motion moving towards greater values of the ac force . accordingly , for a relatively large value of the frequency , above @xmath129 , the window of motion induced by noise disappears . on the other hand , for frequencies slightly smaller than @xmath60 , a new window in absence of noise is obtained . with all these results
, it is clearly established that the unidirectional motion induced by noise occurs only for a narrow window of frequency values .
we will show in the next subsection that this phenomenon seems to be a general feature , since at the cc level the system behaves very much like the dynamics of point particles . in order to understand the behavior observed in the previous section we resort again to the cc approach . as a first step ,
we take only into account one degree of freedom .
although , as discussed above , this framework is inaccurate for describing quantitatively the kink motion on a lattice of inhomogeneities , it does help understand qualitatively most of the features observed in the simulations , without unnecessary analytical complications .
after some algebra ( see appendix a for details ) , with @xmath40 , we find the following expression for the kink center coordinate with noise : @xmath132 with @xmath133 , @xmath134 . for the sake of simplicity we have taken the nonrelativistic approach @xmath135 , for which the noise turns out to be simply additive .
vs driving amplitude @xmath4 for different intensities of the noise and frequency @xmath60 .
continuous line : @xmath48 , dotted line : @xmath119 , dashed line : @xmath120 .
inset : mean kink velocity @xmath136 vs intensity of noise @xmath127 for @xmath137.,width=302 ] figure [ ratchet4 ] presents the results of the numerical integration of eq .
( [ 17 ] ) .
much as we did in the simulations , we calculate the mean velocity using eq .
( [ 16 ] ) , taking up to 500 realizations . from this plot
two main features also observed in the simulations can be seen .
first , smooth curves are obtained for the mean velocity as a function of the amplitude of the ac force , with values that decrease when the noise is increased .
second , new windows appear , and inside them there is a value of the noise intensity for which the velocity reaches a maximum value ( inset in fig . 9 ) . it is thus evident that , in spite of the quantitative differences with the simulations , this simple approach does predict correctly the qualitative behavior of the full system . in order to improve the results presented so far , we have extended the framework to two collective variables . by doing so ( see appendix b ) we arrive at eqs .
( [ b21])-([b22 ] ) with two uncorrelated _ multiplicative _ white noises , meaning that the noises depend on the kink width dynamics . ) -([b22 ] ) of the appendix ) .
mean kink velocity @xmath136 vs driving amplitude @xmath4 for different intensities of the noise .
( a ) @xmath35 ; continuous line : @xmath48 , dotted line : @xmath119 , dashed line : @xmath120 ; ( b ) @xmath60 ; continuous line : @xmath48 , dotted line : @xmath119 , dashed line : @xmath120 .
inset shows mean kink velocity @xmath136 vs intensity of noise @xmath127 for @xmath138 with @xmath60.,title="fig : " ] ) -([b22 ] ) of the appendix ) .
mean kink velocity @xmath136 vs driving amplitude @xmath4 for different intensities of the noise .
( a ) @xmath35 ; continuous line : @xmath48 , dotted line : @xmath119 , dashed line : @xmath120 ; ( b ) @xmath60 ; continuous line : @xmath48 , dotted line : @xmath119 , dashed line : @xmath120 .
inset shows mean kink velocity @xmath136 vs intensity of noise @xmath127 for @xmath138 with @xmath60.,title="fig : " ] the results for this improved approach are collected in fig .
[ ratchet5 ] . comparing with the simulations ( fig .
[ ratchet3 ] ) , we can observe the excellent agreement , with the locations of the windows correctly predicted .
as expected the curves are again smooth , a feature correctly accounted for already in the 1-cc framework . for the frequency @xmath139 ,
a new window is predicted , whose location is also in very good agreement with its corresponding window in the simulations .
these results confirm the importance of considering the kink width dynamics in the framework of the collective coordinates in order to achieve correct quantitative results as compared to the simulations .
we have studied the dynamics of solitons in a lattice of point - like inhomogeneities in two different nonlinear systems , the sg and the @xmath0 models , as good representatives of an entire class of nonlinear models .
we have designed a periodic lattice made of an asymmetric unit cell , where , depending on the lattice parameters , ratchet - like motion of solitons ( rectification ) is observed .
this is one of the few examples proposed so far for soliton ratchets based on spatial inhomogeneity .
building on a preliminary work reported in @xcite , we have analyzed in full detail all the characteristics and features of interest of our model . in order to understand the observed phenomenology , we have developed two cc approaches : one that takes into account only the motion of the kink center , and another one that includes the kink width as a second degree of freedom .
we have seen that whereas both techniques give good qualitative results and allow for a general understanding the phenomenon , only when the kink width enters in the cc description the results are quantitatively correct .
interestingly , the cc approach allows to show that rectification takes place only when the unit cell gives rise to an effective potential of the same length scale as the kink width .
this fact , along with the relevant role played by the width dynamical behavior , makes us conclude that one essential ingredient for observing ratchet - like behavior in this system is the existence of length scale competition @xcite . as a necessary complement of the deterministic study ,
we have considered the influence of thermal noise on the behavior of our model systems . as in the deterministic case , the cc technique yields very good results .
of particular relevance is the appearance , at the 2-cc level , of multiplicative noise in the cc equations , coupled to the kink width .
this is yet another hint to the crucial role played by this degree of freedom in the dynamics of the system .
another important feature is the motion activation induced by noise , that was observed in the simulations and later corroborated in the cc framework . in a wider context
, we believe that the cc approach presented here , and the conclusions about the role of the kink width , are of a much more general applicability .
this is the case , for instance , with the anomalous resonances observed in nonlinear klein - gordon models @xcite , which can only be explained by considering the width degree of freedom . specifically within the context of rectification
, we have previously shown that the existence of ratchet behavior induced by pure asymmetric temporal driving in a homogeneous system , recently demonstrated in experiments in ljj @xcite , is due to the coupling between the translation and the width oscillations @xcite .
based on all these experiences , we believe that the implementation of the 2-cc approach can be very useful in the case of spatially correlated noise @xcite , where the 1-cc approach can not explain the length scale competition between the correlation length and the size of the kink , first suggested in @xcite
. the new formulation would be also necessary in using the cc framework for describing the transport of proteins assisted by a thermal bath provoked by atp molecules hydrolization , where the width is an important quantity to take into account .
on the other hand , it is becoming more and more evident the crucial contribution of internal degrees of freedom in the functioning of molecular motors @xcite . in this respect ,
our results suggests that models including in a natural way this degree of freedom , like in our case , can be the proper descriptions of those phenomena .
size dependent rectification has been also reported in colloids @xcite and rectification through clustering has been observed in granular gases @xcite , which further reinforce our conclusion that the deformation or internal degrees of freedom must be an ingredient of good theoretical description of ratchet phenomena beyond the point particle scenario .
finally , we want to stress that our design of a soliton system with ratchet behavior is a very simple one , that can be implemented in actual experiments and devices such as ljj , for instance .
another field where the current state - of - the - art allows to apply this result is that of engineered molecular motors , demonstrated , e.g. , in @xcite . in this case
, our proposal may apply to the design of biomolecular devices with medical applications . generally speaking
, this type of approach to rectification can be of interest for applications in which it is needed to have a tunable rectifier tailored for a specific regime .
of course , experimental verification of our predictions is needed to ascertain the accuracy of our results .
we hope that this work stimulates experiments in this direction .
indeed the research reported here , opens new perspectives in the design of ratchet devices for more complicated extended nonlinear systems , such as general coupled chains @xcite .
of particular interest in this class are stacked ljj @xcite , although there are many other systems with potential applications in different areas .
work along this line is in progress .
this work has been supported by the ministerio de ciencia y tecnologa of spain through grant bfm2003 - 07749-c05 - 01 ( as ) and by the international research training group ` nonequilibrium phenomena and phase transitions in complex systems ' ( dfg , germany ) .
the authors also want to thanks the comments of n.r . quintero .
in order to apply the gtwa first proposed in @xcite we rewrite eq.([1 ] ) adding gaussian white noise @xmath140 with @xmath141 where @xmath142 , @xmath122 , @xmath143 and @xmath144 is the hamiltonian corresponding to the unperturbed form of eq.([1 ] ) given by @xmath145 as starting point we assume that the solution has the form @xmath146,\ ] ] and therefore by definition of @xmath147 we have that @xmath148.\ ] ] the index @xmath149 refers to the kink shape , but in the following we will omit it for simplicity .
following the procedure in @xcite , inserting @xmath150 , @xmath151 into eqs.([a1])-([a2 ] ) we get the expressions @xmath152 @xmath153 multiplying eq.([a7 ] ) by @xmath154 and eq.([a8 ] ) by @xmath155 , and then subtracting both expressions and integrating we arrive at the following equation @xmath156 whose values for the coefficients and forces are given by @xmath157 where @xmath158 represents the energy of the system , @xmath159 is the hamiltonian density of eq.([a4 ] ) and @xmath160 is the static force due to the external field , equal to zero for the above hamiltonian .
next we consider the sg potential for the system eqs.([a1])-([a2 ] ) for which we assume as solution the ansatz @xmath161=4\arctan\left(\exp\left[\gamma(x - x(t))\right]\right),\ ] ] where @xmath162\right)$ ] is the static kink solution of the sg system , centered in @xmath163 and of width @xmath63 . here @xmath164 where we have put @xmath56 for the sg case .
considering the previous statement for the static force and taking into account @xmath9 from eq.([2 ] ) , we obtain @xmath165 where @xmath166 is the kink mass , @xmath58 is the topological charge and @xmath167 given by @xmath168}\ ] ] is the effective potential . in the non - relativistic limit @xmath135 , @xmath169 .
a representation for the stochastic force @xmath170 can be obtained from the calculation of the variance . in the case of additive noise
it is allowed to make the following assumption @xmath171 hence the correlation function for @xmath170 can be written as @xmath172 for which , taking into account the expression ( [ a2 ] ) , after some algebra we get @xmath173 i.e. , @xmath174 is a white noise with kink diffusion constant @xmath175 as a consequence we obtain a non - additive noise due to the factor @xmath176 , i.e , we have a problem with multiplicative noise . then the equation of motion ( [ a9 ] )
can be rewritten as @xmath177 with @xmath133 , @xmath134 .
the eq.([a15 ] ) in absence of inhomogeneities and noise agrees with the results presented in @xcite .
the other r.h.s .
terms that appear in ( [ a15 ] ) are in correspondence with those already obtained in @xcite in the presence of impurities ( non - relativistic approach ) and gaussian white noise , respectively .
the procedure used here is equivalent to the so - called adiabatic approach by using modified conservation laws @xcite .
in order to get the cc equations we follow a similar procedure as in the previous section but this time we propose a solution with the form inserting eqs.([b1 ] ) and ( [ b2 ] ) in our system eqs.([a1])-([a2 ] ) and then multiplying the first equation by @xmath154 and the second one by @xmath155 ; subtracting both expression and integrating we arrive at the following equation @xmath181 \dot{l } + \int_{-\infty}^{+\infty } dx \frac{\partial \phi}{\partial x } \frac{\partial \psi}{\partial \dot{l } } \ddot{l } \nonumber \\ & & - f^{stat } = \,\ \int_{-\infty}^{+\infty } dx\,\ f(x , t,\phi,\phi_{t } , ... ) \frac{\partial \phi}{\partial x}\qquad \end{aligned}\ ] ] @xmath187 \dot{x } + \int_{-\infty}^{+\infty } dx \frac{\partial \phi}{\partial l } \frac{\partial \psi}{\partial \dot{x } } \ddot{x}+\int_{-\infty}^{+\infty } dx \frac{\partial \phi}{\partial l } \frac{\partial \psi}{\partial \dot{l } } \ddot{l } \nonumber \\ & & - k^{int } = \int_{-\infty}^{+\infty } dx \,\ f(x , t,\phi,\phi_{t } , ... ) \frac{\partial \phi}{\partial l}. \label{b6}\end{aligned}\ ] ] as in the previous section we use the variances of the stochastic forces in order to obtain approximate expressions for it . taking the assumption given by the expression ( [ a12 ] )
we find for ( [ b11 ] ) the correlation function @xmath196 with @xmath200 , @xmath201 , for @xmath202.a feature of particular interest in these new equations is the presence of stochastic forces which are multiplicative white noises dependent on the kink width variable .
the method described here using the technique of projection is equivalent to the variational calculations of the momentum and the energy of the system for perturbed nonlinear klein - gordon systems of the form of eqs.([a1 ] ) and ( [ a2 ] ) and with a hamiltonian of the form of eq.([a4 ] ) ( see @xcite for details ) . another procedure and derivation has been recently presented in @xcite .
j. maddox , nature , * 365 * , 203 ( 1993 ) ; ibid * 368 * , 287 ( 1994 ) ; f. jlicher and j. prost , rev .
phys . * 69 * , 1269 ( 1997 ) .
lee , b. jank , i. dernyi and a.l .
barabasi , nature * 400 * , 337 ( 1999 ) ; j.e .
villegas , s. savelev , f. nori , e.m .
gonzalez , j.v .
anguita , r. garcia , j.l .
vicent , science * 302 * , 1188 ( 2003 ) .
h. linke , w. sheng , a. lfgren , h. xu , p. omling and p.e .
lindelof , europhys . lett .
* 44 * , 341 ( 1998 ) ; h. linke , t.e .
humphrey , a. lfgren , a.o .
sushkov , r. newbury , r.p .
taylor , p. omling , science * 286 * , 2314 ( 1999 ) .
s. flach , y. zolotaryuk , a.e .
miroshnichenko , and m.v .
fistul , phys .
. lett . * 88 * , 184101 ( 2002 ) ; m. salerno and y. zolotaryuk , phys .
e * 65 * , 056603 ( 2002 ) . l. morales - molina , n.r .
quintero , f.g .
mertens , and a. snchez , phys .
lett . * 91 * , 234102 ( 2003 ) . | we investigate the ratchet dynamics of nonlinear klein - gordon kinks in a periodic , asymmetric lattice of point - like inhomogeneities . we explain the underlying rectification mechanism within a collective coordinate framework , which shows that such system behaves as a rocking ratchet for point particles .
careful attention is given to the kink width dynamics and its role in the transport .
we also analyze the robustness of our kink rocking ratchet in the presence of noise .
we show that the noise activates unidirectional motion in a parameter range where such motion is not observed in the noiseless case .
this is subsequently corroborated by the collective variable theory .
an explanation for this new phenomenom is given . |
one of the most intriguing normal properties of the high-@xmath2 ( ht@xmath4 ) cuprates is the so called pseudogap
( pg ) phenomenon .
it is commonly presented in the @xmath5 plane as a line that starts from rather high temperatures ( at small @xmath6 ) and reaches the superconductivity ( sc ) @xmath2 dome below at or above optimal @xmath7 . in a broad sense @xmath6
means the hole concentration in the cuo@xmath3- plane , but more often than not one refers to properties of the sr - doped la@xmath8sr@xmath9cuo@xmath10 the pg feature was seen in numerous experiments ( nmr , tunneling spectra , resistivity etc . ; see for example reviews @xcite ) .
it has been stressed @xcite that the pg temperature is not defined unambiguously .
a widespread view is that the feature comes from some crossover in the electronic density of states ( dos ) .
the main result of the present paper is that after a proper re - arrangement of the experimental data no pg feature exists in the @xmath0cu nuclear spin relaxation time behaviour .
instead , the data show two independent parallel relaxation mechanisms : a temperature independent one that we attribute to stripes caused by the presence of external dopants and an
universal temperature dependent term which turns out to be exactly the same as in the stoichiometric compound ybco 124 .
we attempt below to put the results in the context of a phase separation @xcite .
the decomposition of @xmath11 into two terms , as it will be discussed below in more details , manifests itself in a broad temperature interval above @xmath2 .
it is limited from above by a @xmath12 that depends on the concentration , @xmath6 .
we consider @xmath12 defined in this way as a temperature of a 1st order phase transition , which , however , can not complete itself in spatial coexistence of two phases because of the electroneutrality condition @xcite .
it was already argued in @xcite that such a frustrated 1st order phase transition may actually bear a dynamical character .
the fact that a single resonant frequency for the @xmath0cu nuclear spin is observed in the nmr experiments , confirms this suggestion .
although in what follows , we use the notions of the lattice model @xcite , even purely electronic models @xcite for cuprates may reveal a tendency to phase separation . the basic assumption in @xcite are the following .
at large enough doping holes move between coppers and oxygens .
spins in the system are @xmath13-holes trapped to the cu - sites at the expense of local lattice distorsions .
elastic attractive interactions between these distorsions give rise to a lattice driven frustrated transition below some @xmath12 .
exchange interactions , as in the parent la@xmath14cuo@xmath10 , tend to organize the cu - spins in the antiferromagnetic ( af ) sub - phases .
excess charge of the dopants ions in af regions must be compensated by accumulation of holes in metallic regions .
we now turn to experimental data . in what follows
we address only @xmath15 behaviour because for cuprates af fluctuations prevail over the korringa mechanisms . in fig .
1a we collected data on @xmath16 in lsco from @xcite .
note the following : 1 ) according to @xcite @xmath1 at higher temperatures tends to 2.7 msec@xmath17 _ for all _ sr concentrations , in spite of considerable spread seen in fig .
1a . beginning of deviation from that value could serve us as a definition of @xmath18 ; 2 ) note that dissipation @xmath16 monotonically decreases from small @xmath6 to 0.24 ; 3 ) after an appropriate _ vertical _ offset all curves in fig .
1a collapse onto the @xmath19 dependence of @xmath16 for the `` optimal '' @xmath20 above 50 k ( fig .
we have checked that last tendency works well for ybco ( 6.5 ) doped with ca i.e. , the data for different @xmath21 in y@xmath22ca@xmath23ba@xmath3 cu@xmath24 o@xmath25 @xcite may be put on the top of each other after proper offsets .
this prompts us to verify whether same off - settings of the @xmath16 data apply to a broader group of materials .
the stoichiometric yba@xmath3cu@xmath26o@xmath27 possesses no structural or defect disorder and we adjust all data to the @xmath16 behaviour for this material @xcite .
2 shows that after a vertical shift in @xmath16 all the materials indeed follow the same universal temperature dependence above their @xmath2 and below 300 k. in other words , in this temperature range the nuclear spin relaxation in these cuprates is a sum of contributions from two parallel processes : @xmath28 in eq .
( 1 ) @xmath29 depends on a material and a degree of disorder @xmath30 , but does not depend on temperature , while @xmath31 , depends only on temperature , is the same for all these compounds and coincides with the @xmath16 for the two chains ybco 124 above its @xmath2=62 k. thus , to some surprise the only pseudogap feature in the nmr data that may be discerned in fig.2 is the one for ybco 124 : a change in the temperature regime between 130 and 180 k. it would be tempting to take again this feature as a mark of the ps regime taking place now in the stoichiometric material where doping most definitely comes about as a spill - over of carriers from the cuo - chains into cuo@xmath3 planes .
it is also natural to think that the number of the transferred carriers is not small : actually the low temperature hall effect measurements @xcite show a rapid increase in the number of carriers ( i.e. fermi surface size ) up to one hole per unit cell even in the single layer material like lsco , at the optimal doping @xmath32 0.15 .
recall , however , that little is known for the homogeneous phase ( i.e. above @xmath18 ) . properties of both ybco 124 and the optimally doped lsco ( see @xcite for review ) are unusual and best described in a very broad temperature interval in terms of the marginal fermi liquid @xcite .
we have not found a reliable experiment to define @xmath12 for these compounds and therefore leave the origin of the @xmath33-term for further discussions .
the decomposition ( 1 ) into two parallel dissipation processes show that usual definitions of @xmath12 @xcite have no grounds . in fig.1a
the lsco data with @xmath34 are spread even above 250 k. as a rough estimate for @xmath12 , it is much higher than the sc onset temperature .
3 presents the dependence on @xmath6 for @xmath35 in la@xmath36sr@xmath9cuo@xmath37 the inset provides the offsets
( i.e. @xmath38 terms ) for other materials .
we return to discussion of fig.3 later .
the observation that is central for the following is that in all the materials with non - zero @xmath35 ( see the inset in fig.3 ) incommensurate ( ic ) peaks have been observed in neutron scattering @xcite .
peaks are close to the @xmath39 $ ] point : at @xmath40 $ ] and @xmath41 $ ] @xcite .
we will now look for the connection between these two phenomena .
we first make an attempt to agree on a semi - quantitative level the observed ic magnetic peaks in la@xmath8sr@xmath9cuo@xmath10 with the values of the first term in eq .
we concentrate on la@xmath42sr@xmath43cuo@xmath10 for which the most detailed data are available @xcite . with the notation from @xcite @xmath44where @xmath45 stands for one peak , hyperfine tensor
@xmath46\}^{2}$ ] and for @xmath47 we take near single peak , say @xmath48 $ ] @xmath49 ^{2}}\]]where @xmath50 and @xmath51 and @xmath52 are the correlation lengths in the two proper directions . after integration
the contribution from stripes with @xmath53 along the @xmath6-direction is @xmath54\}^{2}\chi _
{ peak}^{\prime \prime } \]]experimentally @xcite @xmath55 and for @xmath56 @xmath571/4 . assuming the @xmath58 dependence @xcite only for the one of @xmath59 s , @xmath51 and usingfor @xmath60 and @xmath61 the known values @xcite one obtains : @xmath62 msec@xmath17 . with the af correlation length @xmath63 4
this is the correct order of magnitude .
the descending dependence of the offset ( fig.3 ) agrees qualitatively with the behavior of @xmath64 @xcite in eq.(4 ) . for a quantitative description one
need to know the @xmath6-dependence for @xmath65 .
such data in the absolute units are absent yet except aeppli .
another fact that may underlie this behaviour is that with the _
x_-increase buckling in the cuo@xmath3-planes is known to decrease diminishing pinning effects and making the local symmetry of the cuo@xmath3-unit same as in other materials from the class with small offset in fig.3 .
also , the system grows more metallic with a high holes
content 31,33,34 .
next comes the question concerning the origin and the role played by ic peaks and the physics of fluctuations related to them .
discovery of ic spin fluctuations presented a challenge for explaining the nmr results for the oxygen spin relaxation times : hyperfine field
leaks originated by the af incommensurate fluctuations , would considerably increase the oxygen s relaxation rates , but this was not seen experimentally .
slichter ( see in ref.@xcite ) interpreted these contradictions in terms of discommensurations : a periodic array of soliton - like walls separating regions with a short - range af order .
unlike neutrons , the nmr as a local probe , does not feel the overall periodicity .
existence of stripes looks just natural in terms of a static phase separation . at doping the system ( lsco )
must screen the excess charge ( sr@xmath66-ions ) in af regions .
therefore stripes of the af ordered phase must alternate with
domain walls .
the stripe arrangement by itself is nothing but an optimization of the competing coulomb and lattice forces @xcite .
( the phenomena is well known in physics of surface . )
stripes in a dynamical regime need better understanding .
for instance , often the ic peaks are seen by neutrons only at low enough temperatures or for large energy transfer at an inelastic scattering @xcite . at low temperatures
stripes may form a long - range order even in lsco ( at smaller @xmath6 , @xcite ) , breaking the symmetry of the ground state .
a better example of the pinned stripe order is given by nd ( or eu)-doped lsco @xcite .
( la@xmath67nd@xmath68sr@xmath69cuo@xmath10 reproduces all features of la@xmath70sr@xmath9cuo@xmath10 , including sc and same positions for ic peaks at given @xmath6 . )
the transition into the ordered stripe phase is driven by appearance first of the lattice / charge peaks @xcite . at finite temperatures stripes
could be viewed as a new type of excitations above the ordered state . on the other hand in lsco itself
stripes are seen through the inelastic scattering processes for arbitrary low energy transfer even at high temperatures 100 - 300 k @xcite . while the ordered ic ground state sets in only well below , at about @xmath71 30k @xcite .
this example provides the argument against treating
stripes as excitations : at so high a temperature the underlying long - range ic ground state would be already melted .
therefore the two - phase description seems to be closer to reality meaning that in the dynamical regime the af regions get coupled via coulomb forces with the metallic layer .
note that with the further x - increase @xmath64 increases as well and saturates making it meaningless to speak in terms of a strictly monolayer wall already above @xmath72 0.14 , where @xmath73 @xcite .
( note the difference in notations for ic peaks : ( @xmath74 from @xcite equals @xmath75 from @xcite equals @xmath76 from @xcite ) .
fig.3 demonstrates same tendency to increase the share of the metallic fraction with increase of sr - concentration : @xmath29 continues to drop with @xmath6 above 1/8 .
coexistence of a sc and the ic af phases at low temperatures was confirmed recently by the neutron diffraction experiments @xcite for la@xmath8sr@xmath69cuo@xmath10 @xmath77 in the vortex state .
( the coexistence of sc and af formations was found also from the @xmath78sr spectra @xcite ) .
the way of the coexistence of sc and the stripe order in the same sample remains unresolved : one view treats the new stripe symmetry as a superstructure superimposed on the fermi surface that changes the energy spectrum like any sdw / cdw can do it ( e.g. @xcite ) .
another plausible alternative would be a spatially inhomogeneous coexistence of the nonsuperconducting ic af phase and a
metallic phase with strong fluctuations .
we have found that in a temperature interval above @xmath2 and below some @xmath79300 k the nuclear spin relaxation for a broad class of cuprates comes from two independent mechanisms : relaxation on the``stripe``-like excitations that leads to a temperature independent contribution to @xmath16 and an `` universal '' temperature dependent term .
for la@xmath42sr@xmath43cuo@xmath10 we obtained a correct quantitative estimate for the value of the first term .
we concluded from eq.(1 ) and fig.3 that ' ' stripes '' always come about with external doping and may be pinned by structural defects .
we argue that the whole pattern fits well the notion of the dynamical ps into coexisting metallic and ic magnetic phases .
experimentally , it seems that with the temperature decrease dynamical ps acquires the static character with the ic symmetry breaking for af phase dictated by the competition between the lattice and the coulomb forces .
the form of coexistence of the ic magnetism with sc below @xmath2 remains not understood as well as behaviour of stoichiometric cuprates .
the work of l.p.g . was supported by the nhmfl through nsf cooperative agreement dmr-9527035 and the state of florida , that of g.b.t . through the rffr grant n 01 - 02 - 17533 .
et al_. , phys .
lett . * 67 * , 1791 ( 1991 ) ; h.a .
et al . _ ,
nature * 395 * , 580 ( 1998 ) ; m. arai _
et al . _ ,
lett . * 83 * , 608 ( 1999 ) ; a. bianconi , int . j.mod.phys .
b * 14 * , 3289 ( 2000 ) ; p. dai _ et al . _ ,
b * 63 * , 054525 ( 2001 ) . | existing data on @xmath0cu - nuclear spin relaxation reveal two independent relaxation processes : the one that is temperature independent we link to incommensurate peaks seen by neutrons , while the universal temperature dependent contribution coincides with @xmath1 for two - chain ybco 124 . we argue that this new result substitutes for a pseudogap regime in a broad class of high-@xmath2 cuprates and stems from the 1st order phase transition that starts well above the superconductivity @xmath2 but becomes frustrated because of broken electroneutrality in the cuo@xmath3 plane .
superconductivity , pseudogap , magnetic properties , nmr |
electrons in cosmic rays have unique features , complementary to the cosmic - ray nuclear components , because of their low mass and leptonic nature .
high - energy electrons lose energy by synchrotron radiation in the galactic magnetic field and inverse compton scattering with the interstellar photons in the galaxy .
high - energy cosmic - ray electrons can not propagate far from the sources , because the electrons lose rapidly energy with an energy loss rate of the square of energy through these radiative processes .
these processes during the propagations through the galaxy without hadronic interactions simplify modeling of the propagation of electrons compared with other cosmic - ray components such as nucleons .
evidence for non - thermal x - ray emission from supernova remnants ( snrs ) indicate that high - energy electrons in the tev region are accelerated in snrs ( * ? ? ?
these observations strongly suggest that cosmic - ray electrons are accelerated in snrs , and that snrs are the most likely primary sources of cosmic - ray electrons . @xcite
first pointed out that the electron spectrum in the tev region depends on the age and distance of a few local sources .
his proposed concept has been accepted in later calculations of cosmic - ray electrons ( * ? ? ? * and references therein ) .
@xcite suggest that the energy spectrum of cosmic - ray electrons have unique spectral structures in the tev region due to the discrete effect of local sources .
this means that we can identify cosmic - ray electron sources from the electron spectrum in the tev region .
in addition , it is discussed that some dark matter may produce negative electrons and positrons in the energy region of around @xmath6 tev via dark matter annihilations or decaying dark matter ( * ? ? ? * ; * ? ? ? * e.g. ) .
in particular , in the case of mono - energetic electrons from dark matter , although propagation through the galaxy would broaden the line spectrum , the observed electron spectrum could still have distinctive features .
thus , the observations of high - energy electrons bring us unique information about sources and propagation of cosmic rays , and enable us to search for dark matter .
although the cosmic - ray electrons have been observed with many kinds of detectors since 1960 @xcite , most observations are limited below several 100 gev @xcite . among these observations ,
the first - time cosmic - ray electron observation with nuclear emulsions was achieved by @xcite .
they indicated that nuclear emulsions are ideal for the detection of electrons among many background protons because of the excellent imaging capability with a high position resolution of 1 @xmath7 m .
the reason for the difficulty of the electron observations is that the electron flux itself is very low and decreases with energy much more rapidly than that of protons because of the electro - magnetic energy loss .
the electron energy spectra are represented by a power - law function with an index of @xmath8 to @xmath9 , which is steeper spectra than the proton spectra with a power - law index of @xmath10 ( * ? ? ?
* and references therein ) .
the flux of cosmic - ray electrons is @xmath11 % of the protons at 10 gev , and decreases very rapidly with increasing energy to be @xmath12 % of the protons at 1 tev ( * ? ? ? *
e.g. ) . therefore , there are few observations of the electrons in the tev region , since we need a long duration exposure with a detector that has a large geometrical factor , enough thickness , and powerful background rejection powers
. @xcite performed atic-2 balloon experiment in antarctica and reported the energy spectrum in the energy region from 20 gev up to 3 tev , whose instrument contains a deep , fully active , bgo calorimeter of 18 radiation lengths ( r.l . ) .
they indicated an excess of cosmic - ray electrons at energies of @xmath13 gev , compared to a general electron spectrum calculated with the galprop @xcite .
they discussed that the excess may indicate a nearby source of energetic electrons such as the annihilated electrons from dark matter particles . on the other hand , from the independent data analysis of atic-2 @xmath14 atic-4
, @xcite reported the electron spectrum from 30 gev to 1 tev , and indicated that the electron spectrum in the region of the excess includes a fine structure with a number of narrow peaks .
@xcite also observed cosmic - ray electrons from 10 gev to 800 gev by a long duration balloon flight using polar patrol balloon ( ppb ) in antarctica .
the ppb - bets is an imaging calorimeter composed of scintillating - fiber belts and plastic scintillators inserted between lead plates with 9 r.l .
they discussed that the energy spectrum with ppb - bets may indicate a sign of a structure in the several 100 gev region , which is similar with the atic-2 observations , although a single power - law spectrum is acceptable within statistical errors .
@xcite presented the results of cosmic - ray electron observations from 7 gev to 1 tev using about @xmath15 electron candidates detected in the first 12 months on - orbit by the fermi large area telescope ( fermi - lat ) .
their electron spectrum can be described with a power law of @xmath16 with no prominent features , accommodating a slight spectral hardening at around 100 gev and a slight softening above 500 gev .
fermi - lat also searched for anisotropies of electrons from 60 gev to 480 gev with angular scale extending from @xmath17 to @xmath18 , resulting in nul results @xcite .
they indicated that the upper limits for a dipole anisotropy range from @xmath19 % to @xmath20 % .
although the fermi - lat has the large exposures of the electron observations , the detector thickness is insufficient to observe electrons in the tev region . as the result , fermi - lat can not separate electrons and protons one by one , but separated electrons from protons statistically based on monte - carlo simulations and machine learning algorithms .
the h.e.s.s .
ground - based imaging atmospheric cherenkov telescopes measured the electron spectrum in the energy range of 340 gev to 5 tev @xcite .
the h.e.s.s .
data show no indication of a structure in the electron spectrum , but rather a power - law spectrum with a spectral index of @xmath8 which steepens to be around @xmath21 above @xmath11 tev . while h.e.s.s .
team reported electron observations up to several tev , the electron spectrum is provided by indirect observations .
thus , h.e.s.s .
intrinsically has systematic errors on the reconstructed electron spectra arising from uncertainties in the simulation of hadronic interactions , the atmospheric model , and the absolute energy scale .
@xcite reported a statistically significant increase in the positron fraction at energies above @xmath20 gev with the pamela satellite - borne experiment , which is completely inconsistent with standard models describing the secondary production of cosmic rays .
the pamela positron data indicate the existence of primary positron sources such as the annihilation of dark matter particles in vicinity of our galaxy , nearby pulsars , and nearby micro - quasars .
@xcite also presented the negatively charged cosmic - ray electron spectrum between 1 and 625 gev performed by pamela , which is the first time that cosmic - ray negative electrons have been identified separately from positrons above 50 gev .
the negative electron spectrum can be described with a single power law energy dependence with a spectral index of @xmath22 above 30 gev and no significant spectral features .
we have observed high - energy cosmic - ray electrons from 30 gev to 3 tev with emulsion chambers at balloon altitudes , from 1968 to 2001 , accumulating a total exposure of 8.19 [email protected] . in the observations , we have carried out particle identification one event by one event with a proton rejection power larger than @xmath23 in tev region , because of an excellent imaging detector with a position resolution of 1 @xmath7 m , which is one of the outstanding capabilities of the emulsion chambers .
the performance of the emulsion chambers was examined with accelerator beam tests at cern - sps and monte - carlo simulations .
we also estimated the atmospheric electron spectra in a reliable way @xcite , and carried out on - board calibrations by using the flight data . while we previously reported the results from 1968 to 1976 experiments @xcite and some additional publications ( see references of @xcite ) , in this paper we present the final cosmic - ray electron spectrum in the energy range from 30 gev to 3 tev observed with the balloon - borne emulsion chambers up to 2001 , combined with our previous results .
emulsion chambers consists of nuclear emulsion plates , x - ray films , and lead plates ( or tungsten plates in a few chambers ) .
a nuclear emulsion plate is a methacrylate base @xmath24 @xmath7 m thick , double coating of nuclear emulsion with @xmath25 @xmath7 m thickness .
we used fuji et-7b and et-7d for nuclear emulsion .
nuclear emulsion plates are placed under lead plates .
one or two x - ray films are inserted between a lead plate and a nuclear emulsion plate to allow rapid , naked - eye scanning for high - energy cascade showers , which produce dark spots in the films .
figure [ fig : ecc_config ] shows a typical emulsion chamber configuration .
the typical size and thickness of the detector are 40 cm @xmath26 50 cm and 8 cm ( @xmath27 r.l . ) , respectively .
detailed configurations are described in @xcite .
the thickness of one lead plate at the upper layers is 0.5 mm ( @xmath280.09 r.l . ) to identify incident parent particles , determine the incident angles , and investigate the initial shower developments . at the bottom layers , the thickness of one lead plate is 5 mm ( @xmath280.9 r.l . ) and x - ray films are inserted to detect cascade showers , as shown in fig .
[ fig : ecc_config ] . since
high - energy electro - magnetic showers above a few 100 gev leave dark spots on x - ray films , these showers can be detected with the naked eye by scanning the x - ray films .
the corresponding tracks in the adjacent emulsion plate are located by using microscopes .
the detection threshold of the x - ray film is @xmath29 gev for sakura type - n x - ray film used before 1984 , @xmath30 gev for fuji # 200 x - ray film , and @xmath31 gev , @xmath32 gev , @xmath33 gev for screen type x - ray films of fuji g8-rxo , g12-rxo , and gs - rxo used from 1984 to 1988 @xcite .
after 1988 we used screen type x - ray films of hr8-ha30 , hr12-ha30 , and hr16-ha30 .
the sensitivity experiment of screen type x - ray films of hr series were carried out at research center for electron photon science of tohoku university in 2001 using test chambers with multilayers of emulsion plates and x - ray films .
the test chambers were exposed to the 200 mev electron beams .
figure [ fig : x - ray_film ] shows the result of the characteristic curves of different type of x - ray films . since the detection threshold of the net darkness on the x - ray films with naked eye is 0.1 @xcite , the electron densities on the emulsion plates at the x - ray film detection threshold correspond to @xmath34 @xmath2 for hr16-ha30 , @xmath35 @xmath2 for hr12-ha30 , @xmath36 @xmath2 for hr8-ha30 , and @xmath37 @xmath2 for fuji # 200 x - ray films .
these electron densities are compatible with the shower track densities with emulsion chambers at the shower maximum of electrons with energies of 140 gev , 180 gev , 450 gev , and 750 gev , respectively ( see fig . [
fig : ecc_trcv ] ) , which are the detection threshold energies of the x - ray films .
because of the simple configuration of the detector , the geometrical factor ( @xmath38 ) can be estimated very accurately , a difficult task for some electronic detectors .
for the electron observations , the effective geometrical factor is given by @xmath39 where @xmath40 is the upper limit of incident angles and @xmath41 is so - called `` edge effect '' .
since the detector has its edge , some incident electrons near the edge on the top of the detector do not pass through the bottom of the detector .
the edge effect is the efficiency of events that pass through the top and bottom emulsion plates . in the typical case of @xmath42 and @xmath43 m@xmath0
, @xmath44 is 0.39 m@xmath0 sr with @xmath41 of 0.82 for the chamber thickness of 8.0 cm . in appendix
[ sec : ecc_list ] , we summarize the area @xmath45 , edge effect @xmath41 , upper limit of incident angles @xmath46 , and @xmath47 of the emulsion chambers . as shown in appendix [
sec : ecc_list ] , @xmath45 , @xmath41 , @xmath46 , and @xmath47 change depending on the electron energies .
we measure the shower particles within a circle of @xmath48 @xmath7 m radius from shower axis .
this means that we select the shower particles with higher energies , which suffered less multiple scattering in the chamber .
hence , the number of the shower particles selected decreases faster than that of all shower particles .
the shower maximum in emulsion chambers for the shower particles within a circle of @xmath48 @xmath7 m radius appears in @xmath49 r.l . for 1 tev electrons , while the maximum of the total number of shower particles appears in @xmath50 r.l . for 1 tev electrons . as a result
, the energy of higher energy incident electrons can be determined with a thinner detector .
thus the emulsion chamber has the advantages of a wide field of view , small thickness , and lightweight detector , compared to the other instruments . in order to verify the zenith angle dependence of the detection efficiency for the incident electrons , we present the zenith angle distribution of electrons observed with the balloon - borne emulsion chambers in fig . [
fig : zenith ] , which is compared to the expected distribution for primary cosmic - ray electrons . as shown in fig .
[ fig : zenith ] , the zenith angle distribution of electrons is consistent with the expectation .
we have observed cosmic - ray electrons with balloon - borne emulsion chambers in 14 flights between 1968 and 2001 . in order to reject background cosmic rays ,
the emulsion chambers are placed upside down in the balloon gondolas during ascent and descent of the balloons , and are flipped to a normal position during level flight .
the pressure altitude records for each flight correspond to residual atmospheric overburdens in the range from 4.0 g @xmath2 to 9.4 g @xmath2 . in table
[ tab : baln_list ] , we summarize the series of experiments since 1968 , in which the results for 1968 1976 observations were reported in @xcite .
the @xmath47 in table [ tab : baln_list ] present the effective exposure factors for primary electron observations in the energy range above 1 tev within the zenith angle of @xmath51 .
figure [ fig : exposure ] shows the total cumulative effective exposure @xmath47 for primary electrons , which is 8.19 m@xmath0-sr - day in the tev region .
in addition to the electron observations , we have simultaneously observed atmospheric gamma rays , whose results are described in @xcite .
in the balloon observations , we identify electron events among incoming cosmic - ray events and determine the energies . in the following data analysis
, we selected events with incident zenith angle less than 60 degree and which passed all the way from the top to bottom layer of the chambers . high - energy electro - magnetic showers above a few 100 gev are detected with naked - eye scanning of dark spots left on the x - ray films .
the corresponding tracks in the adjacent emulsion plate are located with microscopes , and traced back through the stack to the shower starting points .
as described in section [ sec : detector ] and appendix [ sec : ecc_list ] , since we have improved the x - ray films to detect lower energy electrons , we have used different types of x - ray films that have the different threshold energies . hence , the total cumulative effective exposure @xmath47 depends on electron energies , as shown in fig .
[ fig : exposure ] .
the detection efficiencies are 100 % above the threshold energies and fall off rather rapidly below the thresholds @xcite .
we also confirmed the threshold energy for each balloon flight by using the deviation from a single power - law spectrum of the observed atmospheric gamma rays . in this analysis
, we used the electron events above the threshold energies to derive the electron energy spectrum . in order to detect the electro - magnetic showers below a few 100 gev
, the emulsion plates were directly scanned with microscopes for a part of the 1968 , 1969 , 1970 , 1973 , and 1996 emulsion chambers .
we successfully detected electron events down to 30 gev with the microscope scanning .
the detection efficiency is larger than 95% @xcite .
the microscope scanning is carried out in the smaller rectangular area of the upper emulsion plates . since the shower particles on the bottom emulsion plates are measured in the full area
, the edge effect @xmath41 of the microscope scanning has the larger value than that of the usual shower measurements . in the case of the scanning area within @xmath52 from the detector edge , where @xmath53 is
the thickness of the detector and @xmath46 is the upper limit of incident angles , @xmath41 is @xmath54 .
for the larger scanning area , not within @xmath52 from the detector edge , @xmath41 is smaller than @xmath54 , as shown in table [ tab : ecc_list ] . in emulsion chambers , it is possible to measure the location of shower tracks in each emulsion plate with a precision of @xmath55 @xmath7 m .
the incoming particles such as electrons , gamma - rays , protons , and heavier nuclei are identified by examining the details of shower development , especially around the shower starting points . since electron events start from a single charged track which produces an electron - positron pair within 1 r.l . of the top of the emulsion chamber with about 90 % probability
, they are identified by the existence of a single and a pair track with the spreading angle less than @xmath56 rad at the interaction point , as described in appendix [ sec : rejection ] .
electron events also give the electro - magnetic shower without core structures .
gamma - ray events , which are also a pure electro - magnetic shower , start from a pair with no visible primary track above the shower starting point .
although the incident track of a proton - induced shower shows a single charged track like an electron , proton - induced showers also have many secondaries at the shower starting point and often have multi core structures in the deep layers . even in the case of proton - induced showers with few secondary tracks , it is possible to discriminate the proton - induced showers from electron - induced showers by the differences of the spreading angle between tracks at the shower starting points .
as described in appendix [ sec : rejection ] , the proton rejection power is estimated to be larger than @xmath57 , that is derived to be independent of m.c .
simulation codes and hadron interaction models .
hadron showers of heavier nuclei such as helium are easily distinguished because the grain density of the incident track is larger than a minimum ionizing particle . in emulsion chambers , we can measure the depth of the first electron - positron pair of the electron - induced shower , the so called shower starting point .
the validity of event identification can be checked by comparison of the measured shower starting points with the expected values .
figure [ fig : stp ] presents the shower starting point distributions of the balloon observations for electrons above 400 gev , gamma rays above 300 gev , and protons , compared to the expected distributions . as for the electrons , the shower starting point is compared to the bethe - heitler expectation and the lpm expectation based on migdal s formula ( * ? ? ? * e.g. ) . as shown in fig .
[ fig : stp ] , the shower starting point distributions within 3.0 r.l .
show good agreement with the expectations , whose results suggest the reliability of the particle identification , and the deviation of the proton distribution larger than 3.0 r.l . from the expectation
shows the decrease of the proton detection efficiency .
in particular , the consistent result of the electrons with the lpm expectation strongly suggests the accurate identification of electrons .
electron energies were determined by counting the number of shower tracks in each emulsion plate within a circle of radius 100@xmath7 m centered on the shower axis .
we derived the integral track length from these counted tracks in each layer . the integral track length is theoretically expected to be proportional to the shower energy , as discussed in detail in @xcite .
our chamber structures are slightly different in each flight because of slight differences of lead thicknesses and insertion of different types of x - ray films and phosphoric screen films .
since the differences of the chamber structure affect the integral track lengths , we calculated the shower developments for each chamber one by one using a monte - carlo simulation code called epics @xcite .
epics has been used for cosmic - ray experiments ( * ? ? ?
* ; * ? ? ?
* e.g. ) , and also used for very forward single photon energy spectra from 0.1 tev to 3.6 tev in the large hadron collider forward ( lhcf ) experiments @xcite . the incident electron energies are determined by these track lengths compared with the values estimated from the monte - carlo simulations for each chamber . for the calibration of the detector
, we carried out beam tests of electrons in 2004 at cern - sps .
the detector configuration is same as the balloon - borne emulsion chambers , except for the detector size of 10.0 cm @xmath58 12.5 cm .
results calculated using the epics code were confirmed by emulsion chambers exposed to 50 gev and 200 gev electron beams at cern - sps . in order to evaluate the possible systematic errors at energies greater than 200 gev
, we also compared two independent monte - carlo simulation codes ; epics and geant4 @xcite .
figure [ fig : ecc_trcv ] shows the longitudinal development of the average number of shower tracks from the monte - carlo simulations , compared with the results of 50 gev and 200 gev electron beams . as shown in fig .
[ fig : ecc_trcv ] , the simulations well represent the experimental data .
the differences between the integral track lengths of epics and geant4 are @xmath59 % in the energy range of 30 gev
3 tev , which is negligible small compared to statistical errors of our cosmic - ray electron spectrum as described in section [ sec : spectrum ] . figure [ fig : tl_hist ] shows the energy distributions for 50 gev and 200 gev electron beams .
the determined energies with the simulations for 50 gev and 200 gev electrons are consistent with the experimental data .
the energy resolutions are 14.5 % at 50 gev and 10.6 % at 200 gev , respectively .
figure [ fig : ecc_eneres ] presents the energy resolutions of the simulations , compared with the experimental data .
the energy resolution for the emulsion chamber is well represented by the form of @xmath60^{1/2},\ ] ] where @xmath61 is the electron energy and @xmath62 is the standard deviation of energy determination .
the first term in right - hand side root represents statistics - related fluctuations of the number of shower particles , while the last term represents fluctuations due to shower particles escaping from the finite thickness of the detector . in balloon flight experiments , it is necessary to correct the observed cosmic - ray electron spectrum because of the residual overlying atmosphere .
we corrected energy loss of primary electrons due to bremsstrahlung radiation in the overlying atmosphere .
the average bremsstrahlung energy loss to each electron is given by @xmath63 when the incident electron spectrum is a power - law function with an index of @xmath64 .
here , @xmath65 is the energy of primary electron at the top of atmosphere , @xmath61 is the measured energy in the detector , @xmath66 is the vertical thickness of the overlying atmosphere in radiation lengths , @xmath67 is a zenith angle of the incident electron to the detector , and @xmath68 refers to the function used in electro - magnetic shower theory , is given by as follows : @xmath69 ] described by @xcite . in the case of @xmath70
, @xmath71 is @xmath72 .
for example , in the case of a power - law index of @xmath9 ( @xmath73 ) , an atmospheric thickness of @xmath74 g @xmath2 , and a zenith angle of @xmath75 degree , the energies of incident electrons are reduced by @xmath76 % , and hence the electron flux decreases by @xmath77 % .
this energy loss formula is different from the simple energy loss of @xmath78 , since the energy losses of electrons have broad distributions and the incident electron spectrum is steeply sloped . in the case of @xmath78 with the same parameters ,
the energies are reduced by @xmath79 % and the electron flux decreases by @xmath80 % , which correspond to monochromatic electrons .
in addition to primary cosmic - ray electrons , atmospheric electrons are also produced by hadronic interactions of primary cosmic rays with nuclei in the atmosphere . since
almost all atmospheric electrons are produced via atmospheric gamma rays from neutral pion decay , the atmospheric electron spectrum is estimated by using the simultaneously observed atmospheric gamma - ray spectrum with the emulsion chambers @xcite .
@xcite derived the atmospheric electron spectrum in the upper atmosphere less than @xmath81 g @xmath2 from the observed gamma - ray spectrum using the electro - magnetic shower theory .
their derived atmospheric electron spectrum is substantively free from the uncertainties of the cosmic - ray nuclear spectra and hadronic interaction models .
the contributions of the atmospheric electrons to primary electrons increase with electron energies and with thicknesses of the overlying atmosphere .
table [ tab : elec_flux ] shows the number of atmospheric electrons , which ranges from @xmath82 % to @xmath83 % of the observed electrons .
we observed electrons at each balloon altitude , and derived the cosmic - ray electron spectrum using the following formula : @xmath84 here , @xmath85 is the number of the observed electron events , @xmath86 is the number of atmospheric electrons , @xmath87 is electron detection efficiency , @xmath88 is enhancement factor due to the energy resolution .
the efficiency @xmath87 for detecting electro - magnetic showers in emulsion chambers have been studied to be essentially @xmath48 % above the threshold energy for naked - eye scanning of x - ray films ( * ? ? ?
* and references therein ) .
the detection efficiency for hr16-ha30 was also tested by using emulsion chambers exposed to the 200 gev electron beam at cern - sps . simultaneously with primary electrons , we have also observed atmospheric gamma rays to check the performance of the emulsion chambers in each balloon experiment @xcite .
we also confirmed the detection efficiency of each emulsion chamber from the atmospheric gamma - ray spectra .
the uncertainty of the energy determination has the effect of enhancing the absolute flux of electrons , in particular , for a steep power - law spectrum .
the enhancement factor @xmath88 due to the energy resolution has values from 1.01 to 1.09 , depending on electron energies ( see @xcite in detail ) .
the total number of the observed electrons is 166 events with the balloon - borne emulsion chambers exposed from 1968 to 2001 in the energy range of 30 gev to 3 tev .
after the corrections described above , we derived the primary cosmic - ray electron energy spectrum . figure [ fig : espec ] shows the observed electron spectrum , which is well represented by a power - law function of @xmath89 the flux values and numbers of the electrons in each energy bin are listed in table [ tab : elec_flux ] . compared with our previous electron spectrum in the energy range of @xmath90 gev @xcite ,
the total number of the observed electrons increased threefold the previous result , and the highest energy was extended up to 3 tev .
the cosmic - ray electrons observed with balloon - borne emulsion chambers ( ecc ) extend up to 3 tev with no cut off in the form of a power - law spectrum with an index of @xmath91 . in order to confirm the lower limit of high - energy cut off in tev region , we fitted the observed electron spectrum with an exponentially cut - off power law . with the fixed power - law index of @xmath91 ,
the lower limit ( 90 % c.l . ) of the exponential cut - off energy is @xmath92 tev .
this observed electron spectrum in the energy region below 1 tev is very similar to the electron spectrum by pamela @xcite , and shows agreement with the electron observations by fermi - lat @xcite .
our electron spectrum observed with the emulsion chambers does not exhibit significant spectral excesses .
the result of our electron spectrum compared with the atic electron spectrum @xcite is inconsistent , with a statistical significance level of @xmath93 % ( a reduced @xmath94 value of @xmath95 for d.o.f.@xmath96 ) , although it is acceptable with a statistical significance level of @xmath55 % . above 1 tev region ,
the comparison of our result with the electron spectra by h.e.s.s .
is consistent , with a reduced @xmath94 value of @xmath97 for d.o.f.@xmath98 , while h.e.s.s .
data have large systematic errors @xcite .
we calculated an electron spectrum by galprop code using a standard file , galdef_50p_599278 @xcite .
as shown in fig .
[ fig : espec_models ] , this spectrum is consistent with the electron spectrum observed with the emulsion chambers .
we also compared the observed electron energy spectrum with an electron spectrum calculated by @xcite , in which parameters are set as follows : the diffusion coefficient of @xmath99 cm@xmath0 s@xmath101 at 1 tev , the supernova rate of @xmath102 yr@xmath101 in the galaxy , the electron output energy of @xmath103 erg above 1 gev , the 20 tev cutoff energy of the electron injection spectrum , and the burst - like release at @xmath104 yr after the explosion .
the `` distant component '' in fig .
[ fig : espec_models ] indicates the contributions from continuously distributed distant snrs with the distance larger than 1 kpc or the age older than @xmath23 yr . as shown in fig .
[ fig : espec_models ] , our observed spectrum is consistent with the calculated spectra of the distant component @xmath14 nearby component by @xcite , giving strong evidence for a non - zero flux in the tev region , by the definite identification of electron events one by one .
figure [ fig : espec_models ] also presents that the electron energy spectrum observed with emulsion chambers has the significantly larger flux in the tev region than that of the distant component .
this suggests that nearby electron sources such as snrs exist within the distance of @xmath55 kpc and the age of @xmath23 yr .
we have carried out cosmic - ray electron observations with the balloon - borne emulsion chambers since 1968 .
the emulsion chamber is an excellent imaging detector with a high position resolution of @xmath55 @xmath7 m .
this imaging capability enables the emulsion chamber to identify electrons with the high rejection power against gamma rays and protons , which is the highest proton rejection power of @xmath23 among the existing cosmic - ray electron detectors , and to determine electron energies by using only the central part of the electro - magnetic shower .
this leads for the emulsion chamber to have thin thickness , lower mass , and a wide field of view , compared to the other detectors .
hence , the emulsion chambers , even the comparatively lightweight instruments , successfully observed electrons above several 100 gev in the late 1960s and electrons above 1 tev in the 1980s .
further , these electron observations initiated discussions about the investigation of the propagation mechanisms in the galaxy and the identification of nearby cosmic - ray electron sources .
being recognized for the significance of the electron observations , the high - energy electron observations opened up and have been recently carried out by atic , fermi - lat , h.e.s.s . ,
pamela , and so on . in order to identify nearby cosmic - ray electron sources and
search for dark matter signals , there are also some ongoing and new experiments for the high - energy electron observations such as ams-02 and calet on the international space station ( iss ) @xcite .
ams-02 is ongoing to observe positrons and negative electrons up to 1 tev .
calet is being developed to be installed on the iss , preserving the excellent imaging characteristics of the emulsion chamber .
we sincerely thank the late prof .
j.j.lord and the late prof .
t.taira , who started up this experiment program , brought us to many successful balloon observations , and worked on the development of the experiments .
we are also grateful to a number of collaborators who have carried out together the experiments since 1968 .
we thank the crews of the sanriku balloon center ( sbc ) of isas / jaxa and the nasa columbia scientific balloon facility ( csbf ) for their excellent and successful balloon flights .
we acknowledge the staffs of institute for cosmic ray research ( icrr ) , university of tokyo for their kind support to emulsion experiments .
we also appreciate the excellent work and kind support of the staffs in research center for electron photon science of tohoku university and the h4 beam line of cern - sps .
we are grateful to y. sato for his kind support at cern in the beam tests .
the flux of cosmic - ray electrons is much smaller than that of cosmic - ray protons
. the observed flux ratio of electrons against protons is around 1 % for 10 gev and 0.1 % for 1 tev .
therefore , in order to observe high - energy electrons above 1 tev , the proton rejection power is required to be at least larger than @xmath105 .
we estimate the proton rejection power with the emulsion chamber in the following . while an electron produces the first @xmath106 pair within 2 r.l . with the probabilities of 99.9 % for 100 gev electrons and 99.5 % for 1 tev electrons with the threshold energy of 10 mev for pair production , a proton interfaces hadronically with a mean free path ( m.f.p . ) of about 30 r.l . in lead .
hence , the probability for hadronic interactions of a proton within 2 r.l . in the emulsion chamber is @xmath107 , so that the proton rejection power for the difference between interaction lengths of electrons and protons is about @xmath108 .
secondary neutral pions produced via hadronic interactions of a proton have in total about 30 % of the parent proton energy .
therefore , giving the same shower energy with an electron , the energy of the parent proton should be three times larger than that of the electron . since the integral cosmic - ray proton spectrum has a power - law form of @xmath109 , the effective flux of protons
is reduced to be @xmath110 .
thus , the proton rejection power for the energy shift effect of the proton - induced shower is about 7 .
electrons produce gamma rays via bremsstrahlung radiation , and then the gamma rays produce @xmath106 pairs . therefore , the almost electron - induced shower at the starting point in the emulsion chamber is composed of 3 tracks ( one electron + one @xmath106 pair ) or 5 tracks ( one electron + two @xmath106 pairs ) .
the mean spreading distance @xmath111 of a parent electron and an @xmath106 pair by coulomb scattering is approximately given by @xmath112 where @xmath113 is a traversing thickness of the electron in the material , @xmath114 is a radiation length of the material , @xmath115 is a path length of the electron , @xmath61 is the electron energy , and @xmath116 is the scattering constant of @xmath117 mev . in the case of @xmath118
( i.e. @xmath119 cm for lead ) and @xmath120 cm , the spreading distance @xmath111 is @xmath121 @xmath7 m for the primary electron energy of 100 gev and @xmath122 @xmath7 m for that of 1 tev , taking the electron energy @xmath61 to be a half of the incident electron energy because of the energy loss by bremsstrahlung radiation . among hadronic interactions of high - energy protons with nuclei , possible candidates to be the mimic
electron events of protons are the following events .
the forward produced charged secondary particles are narrowly collimated without the heavily ionizing tracks , which are recorded by low - energy recoil protons and nuclei , and the number of the charged secondary particles is just 1 , 3 or 5 , accompanied with one neutral pion .
in addition , they have no multi - structures in their cascade showers .
in fact , there are such phenomena that the momentum transfer to the target nuclei is relatively low and hence the number of the secondary particles is relatively small , what is called diffraction dissociation . in order to study diffractive coherent production in hadronic interactions of protons with nuclei , the experiments with nuclear emulsions have been performed for 400 gev and 800 gev of proton beams by @xcite and @xcite , respectively .
they selected the events with the number of the secondary charged particles of 1 , 3 or 5 , with the relatively low momentum transfer to the target nuclei , and without the heavily ionizing tracks . according to their results , the fractions of these selected events to the total hadronic interaction events are 3 % for 400 gev protons and 2 % for 800 gev protons .
given that the spreading angle of the forward produced secondary particles is @xmath56 rad , the radial distance between the secondary particles is @xmath81 @xmath7 m with a path length of 1 cm .
since this distance is one order of magnitude larger than the typical distance between a parent electron and an electron - positron pair , the proton events with the spreading angle larger than @xmath56 rad are readily identifiable from the electron events . adding the condition of the spreading angles less than @xmath56 rad , the residual events are less than @xmath123 % of the measured diffraction dissociation events , that is , the fraction of the selected events to the total hadronic interaction events is less than 3 % @xmath264 % @xmath124 % for 400 gev protons and 2 % @xmath264 % @xmath125 % for 800 gev protons . for the electron - like events of protons ,
since there are also further conditions that the selected events should be accompanied with one neutral pion and have no multi - structures in the shower developments , the above fraction gives us just the upper limit .
hence , the proton rejection power with the shower starting point is estimated to be larger than @xmath126% @xmath127 . as described above
, the proton rejection powers are @xmath108 for the difference between interaction lengths of electrons and protons , @xmath128 for the energy shift effect of the proton - induces shower , and @xmath129 for the shower starting point , respectively .
combined these independent proton rejection powers , the total proton rejection power is estimated to be larger than @xmath130 .
thus , for the electron observations with the emulsion chambers , the proton contamination in the tev region is estimated to be less than @xmath11 % .
lrrccr flight & area & time & average altitude & @xmath47 @xmath131 & launch site + & ( m@xmath0 ) & ( min ) & ( g @xmath2 ) & ( m@xmath0 sr s ) & + 1968 & 0.05 & 380 & 6.1 & @xmath132 & harunomachi , japan + 1969 & 0.05 & 267 & 7.1 & @xmath133 & harunomachi , japan + 1970 & 0.05 & 1136 & 6.1 & @xmath134 & sanriku , japan + 1973 & 0.20 & 833 & 8.2 & @xmath135 & sanriku , japan + 1976 & 0.40 & 1526 & 4.0 & @xmath136 & palestine , usa + 1977 & 0.63 & 1760 & 4.5 & @xmath137 & palestine , usa + 1979 & 0.80 & 1680 & 4.9 & @xmath138 & palestine , usa + 1980 & 0.80 & 2029 & 7.8 & @xmath139 & palestine , usa + 1984 & 0.076 & 576 & 9.2 & @xmath140 & sanriku , japan + 1985 & 0.087 & 940 & 9.4 & @xmath141 & sanriku , japan + 1988 & 0.038 & 647 & 7.1 & @xmath142 & uchinoura , japan + 1996 & 0.20 & 2092 & 4.6 & @xmath143 & sanriku , japan + 1998 & 0.20 & 1178 & 5.6 & @xmath144 & sanriku , japan + 1999 & 0.20 & 891 & 5.6 & @xmath145 & sanriku , japan + 2001 & 0.20 & 1108 & 5.5 & @xmath146 & sanriku , japan + + cccccccc 3050 & @xmath147 & @xmath148 & 6 & 0 & 6 & @xmath149 & @xmath150 + 60100 & @xmath151 & @xmath152 & 9 & 0 & 9 & @xmath153 & @xmath154 + 100150 & @xmath155 & @xmath156 & 8 & 1.00 & 7.00 & @xmath157 & @xmath158 + 150200 & @xmath159 & @xmath160 & 7 & 1.43 & 5.57 & @xmath161 & @xmath162 + 200300 & @xmath163 & @xmath164 & 7 & 1.96 & 5.04 & @xmath165 & @xmath166 + 300400 & @xmath167 & @xmath168 & 15 & 4.37 & 10.63 & @xmath169 & @xmath170 + 400600 & @xmath171 & @xmath172 & 35 & 6.44 & 28.56 & @xmath173 & @xmath174 + 600800 & @xmath175 & @xmath176 & 29 & 7.16 & 21.84 & @xmath177 & @xmath178 + 8001000 & @xmath179 & @xmath180 & 20 & 6.54 & 13.46 & @xmath181 & @xmath182 + 10001500 & @xmath183 & @xmath184 & 15 & 7.73 & 7.27 & @xmath185 & @xmath186 + 15003000 & @xmath187 & @xmath184 & 15 & 5.31 & 9.69 & @xmath188 & @xmath189 + 1968 & ( sn ) & & & & & + & & @xmath190 3050gev : & @xmath191 & 1.000 & 45 & @xmath192 + & & @xmath190 60100gev : & @xmath193 & 0.807 & 45 & @xmath194 + & & @xmath190 100150gev : & @xmath193 & 0.807 & 45 & @xmath194 + & & @xmath190150200gev : & @xmath193 & 0.807 & 45 & @xmath194 + & & @xmath190200300gev : & @xmath195 & 0.764 & 45 & @xmath196 + & & @xmath190 300400gev : & @xmath195 & 0.764 & 45 & @xmath196 + & & @xmath190 400600gev : & @xmath195 & 0.764 & 45 & @xmath196 + & & 600800gev : & @xmath197 & 0.680 & 60 & @xmath198 + & & 8001000gev : & @xmath197 & 0.680 & 60 & @xmath198 + & & 10001500gev : & @xmath197 & 0.680 & 60 & @xmath198 + & & 15003000gev : & @xmath197 & 0.680 & 60 & @xmath198 + 1969 & ( sn ) & & & & & + & & 3050gev : & & & & + & & 60100gev : & & & & + & & 100150gev : & & & & + & & 150200gev : & & & & + & & @xmath190 200300gev : & @xmath199 & 0.789 & 45 & @xmath200 + & & @xmath190 300400gev : & @xmath199 & 0.789 & 45 & @xmath200 + & & @xmath190 400600gev : & @xmath199 & 0.789 & 45 & @xmath200 + & & 600800gev : & @xmath197 & 0.680 & 60 & @xmath201 + & & 8001000gev : & @xmath197 & 0.680 & 60 & @xmath201 + & & 10001500gev : & @xmath197 & 0.680 & 60 & @xmath201 + & & 15003000gev : & @xmath197 & 0.680 & 60 & @xmath201 + 1970 & ( sn ) & & & & & + & & 3050gev : & & & & + & & 60100gev : & & & & + & & 100150gev : & & & & + & & 150200gev : & & & & + & & @xmath190 200300gev : & @xmath202 & 0.870 & 45 & @xmath203 + & & @xmath190 300400gev : & @xmath202 & 0.870 & 45 & @xmath203 + & & @xmath190 400600gev : & @xmath202 & 0.870 & 45 & @xmath203 + & & 600800gev : & @xmath197 & 0.680 & 60 & @xmath204 + & & 8001000gev : & @xmath197 & 0.680 & 60 & @xmath204 + & & 10001500gev : & @xmath197 & 0.680 & 60 & @xmath204 + & & 15003000gev : & @xmath197 & 0.680 & 60 & @xmath204 + 1973 & ( sn ) & & & & & + & & 3050gev : & & & & + & & 60100gev : & & & & + & & @xmath190 100150gev : & @xmath205 & 1.000 & 45 & @xmath206 + & & @xmath190 150200gev : & @xmath205 & 1.000 & 45 & @xmath206 + & & @xmath190 200300gev : & @xmath207 & 0.873 & 45 & @xmath208 + & & @xmath190 300400gev : & @xmath207 & 0.873 & 45 & @xmath208 + & & @xmath190 400600gev : & @xmath207 & 0.873 & 45 & @xmath208 + & & 600800gev : & @xmath209 & 0.821 & 60 & @xmath210 + & & 8001000gev : & @xmath209 & 0.821 & 60 & @xmath210 + & & 10001500gev : & @xmath209 & 0.821 & 60 & @xmath210 + & & 15003000gev : & @xmath209 & 0.821 & 60 & @xmath210 + 1976 & ( sn ) & & & & & + & & 3050gev : & & & & + & & 60100gev : & & & & + & & 100150gev : & & & & + & & 150200gev : & & & & + & & 200300gev : & & & & + & & 300400gev : & & & & + & & 400600gev : & & & & + & & 600800gev : & @xmath209 & 0.821 & 60 & @xmath211 + & & 8001000gev : & @xmath212 & 0.821 & 60 & @xmath213 + & & 10001500gev : & @xmath212 & 0.821 & 60 & @xmath213 + & & 15003000gev : & @xmath212 & 0.821 & 60 & @xmath213 + 1977 & ( sn ) & & & & + & & 3050gev : & & & & + & & 60100gev : & & & & + & & 100150gev : & & & & + & & 150200gev : & & & & + & & 200300gev : & & & & + & & 300400gev : & & & & + & & 400600gev : & & & & + & & 600800gev : & @xmath214 & 0.810 & 60 & @xmath215 + & & 8001000gev : & @xmath216 & 0.810 & 60 & @xmath217 + & & 10001500gev : & @xmath218 & 0.810 & 60 & @xmath219 + & & 15003000gev : & @xmath218 & 0.810 & 60 & @xmath219 + 1979 & ( sn ) & & & & & + & & 3050gev : & & & & + & & 60100gev : & & & & + & & 100150gev : & & & & + & & 150200gev : & & & & + & & 200300gev : & & & & + & & 300400gev : & & & & + & & 400600gev : & & & & + & & 600800gev : & @xmath209 & 0.810 & 60 & @xmath220 + & & 8001000gev : & @xmath221 & 0.810 & 60 & @xmath222 + & & 10001500gev : & @xmath223 & 0.810 & 60 & @xmath224 + & & 15003000gev : & @xmath223 & 0.810 & 60 & @xmath224 + 1980 & ( sn ) & & & & & + & & 3050gev : & & & & + & & 60100gev : & & & & + & & 100150gev : & & & & + & & 150200gev : & & & & + & & 200300gev : & & & & + & & 300400gev : & & & & + & & 400600gev : & & & & + & & 600800gev : & @xmath225 & 0.821 & 60 & @xmath226 + & & 8001000gev : & @xmath223 & 0.821 & 60 & @xmath227 + & & 10001500gev : & @xmath223 & 0.821 & 60 & @xmath227 + & & 15003000gev : & @xmath223 & 0.821 & 60 & @xmath227 + 1984 & ( sn ) , ( g8 ) & & & & & + & & 3050gev : & & & & + & & 60100gev : & & & & + & & 100150gev : & & & & + & & 150200gev : & & & & + & & 200300gev : & & & & + & & 300400gev : & @xmath228 & 0.859 & 60 & @xmath229 + & & 400600gev : & @xmath230 & 0.859 & 60 & @xmath231 + & & 600800gev : & @xmath230 & 0.859 & 60 & @xmath231 + & & 8001000gev : & @xmath230 & 0.859 & 60 & @xmath231 + & & 10001500gev : & @xmath230 & 0.859 & 60 & @xmath231 + & & 15003000gev : & @xmath230 & 0.859 & 60 & @xmath231 + 1985 & ( f ) , ( g8 ) & & & & & + & & 3050gev : & & & & + & & 60100gev : & & & & + & & 100150gev : & & & & + & & 150200gev : & & & & + & & 200300gev : & & & & + & & 300400gev : & @xmath230 & 0.859 & 60 & @xmath232 + & & 400600gev : & @xmath230 & 0.859 & 60 & @xmath232 + & & 600800gev : & @xmath233 & 0.859 & 60 & @xmath234 + & & 8001000gev : & @xmath233 & 0.859 & 60 & @xmath234 + & & 10001500gev : & @xmath233 & 0.859 & 60 & @xmath234 + & & 15003000gev : & @xmath233 & 0.859 & 60 & @xmath234 + 1988 & ( f ) , ( g12 ) , ( gs ) , ( g8 ) & & & & & + & & 3050gev : & & & & + & & 60100gev : & & & & + & & 100150gev : & & & & + & & 150200gev : & @xmath228 & 0.846 & 60 & @xmath235 + & & 200300gev : & @xmath228 & 0.846 & 60 & @xmath235 + & & 300400gev : & @xmath228 & 0.846 & 60 & @xmath235 + & & 400600gev : & @xmath228 & 0.846 & 60 & @xmath235 + & & 600800gev : & @xmath228 & 0.846 & 60 & @xmath235 + & & 8001000gev : & @xmath228 & 0.846 & 60 & @xmath235 + & & 10001500gev : & @xmath228 & 0.846 & 60 & @xmath235 + & & 15003000gev : & @xmath228 & 0.846 & 60 & @xmath235 + 1996 & ( f ) , ( h8 ) & & & & & + & & 3050gev : & & & & + & & 60100gev : & & & & + & & 100150gev : & & & & + & & @xmath190 150200gev : & @xmath236 & 1.000 & 45 & @xmath237 + & & @xmath190 200300gev : & @xmath236 & 1.000 & 45 & @xmath237 + & & @xmath190 300400gev : & @xmath236 & 1.000 & 45 & @xmath237 + & & 400600gev : & @xmath209 & 0.824 & 60 & @xmath238 + & & 600800gev : & @xmath209 & 0.824 & 60 & @xmath238 + & & 8001000gev : & @xmath209 & 0.824 & 60 & @xmath238 + & & 10001500gev : & @xmath209 & 0.824 & 60 & @xmath238 + & & 15003000gev : & @xmath209 & 0.824 & 60 & @xmath238 + 1998 & ( f ) , ( h12 ) , ( h8 ) & & & & & + & & 3050gev : & & & & + & & 60100gev : & & & & + & & 100150gev : & & & & + & & 150200gev : & & & & + & & 200300gev : & & & & + & & 300400gev : & @xmath209 & 0.819 & 60 & @xmath239 + & & 400600gev : & @xmath209 & 0.819 & 60 & @xmath239 + & & 600800gev : & @xmath209 & 0.819 & 60 & @xmath239 + & & 8001000gev : & @xmath209 & 0.819 & 60 & @xmath239 + & & 10001500gev : & @xmath209 & 0.819 & 60 & @xmath239 + & & 15003000gev : & @xmath209 & 0.819 & 60 & @xmath239 + 1999 & ( f ) , ( h16 ) , ( h12 ) , ( h8 ) & & & & & + & & 3050gev : & & & & + & & 60100gev : & & & & + & & 100150gev : & & & & + & & 150200gev : & & & & + & & 200300gev : & & & & + & & 300400gev : & & & & + & & 400600gev : & @xmath240 & 0.796 & 60 & @xmath241 + & & 600800gev : & @xmath209 & 0.796 & 60 & @xmath242 + & & 8001000gev : & @xmath209 & 0.796 & 60 & @xmath242 + & & 10001500gev : & @xmath209 & 0.796 & 60 & @xmath242 + & & 15003000gev : & @xmath209 & 0.796 & 60 & @xmath242 + 2001 & ( f ) , ( h16 ) , ( h8 ) & & & & & + & & 3050gev : & & & & + & & 60100gev : & & & & + & & 100150gev : & & & & + & & 150200gev : & & & & + & & 200300gev : & & & & + & & 300400gev : & & & & + & & 400600gev : & @xmath209 & 0.796 & 60 & @xmath243 + & & 600800gev : & @xmath209 & 0.796 & 60 & @xmath243 + & & 8001000gev : & @xmath209 & 0.796 & 60 & @xmath243 + & & 10001500gev : & @xmath209 & 0.796 & 60 & @xmath243 + & & 15003000gev : & @xmath209 & 0.796 & 60 & @xmath243 + | we have performed a series of cosmic - ray electron observations using the balloon - borne emulsion chambers since 1968 .
while we previously reported the results from subsets of the exposures , the final results of the total exposures up to 2001 are presented here .
our successive experiments have yielded the total exposure of 8.19 m@xmath0 sr day at the altitudes of @xmath1 g @xmath2 .
the performance of the emulsion chambers was examined by accelerator beam tests and monte - carlo simulations , and the on - board calibrations were carried out by using the flight data . in this work we present the cosmic - ray electron spectrum in the energy range from 30 gev to 3 tev at the top of the atmosphere , which is well represented by a power - law function with an index of @xmath3 .
the observed data can be also interpreted in terms of diffusive propagation models .
the evidence of cosmic - ray electrons up to 3 tev suggests the existence of cosmic - ray electron sources at distances within @xmath4 kpc and times within @xmath5 yr ago . |
while the problem of electroweak symmetry breaking can be solved in the standard model ( sm ) by introducing one higgs boson , the minimal supersymmetric standard model ( mssm ) requires five physical higgses : a light cp - even ( @xmath6 ) , a heavy cp - even ( @xmath7 ) , a heavy cp - odd ( @xmath8 ) and two charged higgs bosons ( @xmath9 ) .
therefore , the discovery of heavy neutral higgs bosons would be a major breakthrough in verifying the supersymmetric nature of the fundamental theory , which is one of the main physics goals of the large hadron collider project .
+ the most promising channel to discover the heavy susy higgses is the @xmath10 @xcite channel , where both the leptonic and hadronic decays of the tau can be exploited .
this channel has been shown to cover large parts of the intermediate and high @xmath11 region of the mssm parameter space for an integrated luminosity of 30 @xmath12 . for low values of @xmath11 ,
the coupling of the higgs bosons to taus is not sufficiently enhanced and therefore this region is inaccessible for the @xmath13 channel .
+ in all studies of the sm channels ( meaning that the susy higgses decay into standard model particles ) , it is assumed that sparticles are too heavy to participate in the decay process .
one should ask what would happen if some of the sparticles would be light and the decays of higgs bosons into these susy particles would be kinematically allowed .
indeed , the existence of light neutralinos ( @xmath14 ) , charginos ( @xmath15 ) and sleptons ( @xmath16 ) seems favoured by a large number of supersymmetric models in order to explain electroweak symmetry breaking without large fine - tuning @xcite .
also recent experimental results ( precision measurements at lep2 @xcite , muon @xmath17 @xcite ) may point towards the existence of light gauginos and sleptons .
+ light susy particles may jeopardize the higgs discovery potential of the sm channels , since their presence can drastically decrease the branching ratios of the higgses into sm particles .
furthermore , pair and cascade production of light sparticles becomes an extra background to the higgs searches . on the other hand ,
higgs bosons decaying into sparticles might open new possibilities to explore regions of parameter space where sm decays would not be accessible @xcite . in this note
we report on a study of this type of decay with the cms detector .
we will focus on the decay of the heavy neutral higgses @xmath7 and @xmath8 into two next - to - lightest neutralinos , with each of the neutralinos in turn decaying as @xmath1 , i.e. into two ( isolated ) leptons + @xmath2 , so we get @xmath18 this results in a clear four lepton final state signature .
we will show that , as is often the case for supersymmetric channels , susy backgrounds are more difficult to suppress than the sm backgrounds . of the latter , basically only @xmath19 survives after requiring four isolated leptons .
of the susy backgrounds , sneutrino pair production and sparticle cascade decay production of neutralinos are the most dangerous processes . using a set of selection criteria as described in section 5
, we can clearly distinguish the signal from the background in the intermediate mass range 230 gev @xmath20 @xmath21 @xmath20 450 gev and for low and intermediate values of @xmath11 , depending on the values of the other mssm parameters .
the remainder of this note is organised as follows : first we study the behaviour of the relevant branching ratios .
then we describe the event generation , the signal versus background discrimination methods , and the discovery potential of the channel in the @xmath21 - @xmath11 plane . as a next step
we investigate the effects of varying the other mssm parameter values . in the last section
the results are summarized .
the main difficulty in studying decay modes involving supersymmetric particles is the large amount of free parameters in the mssm .
therefore most studies are carried out in the msugra or gmsb context in order to reduce the number of free parameters ; we will however stick to the more general mssm framework , to avoid too many model dependent assumptions . as free parameters ,
we take the mass of the cp - odd higgs @xmath21 , the higgs vev ratio @xmath5 , the higgsino mass parameter @xmath22 , the bino mass parameter @xmath23 , the wino mass parameter @xmath24 , the slepton mass @xmath25 and the squark / gluino mass @xmath26 . as a starting point for our studies
, we will adopt the following framework : h & + & + * we consider light neutralinos and charginos , above the lep2 limits . initially , we fix @xmath23 at 60 gev , and using the renormalisation group relation @xmath24 @xmath27 2 @xmath23 , we can set @xmath24= 120 gev .
we take @xmath28 .
this large @xmath22 scenario is favoured in models where @xmath29 is the dark matter candidate , like msugra . in low @xmath22 scenarios , the decay of @xmath0 into leptons
will be strongly suppressed . for large values of @xmath30
, @xmath0 is rather wino and @xmath29 is bino - like .
therefore it approximately holds that @xmath31 and @xmath32 .
the effects of varying these parameters will be discussed later on .
* we also take sleptons to be light . in the most favourable case
they would be lighter than @xmath0 , thereby allowing two - body decays into leptons .
we will consider two scenarios : @xmath25 @xmath33 @xmath34 , where real decays of neutralinos into sleptons are allowed and @xmath25 @xmath35 @xmath34 , where only the virtual exchange is possible . * the masses of squarks and gluinos are kept at the 1 tev scale . in the mssm , it is natural that these sparticles are heavier than neutralinos and sleptons . in section 7 , we will investigate the effect of lowering the masses of squarks and gluinos .
these parameter values and domains for @xmath22 , @xmath23 , @xmath24 , @xmath25 and @xmath26 will be used as default throughout this note .
the exact values for @xmath22 and @xmath25 will be chosen after analysing and optimizing the @xmath36 cross sections through the mssm parameter space .
after establishing the visibility in this optimal point , we will scan the area in @xmath37 around it to see how far the discovery region reaches .
effects of varying the initial susy parameter values will be discussed .
in order to determine the regions in mssm parameter space where sparticle decay modes may be accessible , we will first discuss the behaviour of the relevant branching ratios .
the package hdecay @xcite is used to study the supersymmetric decay modes @xmath38 of the heavy higgses . the @xmath7 and @xmath8 couple preferably to mixtures of gauginos and higgsinos .
the dominant mssm parameters controlling this process are @xmath21 , @xmath39 , @xmath24 and @xmath22
. + figs . [
fig : hd1 ] and [ fig : hd2 ] show the decays of @xmath7 and @xmath8 into neutralinos and charginos , in the case where @xmath23 = 60 gev@xmath40 = 120 gev , @xmath22 = -500 gev and @xmath11 = 5 . for @xmath41 @xmath20 500 gev
, the probability for the heavy higgses to decay into susy particles can be as high as 20 % . in this mass region ,
the decay mode to @xmath42 has the highest branching ratio ( br ) , however it produces a final state with only two leptons
. this mode would be in competition with numerous sm and susy backgrounds .
the second best sparticle mode is @xmath43 .
this channel can provide four leptons and is thus a priori more appropriate for obtaining a good signal to background ratio .
the @xmath43 threshold is determined by our choice of @xmath24 ( @xmath44 ) = 120 gev ; the fall in br at @xmath41 @xmath35 350 gev is caused by the opening of the @xmath45 mode .
the partial decay widths into supersymmetric particles remain the same as for lower values of @xmath41 , but due to the opening of the @xmath45 mode , the total decay width increases and the branching ratio into charginos / neutralinos decreases . for values of @xmath41 @xmath33 350 gev ,
the br of the cp - odd higgs ( @xmath8 ) into gauginos is substantially higher than in the cp - even ( @xmath7 ) case .
this is due to the fact that for the cp - even higgs more couplings to sm particles are allowed , thus leading to a larger total decay width and smaller br s to sparticles . for high values of @xmath21 the br s
are about the same for @xmath7 and @xmath8 since one reaches the decoupling regime . also for higher masses ,
other neutralino modes like @xmath46 , @xmath47 or @xmath48 may open up which will contribute to the four lepton signal .
the next - to - lightest neutralino @xmath0 will decay into two fermions and the lightest neutralino : @xmath49 .
these fermions will most often be quarks , leading to two jets and missing @xmath50 in the final state . to obtain a clean signature ,
we will only focus on the case where the neutralino decays into two leptons @xmath51 , where @xmath52 = @xmath53 or @xmath22 .
this process is determined by the bino , wino and higgsino mass parameters @xmath23 , @xmath24 , @xmath22 , by @xmath11 and by the slepton masses @xmath25 .
if sleptons are heavier than the @xmath0 , and as long as direct decays into a @xmath54 boson are not allowed ( or suppressed ) , only three - body decays @xmath51 will contribute .
these decays are mediated by virtual slepton and @xmath54 exchange @xcite .
therefore it is more favourable to have light sleptons in order to have larger br s .
the decay branching ratios can be rather sensitive to the mssm parameters due to the fact that the @xmath54 and slepton exchange amplitudes may interfere constructively as well as destructively .
ht & + & + in fig .
[ fig : nd1 ] we show the br as a function of @xmath11 .
sleptons ( including staus ) are taken at 250 gev and @xmath22 = -500 gev .
@xmath0 is then rather wino and @xmath29 is bino - dominated . because of this , their coupling to the @xmath54 is dynamically suppressed , and the @xmath55 branching ratio will depend strongly on slepton masses . + if sleptons are lighter than the @xmath0 , direct two - body decays of the neutralino into a slepton - lepton pair are allowed , which may lead to large branching ratios . in fig .
[ fig : nd2 ] the evolution of the br with @xmath11 is shown for @xmath24 = 200 gev , @xmath22 = 300 gev , @xmath56 = 200 gev , @xmath57 = 150 gev .
this is however only valid in a rather limited region of the mssm parameter space , since often sneutrinos will be lighter than sleptons , causing the neutralinos to decay purely into invisible particles .
+ the fall of br(@xmath51 ) with @xmath5 in fig .
[ fig : nd2 ] is compensated by a rise in br(@xmath58 ) .
this means that allowing taus in the final state could possibly extend our discovery reach towards higher @xmath11 values .
however , taus decay into leptons ( @xmath53 , @xmath22 ) in only @xmath435% of the cases , whilst the hadronic decay modes have detection efficiencies of @xmath430% @xcite .
this , together with the fact there are up to four taus in the final state , makes that there is only a limited hope for a large improvement by including taus in the final state , but a dedicated study is needed .
the signal events are generated with spythia @xcite . for low @xmath11 values ( @xmath59 ) , the gluon - gluon fusion mechanism
@xmath60 dominates the production . due to the large coupling of the higgses to @xmath61 ,
the associated production @xmath62 dominates for @xmath11 @xmath35 5 @xcite .
the cp - odd higgs is produced more than the cp - even one because the @xmath63 coupling is directly proportional to @xmath5 , whilst the @xmath64 coupling is proportional to @xmath65 . in the decoupling regime ( i.e. high values of @xmath21 ) , both couplings become equal . besides these two main processes , we also included the @xmath66 fusion and higgsstrahlung processes .
we scanned the @xmath67 cross section in the ( @xmath21 , @xmath11 ) plane for different values of @xmath22 and @xmath25 .
@xmath23 and @xmath24 were initially kept on 60 and 120 gev respectively , and no direct decays of neutralinos in sleptons were allowed . in figs . [ fig : sc1 ] and [ fig : sc2 ] , the plot of @xmath68 br for @xmath22 = -500 gev and @xmath25 = 250 gev is shown . values of @xmath11 @xmath20 30 - 40 and @xmath21 @xmath20 400 - 500 gev seem to be favoured .
one also notices that the pseudoscalar higgs gives much higher cross sections than the scalar one .
h & + & + for the background processes , pythia 6.136 @xcite was used with a few bugs fixed both in the susy and general code .
the following sm backgrounds giving rise to four ( real or fake ) leptons in the final state have been generated : @xmath19 , @xmath69 , @xmath70 , @xmath71 , @xmath72 and @xmath45 .
decays of @xmath73 into @xmath74 s have been included , since they might be dangerous due to their non - zero @xmath2 .
for the susy backgrounds , we generated all pair production processes involving squarks , gluinos , sleptons , charginos and neutralinos .
+ + the cms detector response is simulated using the cmsjet fast monte carlo @xcite .
the effects of pile - up at high luminosity running of lhc have not been included yet , but are expected to be minor in the four - lepton final state .
in order to obtain a clear signal , we will have to discriminate between the signal events and background events that contain a similar four lepton final state .
two categories of background have to be considered : standard model processes and susy backgrounds .
+ the main sm backgrounds are @xmath19 and @xmath45 production .
they are dangerous because of their large cross sections at the lhc . in order to distinguish between events coming from the signal and from the sm background , we apply the following selection criteria : * we require two pairs of isolated leptons with opposite sign and same flavour , with a @xmath75 larger than 10 gev and within @xmath76 @xmath33 2.4 .
the isolation criterion demands that there are no charged particles with @xmath75 @xmath35 1.5 gev in a cone of r = 0.3 rad around each lepton track , and that the sum of the transverse energy in the crystal towers between r = 0.05 and r = 0.3 rad is smaller than 3 gev . *
all dilepton pairs of opposite sign and same flavour that have an invariant mass in the range @xmath77 gev are rejected ( z veto ) . demanding four tightly isolated leptons with a transverse momentum higher than 10 gev is a powerful requirement in fighting the @xmath45 and @xmath72 background .
an explicit @xmath73 veto eliminates the @xmath19 , @xmath78 , @xmath70 production and all other backgrounds containing a @xmath73 boson .
furthermore , to reduce @xmath19 we also require a minimal missing transverse energy of 20 gev .
@xmath19 events where one of the @xmath73 s decays into taus , with the taus decaying leptonically , can however pass this criterion .
+ + the susy background is more complex .
squark / gluino production is characterised by a large jet multiplicity ( 5 jets on average ) , a significant @xmath2 ( @xmath79 100 gev ) and jet transverse momenta that are large compared to the expectations for the signal .
selecting events with few , rather soft jets ( e.g. @xmath80 2 jets , @xmath50 of the hardest jet below 100 gev ) and with @xmath2 @xmath33 130 gev allows us to eliminate most of these events .
the @xmath81 threshold can be lowered to 50 gev if necessary .
the squark / gluino - gaugino associated production can be eliminated this way too .
if we assume @xmath82 = 1000 gev as in our default scenario , no squark / gluino events will survive the selection . in paragraph 7 , the effects of lighter masses will be discussed .
+ slepton - slepton production predominantly ends up in a 2-lepton final state .
sneutrino - sneutrino production remains however as the dominant susy background .
it could possibly be distinguished from the signal because of larger @xmath2 and larger @xmath83 of the leptons , as sneutrinos either decay into @xmath84 + @xmath85 ( leading to extra @xmath2 ) or into @xmath86 + @xmath87 ( leading to harder leptons ) .
+ pair production of heavier neutralinos and charginos will lead to more and harder jets and will often contain @xmath73 bosons in the final state .
direct @xmath0-@xmath0 production gives the same signature as the signal , but the production cross section is much smaller due to the strongly suppressed coupling of gauginos to the @xmath73/@xmath88 intermediate state .
+ + in figures [ fig : sel7 ] - [ fig : sel2 ] , the distributions of the different kinematical variables for the signal and the total background ( sm + susy ) are plotted .
the parameters of the considered case are : @xmath21 = 350 gev , @xmath11 = 5 , @xmath23 = 60 gev , @xmath24 = 120 gev , @xmath22 = -500 gev , @xmath25 = 250 gev , @xmath82 = 1000 gev .
the dark shaded ( blue ) area is the part of the spectrum that is retained in the event selection .
hp & + & + hp & + & + hp & + & + hp & + & + htp & + & + + in view of these distributions , we will apply the following search strategy : events are selected with @xmath2 smaller than 130 gev ( to suppress the susy background ) , but larger than 20 gev ( to suppress zz background ) .
the @xmath75 of the hardest lepton should be less than 80 gev .
the @xmath50 of the harderst jet in the event is taken smaller than 100 gev .
in addition , we could also make a jet multiplicity requirement ( @xmath80 2 jets ) , but this seems to be needed only if squarks and/or gluinos would be light ( cfr .
paragraph 7 ) .
the four lepton invariant mass of the signal events should not exceed @xmath21 - @xmath89 .
if the mass of the lightest neutralino is approximately known at the time of the analysis , one could set a limit at @xmath90 @xmath91 230 gev .
+ the number of signal and background events remaining after applying this selection is , for the considered case , given in table 1 .
[ tab:1 ] .number of events after successive cuts ( at 100 @xmath12 ) .
as parameters were used : @xmath21 = 350 gev , @xmath11 = 5 , @xmath23 = 60 gev , @xmath24 = 120 gev , @xmath22 = -500 gev , @xmath25 = 250 gev , @xmath82 = 1000 gev . [ cols="<,^,^,^,^,^,^",options="header " , ]
the sparticle decay modes of the heavy neutral susy higgs bosons have been investigated .
+ the channel @xmath92 ( @xmath52 = @xmath53 , @xmath22 ) seems the most promising . in a rather large region of the mssm parameter space ,
a clean signal can be observed by selecting events with 4 isolated leptons in the final state .
the main backgrounds are @xmath19 and sparticle pair production ( sneutrino , neutralino ) , but they can be sufficiently suppressed using appropriate selection criteria .
extra backgrounds due to light squark / gluino production can also be kept under control by applying additional cuts .
+ in the most common case where direct decays of neutralinos to sleptons are not allowed , the @xmath93 channel seems to provide a detectable signal in the region between @xmath21 @xmath27 230 and 450 gev and for @xmath11 @xmath20 40 ( at 100 @xmath12 ) , in a scenario where @xmath24 @xmath27 120 gev , @xmath22 @xmath27 -500 gev and @xmath25 @xmath27 250 gev .
+ since the branching ratio of the @xmath94 into four leptons is determined by the interplay between a number of mssm parameters , the observability will also depend strongly on the values of @xmath22 , @xmath25 , @xmath23 and @xmath24 .
large values of @xmath30 and low values of @xmath25 are favourable since they enhance the decay rate of the neutralinos into leptons .
+ + motivated by the low @xmath11 discovery potential of the @xmath95 channel , we also plan a similar study of the sparticle decay modes of the charged higgs bosons @xmath9 .
the authors would like to thank abdel djouadi for helpful discussions .
9999999 r. kinnunen and d. denegri , cms note-1999/037 m. bastero - gil , g.kane and s. king , phys . rev .
* b474 * ( 2000 ) 103 - 112 g. altarelli et al .
, jhep 0106 ( 2001 ) 018 s. komine et al .
, phys.lett . * b506 * ( 2001 ) 93 - 98 a. nikitenko and r. kinnunen , cms note-2001/031 h. baer et al .
, phys . rev .
* d50 * ( 1994 ) 316 - 324 a. djouadi , j. kalinowski and m. spira , hdecay , comput .
phys . comm .
* 108 * ( 1998 ) 5 h. baer and x. tata , phys
* d47 * ( 1993 ) 2739 - 2745 s. mrenna , spythia , a supersymmetric extention of pythia , anl - hep - pr-96 - 63 d.dicus and s. willenbrock , phys . rev .
* d39 * ( 1989 ) 751 t. sjostrand , comp .
* 82 * ( 1994 ) 74 s. abdullin , a. khanov and n. stepanov , cmsjet , cms note/94 - 180 ; + _ http://cmsdoc.cern.ch/@xmath4abdullin/cmsjet.html _ | we discuss the possibilities to observe the decays of heavy susy higgs bosons into supersymmetric particles at the lhc .
such an observation would be of interest either in a discovery search if sparticle modes are the dominant ones , or in a study of additional decay modes , bringing information on the susy scenario potentially at work .
we will focus on the most promising channel where the heavy neutral higgses decay into a pair of next - to - lightest neutralinos @xmath0 , followed by @xmath1 , thus leading to four isolated leptons + @xmath2 as the main final state signature .
a study with the cms detector shows that the background ( sm + susy ) can be sufficiently suppressed and that in the mass region between @xmath3 @xmath4 230 and 450 gev , for low and intermediate values of @xmath5 , the signal would be visible provided neutralinos and sleptons are light enough .
filip moortgat salavat abdullin daniel denegri |
quantum cryptography has brought us new ways of exchanging a secret key between two users ( known as alice and bob ) .
the security of such quantum key distribution ( qkd ) methods is based on a very basic rule of nature and quantum mechanics the `` no - cloning '' principle .
the first qkd protocol was suggested in a seminal paper by bennett and brassard @xcite in 1984 , and is now known as bb84 . during recent years
many security analyses were published @xcite which proved the information - theoretical security of the bb84 scheme against the most general attack by an unlimited adversary ( known as eve ) , who has full control over the quantum channel .
those security proofs are limited as they always consider a theoretical qkd that uses perfect qubits .
although these security proofs do take errors into account , and the protocols use error correction and privacy amplification ( to compensate for these errors and for reducing any partial knowledge that eve might have ) , in general , they avoid security issues that arise from the implementation of qubits in the _
real world_. a pivotal paper by brassard , ltkenhaus , mor , and sanders @xcite presented the `` photon number splitting ( pns ) attack '' and exposed a security flaw in experimental and practical qkd : one must take into account the fact that alice does not generate perfect qubits ( 2 basis - states of a single photon ) , but , instead , generates states that reside in an enlarged hilbert space ( we call it `` quantum space '' here ) , of six dimensions .
the reason for that discrepancy in the size of the used quantum space is that each electromagnetic pulse that alice generates contains ( in addition to the two dimensions spanned by the single - photon states ) also a vacuum state and three 2-photon states , and these are extremely useful to the eavesdropper .
that paper proved that , in contrast to what was assumed in previous papers , eve can make use of the enlarged space , and get a lot of information on the secret key , sometimes even full information , without inducing any noise .
many attacks on the practical protocols then followed ( e.g. , @xcite ) , based on extensions of the quantum spaces , exploring various additional security flaws ; other papers @xcite suggested possible ways to overcome such attacks . on the one hand , several security proofs , considering specific imperfections , were given for the bb84 protocol @xcite . yet on the other hand , it is generally impossible now to prove the security of a practical protocol , since _ a general framework _ that considers such realistic qkd protocols , _ and _ the possible attacks on such protocols , is still missing .
we show that the pns attack , and actually all attacks directed at the channel , are various special cases of a general attack that we define here , the _ quantum - space attack _ ( qsa ) .
the qsa generalizes existing attacks and also offers novel attacks .
the qsa is based on the fact that the `` qubits '' manipulated in the qkd protocol actually reside in a larger hilbert space , and this enlarged space _ can be assessed_. although this enlarged space is not fully accessible to the legitimate users , they can still analyze it , and learn what a fully powerful eavesdropper can do .
we believe that this assessment of the enlarged `` quantum space of the protocol '' is a vital step on the way to proving or disproving the unconditional security of practical qkd schemes .
we focus on schemes in which the quantum communication is uni - directional , namely , from alice s laboratory ( lab ) to bob s lab .
we consider an adversary that can attack all the quantum states that come out of alice s lab , and all the quantum states that go into bob s lab .
the paper is organized as follows : definitions of the quantum spaces involved in the realization of a protocol , and of the `` quantum space of the protocol '' , are presented and discussed in section [ sec : qsop ] .
the `` quantum - space attack '' is defined and discussed in section [ sec : qsa ] . using the general framework when the information carriers are photons is discussed in section [ sec : qsaphotonicworld ] .
next , in section [ sec : knownqsa ] we show that the best known attacks on practical qkd are special cases of the qsa .
section [ sec : interferobb84 ] demonstrates and analyzes a novel qsa on an interferometric implementation of the bb84 and the six - state qkd protocols .
last , we discuss a few subtleties and open problems for future research in section [ sec : conclusion ] .
we would like to emphasize that our ( crypt)analysis presents the difficulty of proving unconditional security for practical qkd setups , yet also provides an important ( probably even vital ) step in that direction .
the quantum space attack ( qsa ) is the most general attack on the quantum channel that connects alice to bob .
it can be applied to any realistic qkd protocol , yet here we focus on uni - directional schemes and on implementations of the bb84 protocol and the six - state protocol .
we need to have a proper model of the protocol in order to understand the hilbert space that an unlimited eve can attack .
this space has never been analyzed before except for specific cases .
our main finding is a proper description of this space , which allows , for the first time , defining the most general eavesdropping attack on the channel .
we start with a model of a practical `` qubit '' , continue with understanding the spaces used by alice and bob , and end by defining the relevant space , the _ quantum space of the protocol _ ( qsop ) , used by eve to attack the protocol .
the attacks on the qsop are what we call _ quantum - space attacks_. in most qkd protocols , alice sends bob qubits , namely , states of 2 dimensional quantum spaces ( @xmath0 ) .
a realistic view should take into account any deviation from theory , caused by alice s equipment .
for example , alice might encode the qubit via a polarized photon : @xmath1 via a photon polarized horizontally , and @xmath2 polarized vertically .
this can be written using fock notation are called fock states , see section [ sec : qsaphotonicworld ] . ] as @xmath3 where @xmath4 ( @xmath5 ) represents the number of horizontal ( vertical ) photons ; then @xmath6 and @xmath7 .
when alice s photon is lost within her equipment ( or during the transmission ) , bob gets the state @xmath8 , so that alice s realistic space becomes @xmath9 .
alice might send multiple photons and then @xmath10 is of higher dimension , see section [ sec : alicerealphotonic ] .
[ def : ha ] * alice s realistic space , @xmath10 , * is the minimal space containing the actual quantum states sent by alice to bob during the qkd protocol . in the bb84 protocol , alice sends qubits in two , etc . ]
fixed conjugate bases .
theoretically , alice randomly chooses a basis and a bit value and sends the chosen bit encoded in the appropriate chosen basis as a state in @xmath0 ( e.g. @xmath1,@xmath2 , @xmath11 , and @xmath12 ) .
to a better approximation , the states sent by alice are four different states @xmath13 ( @xmath14 ) in her realistic space @xmath10 , spanned by these four states .
this space @xmath10 is of dimension @xmath15 , commonly between 2 and 4 , depending on the specific implementation .
as practical instruments often diverse from theory , alice might send quite different states . as an extreme example , see the _ tagging attack _
( section [ sec : tagasqsa ] ) , which is based on the fact that alice s space could contain more than just these four theoretical states , so that @xmath16 is possible .
bob commonly receives one of several possible states @xmath17 sent by alice , and measures it .
the most general measurement bob can perform is to add an ancilla , perform a unitary transformation on the joint system , perform a complete measurement , and potentially `` forget '' some of the outcomes . however ,
once alice s space is larger than @xmath0 , the extra dimensions provided by alice could be used by bob for his measurement , _ instead of _ adding an ancilla . interestingly , by his measurement bob might be _ extending _ the space vulnerable to eve s attack well beyond @xmath10 .
this is possible since in many cases the realistic space , @xmath10 , is embedded inside a larger space @xmath18 .
[ def : m ] the space @xmath18 is the space in which @xmath10 is embedded , @xmath19 . the space @xmath18 is the actual space available for alice and an eavesdropper . due to the presence of an eavesdropper ,
bob s choice whether to add an ancilla or to use the extended space @xmath18 is vital for security analysis . in the first case
the ancilla is added by bob , inside his lab , while in the second it is controlled by alice , transferred through the quantum channel and exposed to eve s deeds .
eve might attack the extended space @xmath18 , and thus have a different effect on bob , considering his measurement method .
for example , suppose alice sends two non - orthogonal states of a qubit , @xmath20 and @xmath21 , with a fixed and known angle @xmath22 .
bob would like to distinguish between them , while allowing inconclusive results sometimes , but no errors @xcite .
bob can add the ancilla @xmath23 and perform the following transformation @xmath24 : @xmath25 where @xmath26 .
this operation leads to a conclusive result with probability @xmath27 ( when the measured ancilla is @xmath28 ) , and inconclusive result otherwise .
it is simple to see that the same measurement can be done , _ without the use of an ancilla _
, if the states @xmath29 and @xmath30 are embedded at alice s lab in a larger space @xmath18 , e.g. @xmath31 , using bob s transformation @xmath32 in the general case , the space @xmath18 might be very large , even infinite .
bob might use only parts of it , for his measurements .
a complication in performing security analysis is due to bob s option to _ both _ use an ancilla and extend the space used by alice .
our analysis in the following sections starts with the space extension only ( sections [ sec : hb][sec : hp ] ) , and later on deals with the general case ( sections [ sec : hb+anc][sec : hp+anc ] ) .
let us formulate the spaces involved in the protocol , as described above .
assume alice uses the space @xmath10 according to definition [ def : ha ] , which is embedded in a ( potentially larger ) space @xmath18 .
ideally , in the bb84 protocol , bob would like to measure just the states in @xmath10 , but in practice he usually can not do so .
each one of alice s states @xmath33 is transformed by bob s equipment into some pure . for the notion of mixed states or quantum mixture see @xcite .
] state @xmath34 .
the space which is spanned by those states contains all the information about alice s states @xmath35 .
more important , bob might be measuring un - needed subspaces of @xmath18 which alice s states do not span .
for instance , examine the case where bob uses detectors to measure the fock states @xmath36 and @xmath37 .
bob is usually able to distinguish a loss ( the state @xmath8 ) or an error ( e.g. @xmath38 , one horizontal photon and one vertical photon ) , from the two desired states , but he can not distinguish between other states containing multiple photons .
this means that bob measures a much larger subspace of the entire space @xmath18 , but ( inevitably ) interprets outcomes outside @xmath10 as legitimate states ; e.g. the states @xmath39 , @xmath40 , etc .
are ( mistakenly ) interpreted as @xmath36 .
see further discussion in section [ sec : photonicextension ] .
we denote bob s setup ( beam splitters , phase shifters , etc . ) by the unitary operation @xmath41 , followed by a measurement ; all these operations are operating on the space @xmath18 ( or parts of it ) .
bob might have several different setups ( e.g. a different setup for the @xmath42-basis and for the @xmath43-basis ) .
let @xmath44 be the set of unitary transformations in all bob s setups .
[ def : hb ] * [ this definition is temporary . ] * given a specific setup - transformation @xmath45 , let @xmath46 be the subsystem actually measured by bob , having @xmath47 basis states @xmath48 .
the set of * bob s measured spaces * is the set @xmath49 of @xmath50 spaces .
we have already seen that bob might be measuring un - needed dimensions . on the other hand he might not measure certain subspaces of @xmath18 , even when alice s state might reach there . in either case , the deviation is commonly due to limitations of bob s equipment .
the `` quantum space of the protocol '' ( qsop ) is in fact alice s _ extended _ space , taking into consideration its _ extensions _ due to bob s measurements .
the security analysis of a protocol depends on the space @xmath51 defined below .
[ def : hb-1 ] * [ this definition is temporary . ] * * the reversed space @xmath51 * is the hilbert space spanned by the states @xmath52 , for each possible setup @xmath53 , and for each basis state @xmath54 of the appropriate @xmath55 .
the space @xmath51 usually resides in a larger space than @xmath10 .
for instance , using photons , the ideal space @xmath10 consists of two modes with 2 basis states , see section [ sec : qsaphotonicworld ] .
now @xmath51 could have an infinite space in each mode , but also could have more modes . in order to derive the quantum space of the protocol we need to define the way alice s space
is extended according to @xmath51 , for this simple case where bob does not add an ancilla . in this case
, the space @xmath51 simply extends alice s space to yield the qsop via @xmath56 . formally speaking [ def : hp ]
* [ this definition is temporary . ] * * the quantum space of the protocol * , @xmath57 , is the space spanned by the basis states of the space @xmath10 and the basis states of the space @xmath51 . if alice s realistic space is fully measured by bob s detection process , then @xmath10 is a subspace of @xmath51 , hence @xmath58 . in the general case , one must consider bob s option to add an ancilla during his measurement process .
this addition causes a considerable difficulty in analyzing a protocol , however it is often an inherent part of the protocol , and can not be avoided .
we denote the added ancilla as the state @xmath59 that resides in the space @xmath60 .
[ def : m+anc ] @xmath61 is the space that includes the physical space used by alice as defined in definition [ def : m ] , in addition to bob s ancilla , @xmath62 .
bob measures a subspace of the space @xmath61 , so the ( permanent ) definitions of his measured spaces @xmath63 and the reversed space @xmath51 should be modified accordingly .
[ def : hb+anc ] given a specific setup - transformation @xmath45 let @xmath64 be the subsystem actually measured by bob , having @xmath47 basis states @xmath48 .
the set of * bob s measured spaces * , is the set @xmath49 of @xmath50 spaces .
the quantum space of the protocol is still alice s _ extended _ space , while considering its _ extensions _ due to bob s measurements .
yet , the added ancilla makes things much more complex .
the security analysis of a protocol depends now _ not _ on the space @xmath51 defined below , but on a ( potentially _ much larger _ ) space obtained from it by tracing - out bob s ancilla .
as before , we first define the reversed space .
[ def : hb-1+anc ] * the reversed space @xmath51 * is the hilbert space spanned by the states @xmath52 , for each possible setup @xmath53 , and for each basis state @xmath54 of the appropriate @xmath65 .
once a basis state of one of bob s measured spaces @xmath54 is reversed by @xmath66 we result with a state that might , partially , reside in bob s ancillary space @xmath60 .
since eve has no access to this space it must be traced - out ( separated out ) , for deriving the qsop .
let us redefine the qsop given the addition of the ancilla : [ def : hp+anc ] * the quantum space of the protocol , @xmath57 * , is the space spanned by * ( a ) * the basis states of the space @xmath10 ; and * ( b ) * the states @xmath67 $ ] , ( namely , after tracing out bob ) , for each possible setup @xmath53 , and for each basis state
@xmath54 of the appropriate space @xmath63 . whenever @xmath41 entangles bob s ancilla with the system sent from alice , tracing out bob s ancilla after performing @xmath68 might cause an increase of the qsop to the dimension of bob s ancillary space . for instance , assume alice s state is embedded in an @xmath69-qubit space to which bob adds an ancilla of @xmath69-qubits and performs a unitary transformation @xmath24 , such that for one state measured by bob , @xmath70 . tracing out bob from this state yields the maximally mixed state @xmath71 , so that in this example the whole @xmath69-qubits space is spanned .
when alice and bob use qubits , in theoretical qkd , eve can attack the protocol in many ways . in her simplest attack , the so - called `` measure - resend attack '' ,
eve performs any measurement ( of her choice ) on the qubit , and accordingly decides what to send to bob .
a generalization of that attack is the `` translucent attack '' , in which eve attaches an ancilla , in an initial state @xmath72 ( and in any dimension she likes ) , and entangles the ancilla and alice s qubit , using @xmath73 where @xmath74 is a basis for alice s qubit , and eve s states after the unitary transformation are @xmath75 . using this transformation
one can define the most general `` individual - particle attack '' @xcite , and also the most general `` collective attack '' @xcite . in the individual - particle attack eve
delays the measurement of her ancilla till after learning anything she can about the qubit ( e.g. , its basis ) , while in the collective attack eve delays her measurements further till she learns anything she can about _ all _ the qubits ( e.g. , how the final key is generated from the obtained string of shared bits ) , so she attacks directly the _
final key_. the most general attack that eve could perform on the channel is to attack all those qubits transmitted from alice to bob , using _
one _ large ancilla .
this is the `` joint attack '' .
security , in case eve tries to learn a maximal information on the final key , was proven in @xcite via various methods .
the attack s unitary transformation is written as before , but with @xmath76 a binary string of @xmath69 bits , and so is @xmath77 , @xmath78 . by replacing the qubit space @xmath0 by alice s realistic `` qubit '' in the space @xmath10 , and by defining eve s attack on the entire space of the protocol @xmath57 , we can generalize each of the known attacks on theoretical qkd to a `` quantum space attack '' ( qsa ) .
we can easily define now eve s most general _ individual - transmission qsa _ on a realistic `` qubit '' , which generalizes the individual - particle attack earlier described .
eve prepares an ancilla in a state @xmath79 , and attaches it to alice s state , but actually her ancilla is now attached to the entire qsop .
eve performs a unitary transformation @xmath80 on the joint state .
if eve s attack is only on @xmath10 , we write the resulting transformation on any basis state of @xmath10 , @xmath74 , as @xmath81 , where the sum is over the dimension of @xmath10 .
the photon - number - splitting attack ( see section [ sec : pnsasqsa ] ) is an example for such an attack .
the most general individual - transmission qsa is based on a translucent qsa on the qsop , @xmath82 where the sum is over the dimension of @xmath57 .
the subsystem in @xmath57 is then sent to bob while the rest ( the subsystem @xmath83 ) is kept by eve .
we write the transformation on any basis state of @xmath57 , @xmath84 , but note that it is sufficient to define the transformation on the different states in @xmath10 , namely for all states of the form @xmath74 , since other states of the qsop are never sent by alice ( any other additional subsystem of the qsop is necessarily at a known state when it enters eve s transformation ) .
attacks that are more general than the _ individual transmission qsa _ , the _ collective qsa _ and the _ joint qsa _ , can now be defined accordingly . in the most general collective qsa ,
eve performs the above translucent qsa on many ( say , @xmath69 ) realistic `` qubits '' ( potentially a different attack on each one , if she likes ) , waits till she gets all data regarding the generation of the final key , and she then measures all the ancillas together , to obtain the optimal information on the final key or the final secret .
the most general attack that eve could perform on the channel is to attack all those realistic `` qubits '' transmitted from alice to bob , using _
one _ large ancilla .
this is the `` joint qsa '' . the attack s unitary transformation is written as before , but with @xmath76 a string of @xmath69 digits rather than a single digit ( digits of the relevant dimension of @xmath57 ) , and so is @xmath77 , @xmath85 eve measures the ancilla , after learning all classical information , to obtain the optimal information on the final key or the final secret . as before ,
it is sufficient to define the transformation on the different input states from @xmath86 .
we would like to emphasize several issues : 1. when analyzing specific attacks , or when trying to obtain a limited security result , it is always legitimate to restrict the analysis to the relevant ( smaller ) subspace of the qsop , for simplicity , e.g. , to @xmath10 , or to @xmath51 , etc .
2. any bi - directional protocol will have a much more complicated qsop , thus it might be extremely difficult to analyze any type of qsa ( even the simplest ones ) on such protocols .
this remark is especially important since bi - directional protocols play a very important role in qkd , since they appear in many interesting protocols such as the plug - and - play @xcite , the ping - pong @xcite , and the classical bob @xcite protocols .
specifically they provided ( via the plug - and - play ) the only commerical qkd so far @xcite .
3. it is well known that the collective or joint attack is only finished after eve gets all quantum and classical information , since she delays her measurements till then @xcite ; if she expects more information , she better wait and attack the final secret rather than the final key ; it is important to notice that if the key will be used to encode quantum information ( say , qubits ) then the quantum - space of the protocol will require a modification , potentially a major one ; it is interesting to study if this new notion of qsop has an influence on analysis of such usage of the key as done ( for the ideal qubits ) in @xcite .
since most of the practical qkd experiments and products are done using photons , in this section we demonstrate our qsop and qsa definitions and methods via photons .
our analysis uses the fock - space notations for describing photonic quantum spaces . for clarity , states written using the fock notation are denoted with the superscript ` f ' , e.g. @xmath87 , @xmath88 , and @xmath89 .
a photon can not be treated as a quantum system in a straightforward way .
for instance , unlike dust particles or grains of sand , photons are indistinguishable particles , meaning that when a couple of photons are interacting , one can not define the evolution of the specific particle , but rather describe the whole system .
let us examine a cavity , for instance .
it can contain photons of specific wavelengthes ( @xmath90 , @xmath91 , etc . ) and the energy of a photon of wavelength @xmath92 is directly proportional to @xmath93 . while one can not distinguish between photons of the same wavelength
, one can distinguish between photons of different wavelengths .
therefore , it is convenient to define distinguishable `` photonic modes '' , such that each wavelength corresponds to a specific mode ( so a mode inside a cavity can be denoted by its wavelength ) , and then count the number of photons in each mode . if a single photon in a specific mode carries some unit of energy , then @xmath69 such photons of the same wavelength carry @xmath69 times that energy .
if the cavity is at its ground ( minimal ) energy level , we say that there are `` no photons '' in the cavity and denote the state as @xmath87the vacuum state . the convention is to denote only those modes that are potentially populated , so if we can find @xmath69 photons in one mode , and no photons in any other mode , we write , @xmath94 . if two modes are populated by @xmath95 and @xmath96 photons , and all other modes are surely empty , we write @xmath97 ( or @xmath98 ) .
when there is no danger of confusion , and the number of photons per mode is small ( smaller than ten ) , we just write @xmath99 for @xmath100 photons in one mode and @xmath69 in the other .
in addition to its wavelength , a photon also has a property called polarization , and a basis for that property is , for instance , the horizontal and vertical polarizations mentioned earlier .
thus , two modes ( in a cavity ) can also have the same energy , but different polarizations . outside
a cavity photons travel with the speed of light , say from alice to bob , yet modes can still be described , e.g. , by using `` pulses '' of light @xcite .
the modes can then be distinguished by different directions of the light beams ( or by different paths ) , or by the timing of pulses ( these modes are denoted by non - overlapping time - bins ) , or by orthogonal polarizations . a proper description of a photonic qubit is commonly based on using two modes ` @xmath101 ' and ` @xmath102 ' which are populated by exactly a single photon , namely , a photon in mode @xmath101 , so the state is @xmath103 , or a photon in mode @xmath102 , so the state is @xmath104 .
however , a quantum space that consists of a single given photonic mode ` @xmath101 ' is not restricted to a single photon , and can be populated by any number of photons . a basis for this space is @xmath105 with @xmath106 , so that the quantum space is infinitely large , @xmath107 .
theoretically , a general state in this space is can be written as the superposition @xmath108 , with @xmath109 , @xmath110 .
similarly , a quantum space that consists of two photonic modes has the basis states @xmath97 , for @xmath111 and a general state is of the form @xmath112 with @xmath113 , @xmath114 .
this quantum space is described as a tensor product of two `` systems '' @xmath115 . using _
exactly _ two photons in two different ( and orthogonal ) modes assists in clarifying the difference between photons and dust particles ( or grains of sand ) : due to the indistiguishability of photons , only 3 different states can exist ( instead of 4 ) : @xmath116 , @xmath117 and @xmath118 .
the last state has one photon in mode ` @xmath101 ' and another photon in ` @xmath102 ' , however , exchanging the photons is meaningless since one can never tell one photon from another .
a realistic model of a photon source ( in a specific mode ) is of a coherent pulse ( a poissonian distribution ) @xmath119 including terms that describe the possibility of emitting any number @xmath69 of photons . as the number of photons increases beyond some number , the probability decreases , so it is common to neglect the higher orders . in qkd , experimentalists commonly use a `` weak '' coherent state ( such that @xmath120 ) and then terms with @xmath121 can usually be neglected . there is also a lot of research about sources that emit ( to a good approximation ) single photons , and then , again , terms with @xmath121 can usually be neglected .
while the theoretical qubit lives in @xmath0 , a realistic view defines the space actually used by alice to be much larger .
the possibility to emit empty pulses increases alice s realistic space into @xmath9 , due to the vacuum state @xmath122 . when alice sends a qubit using two modes , using a weak coherent state ( or a `` single - photon '' source ) , her realistic space , @xmath10 , is embedded in @xmath123 .
terms containing more than two photons can be neglected , so these are excluded from alice s space @xmath10 .
the appropriate realistic quantum space of alice , @xmath10 , is now a quhexit : the six - dimensional space spanned by @xmath124 , @xmath125 , @xmath126 , @xmath127 , @xmath128 , @xmath129 . the pns attack demonstrated in section [ sec : pnsasqsa ] , is based on attacking this 6 dimensional space @xmath10 .
note also that terms with more than two photons still appear in @xmath18 , and thus could potentially appear in the qsop ( and then used by eve ) . at times ,
alice s realistic space is even larger , due to extra modes that are sent through the channel , and are not meant to be a part of the protocol .
these extra modes might severely compromise the security of the protocol , since they might carry some vital information about the protocol .
a specific qsa based on that flaw is the `` tagging attack '' ( section [ sec : tagasqsa ] ) . note that even if alice uses exactly two modes , the quantum space @xmath18 where @xmath10 is embedded , certainly contains other modes as well .
let us discuss bob s measurement of photonic spaces .
there are ( mainly ) two types of detectors that can be used .
the common detector can not distinguish a single photon from more than one photon ( these kind of detectors are known as _ threshold detectors _ ) .
the hilbert space where bob s measurement is defined is infinite by some large number @xmath130 . ] , since a click in the detector tells bob that the number of photons occupying the mode is `` not zero '' i.e.the detector clicks when @xmath94 is detected , for @xmath131 .
this means that bob measures the state @xmath87 , or he measures @xmath132 , @xmath133 , @xmath134 but then `` forgets '' how many photons were detected .
bob might severely compromise the security , since he inevitably interprets a measurement of a state containing multiple photons as the `` legal '' state that contains only a single photon .
an attack based on a similar limitation is the `` trojan - pony '' attack described below , in section [ subsec : trojan ] . in order to avoid false interpretations of the photon number reaching the detector
, bob could use an enhanced type of detector known as the _ photon - number resolving detector _ or a _ counter _ ( which is still under development ) .
this device distinguishes a single photon from @xmath135 photons , hence any eavesdropping attempt that generates multi - photon states can potentially be noticed by bob .
a much enhanced security can be achieved now , although the qsop is infinite also in this case , due to identifying correctly the legitimate state @xmath132 , from various legitimate states .
the number of modes in the qsop depends on bob s detectors as well .
bob commonly increases the number of measured modes by `` opening '' his detector for more time - bin modes or more frequency modes .
for instance , suppose bob is using a detector whose detection time - window is quite larger than the width of the pulse used in the protocol , since he does not know when exactly alice s pulse might arrive .
the result is an extension of the space used by alice , so that the qsop includes the subspace of @xmath18 that contains all these measured modes .
when a single detector is used to measure more than one mode _ without distinguishing them _ , the impact on the security might be severe , see the `` fake state '' attack ( section [ sec : fakeasqsa ] ) .
in addition to the known attacks described in the following subsection , a new qsa is analyzed in section [ sec : interferobb84 ] , where we examine the more general case of qsa , in which bob adds an ancilla during the process .
all known attacks can be considered as special cases of the quantum - space attack . in this section
we show a description of several such attacks using qsa terms . for each and every attack we briefly describe the specific protocol used , the quantum space of the protocol , and a realization of the attack as a qsa .
* the protocol . * consider a bb84 protocol , where alice uses a `` weak pulse '' laser to send photons in two modes corresponding to the vertical and horizontal polarizations when using the @xmath42 basis ( the diagonal polarizations then relate to using the @xmath43 basis ) .
bob uses a device called a pockel cell to rotate the polarization ( by @xmath136 ) for measuring the @xmath43 basis , or performs no rotation if measuring the @xmath42 basis .
the measurement of the state is then done using two detectors and a `` polarization beam splitter '' that passes the first mode to one detector and the second mode to the other detector ( for a survey of polarization - based qkd experiments , see @xcite ) .
* the quantum space of the protocol . * every pulse sent by alice is in one of four states , each in a superposition of the 6 orthogonal states @xmath137 , @xmath125 , @xmath138,@xmath127,@xmath128 , @xmath139 , where the space used by alice is @xmath140 .
bob uses two setups , @xmath141 for the @xmath42 basis , and @xmath142 for the @xmath43 basis , which is more complex and described in appendix [ app : polu ] .
the detectors used by bob can not distinguish between modes having single photon and multiple photons .
each one of his two detectors measures the basis elements @xmath143 for @xmath144 ( of the specific mode directed to that specific detector ) , where bob interprets the states @xmath143 with @xmath145 as measuring the state @xmath132 of the same mode .
bob s measured space @xmath146 is thus infinite and spanned by the states @xmath147 for @xmath148 .
the qsop @xmath57 is equal to @xmath149 ( @xmath150 ) since performing @xmath151 does not change the dimensionality of the spanned space ( in both setups ) . *
the attack . * eve measures the number of photons in the pulse , using non - demolition measurement . if she finds that the number of photons is @xmath152 , she blocks the pulse and generates a loss . in the case
she finds that the pulse consists of 2 photons , she splits one photon out of the pulse and sends it to bob , keeping the other photon until the bases are revealed , thus getting full information of the key - bit .
eve sends the eavesdropped qubits to bob via a lossless channel so that bob will not notice the enhanced loss - rate . as is common in experimental qkd , bob is willing to accept a high loss - rate ( he does not count losses as errors ) , since most of alice s pulses are empty .
see the precise mathematical description of this attack in appendix [ app : mathpns ] .
* the protocol . * consider a bb84 qkd protocol in which alice sends an enlarged state rather than a qubit .
this state contains , besides the information qubit , a _ tag _ giving eve some information about the bit .
the tag can , for example , tell eve the basis being used by alice . for a potentially realistic example , let the tag be an additional qutrit indicating if alice used the @xmath43-basis , or the @xmath42-basis , or whether the basis is _ unknown _ : whenever alice switches basis , a single photon comes out of her lab prior to the qubit - carrying pulse , telling the basis , say using the states @xmath153 and @xmath154 , and when there is no change of basis , what comes out prior to the qubit is just the vacuum @xmath155 . * the quantum space of the protocol . * in this example
, alice is using the space @xmath156 .
bob , unaware of the enlarged space used by alice , expects and receives only the subspace @xmath0 .
we assume that bob ideally measures this space with a single setup @xmath157 , therefore @xmath158 .
since bob s setup does not change the space , @xmath159 as well . however , the tag is of a much use to eve , and indeed the qsop following definition [ def : hp ] , defined to be @xmath160 . * the attack .
* eve uses the tag in order to retrieve information about the qubit without inducing error ( e.g.via cloning the qubit in the proper basis ) .
the attack is then an intercept - resend qsa .
we mention that this attack is very similar to a side - channel cryptanalysis of classic cryptosystems .
12 pt * a short summery . * it can be seen that the pns attack described above is actually a special case of the tagging attack , where the _ tag _ in that case is in fact another copy of the transmitted qubit .
this copy is kept by eve until the bases are revealed , then it can be measured so the the key - bit value is exposed with certainty .
both those qsa attacks are based on the fact that alice ( realistic ) space is larger than the theoretical one .
although in the pns example , the qsop is further extended due to bob s measurement , the attack is not based on that extension but on the fact that @xmath10 is larger than @xmath0 . in the following attacks
bob s measurements cause the enlargement of the qsop , allowing eve to exploit the larger qsop for her attack . in trojan - pony attacks
eve modifies the state sent to bob in a way that gives her information .
in contrast to a `` trojan - horse '' that goes in - and - out of bob s lab , the `` pony '' only goes in , therefore , it is not considered an attack on the lab , but only on the channel .
we present here an interesting example @xcite . *
the protocol .
* assume a polarization - encoded bb84 protocol , in which alice is ideal , namely , sending perfect qubits ( @xmath161 ) .
however , bob uses realistic threshold detectors that suffer from losses and dark counts , and that can not distinguish between one photon and @xmath162 photons for @xmath163 . in order to be able to `` prove '' security , for a longer distance of transmission bob wants to keep the error - rate low although the increase of dark counts impact with the distance @xcite .
therefore , bob assumes that eve has no control over dark counts , and whenever both detectors click , alice and bob agree to consider it as _ a loss _ since it is outside of eve s control ( i.e. the qsop is falsely considered to be @xmath0 ) .
namely , they assume that _ an error _ occurs only when bob measures in the right basis , and only one detector clicks , ( which is the detector corresponding to the wrong bit - value ) .
* the quantum space of the protocol . * same as in section [ sec : pnsasqsa ] , bob s measured spaces @xmath149 , @xmath164 , the reversed space @xmath51 as well as the qsop @xmath57 , are merely the spaces describing two modes ( with up to @xmath130 photons ) , @xmath165 .
bob s detectors can not distinguish between receiving a single - photon pulse from a multi - photon pulse , so his measurement is properly described as a projection of the received state onto the space containing @xmath166 followed by `` forgetting '' the exact result , and keeping only one of three results : `` @xmath167 detector-1 clicks '' , `` @xmath168 detector-2 clicks '' , and else it is @xmath169 , a `` loss '' . in formal , _ generalized - measurements _ language ( called povm , see @xcite ) these three possible results are written as : @xmath170 , @xmath171 , @xmath172 , and their sum is the identity matrix . * the attack . * eve s attack is the following : ( a ) randomly choose a basis ( b ) measure the arriving qubit in that specific chosen basis ( c ) send bob @xmath100-photons identical to the measured qubit , where @xmath173 . obviously , when eve chooses the same basis as alice and bob then bob measures
the exact value sent by alice , and eve gets full information .
otherwise , both of his detectors click , implying a `` loss '' , except for a negligible probability , @xmath174 , thus eve induces no errors .
the main observation of this measure - resend qsa is that treating a count of more than a single photon as a loss , rather than as an error , is usually not justified .
a second conclusion is that letting bob use counters instead of threshold detectors ( to distinguish a single photon from multiple photons ) , together with treating any count of more than one photon as an error , could be vital for proving security against qsa .
the price is that dark counts put severe restrictions on the distance to which communication can still be considered secure , as suggested already by @xcite . *
the protocol .
* in this example , we examine a polarization encoded bb84 protocol , and an ideal alice ( @xmath175 ) .
this time bob s detectors are imperfect so that their detection windows do not fully overlap , meaning that there exist times in which one detector is blocked ( or it has a low efficiency ) , while the other detector is still regularly active .
thus , if eve can control the precise timing of the pulse , she can control whether the photon will be detected or lost .
the setup is built four detectors and a rotating mirror ( since bob does not want to spend money on a pockel cell ( polarization rotator ) , he actually uses 2 fixed different setups ) .
using the rotating mirror bob sends the photon into a detection setup for basis @xmath42 or a detection setup for basis @xmath43 .
suppose the two detection setups use slightly different detectors , or slightly different delay lines , or slightly different shutters , and eve is aware of this ( or had learnt it during her past attacks on the system ) . for simplicity
, we model the non - overlapping detection windows , as additional two modes , one slightly prior to alice s intended mode ( the pulse ) , and one right after it . * the quantum space of the protocol . *
the original qubit is sent in a specific time - bin @xmath176 ( namely , @xmath161 ) .
the setup @xmath177 is a set of two detectors and a polarized beam splitter , separating the horizontal and the vertical modes to the detectors , where @xmath178 separate the diagonal modes into a set of two ( different ) detectors .
let the detectors for one basis , say @xmath42 , be able to measure a pulse arriving at @xmath176 or @xmath179 , while the detectors for the other basis ( @xmath43 ) measure pulses arriving at @xmath180 or @xmath176 . for simplicity ,
we degenerate the space to contain one or less photons . ] , so that @xmath149 is @xmath181 , i.e. two possible time - bins consisting each of two ( polarization ) modes of one or less photons .
the measured space of the @xmath43-setup has two possible time - bins and two possible polarization modes , thus @xmath182 as well , however , the two time - bins for this setup are @xmath176 and @xmath179 .
following definition [ def : hb-1 ] we get that that the reversed space @xmath51 contains three time - bins ( @xmath180 , @xmath176 and @xmath179 ) with two polarization modes in each , therefore @xmath183 , under the single - photon assumption .
the qsop , following definition [ def : hp ] equals @xmath51 since @xmath184 . * the attack . *
eve exploit the larger space by sending `` fake '' states using the external time bins ( @xmath180 and @xmath179 ) .
eve randomly chooses a basis , measures the qubit sent by alice , and sends bob the same polarization state she found , but at @xmath180 if she have used the @xmath43 basis , or at @xmath185 if she have used the @xmath42 basis . since no ancilla is kept by eve , this is an intercept - resend qsa .
bob will get the same result as eve if he uses the same basis , or _ a loss _ otherwise .
the mathematical description of the attack is as follows : eve can generate superpositions of states of the form @xmath186 , where the index @xmath187 denotes this mode has vertical or horizontal polarization , and its subscript denotes the time - bin of the mode .
eve s measure - resend attack is described as measuring alice s qubit in the @xmath43 basis , creating a new copy of the measured qubit , and performing the transformation @xmath188 ; @xmath189 or as performing a measurement in the @xmath42 basis , and performing the transformation @xmath190 ; @xmath191 on the generated copy .
12 pt * a short summery * we see that eve can `` force '' a desired value ( or a loss ) on bob , thus gaining all the information while inducing no errors ( but increasing the loss rate ) .
bob can use a shutter to block the irrelevant time - bins but such a shutter could generate a similar problem in the frequency domain .
this attack is actually a special case of the trojan - pony attack , in which the imperfections of bob s detectors allow eve to send states that will be un - noticed unless the measured basis equals to eve s chosen basis .
in order to demonstrate the power of qsa , and to see its advantages , this section presents a partial security analysis of some interferometric bb84 and 6-state schemes .
interferometric schemes are more common than any other type of implementation in qkd experiments @xcite and products @xcite . in this section
we define the specific equipment used by bob , and we formulate @xmath41 and bob s measurements .
we then find the spaces @xmath10 , @xmath63 , @xmath51 and the qsop , @xmath57 . finally , we demonstrate a novel attack which is found to be very successful against a specific variant of the bb84 interferometric scheme ; this specific qsa , which we call the `` reversed - space attack '' , is designed using the tools developed in sections [ sec : qsop ] and [ sec : qsa ] .
we begin with a description of interferometric ( bb84 and six - state ) schemes , which is based on sending phase - encoded qubits arriving in two time - separated modes @xcite .
alice encodes her qubit using two time - bins @xmath192 and @xmath193 , where a photon in the first mode , @xmath194 , represents the state @xmath1 , and a photon in the other mode , @xmath195 , represents @xmath2 .
the bb84 protocol of @xcite ( and many others ) uses the @xmath43 and @xmath196 bases , meaning that alice ( ideally ) sends one of the following four states : @xmath197 ; @xmath198 ; @xmath199 ; and @xmath200 .
bob uses an interferometer built from two beam splitters with one short path and one long path ( figure [ fig : lab - xy ] ) . a pulse of light travels through the short arm of the interferometer in @xmath201 seconds , and through the long arm in @xmath202 seconds , where @xmath203 is also _ precisely _ the time separation between the two arriving modes of the qubit , @xmath204 . a controlled phase shifter @xmath205 ,
is placed in the long arm of the interferometer .
it performs a phase shift by a given phase @xmath206 , i.e. @xmath207 . the phase shifter is set to @xmath208 ( @xmath209 ) when bob measures the @xmath43 ( @xmath196 ) basis . and @xmath196 basis .
( a ) alice sends a qubit ; ( b ) vacuum states are added in the interferometer ; ( c ) , ( d ) beam - splitters ; ( e ) phase shifter @xmath205 . ] each beam splitter interferes two input arms ( modes 1 , 2 ) into two output arms ( modes 3 , 4 ) , in the following way ( for a single photon ) : @xmath210 , and @xmath211 .
the photon is transmitted / reflected with a probability of @xmath212 ; the transmitted part keeps the same phase as the incoming photon , while the reflected part gets an extra phase of @xmath213 , if it carries a single photon .
when a single mode , carrying at least a single photon , enters a beam splitter from one arm , and nothing enters the other input arm , we must consider the other entry to be an additional mode ( an ancilla ) in a vacuum state .
when a single mode ( carrying one or more photons ) enters the interferometer at time @xmath192 , see figure [ fig : lab - xy ] , it yields two modes at time @xmath176 due to traveling through the short arm , and two modes at time @xmath179 due to traveling through the long arm .
those four output modes are : times @xmath176 , @xmath179 in the ` @xmath214 ' ( straight ) arm of the interferometer , and times @xmath176 , @xmath179 in the ` @xmath215 ' ( down ) arm .
a basis state in this fock space is then @xmath216 . in the case of having that single mode carrying exactly a single photon , the transformation , which requires three additional empty ancillas .
] , is @xmath217 . note that a pulse which is sent at a different time ( say , @xmath218 ) results in the same output state , but with the appropriate delays , i.e. @xmath219 where the resulting state is defined in the fock space whose basis states are @xmath220 .
let us now examine any superposition of two modes ( @xmath192 and @xmath193 ) that enter the interferometer one after the other , with exactly the same time difference @xmath203 as the difference lengths of the arms .
the state evolves in the following way ( see appendix [ app : modesevo ] ) : @xmath221 describing the evolution for any possible bb84 state sent by alice ( @xmath222 , @xmath223 , @xmath224 , @xmath225 determined by the value of @xmath226 , @xmath227 , @xmath228 , @xmath229 respectively , when @xmath230 ) . as a result of this precise timing
, these two modes are transformed into a superposition of 6 possible modes ( and not 8 modes ) at the outputs , due to interference at the second beam splitter .
only four vacuum - states ancillas ( and not six ) are required for that process .
the resulting 6 modes are @xmath176 , @xmath179 , @xmath231 in the ` @xmath214 ' arm and in the ` @xmath215 ' arm of the interferometer .
denote this fock space as @xmath232 , with basis elements @xmath233 .
the measurement is performed as follows : bob opens his detectors at time @xmath179 in both output arms of the interferometer .
a click in the `` down '' direction means measuring the bit - value @xmath234 , while a click in the `` straight '' direction means @xmath235 .
the other modes are commonly considered as a loss ( they are not measured ) since they give an inconclusive result regarding the original qubit .
we refer this bb84 variant as `` @xmath236-bb84 '' .
one might want to use the @xmath42 basis in his qkd protocol ( using @xmath226 , and @xmath237 or @xmath238 ) , for instance , in order to avoid the need for a controlled phase shifter or for another equipment - related reason , or in order to perform `` qkd with classical bob '' @xcite .
a potentially more important reason might be to perform the 6-state qkd @xcite protocol , due to its improved immunity against errors ( 27.4% errors versus only 20% in bb84 @xcite ) .
a possible and easy to implement variant for realizing a measurement in the @xmath42 basis is the following : bob uses the setup @xmath239 ( i.e. he sets @xmath205 to @xmath240 ) , and opens his detectors at times @xmath176 and @xmath231 , corresponding to the bit - values @xmath234 and @xmath235 respectively ( see equation ( [ eqn : interf_evu ] ) ) . unfortunately
, technological limitations , e.g. of telecommunication wavelength ( ir ) detectors , might make it difficult for bob to open his detectors for more than a single detection window per pulse .
bob could perform a measurement of _ just _ the states @xmath241 , opening the @xmath215 arm detector at time @xmath176 ( to measure @xmath1 ) and the @xmath214 arm detector at time @xmath231 ( to measure @xmath2 ) .
we refer this variant as `` @xmath242-six - state '' .
we assume alice to be almost ideal , having the realistic space @xmath243 ( a qubit or a vacuum state ) , using two time - bin modes .
as we have seen , four ancillary modes in vacuum states are added to each transmission .
therefore , the interferometer setups @xmath239 and @xmath244 transform the 2-mode states of @xmath10 into a subspace that resides in the 6 modes space @xmath232 . for simplicity
, we assume that eve does not generate @xmath69-photon states , with @xmath135 , so we can ignore high photon numbers in the @xmath232 space , this assumption is not legitimate when proving unconditional security of a protocol . ] .
therefore , we redefine @xmath245 , the space spanned by the vacuum , and the six single - photon terms in each of the above modes . using the @xmath43 and @xmath196 bases , bob measures only time - bin @xmath179 , so his actual measured spaces consist of two modes : time - bin @xmath179 in the ` @xmath214 ' arm and the ` @xmath215 ' arm . in that case , the measured spaces are @xmath246 , spanned by the states @xmath247 , @xmath248 , @xmath249 .
when bob uses the @xmath42 basis , he measures two different modes , so @xmath149 is spanned by the states @xmath247 , @xmath250 , @xmath251 .
let us define the appropriate space @xmath51 for the 6-state protocol , according to definition [ def : hb-1+anc ] .
the space @xmath51 is spanned by the states given by performing @xmath252 on @xmath253 , @xmath254 , @xmath255 , as well as the states given by performing @xmath256 on @xmath253 , @xmath257 , @xmath258 .
interestingly , once applying @xmath151 , the resulting states are embedded in an 8-mode space defined by the two incoming arms of the interferometer , ` @xmath101 ' ( from alice ) and ` @xmath102 ' ( from bob ) , at time bins @xmath259 , @xmath260 , @xmath261 , and @xmath262 .
the basis states of @xmath51 are listed in appendix [ app : interf_hb-1_basis ] . following definition [ def : hp+anc ] , the qsop @xmath57 of this implementation for the 6-state protocol , is the subsystem of @xmath51 which is _ controlled _ by eve .
it is spanned by the 8-mode states spanning @xmath51 after tracing out bob .
the space that contains those `` traced - out '' states has only four modes that are controlled by eve , specifically , input ` @xmath101 ' of the interferometer at times @xmath259 to @xmath263 , having a basis state of the form @xmath264 . given the single - photon restriction , we get @xmath265 , namely , the space spanned by the vacuum state , and a single photon in each of the four modes , i.e. @xmath266 , @xmath267 , @xmath268 , @xmath269 , @xmath270 . this same result is obtained also if bob measures all the six modes in @xmath232 .
bob might want to see how the basis states of the 4-mode qsop , @xmath57 , evolve through the interferometer in order to place detectors on the resulting modes , which will be used to identify eve s attack .
it is interesting to note , that those basis states result in _ 10 different non - empty modes ( ! ) _ .
if bob measures all these modes , he _ increases _ the qsop , and maybe allows eve to attack a larger space , and so on and so forth .
therefore , in order to perform a security analysis , one must first fix the scheme and only then assess the qsop . otherwise , a `` ping - pong '' effect might increase the spaces dimensions to infinity . a similar ,
yet reversed logic , hints that it could actually be better for bob , in terms of the simplicity of the analysis for the `` @xmath236-bb84 '' scheme , to measure _ just _ the two modes at @xmath179 ( i.e. the space spanned by @xmath271 ) , thus reducing the qsop to a 2-mode space , @xmath272 , see appendix [ app : qsop2modesintf ] .
although eve is allowed to attack a larger space than this two - mode @xmath57 , she has no advantage in doing so : pulses that enter the interferometer on different modes ( i.e. other time - bins than @xmath192 and @xmath193 ) , never interfere with the output pulses of time - bin @xmath179 measured by bob .
therefore , state occupying different modes can not be distinguished from the states in which those modes are empty . consider a bb84 variant in which bob uses only the @xmath43 and the @xmath42 bases , using a single interferometer , where the @xmath42-basis measurement is performed according to the description in the last few lines of section [ sec : xy - setup ] .
we refer this variant as `` @xmath273-bb84 '' .
the qsop of this scheme , @xmath57 is the space described above for the `` @xmath242-six - state '' protocol .
the following attack @xmath274 which we call `` the reversed - space attack '' , allows eve to acquire information about the transmitted qubits , without inducing _ any _ errors .
the states @xmath275 denote eve s ancilla which is not necessarily a photonic system .
the state @xmath276 and @xmath277 are the regular states send by alice , where we added the relevant extension of @xmath10 in @xmath57 .
when @xmath278 is sent by alice , the attacked state @xmath279 reaches bob s interferometer , and interferes in a way such that it can never reach bob s detector at time @xmath231 , i.e. @xmath280 . although the attacked state @xmath281 reaches modes that alice s original state @xmath278 can never reach , bob never measures those modes , and can not notice the attack .
a similar argument applies when alice sends @xmath282 . as for the @xmath43 basis , etc . ]
, this attack satisfies @xmath283
@xmath284 the first element in the sum results in the desired interference in bob s lab , while the second is not measured by bob s detectors at time @xmath179 . by letting eve s probes @xmath285 and @xmath286 be orthogonal states , eve gets a lot of information while inducing no errors at all . yet
, we find that eve is increasing the loss rate by this attack to 87.5% , but a very high loss rate is anyhow expected by bob ( as explained in the analysis of the pns @xcite and the tagging @xcite attacks ) . in conclusion
, this attack demonstrates the risk of using various setups without giving full security analysis for the _ specific _ setup .
we are not familiar with any other security analysis that takes into account the enlarged space generated by the inverse - transformation of bob s space .
in this paper we have defined the qsa , a novel attack that generalizes all currently known attacks on the channel .
this new attack brings a new method for performing security analysis of protocols .
the attack is based on a realistic view of the quantum spaces involved , and in particular , the spaces that become larger than the theoretical ones , due to practical considerations .
although this paper is explicitly focused on the case of uni - directional implementations of a few schemes , its main observations and methods apply to any uni - directional qkd protocol , to bi - directional qkd protocols , and maybe also to any realistic quantum cryptography scheme beyond qkd .
the main conclusion of this research is that the quantum space which is attacked by eve can be assessed , given a proper understanding of the experimental limitations .
this assessment requires a novel cryptanalysis formalism analyzing the states generated in alice s lab , as well as the states that are to be measured by bob ( assessing them as if they go backwards in time from bob s lab ) ; this type of analysis resembles the two - time formalism in quantum theory @xcite .
open problems for further theoretical research include : 1. generalization of the qsa to other conventional protocols ( such as the two - state protocol , epr - based protocols , d - level protocols , etc . )
; such a generalization should be rather straightforward .
2. proving unconditional security ( or more limited security results such as `` robustness '' @xcite ) against various qsas .
this is especially important for the interferometric setup , where the qsop is much larger than alice s six - dimensional space ( the one spanned by @xmath287 ) .
3. describing the qsa for more complex protocols , such as two - way protocols @xcite in which the quantum communication is bi - directional , and protocols which use a larger set of states such as data - rejected protocols @xcite or decoy - state protocols @xcite . 4. extend the analysis and results to composable qkd @xcite .
5(a). in some cases , if bob uses `` counters '' and treats various measurement outcomes as errors , the effective qsop relevant for proving security is potentially _ much smaller _ than the qsop defined here .
5(b). adding counters on more modes increases the qsop defined here , but might allow analysis of a smaller `` attack s qsop '' , if those counters are used to identify eve s attack .
more generally , the connection between the way bob interprets his measured outcomes , and the `` attack s qsop '' is yet to be further analyzed .
[ [ acknowledgments . ] ] * acknowledgments . * + + + + + + + + + + + + + + + + + + we thank michel boyer , dan kenigsberg and hoi - kwong lo for helpful remarks . 10 d. z. albert , y. aharonov , and s. damato .
curious new statistical prediction of quantum mechanics .
, 54(1):57 , jan 1985 .
s. m. barnett , b. huttner , and s. j. d. phoenix . . ,
40:25012513 , dec .
1993 . h. bechmann - pasquinucci and n. gisin . .
, 59(6):42384248 , jun . 1999 .
m. ben - or , m. horodecki , d. w. leung , d. mayers , and j. oppenheim .
the universal composable security of quantum key distribution . in _
tcc 2005 : second theory of cryptography conference _ , pages 386406 , jan . 2005 . c. h. bennett and g. brassard . .
, pages 175179 , dec . 1984 .
e. biham , m. boyer , p. o. boykin , t. mor , and v. p. roychowdhury .
a proof of the security of quantum key distribution . in _ proceedings of the 32nd annual acm symposium on theory of computing ( stoc ) _ , pages 715724 , new york , 2000 .
acm press .
e. biham , m. boyer , p. o. boykin , t. mor , and v. p. roychowdhury .
a proof of the security of quantum key distribution .
, 19(4):381439 , 2006 .
e. biham , m. boyer , g. brassard , j. van de graaf , and t. mor . . , 34:372388 ,
e. biham and t. mor .
security of quantum cryptography against collective attacks .
, 78(11):22562259 , mar 1997 .
k. bostrm and t. felbinger .
deterministic secure direct communication using entanglement .
, 89(18):187902 , oct 2002 .
m. boyer , d. kenigsberg , and t. mor . quantum key distribution with classical bob .
arxiv quantum physics e - prints , 2007 .
quant - ph/0703107 .
g. brassard , n. ltkenhaus , t. mor , and b. c. sanders . .
eurocrypt 2000 : international conference on the theory and application of cryptographic techniques _
, lncs 1807:289299 , 2000 g. brassard , n. ltkenhaus , t. mor , and b. c. sanders . . , 85:13301333 ,
k. j. blow , r. loudon , s. phoenix and t. j. shepherd .
, 42(7):41024114 , oct . 1990 . d. bru . . , 81:30183021 ,
h. f. chau . .
, 66(6):060302 , dec .
_ for different ( slightly smaller ) numbers , see @xcite .
_ m. dusek , n. lutkenhaus , and m. hendrych . .
arxiv quantum physics e - prints , jan .
quant - ph/0601207 . c. elliott , d. pearson , and g. troxel .
quantum cryptography in practice . in _
sigcomm 03 : proceedings of the 2003 conference on applications , technologies , architectures , and protocols for computer communications _ , pages 227238 , new york , ny , usa , 2003 .
acm press .
a. ekert , b. huttner , g. palma and a. peres .
eavesdropping on quantum - cryptographical systems .
, 50(2):10471056 , aug .
1994 . c. fuchs , n. gisin , r. griffiths , c.s .
niu and a. peres .
optimal eavesdropping in quantum cryptography .
i. information bound and optimal strategy .
, 56(2):11631172 , aug . 1997 .
n. gisin , s. fasel , b. kraus , h. zbinden , and g. ribordy . .
, 73(2):022320+ , feb .
n. gisin , b. kraus , and r. renner .
lower and upper bounds on the secret key rate for qkd protocols using one way classical communication .
, 95:080501 , 2005 .
n. gisin , g. ribordy , w. tittel , and h. zbinden . . ,
74:145195 , jan .
d. gottesman , h .- k .
lo , n. ltkenhaus , and j. preskill .
security of quantum key distribution with imperfect devices .
, 5:325360 , 2004 .
quantum key distribution with high loss : toward global secure communication .
, 91(5):057901 , aug .
hwang , i .-
lim , and j .- w .
no - clicking event in quantum key distribution .
arxiv quantum physics e - prints , 2004 .
quant - ph/0412206 .
h. inamori , n. ltkenhaus , and d. mayers . .
, 41:599627 , mar .
lo and h. f. chau . unconditional security of quantum key distribution over arbitrarily long distances . , 283:20502056 , 1999 .
, 1(2):8194 , aug .
lo , x. ma , and k. chen . . ,
94(23):230504+ , jun . 2005 .
v. makarov , a. anisimov , and j. skaar .
effects of detector efficiency mismatch on security of quantum cryptosystems . , 74:022313 , 2006 . c. marand and p. townsend , 20:16951697 , aug .
a. muller , t. herzog , b. huttner , w. tittel , h. zbinden and n. gisin . .
, 70:793395 , feb 1997 . v. makarov and d. r. hjelme . . , 52:691705 ,
may 2005 .
d. mayers .
unconditional security in quantum cryptography .
, 48(3):351406 , 2001 , _ based on @xcite_. a. niederberger , v. scarani , and n. gisin .
photon - number - splitting versus cloning attacks in practical implementations of the bennett - brassard 1984 protocol for quantum cryptography .
, 71:042316 , 2005 .
m. a. nielsen and i. l. chuang . .
cambridge university press , cambridge , uk , 2000 .
a. peres . .
, 128:19 , mar 1988 .
a. peres . .
kluwer , dordrecht , 1993 .
v. scarani , a. acn , g. ridbory and n. gisin .
quantum cryptography robust against photon number splitting attacks for weak laser pulse implementations .
, 92(5):057901+ , feb .
m. scully and m. s. zubairy . .
cambridge university press , cambridge , united kingdom , 1997 .
p. w. shor and j. preskill .
simple proof of security of the bb84 quantum key distribution protocol .
, 85:441444 , 2000 , _ based on @xcite_. p. d. townsend . . , 30:809811
l. vaidman , y. aharonov , and d. z. albert . how to ascertain the values of @xmath288 , @xmath289 , and @xmath290 of a spin-1/2 particle . , 58(14):13851387 , apr .
x. b. wang . .
, 94:230503 , jun . 2005 .
, pages 6775 , las vegas , navada , united states , 1995 .
z. l. yuan , a. w. sharpe , and a. j. shields . .
, 90:1118+ , jan . 2007 .
the pns attack can be realized using ( an infinite set of ) polarization independent beams splitters .
eve uses a beam splitter to split photons from alice s state . using a non - demolition measurement eve measures the number of photons in one output of the beam splitter , and repeat the splitting until she acquires exactly one photon .
formally @xmath80 is defined : @xmath291 whenever alice sends a pulse with two photons of the same polarization , eve and bob end up , each , with having a single photon of the original polarization .
eve s pns attack for a pulse of 2 photons , gives eve full information while inducing no errors .
according to its definition it is trivial to verify the attack for the horizontal and vertical polarizations @xmath292 and @xmath293 ( where @xmath294 means @xmath162 photons having polarization @xmath295 ) . using the standard creation and annihilation operators ( @xmath296 and @xmath101 ) ,
we can write the state of two photons in the diagonal polarization ( @xmath43 basis ) : @xmath297 , similarly @xmath298 .
@xmath299 @xmath300 which completes the proof .
a polarization based qkd protocol makes a use of a pockel cell ( @xmath239 ) , rotating the polarization of the photons going through it . for a single photon ,
its action is trivial , @xmath301 for a state that contains multiple photons , the transformation is not intuitive , and most simply defined using the creation and annihilation operators . in a somewhat simplified way , the pokcel cell can be considered as performing @xmath302 and @xmath303 , so that a state is transformed in the following way @xmath304
in order to simplify the analysis ( a simplification that is not allowed when proving the full security of a scheme ) we look at the ideal case in which exactly one photon ( or none ) is sent by alice .
the basis states are then the vacuum @xmath305 , and the six states ( that we denote for simplicity by ) @xmath306 ; @xmath307 ; @xmath308 ; @xmath309 ; @xmath310 and @xmath311 .
the full transformation of a single photon pulse through the interferometer is given by equation .
alice sends photons at time bins @xmath192 and @xmath193 only , so the interferometer transformation on alice s basis states is @xmath312 , and @xmath313 where @xmath314 denotes ancilla added during the process ) are originated by alice extended space @xmath57 and by bob ( @xmath60 ) . performing @xmath151 reveals the exact origin of those ancillas . ] .
equation [ eqnb+-on01 ] can be used to describe the interferometer effect on a general qubit , shown in equation .
the states sent by alice during the `` @xmath236-bb84 '' protocol evolve in the interferometer as follows : @xmath315 bob can distinguish the computation basis elements of bases @xmath43 and @xmath196 , measuring time - bin @xmath179 , i.e. the states @xmath316 for @xmath317 and @xmath318 for @xmath319 in the measured basis .
other states give bob no information about the state sent by alice .
let bob be using interferometric setups @xmath239 and measuring 6 modes ( corresponding the space with a basis state @xmath320 ) with one or less photons . following definition [ def : hb-1+anc ] , the states spanning the space @xmath51 can be derived using equation ( adjusted to the appropriate space ) : @xmath321 defined over the space @xmath322 with basis state @xmath323 .
note that performing @xmath151 requires an additional ancilla , since the modes number increases from six to eight .
assume bob measures only time - bin @xmath179 in both output arms of the interferometer , i.e. the measured space is @xmath232 subspace spanned by @xmath271 . assuming a single - photon restriction , the reversed space , of that measured space that is spanned by : @xmath324 as can be verified using equation .
the space @xmath51 is embedded in a 4-mode space @xmath325 , having the basis element @xmath326 , i.e. alice modes at times @xmath192 and @xmath193 and bob s added ancillary modes at times @xmath192 and @xmath193 respectively .
the resulting six states span a 4-dimensional space , i.e. @xmath327 .
the qsop in this special case is @xmath328 , spanned by @xmath329 with one or less photons . | theoretical quantum key distribution ( qkd ) protocols commonly rely on the use of qubits ( quantum bits ) . in reality , however , due to practical limitations , the legitimate users are forced to employ a larger quantum ( hilbert ) space , say a quhexit ( quantum six - dimensional ) space , or even a much larger quantum hilbert space .
various specific attacks exploit of these limitations .
although security can still be proved in some very special cases , a general framework that considers such realistic qkd protocols , _ as well as _
attacks on such protocols , is still missing .
we describe a general method of attacking realistic qkd protocols , which we call the ` quantum - space attack ' .
the description is based on assessing the enlarged quantum space actually used by a protocol , the ` quantum space of the protocol ' .
we demonstrate these new methods by classifying various ( known ) recent attacks against several qkd schemes , and by analyzing a novel attack on interferometry - based qkd . |
an important issue in the investigation of the dynamics of competing self - reproducing macromolecules , whose paradigm is eigen s quasispecies model @xcite , is the effect of the finite size of the population on the error threshold phenomenon that limits the length of the molecules @xcite .
the quasispecies model was originally formulated as a deterministic kinetic theory described by a set of ordinary differential equations for the concentrations of the different types of molecules that compose the population . such formulation , however , is valid only in the limit where the total number of molecules @xmath0 goes to infinity .
more pointedly , in this model a molecule is represented by a string of @xmath1 digits @xmath2 , with the variables @xmath3 allowed to take on @xmath4 different values , each of which representing a different type of monomer used to build the molecule . for sake of simplicity ,
in this paper we will consider only binary strings , i.e. , @xmath5 .
the concentrations @xmath6 of molecules of type @xmath7 evolve in time according to the following differential equations @xcite @xmath8 x_i \ ; , \ ] ] where the constants @xmath9 stand for the death probability of molecules of type @xmath10 , and @xmath11 is a dilution flux that keeps the total concentration constant .
this flux introduces a nonlinearity in ( [ ode ] ) , and is determined by the condition @xmath12 .
the elements of the replication matrix @xmath13 depend on the replication rate or fitness @xmath14 of the molecules of type @xmath10 as well as on the hamming distance @xmath15 between strings @xmath10 and @xmath16 .
they are given by @xmath17 and @xmath18 where @xmath19 is the single - digit replication accuracy , which is assumed to be the same for all digits .
henceforth we will set @xmath20 for all @xmath10 .
the quasispecies concept is illustrated more neatly for the single - sharp - peak replication landscape , in which we ascribe the replication rate @xmath21 to the so - called master string @xmath22 , and the replication rate @xmath23 to the remaining strings . in this context , the parameter @xmath24 is termed selective advantage of the master string . as the error rate @xmath25 increases , two distinct regimes are observed in the population composition : the _ quasispecies _ regime characterized by the master string and its close neighbors , and the _ uniform _ regime where the @xmath26 strings appear in the same proportion .
the transition between these regimes takes place at the error threshold @xmath27 , whose value depends on the parameters @xmath1 and @xmath24 @xcite .
a genuine thermodynamic order - disorder phase transition occurs in the limit @xmath28 only @xcite .
we must note , however , that standard statistical mechanics tools developed to study the surface equilibrium properties of lattice systems can be used to investigate the finite @xmath1 case as well @xcite .
moreover , the complete analytical solution of the single - sharp - peak replication landscape has been found recently by mapping the stationary solution of the kinetic equations ( [ ode ] ) into a polymer localization problem @xcite .
closely related to our approach to the quasispecies evolution of finite populations is the population genetics formulation of the deterministic quasispecies model proposed recently @xcite . in that formulation
it is assumed that the molecules are characterized solely by the number of monomers @xmath23 they have , regardless of the particular positions of these monomers inside the molecules .
hence there are only @xmath29 different types of molecules which are labeled by the integer @xmath30 .
this assumption is not so far - fetched since the feature that distinguishes the molecules is their replication rates @xmath14 , which in most analyses have been chosen to depend on @xmath31 only , i.e. , @xmath32 @xcite .
furthermore , denoting the frequency of monomers @xmath23 in generation @xmath33 by @xmath34 , it is assumed that the molecule frequencies @xmath35 are given by the binomial distribution @xmath36 for @xmath37 .
thus , in each generation the monomers are sampled with replacement from a pool containing monomers @xmath23 and @xmath38 in the proportions @xmath34 and @xmath39 , respectively .
this amounts to neglecting linkage disequilibrium , i.e. , in each generation the molecule frequencies are random combinations of the constituent monomers @xcite . with the two assumptions presented above a simple recursion relation for the monomer frequency @xmath34
can be readily derived @xcite . to take into account the effect of finite @xmath0
, the deterministic kinetic formulation must be replaced by a stochastic formulation based on a master equation for the probability distribution of the number of different types of molecules in the population @xcite .
however , the extreme approximations used to derive results from that master equation or from related langevin equations @xcite have hindered the analysis of the error threshold for finite populations .
an alternative approach to study stochastic chemical reaction networks is the algorithm proposed by gillespie @xcite , which has been successfully employed to simulate numerically the quasispecies model , providing thus a base line for analytical investigations @xcite .
the goal of this work is to propose an analytical framework to investigate the quasispecies evolution of finite populations .
more specifically , we will focus on the evolution of the molecule frequencies @xmath35 for @xmath40 and , since for finite @xmath0 these frequencies are random variables , we will derive a recursion relation for the average values @xmath41 .
although we will concentrate mainly on the dependence of the error threshold on the population size @xmath0 , the formalism presented in the sequel can be applied to study a variety of fascinating phenomena related to the finitude of the population , such as mutational meltdown @xcite and punctuated equilibria or stasis @xcite , to mention only a few . moreover , since modern theories of integration of information in pre - biotic systems involve the compartmentation of a small number of molecules ( typically 10 to 100 ) @xcite , the understanding of the effects of the error propagation in finite populations has become an important issue to the theories of the origin of life .
in each generation the population is described by the vector @xmath42 where @xmath43 is the number of molecules of type @xmath31 , so that @xmath44 . similarly to the deterministic case @xcite , we have to resort to a simplifying assumption to relate the molecule frequencies @xmath45 to the vector @xmath46 .
in particular , in generation @xmath33 we consider a molecule pool containing the different molecule types in the proportions @xmath47 , so that @xmath48 is distributed by the multinomial distribution @xmath49^{n_0 } \left [ \pi_1 ( t ) \right]^{n_1 } \ldots \left [ \pi_\nu ( t ) \right ] ^{n_\nu } .\ ] ] hence in each generation the molecules are sampled with replacement from the molecule pool . in this sense
, in each generation the population is a random combination of the constituent molecules , which amounts to neglecting linkage disequilibrium at the population level .
more pointedly , the population composed of the offspring of the molecules present in the generation @xmath50 is destroyed and its molecule frequency @xmath35 used to create an entire new population according to ( [ p_n ] ) .
although this procedure destroys the correlations between the molecules , it does not cause any significant loss of genetic information since the fitness of the molecules depend only on the number of monomers @xmath23 they have , which , in the average , is not affected by the procedure .
the changes in the population composition @xmath46 are due to the driving of natural selection , modeled by the replication rate @xmath51 , and to mutations , modeled by the error rate per digit @xmath25 . following the prescription used in the implementation of the standard genetic algorithm @xcite
, we consider first the effect of natural selection and then the effect of mutations . as usual
we assume that the number of offspring that each molecule contributes to the new generation is proportional to its relative replication rate which , for molecules of type @xmath31 , is defined by @xmath52 thus the population composition after selection is described by the random vector @xmath53 which is distributed according to the conditional probability distribution @xmath54^{n_0 ' } \left [ w_1 \left ( { \bf n } \right ) \right ] ^{n_1 ' } \ldots \left [ w_\nu \left ( { \bf n } \right ) \right ] ^{n_\nu ' } .\ ] ] next we consider the changes in @xmath55 due to mutations .
after mutation , the population is described by @xmath56 whose components are written as @xmath57 where the integer @xmath58 stands for the number of molecules of type @xmath59 that have mutated to a molecule of type @xmath31 .
clearly , @xmath60 .
we note that the probability of mutation from a molecule of type @xmath59 to a molecule of type @xmath31 is given by @xmath61 where @xmath62 and @xmath63 .
the population is more conveniently described by the set @xmath64 rather than by @xmath65 .
in fact , given @xmath66 the conditional probability distribution of @xmath67 is again a multinomial @xmath68 for @xmath69 . in this framework
the frequency of molecules of type @xmath31 in the next generation @xmath70 is given simply by @xmath71 .
this frequency is used to generate the new population of @xmath0 molecules of length @xmath1 according to the distribution ( [ p_n ] ) .
the procedure is then repeated again .
we have run simulations for the single - sharp - peak replication landscape using the procedure described above , which neglects linkage disequilibrium at the population level , as well as the standard genetic algorithm @xcite , in which the correlations between consecutive generations are maintained .
we have focused on the effect of the error rate @xmath25 on the normalized mean hamming distance @xmath72 between the master string and the whole population in the stationary regime .
this quantity is given by the fraction of monomers @xmath38 in the entire population , i.e. , @xmath73 . in figs .
( [ f1 ] ) and ( [ f2 ] ) we present the results of the simulations for @xmath72 and its standard deviation @xmath74 , respectively , as functions of the error rate @xmath25 .
the initial population is set with @xmath75 and @xmath76 for @xmath77 , and it is left to evolve for @xmath78 generations .
no significant differences were found for longer runs or for different choices of the initial molecular frequencies .
each data point involves two kinds of average : for each run we average over the mean hamming distance in the last @xmath79 generations ; this value is then averaged over @xmath80 runs .
we note that even if the populations are identical in the initial generation , the random character of the transitions @xmath81 will make them distinct in the next generation .
it is clear from these results that , as @xmath0 increases , the quantitative effects of assumption ( [ p_n ] ) become less significant .
moreover , the dependence of @xmath72 and @xmath74 on the error rate is qualitatively the same for both algorithms .
to derive an analytical recursion relation for the average molecular frequencies @xmath82 we consider the following approximate procedure , akin to the annealed approximation of the statistical mechanics of disordered systems , which facilitates greatly the analysis : instead of averaging over the populations only after the stationary regime is reached , we perform this average in each generation .
the result obtained @xmath83 is then used to build the new populations .
of course , in doing so we neglect the fluctuations of @xmath35 for the different runs . within this framework the average frequency of molecules of type @xmath31 in generation @xmath84 is written as @xmath85 using @xmath86 and @xmath87 we rewrite ( [ av_1 ] ) as @xmath88 noting that @xmath89 and @xmath90 , we can easily verify that the normalization condition @xmath91 is satisfied . to proceed
further we must specify the replication rate @xmath51 . in the case of the single - sharp - peak replication landscape ,
i.e. , @xmath92 and @xmath93 for @xmath77 , the summations over @xmath94 can be readily carried out .
the final result is @xmath95^n + \sum_{n=0}^{n-1 } b_n \frac { \sum_{r = 0}^{\nu - 1 } \overline{\pi}_r ( t ) \left [ m_{pr } + a \frac{r}{1-r } m_{p \nu } \right ] } { 1 + r \left ( a -1 \right ) } \ ] ] for @xmath96 . here
@xmath97^n \left [ 1 - \overline{\pi}_\nu ( t ) \right ] ^{n-1-n } , \ ] ] and @xmath98 .
thus , given the initial average molecular frequencies @xmath99 for @xmath100 , equations ( [ rec ] ) are iterated till the stationary regime is reached .
before we proceed on the analysis of the stationary solutions of the recursion equations ( [ rec ] ) , some comments regarding the definition of the error threshold are in order .
a popular definition of error threshold is the error rate at which the master frequency @xmath101 vanishes @xcite .
the problem with this definition is that , even in the deterministic limit @xmath102 , @xmath101 never vanishes for _ finite _ @xmath1 .
the vanishing of the master frequency is an artifact of neglecting reverse mutations @xcite , which can be justified in the limit @xmath28 only . in particular , for the single - sharp - peak replication landscape this prescription yields a very simple equation for the replication accuracy at the threshold in the deterministic regime @xcite , @xmath103 we must emphasize that for finite @xmath1 this equation is an approximation only .
a more appropriate definition of the error threshold , which is useful for the finite @xmath0 case as well , is obtained by considering the statistical properties of the entire molecular population @xcite . in particular , we focus on the normalized mean hamming distance @xmath72 between the master sequence and the entire population and define the error threshold as the error rate at which the standard deviation @xmath74 is maximal @xcite . in figs .
[ f1 ] and [ f2 ] we show the theoretical predictions for @xmath72 and @xmath74 using the steady - state solution of the recursion equations ( [ rec ] ) .
as expected , the effects of the fluctuations in @xmath35 for the different runs are stronger for small @xmath0 and hence our analytical approximation yields very poor results in this case , although it reproduces quite well the qualitative behavior pattern of the quantities measured .
however , already for @xmath104 there is a good agreement between the theoretical predictions for @xmath72 and the simulation results , provided that @xmath25 is not too near the threshold transition .
rather surprisingly , that agreement is better for the standard genetic algorithm .
as expected , however , the theoretical predictions for the standard deviation @xmath74 are very poor since our approximation scheme neglects the fluctuations in @xmath45 between the different runs , which are directly measured by @xmath74 .
nevertheless , the qualitative features of the simulation results are again well described by the theoretical curves .
we note , in particular , the abrupt increase of @xmath74 as the error threshold is approached from below and the slow decay as the error rate increases further .
the agreement between theory and simulation becomes better as @xmath0 increases . in fig .
[ f3 ] we show the replication accuracy at the threshold @xmath105 as a function of the reciprocal of the population size .
the increase of @xmath105 with decreasing @xmath0 is expected since the fluctuations become stronger for small @xmath0 and so the replication must be more accurate in order to keep the master string in the population .
in particular , for large @xmath0 we find that @xmath105 increases linearly with @xmath106 .
this result is in disagreement with the predictions of the birth and death model of error threshold proposed by nowak and schuster , which predicts that @xmath105 increases with @xmath107 for large @xmath0 @xcite .
we note that , despite the claim of those authors , it is not possible to discern whether @xmath105 increases with @xmath106 or @xmath107 from their numerical data obtained using gillespie s algorithm @xcite .
another interesting phenomenon , termed stochastic escape , is the loss of the master string in a finite population @xcite . in the limit @xmath28 this loss becomes irreversible , since no reverse mutation will be able to restore the master string .
we can easily derive a lower bound to the probability that the master string is absent from the population using the inequality @xmath108 which follows trivially from the fact that @xmath109 .
using @xmath110 , we can find the replication accuracy @xmath111 such that the condition @xmath112 is satisfied for fixed @xmath1 and @xmath24 .
clearly , for @xmath113 the probability that the master string is absent from the population is nonzero .
however , since this probability may be nonzero for @xmath114 as well , @xmath115 gives only a lower bound to the value of the replication accuracy below which the stochastic escape phenomenon actually takes place .
this bound is presented in fig .
[ f3 ] as a function of the reciprocal of the population size . in the limit @xmath102
we find @xmath116 for finite @xmath1 , since in the deterministic regime @xmath117 is bounded by @xmath118 .
it is interesting that for @xmath0 not too large we find @xmath119 so that the master string is likely to be absent from the quasispecies for replication accuracies in the range @xmath120 .
we turn now to the analysis of the deterministic regime , @xmath102 . in this case , the sum in eq .
( [ rec ] ) is dominated by the closest integer to @xmath121 , so that @xmath122 and the recursion equations ( [ rec ] ) reduce to @xmath123 for @xmath124 . to appreciate the relevance of the present formulation of the quasispecies model , we compare it with the exact solution of the kinetic equations ( [ ode ] ) for finite @xmath1 .
in fact , as pointed out by swetina and schuster @xcite , for the type of replication landscape considered in this paper , the @xmath26 molecular concentrations @xmath6 can also be grouped into @xmath125 distinct classes according to the number of monomers 1 that compose the molecules .
this procedure allows the description of the chemical kinetics by only @xmath126 coupled first - order differential equations .
in particular , for the single - sharp peak landscape the concentrations of molecules in class @xmath127 , denoted by @xmath128 with @xmath129 , obey the differential equations @xcite @xmath130 .\ ] ] it is clear then that both models ( [ det ] ) and ( [ sw ] ) possess the same stationary state .
this very interesting finding indicates that in the quasispecies model there is no linkage disequilibrium at the population level in the stationary regime , i.e. , the population is a random combination of the constituent molecules .
in fact , this results holds true for any choice of the replication landscape @xmath51 , as can be easily verified by taking the limit @xmath102 in eq .
( [ av_2 ] ) .
it is also interesting to compare the deterministic limit of our model ( [ det ] ) with the population genetics approach to the deterministic quasispecies model @xcite .
this can easily be done by writing down a recursion equation for the frequency of monomers 1 in the population @xmath131 , namely , @xmath132 clearly , this equation is useless since one must solve the general recursion equations ( [ det ] ) in order to find @xmath133 .
however , the population genetics formulation makes use of the binomial assumption ( [ binomial ] ) to set @xmath134 so that the recursion equation ( [ p_d ] ) will involve the monomer frequencies only @xcite . in fig .
[ f4 ] we present the logarithm of the replication accuracy at the threshold @xmath135 as a function of the logarithm of the selective advantage @xmath136 for several values of @xmath1 for the deterministic case .
there is a good agreement with ( [ simple ] ) for @xmath137 and , as expected , this agreement becomes better as @xmath1 increases .
we have also verified that the prediction of the population genetics approach @xcite yields a very poor approximation for the location of the error threshold .
furthermore , we have verified that @xmath138 departs significantly from a binomial distribution only near the threshold transition .
aside this region , the population genetics approach provides a reliable and concise description of the deterministic quasispecies model .
in this paper we have proposed a simple analytical model , based on the neglect of linkage disequilibrium , to study the error propagation in the quasispecies evolution of finite populations . in particular , our finding that in the deterministic regime this model yields exactly the same stationary state of the original kinetics model @xcite implies that the steady - state molecular population of the deterministic quasispecies model is a random assembly of the component molecules .
some comments regarding the comparison of our approach with previous population theoretical analyses of the finite @xmath0 quasispecies model @xcite are in order .
these works provide approximate formalisms to study the evolution of finite populations on a multiplicative single - peak fitness landscape without neglecting linkage disequilibrium .
interestingly , for this fitness landscape , which is given by @xmath139 with @xmath140 , the binomial assumption ( [ binomial ] ) yields the exact solution to the deterministic equations for @xmath45 @xcite .
moreover , in the weak selection limit ( @xmath141 ) the @xmath106 corrections to the deterministic value of the mean hamming distance between the master sequence and the whole population can be calculated analytically @xcite .
an alternative formalism concentrates on the evolution of the ensemble average of the first cumulants of the distribution of fitness in the population @xcite .
it is not clear , however , whether these quantities can be related to the more natural measures of the population composition , namely , @xmath72 and @xmath74 .
we note that the location of the error threshold is not addressed in these works
. an important open question , which can be answered through intensive numerical simulations only , is the dependence of the replication accuracy at the threshold on the population size for large populations .
in fact , the numerical data existent in the literature do not allow to distinguish between the @xmath106 dependence predicted by our model and the @xmath107 dependence predicted by the birth and dead model @xcite .
we think that simulations based on genetic algorithms rather than on gillespie s algorithm may prove more effective to address this issue . | a simple analytical framework to study the molecular quasispecies evolution of finite populations is proposed , in which the population is assumed to be a random combination of the constituent molecules in each generation , i.e. , linkage disequilibrium at the population level is neglected . in particular , for the single - sharp - peak replication landscape we investigate the dependence of the error threshold on the population size and find that the replication accuracy at the threshold increases linearly with the reciprocal of the population size for sufficiently large populations .
furthermore , in the deterministic limit our formulation yields the exact steady - state of the quasispecies model , indicating then that the population composition is a random combination of the molecules . |
n 157b ( henize , 1956 ) is a crab - type supernova remnant ( snr ) just 7 arcmin from the center of 30 doradus ( bode , 1801 ) in the large magellanic cloud ( lmc ) .
it contains a 16-ms x - ray pulsar undetected at any other wavelength ( marshall et al .
there is bright non - thermal x - ray emission with structure on arcsec scales just around the pulsar with an extended feature off to the northwest ( wang and gotthelf , 1998a , 1998b ; wang et al .
there is bright non - thermal radio emission from the extended feature but not at the pulsar location ( lazendic et al .
we shall call the extended emission region the pulsar wind nebula ( pwn ) .
the overall struture suggests that the pulsar is moving toward the southeast .
there is also extended radio emission toward the south that gives a hint of a shell , suggesting that the remnant may be in transition to becoming a composite remnant with a shell and a pulsar wind nebula .
the differences in the radio and x - ray structures plus the apparent large motion of the pulsar make this snr unusual .
we shall describe its properties and then discuss the implications of the data .
figures 1 and 2 show the simularities of the radio and x - ray emission of the pwn component of n 157b but a striking difference toward the pulsar .
the radio emission in the pulsar wind component sits on a plateau of the rest of the remnant .
fine structure in the pwn appears very similar in both wavelength ranges although the radio emission extends further northwest .
this structure probably represents clumpiness in the interstellar medium or in pre - explosion mass loss of the progenitor .
the peak in the x - ray emission in the compact source around the pulsar is 13 times the brightness of the peak in the pwn but in the radio there is nothing seen above the residual brightness of the pwn .
the pulsar lies at @xmath1 and @xmath2 ( wang and gotthelf , 1988b ) about 16@xmath3 out from the center along the se axis of the tail .
the pulsar is about 1@xmath3 closer to the center of the tail than the peak of the x - ray emission . .
the thin line shows the location of slices shown in figure 3 and the cross is the position of the pulsar.,width=340 ] wide centered on the radio peak and the ones through the pulsar position are 5@xmath3 wide.,width=340 ] wide centered on the radio peak and the ones through the pulsar position are 5@xmath3 wide.,width=340 ] further details of the emission are revealed by the slices presented in figures 3 and 4 .
the emission from the pwn is clearly more extended in all directions at radio wavelengths than at x - ray ones .
the overall symmetry is the same , however , with an approximately elliptical shape centered about a point at @xmath4 and @xmath5 .
the radio emission falls off uniformly out to a semi - major axis extent of 20@xmath3 in the se - nw direction and 10@xmath3 in the ne - sw direction .
there is no sign of the pulsar or any enhancement in the radio emission , @xmath6 mjy beam@xmath7 , at the pulsar s position of @xmath1 and @xmath2 ( wang and gotthelf , 1988b ) .
the non - thermal x - rays around the pulsar position , on the other hand , show a strong approximately elliptical component , about 7@xmath3 @xmath8 3@xmath3 with its long axis perpendicular to the long axis of the pwn tail .
wang and gotthelf ( 1998 ) suggested that this small source could be a bow - shock from the particles leaving the moving pulsar .
we shall henceforth call that structure the shock region . from the inner ( nw ) edge of this shock ,
the x - ray emission first decreases somewhat and then increases gradually toward the radio center of the pwn but peaks 3@xmath3 before the radio and then falls sharply toward the northwest . to compare the actual brightnesses of the features , we show their spectra in figure 5 .
the squares represent the integrated values for the radio emission of the entire snr .
they give a spectral index , @xmath9 , of @xmath10 , where the flux density @xmath11 @xmath12 @xmath13 ( lazendic et al . 2000 ) .
xmm - newton spectra , that can not resolve angular detail , show that most of the x - ray emission from the snr has a steep power - law spectrum with @xmath14 although some thermal emission is present as well ( dennerl et al .
they do not give a value for the actual x - ray flux .
for the pwn , the lower frequency radio data do not have sufficient resolution for a good separation of the components so we report only the 4.8- and 8.6-ghz results ( lazendic et al .
the spectral index for the pwn is more uncertain because of the snr background
. the value of @xmath15 could easily be the same as that of the whole snr .
the error of the spectral index for this fit to only the two data points tries to take into account the uncertainty in evaluation of the background .
we can not determine the radio spectrum of the shock region because it is not detected .
we do show the upper limits for its flux density at the two radio frequencies .
the x - ray spectra are from the paper by wang and gotthelf ( 1998b ) .
their formal fits give values of @xmath16 for the pwn and @xmath17 for the bow - shock .
realizing that these errors are only for the formal fits to the data , we suggest that the slopes of both components could be the same but that of the pwn can not be much steeper than that of the shock region .
for analysis , we divide the snr into three parts as outlined by wang and gotthelf ( 1998a ) and discussed above : the elliptical pulsar wind nebula extending northwest from the pulsar with major and minor axes of @xmath18 pc ; the brignt shock region ( @xmath19 pc ) centered just outside the pulsar ; and the entire snr , about @xmath20 pc across , which extends well beyond the images in figures 1 and 2 . assuming that the pulsar has been moving southeastward from an explosion site at the radio peak , we can estimate its speed using the characteristic spin - down age of 5000 years ( marshall et al .
1998 ; wang and gotthelf , 1998b ) . to have moved 4 pc ,
the pulsar has a speed of 800 km sec@xmath7 , large but not impossible for a pulsar ( arzoumanian et al .
because the pwn emission falls off in all directions from its center , including toward the pulsar , we are led to the conclusion that there was a brief period of particle injection by the pulsar , lasting up to about 1000 years after the explosion occurred .
that period would account for the @xmath21 3/4 pc shift of the x - ray peak from the radio one and also allow for the apparent aging of the electrons on the outer edge of the pwn relative to the center . as higher energy electrons decay more rapidly from their synchrotron emission , those further from the pulsar should be the oldest and thus have reduced x - ray emission relative to the radio and steeper spectra ( pacini and salvati , 1973 ; reynolds and chevalier , 1984 ) .
the particle injection and/or stimulation can not have lasted much longer , however , or the emission in the pwn would still be bright toward the southeast between the pwn peak and the current location of the pulsar .
the emission from the shock region just around the pulsar appears to have arisen from a second injection of particles .
the x - ray brightness rises so drastically just there and appears to sit on top of the decreasing pwn emission in that direction .
the radio emission is below detection level .
we also note that if the injection of particles has been continuous , the x - ray spectrum of the region closest to the pulsar should be flatter ( harder ) than that of the region further from the current pulsar position , but that is not the case .
thus the newer particles around the pulsar must have been injected with a different spectrum or be interacting with a very different medium that those injected earlier near the center of the overall pwn .
finally , the center of this shock source is slightly ( about 1/4 pc ) outside the pulsar position .
these conditions lead us to believe that the object is indeed a shock region caused by the supersonic motion of the particles coming from the pulsar moving at about 800 km sec@xmath7 plus their injection speed and perhaps interacting with a reverse shock from the intitial blast wave . without a more accurate separation of the thermal component of the x - ray emission in that region to evaluate the temperature and density of the gas , we can not determine the actual sound speed but the relative motion could be enough to generate a strong shock wave and compressed magnetic fields across which to accelerate particles to even greater energies than they get from the pulsar . the pulsar may have also encountered a density enhancement which will further increase the total emission to produce the high observed x - ray brightness .
indications of a clumpy medium around n 157b include many h@xmath9 filaments and a very low polarization of the radio emission , presumably caused by significant faraday rotation ( lazendic et al .
the other lumps in the pwn may represent temporary increases due to previous brief episodes of acceleration near density enhancements .
we do nt know the radio spectral index of the shock region so we do nt know if the energy injection in that spectral range is the same for both episodes . to get an approximate ratio of radio to x - ray luminosities , we will assume that the radio spectrum of the shock region is the same as the rest of the snr , @xmath22 .
extrapolation of any value steeper than about @xmath230.3 would require less x - ray emission than observed . with the adopted spectral index , we calculate an upper limit to the radio luminosity between 10@xmath24 and 10@xmath25 hz of @xmath26 erg sec@xmath7 or about 1/2000 of the x - ray luminosity in the 0.2 4 kev band ( wang et al .
this upper limit is a low ratio for either a pulsar wind alone , e.g. 0540 - 693 has a ratio of radio to x - ray luminosity of 1/120 , or for an snr with shock generated x - ray emission , e.g. ad1006 has a ratio of about 1/100 ( taken from allen et al .
perhaps the combined processes may create a harder spectrum at low energies but then a sharper break at the transition between the emission in the two frequency ranges . while still qualitative , this description of multiple episodes of particle injection from the pulsar is the only one we can come up with to explain all the data . better radio resolution and sensitivity plus improved spectra in all wavelength bands , including a detection of infrared and optical continuum emission , will be important for checking these ideas .
we thank paul plucinsky , steve reynolds , brian fields , you - hua chu , douglas bock , martin guerrero , and pat slane for help and ideas .
this research is supported by nasa grant nag 5 - 11159 . | the supernova remnant n 157b contains a pulsar and three distinct synchrotron components with rather unusual properties .
1 ) a somewhat irregular elliptical pulsar wind nebula ( pwn ) visible in both x - ray and radio wavelengths . the nebula is quite symmetrical with an extent of about 10 @xmath0 5 parsecs but offset along the long axis by about 4 pc from the pulsar position .
it is apparently the result of a short - lived injection of energetic particles , perhaps starting at the time of explosion .
2 ) a very bright x - ray shock region located just outside the pulsar position in the edge of the pwn . this is undetected in the radio .
we attribute this to a new burst of particles from the pulsar suggesting there are multiple episodes rather than continuous injection .
3 ) the beginning of a radio synchrotron shell on the southern side of the snr where thermal x - rays appear to arise suggesting that n 157b is starting to become a composite snr . |
the equation of state ( eos ) has attracted a lot of attention recently for asymmetric nuclear matter , which can be described approximately by the parabolic law @xmath4 here @xmath5 is the isospin asymmetry , @xmath6 the energy per nucleon for symmetric nuclear matter , and @xmath7 the bulk symmetry energy .
the symmetry energy term @xmath8 is very important for understanding many interesting astrophysical phenomena ( see , _ e.g. _
@xcite ) , but so far results in large uncertainties : _
e.g. _ , the symmetry energy calculated with different kinds of parameter sets ( skyrme or gogny type ) are largely divergent @xcite ) , especially at high densities , and for some cases , _
i.e. _ , when the density is higher than three times of the normal density , even a negative symmetry energy can be obtained . therefore , acquiring the more accurate knowledge of the symmetry energy , and the isospin asymmetry , becomes one of the main goals in nuclear physics at present .
the recently available facilities of rare - isotope beams provide the opportunities to study the dynamical evolution of nuclear systems with a large range of isospin asymmetries , which increases the domain over which a spatially uniform local isospin asymmetry @xmath9 may be achieved . in order to obtain the information about the symmetry potential at high density
, the beam energy required has to be of higher than several hundreds of mev per nucleon , but then the isospin effects on heavy ion collisions would become negligible and are usually not considered .
however , in some special cases , like the one around the particle emission threshold , it is found that the symmetry potential affects the particle production @xcite , especially , the ratio between the number of produced negatively and positively charged particles may depend sensitively on the density dependence of the symmetry potential .
li @xcite found that , in an isospin - dependent hadronic transport model , the @xmath10 ratio , as well as the neutron - proton differential collective flow , were sensitive to the behavior of the nuclear symmetry potential at high densities .
more recently , within the framework of relativistic landau vlasov transport method , gaitanos et al .
@xcite found that when the beam energy was higher than 2 agev the sensitivity of @xmath11 ratio to the form of the symmetry potential is largely reduced at high densities . in this paper , we attempt a further investigation of the energy dependence of the sensitivity of the @xmath10 production ratio to the form of the symmetry potential in the urqmd model . furthermore , we try to explore a new candidate in terms of the @xmath3 ratio for probing the symmetry potential in high density matter , which is in addition to the @xmath12 production ratio .
the production of @xmath13 and @xmath14 hyperons is closely related to the neutron - proton asymmetry of the projectile - target system , which means that the symmetry potential of nuclear matter will affect the @xmath3 ratio .
consequently , the @xmath3 ratio in heavy ion collisions at high energies may also carry the information about the density dependence of the symmetry potential of nuclear matter .
furthermore , the isospin - dependent part of the single - particle potential of @xmath1 hyperon in a nuclear medium , the , so called , lane potential , depends on the isospin asymmetry of nuclear matter , which might also influence the @xmath3 ratio . in turn , this study of @xmath3 ratio might provide us with the information about the isospin - dependent part of the @xmath1 hyperon single - particle potential .
in this paper , the study of @xmath1 hyperon production is limited near its threshold ( 1.79 gev for @xmath1 hyperon production in free space through a process of the type @xmath15 reaction ) in order to observe the effect of the symmetry potential on the @xmath3 ratio . specifically , we consider the neutron - rich system @xmath16 and the nearly isospin - symmetric system @xmath17 at three different beam energies of @xmath18 ( the sub - threshold energy ) , @xmath19 , and @xmath20 gev . for our calculations ,
the urqmd model @xcite , version 1.3 , is adopted , using the hard skyrme - type eos for reactions with beam energies @xmath21 gev .
we find that most of the urqmd model calculations can simultaneously reproduce many experimental measurements , which offers a good platform for studying the isospin effects at sis energies .
the paper is arranged as follows . in section
ii , we give our method of including the isospin - dependent part of the mean field in the urqmd transport model . in section iii ,
the numerical results of pion and @xmath1 hyperon production and the corresponding ratios between the negatively and positively charged particles are presented .
finally , in section iv , a brief summary and discussion are given .
since the isospin dependence of nucleon - nucleon interaction has been introduced explicitly in the urqmd model , in order to study the isospin effects in heavy ion collisions , we have to introduce the symmetry potential into the mean field . in urqmd model , the skyrme and the yukawa potentials are included in the iso - scalar part of the mean field , where the yukawa parameter is related to skyrme parameters .
in infinite nuclear matter , the contribution of yukawa potential to the total energy acts like the two - body skyrme contribution @xcite . the coulomb potential is also implemented explicitly .
similarly , a symmetry potential should also be included in the mean - field part .
the bulk symmetry energy @xmath22 in eq .
( [ ieos ] ) can be expressed as @xmath23 where @xmath24 is the symmetry energy at the normal density and @xmath25 the reduced density . in this paper , we take @xmath26 mev and , in order to mimic the strong variation of the density dependence of the symmetry energy at high densities , we adopt the form of @xmath27 as used in @xcite : @xmath28 here @xmath29 is the reduced critical density .
note that when @xmath30 , the symmetry potential energy will be negative .
similarly , following @xcite , we take @xmath31 for stiff symmetry potential ( stiff-sym.pot . ) , namely , the @xmath32 , and @xmath33 for soft symmetry potential ( soft-sym.pot . ) , namely , the @xmath34 .
apparently , @xmath32 and @xmath34 give the two extremes of the symmetry energy at high densities , as is illustrated in the following for the chemical potential .
the neutron and proton chemical potentials @xmath35 , contributed only by the symmetry potential energy , are shown in fig .
[ fig1 ] . , using both the stiff-sym.pot . @xmath32 and soft-sym.pot .
@xmath34 . here
@xmath36.,scaledwidth=80.0% ] we notice in fig .
[ fig1 ] that for @xmath37 the difference between the neutron ( and so also proton ) chemical potentials calculated with @xmath32 and @xmath34 is small but becomes large at high densities .
for @xmath38 , the @xmath39 becomes negative for @xmath34 .
furthermore , the curves of @xmath39 for @xmath32 and @xmath34 cross each other at @xmath40 ( called , crossing point ) and those of @xmath41 at @xmath42 .
apparently , such different behaviors of the @xmath35 , calculated with @xmath32 and @xmath34 , will strongly influence the motion of protons and neutrons and so also the time evolution of the proton and neutron density distributions , which will further influence the ratios of produced particles with different charges . for simplicity , the isospin - independent part of the mean field for resonances and hyperons is taken here to be the same as that of the nucleon .
this simplification is quite adventurous , but we think that it should not alter the final conclusion about the influence of the symmetry potential on the ratios of emitted negatively versus positively charged particles .
the symmetry potentials of the resonances [ n@xmath43(1440 ) and @xmath44(1232 ) ] and hyperons [ @xmath45 and @xmath1 ] are also introduced in the calculations , in addition to that of the nucleon .
the symmetry potential for the resonances is obtained through the constants of isospin coupling ( the clebsch - gordan coefficients ) in the process of @xmath47 [ or @xmath48 @xmath49 . for hyperons ,
the single - particle potential in a spin - saturated nuclear matter can be expressed as @xmath50 assuming charge independence of the baryon - baryon interaction @xcite . here
@xmath51 and @xmath52 represent the isospin - independent and -dependent parts .
the @xmath52 is the lane potential , which is known to be important for the structure of the @xmath1 hyper - nuclear state@xcite .
this expression has the same form as the single - particle potential of nucleons , up to the first order in @xmath53 .
however , the value of @xmath54 , even its sign , is still a matter of argument : within the relativistic mean - field theory ( see , _ e.g. _ , @xcite ) , the symmetry potential of the octet of baryons is described via the coupling between the baryon and @xmath55 meson . for hyperons ( @xmath56 ) ,
the isospin dependent part is determined by the coupling constant @xmath57 and their isospin . since
@xmath57 is taken to be @xmath58 , as well as @xmath59 @xcite , we simply take the symmetry potential of @xmath60 hyperon to be proton - like and neutron - like , according to eq .
( [ vsigma ] ) .
however , the symmetry potential of the excited states of hyperon is not considered , for lack of information .
combining both the resonances and hyperons , we express the symmetry potential in an unified form , which reads as @xmath61 where the values of @xmath62 and @xmath63 for different baryons ( @xmath64 ) are listed in table [ tab1 ] . from this table
we can see that the symmetry potentials of @xmath65 and @xmath66 are neutron - like and those of @xmath67 and @xmath68 are proton - like .
on the other hand , the symmetry potentials of @xmath69 , @xmath70 and @xmath71(1440 ) are a mixture of the neutron and proton symmetry potentials .
since the value of lane potential is very uncertain , in order to make comparisons , we have also investigated the cases when the symmetry potential of hyperons ( and also of resonances ) is switched off .
.the values of @xmath62 and @xmath63 for isospin - dependent potentials of different baryons . [ cols="<,^,^,<,^,^",options="header " , ] [ tab1 ]
in the urqmd model , the total cross section depends on the isospins of colliding particles , their flavour and the center - of - mass energy . if high quality experimental data on the respective cross sections ( the neutron - proton , proton - proton elastic scattering cross sections , _ etc .
_ ) exist , a phenomenological fit to the respective data is adopted .
if no data are available , the additive quark model ( aqm ) and the detailed balance arguments are used .
the details of the determination of the elementary cross sections can be found in ref .
@xcite .
first of all , we check the pion and @xmath1 hyperon production when the symmetry potential is not taken into account . in the version 1.3 of urqmd model , compared to the version 1.2 ,
a channel of @xmath72 is implemented for the improvement of the kaon production at low energies .
the introduction of this channel may also influence the yields of hyperons . in fig .
[ fig2 ] , we show the yields of pions , @xmath45 , and @xmath1 hyperons in the reaction @xmath16 at @xmath73 gev and @xmath74 fm , calculated by using both the versions 1.2 and 1.3 of urqmd , and with and without iso - scalar part of the mean field .
the eos0(v1.2 ) and eos0(v1.3 ) denote the results of the versions 1.2 and 1.3 without the mean field , and the eos1(v1.3 ) denotes the results of the version 1.3 with a mean field of hard skyrme - type force . the freeze - out time is taken to be @xmath75 and for each case we calculate tens of thousand events for good statistics . in fig .
[ fig2 ] ( lower panel ) we find that the effect of introducing the process of @xmath72 and the mean field on @xmath0 production is almost negligible , but its effect on the hyperon production is rather large ( see upper panels in fig .
[ fig2 ] ) .
the yield of @xmath1 hyperon is enhanced by about @xmath76 whereas the yield of @xmath45 hyperon is suppressed by about @xmath77 with the version 1.3 , which together suppresses the @xmath78 yields by about @xmath79 . when the mean field is switched on , the suppression effect on the yields of hyperons
is enhanced by about @xmath80 , the same as in @xcite . in refs .
@xcite , it was pointed out that the yields of hyperons within the cascade model were always overestimated .
this means that our suppressed yields of the hyperon production , with mean - field correction , could be in the right direction of the experiments at these energies .
, @xmath45 , @xmath1 , and ( @xmath78 ) yields with different conditions in the @xmath16 reaction at @xmath73 gev and impact parameter @xmath74 fm .
eos0 and eos1 refer , respectively , to without and with isoscalar part of the mean field , v1.2 and v1.3 being the version 1.2 and version 1.3 of urqmd ( see text).,scaledwidth=70.0% ] in fig .
[ fig3 ] , we show the fractions of the yields of pions and @xmath1 hyperons produced in different baryon density regions , for cases of @xmath81 and @xmath20 gev . here
the percentage of the contributions in the individual density regions to the total yield is plotted . in general , pions and @xmath1 hyperons
are mainly produced in the high density region ( @xmath82 ) .
our calculations at both the energies show that more than @xmath83 of pions and @xmath1 hyperons are produced in the region of @xmath82 . among them , more than @xmath84 is produced in the region of @xmath85 , and the remaining is produced in the region of @xmath86 .
this happens because the phase - space available at such higher densities is quite restricted .
furthermore , with the energy increasing from @xmath18 gev to @xmath20 gev , the fractions of the yields of @xmath0 ( and @xmath1 ) produced in the density region of @xmath85 decrease while those at other densities increase . comparing the pion and @xmath1 production , we find that at both energies the fraction of pions produced is larger than that of @xmath1 hyperons for densities @xmath87 , while the situation is just opposite for densities @xmath88 . and @xmath1 at different densities for @xmath16 reactions at @xmath81 and @xmath20 gev and @xmath74 fm.,scaledwidth=80.0% ] figs .
[ fig4 ] ( a ) and ( b ) show the percentage of the contributions from the relevant baryon - baryon ( b - b ) , meson - baryon ( m - b ) inelastic scattering , and the resonance ( res ) decay processes to the total @xmath1 hyperon production and annihilation at ( a ) @xmath81 gev and ( b ) @xmath89 gev .
it should be pointed out that in these results , the contributions from the @xmath90 in @xmath91 and the @xmath92 in @xmath93 are added to both the @xmath1 hyperon production and annihilation processes . here ,
and below , `` x '' represents the products ( including multi - hadrons ) other than @xmath1 . from these two plots
, we find that the most important channels for @xmath1 production and annihilation are @xmath93 , and the res ( @xmath94 , and @xmath95 ) at the two energies studied . figs .
[ fig4 ] ( c ) and ( d ) show the time evolution of the average number of @xmath1 produced and annihilated per unit time , @xmath96 , through the processes of @xmath97 , @xmath94 , and @xmath98 at @xmath81 gev and @xmath89 gev , respectively . here , the unit time is taken to be @xmath99 . for @xmath1 production , at the early stage ,
the channel @xmath100 is the most important one .
the @xmath96 for @xmath93 , particularly at @xmath89 gev , is very much pronounced at the early reaction time , but then it decreases , even faster than for the case of @xmath18 gev .
the @xmath96 for @xmath101 is reduced to @xmath102 of its highest value at @xmath103 for the case of @xmath89 gev but at @xmath104 for the case of @xmath81 gev . at the late reaction stage
, @xmath1 is mainly produced from the decay of baryon resonances ( i.e. @xmath105 or @xmath106 ) , which then continues for a longer time .
concerning the @xmath96-value for the @xmath1 annihilating channel of @xmath107 , it is much smaller compared to the case of the reverse process of @xmath94 at @xmath81 gev , while it remains comparable at @xmath89 gev .
the reason for this behavior is that a larger number of @xmath1 s are produced through @xmath101 at @xmath89 gev than at @xmath81 gev , which leads to a stronger annihilation of @xmath1 s at @xmath89 gev .
thus , from figs . [ fig4 ] ( c ) and ( d ) we may conclude that @xmath1 hyperons are mainly produced during @xmath108 fm / c for the case of @xmath89 gev but continues up to @xmath109fm / c for the case of @xmath81 gev .
[ fig5 ] shows the time evolution of the average density in the central reaction zone of @xmath110 fm at @xmath81 and @xmath20 gev .
one sees that the average density in the region of @xmath110 fm reduces to the normal density at @xmath111 fm / c for the case of @xmath89 gev , but for @xmath81 gev at @xmath112 .
compared to the case when @xmath96 for the process @xmath93 reduces to @xmath102 of its highest value , this result at high energy is similar but is somewhat longer at the lower energy . in other word , figs .
[ fig4 ] and [ fig5 ] show that , compared to the time when the system stays at high densities ( @xmath82 ) , the time when most of the @xmath1 hyperons are produced is similar at @xmath89 gev but is somewhat _ shorter _ at @xmath81 gev
. production or annihilation plotted at ( a ) @xmath81 gev and ( b ) @xmath20 gev .
the time evolution of d@xmath113/dt of several important @xmath1 production and annihilation processes is shown at ( c ) @xmath81 gev and ( d ) @xmath89 gev.,scaledwidth=100.0% ] and @xmath20 gev.,scaledwidth=80.0% ] similar to figs .
[ fig4 ] ( a ) and ( b ) , fig .
[ fig6 ] shows the accumulated number of @xmath0 s produced and annihilated via the different processes . unlike @xmath1
, pion is mainly produced via the baryon decay , where the most important process is @xmath44 decay .
the contribution from other channels , such as @xmath114 , @xmath115 , @xmath116 decay , as well as @xmath117 decay , becomes visible only for the case of @xmath20 gev .
the obvious difference between the two cases of @xmath81 gev and @xmath20 gev is that the _ fraction _ of @xmath0 produced via @xmath44 decay is much smaller for the higher energy case .
a similar scenario occurs for @xmath0 annihilation . produced and annihilated through different channels for @xmath118 at @xmath81 gev ( upper plot ) and @xmath20 gev ( lower plot ) . for a comparison between the two plots , the number in lower plot is drawn as half of the real quantity .
, scaledwidth=90.0% ] in this section , we mainly explore the effect of the symmetry potential on the ratios between the yields of negatively and positively charged pions and @xmath1 hyperons . in fig .
[ fig7 ] , we show the time evolution of the @xmath2 ratios ( left - hand side ) and the @xmath3 ratios ( right - hand side ) calculated with @xmath32 and @xmath34 for the reaction @xmath16 at @xmath81 , @xmath19 , @xmath20 gev and @xmath74 fm , and @xmath17 at @xmath89 gev and @xmath74 fm . the symmetry potentials of all the particles mentioned in sec .
ii are considered . here , and below , the @xmath1 resonances are also included in order to improve the statistics of calculated quantities at the early stage of the reaction .
also , at the freeze - out time , all the unstable particles are considered to decay . for the pions
, one can see from the left plot in fig .
[ fig7 ] that at @xmath81 gev the @xmath2 ratio calculated with the soft symmetry potential ( @xmath34 ) is enhanced compared with the stiff one ( @xmath32 ) , which is the same as was found in @xcite .
this happens because , at this energy , pions are mainly produced by @xmath44 decay ( refer to fig .
[ fig6 ] ) .
furthermore , we also find that the difference in the ratios of @xmath2 calculated with different symmetry potentials @xmath32 and @xmath34 is reduced with the increase of the beam energy , like in ref .
the reason for this insensitivity at high energies is that more @xmath0 production and annihilation channels are involved at high energies and then the @xmath0 production and annihilation via @xmath44-decay also become less important . in @xcite
, it is mentioned that the recent fopi data show this above noted tendency .
thus , we could take the ratio @xmath2 at high energies ( as is the case with @xmath89 gev studied in this work ) to have become insensitive to the symmetry potential .
( left ) and @xmath3 ( right ) for the collisions @xmath16 ( @xmath81 , @xmath19 , and @xmath20 gev ; @xmath74 fm ) and @xmath17 ( @xmath89 gev ; @xmath74 fm ) , calculated with the different symmetry potentials @xmath32 and @xmath34.,scaledwidth=80.0% ] next , for @xmath1 hyperons , from fig .
[ fig7 ] ( right plot ) , firstly , one sees that the @xmath3 ratio is sensitive to the density dependence of the symmetry potential for neutron - rich @xmath16 collisions , but insensitive to that for the nearly symmetric @xmath17 collisions . for @xmath16 at @xmath81 gev , the @xmath3 ratio calculated with the stiff symmetry potential is higher than the one with the soft symmetry potential . as the beam energy increases
, the @xmath3 ratio falls and the difference between the @xmath3 ratios calculated with @xmath32 and @xmath34 reduces strongly .
as the beam energy increases further , at @xmath89 gev the @xmath3 ratio falls further but the difference between the @xmath3 ratios calculated with @xmath32 and @xmath34 appears again , the @xmath3 ratio with soft symmetry potential now becoming higher than that with the stiff one . in sequel
, it might be interesting to investigate why the behavior of @xmath3 and @xmath2 ratios is so different , as far as the influence of the density dependence of the symmetry potential is concerned .
one basic difference is that , like nucleons , @xmath60 hyperons are under the influence of the mean field produced by the surrounding nucleons , as soon as they are produced .
the symmetry potential of hyperons also play an important dynamic role and results in a strong effect on the ratio of the negatively to positively charged @xmath1 hyperons .
thus , we further investigate the @xmath3 and @xmath2 production ratios when the symmetry potential of @xmath1 hyperons and resonances , except nucleons , is switched off . a very small difference for the @xmath2 ratio , but a large difference for @xmath3 ratio
is found with the switching on and off of the symmetry potential of @xmath1 and resonances .
our results for @xmath3 ratio are plotted in fig .
two cases are demonstrated : 1 ) only the symmetry potential of nucleons is considered ; and 2 ) only the @xmath1 symmetry potential is not considered . from fig .
[ fig8 ] one sees that the @xmath3 ratio at @xmath81 gev with soft symmetry potential is higher than with the stiff symmetry potential , no matter whether the symmetry potential of @xmath117 , @xmath44 , and @xmath45 is introduced or not . as the incident energy
is increased to @xmath20 gev , the sensitivity to the density dependence of the symmetry potential is almost lost , like for the case of @xmath2 ratio ( not shown here ) .
this means that the different behavior of the @xmath3 ratio , with respect to the @xmath2 ratio in figs .
[ fig7 ] , is due to the introduction of the symmetry potential of @xmath1 hyperon .
this can be understood as follows : in fig .
[ fig4 ] , it was shown that one of the most important channel for @xmath1 production is @xmath97 , where the most important channel is @xmath119 .
there are primarily 6 processes for @xmath119 , namely , 1 ) @xmath120 , 2 ) @xmath121 , 3 ) @xmath122 , 4 ) @xmath123 , 5 ) @xmath124 , 6 ) @xmath125 .
the 1 ) and 5 ) are not relevant here ; 2 ) and 6 ) are relevant to the @xmath14 production and 3 ) and 4 ) are relevant to the @xmath13 production .
thus , one can draw the conclusion that the @xmath3 ratio should be proportional to the @xmath2 ratio , and hence , the @xmath3 ratio has the same behavior as the @xmath2 ratio when the symmetry potential of @xmath1 is not taken into account . for @xmath32 and @xmath34 , with the symmetry potentials of different baryons switched on or off.,scaledwidth=80.0% ]
in summary , based on the urqmd model ( version 1.3 ) , we have investigated the influence of the symmetry potential on the ratios between the negatively and positively charged pions and @xmath1 hyperons , in the nearly central collisions @xmath16 and @xmath17 at @xmath18 , @xmath19 and @xmath20 gev energies . in order to find sensitive probes to the behavior of the symmetry potential at high - density nuclear matter ,
two different forms of the density dependence of symmetry potential in the mean field are considered .
the obvious dynamical effect of the symmetry potential is found on the neutron - rich reaction @xmath16 and not on the nearly isospin - symmetric reaction @xmath17 .
the effect of the symmetry potential on the @xmath2 ratio in @xmath16 at @xmath81 gev is similar to that found in @xcite , namely , the @xmath2 ratio calculated with the soft symmetry potential is higher than that with the stiff one , but at higher energies , like @xmath89 gev , it disappears .
this is explained as follows : at @xmath81 gev the most important channel for the production of pions is @xmath44 decay , while at @xmath89 gev other channels also play important role and the contribution from @xmath44 decay is largely reduced .
the situation about the effect of the symmetry potential on @xmath3 is more complicated because @xmath1 hyperon itself also experiences a mean field of nuclear medium as soon as it is produced .
when the symmetry potential of @xmath1 hyperons is not taken into account , a behavior similar to that of @xmath2 ratio is obtained , i.e. , the @xmath3 ratio calculated with the soft symmetry potential is higher than that with the stiff one at @xmath81 gev and the sensitivity to the symmetry potential disappears at @xmath89 gev .
as soon as the symmetry potential of @xmath1 is introduced , the motions of @xmath66 and @xmath68 are also governed by the symmetry potential , in addition to the iso - scalar part of the single - particle potential .
the density dependence of @xmath126 ( see fig .
[ fig1 ] , where @xmath126 of @xmath66 is similar to that of neutron and that of @xmath68 is similar to that of proton ) drives @xmath66 ( @xmath68 ) to high ( low ) density area for soft symmetry potential and to low ( high ) density area for stiff symmetry potential , when the density is higher than the crossing density , and vice versa when the density is lower than the crossing density , though the effect is then much smaller .
simultaneously , @xmath66 ( @xmath68 ) with soft symmetry potential possesses lower ( higher ) symmetry potential energy than that with stiff symmetry potential , when density is higher than the crossing density , and vice versa when density is lower than the crossing density . for the case of @xmath81 gev ,
the time duration at high densities ( @xmath127 ) is much longer than the time when the most of @xmath1 hyperons are produced .
therefore , the situation in this case is mostly like that of the case when the density is higher than the crossing density , i.e. , @xmath66 hyperons move to the high density area and @xmath68 hyperons move to the low density area for the soft symmetry potential .
thus , with the soft symmetry potential , the annihilation of @xmath66 is enhanced and that of @xmath68 is reduced , compared with those with stiff symmetry potential . furthermore , the single - particle energy of @xmath66 with the stiff symmetry potential is higher than that with the soft symmetry potential , which leads to more @xmath66 hyperons emitted for the stiff symmetry potential case than for the soft one .
finally , the @xmath3 ratio with the stiff symmetry potential may exceed the one with that of the soft symmetry potential . on the other hand , for the case of @xmath89 gev , the duration time at high density is much shorter , as is also the case for the @xmath1 hyperons produced at high densities .
also , the nuclear density reduces quickly and the most of @xmath1 hyperons after their production will experience the situation with the density being lower than the crossing density .
consequently , the @xmath3 ratio with the soft symmetry potential may exceed as compared to that with the stiff symmetry potential .
this kind of energy dependence of the behavior of the @xmath3 ratio with respect to the symmetry potential is completely due to the dynamical effect of the symmetry potential of @xmath1 in nuclear medium , which allows us to understand why the behavior of the @xmath3 ratio is different from the @xmath2 ratio .
however , there exists a large uncertainty about the single - particle potential of @xmath1 , especially the symmetry potential , so that the results presented here may not be accurate quantitatively . however , all the features about the energy dependence of the relative values of the @xmath3 ratios , corresponding to the different forms of the density dependence of the symmetry potential , should not change . these features are also useful for us to extract the information about the symmetry potential of @xmath1 hyperon .
this work is supported by the national natural science foundation of china under grant nos .
10175093 and 10235030 , major state basic research development program under contract no .
g20000774 , the knowledge innovation project of the chinese academy of sciences under grant no .
kjcx2-sw - n02 , the cask.c .
wong post - doctors research award fund , and the alexander von humboldt foundation , germany . 99 b.a .
et al . _ ,
. j. phys . *
e7 * , 147 ( 1998 ) .
+ b. alex brown , phys . rev
. lett . * 85 * , 5296 ( 2000 ) .
+ j. margueron , j. navarro , n. van giai , and w. jiang , nucl - th/0110026 .
ko and g.q .
li , j. phys .
* g22 * , 225 ( 1996 )
lee , and g.e .
brown , nucl .
phys . a * 625 * , 372 ( 1997 ) .
+ w. cassing , _ et al .
_ , phys . rep . *
308 * , 65 ( 1999 ) .
+ c. fuchs , prog .
. phys . * 53 * 113 ( 2004 ) .
li , phys . rev
. lett . * 88 * , 192701 ( 2002 ) .
li , nucl .
a * 722 * , 209 ( 2003 ) .
+ t. gaitanos , m. di toro , g. ferini , m. colonna , and h.h .
wolter , nucl - th/0402041 .
+ t. gaitanos , m. di toro , s. typel , v. baran , c. fuchs , v. greco , and h.h .
wolter , nucl .
a * 732 * 24 ( 2004 ) .
et al . _ ,
prog . part .
. phys . * 41 * , 225 ( 1998 ) .
+ m. bleicher , _ et al .
_ , j. phys .
g : nucl . part .
phys . * 25 * , 1859 ( 1999 ) .
+ h. weber , e.l .
bratkovskaya , w. cassing , and h. stcker , phys .
c * 67 * , 014904 ( 2003 ) .
+ m. reiter , e.l .
bratkovskaya , m. bleicher , w. bauer , w. cassing , h. weber , and h. stcker , nucl .
a * 722 * , 142 ( 2003 )
. + a.m. lane , nucl .
a * 128 * , 256 ( 1969 ) .
+ j. dabrowski , phys .
c * 60 * , 025205 ( 1999 ) .
+ a. gal and d.j .
millener , phys .
b * 138 * , 337 ( 1984 ) .
+ t. harada , s. shinmura , y. akaishi , and h. tanaka , nucl .
phys . a * 507 * , 715 ( 1990 ) .
glendenning and s.a .
moszkowski , phys .
* 67 * , 2414 ( 1991 ) .
+ j. schaffner and i.n .
mishustin , phys .
c * 53 * , 1416 ( 1996 ) . | based on the urqmd model , we have investigated the influence of the symmetry potential on the negatively and positively charged @xmath0 and @xmath1 hyperon production ratios in heavy ion collisions at the sis energies .
we find that , in addition to @xmath2 ratio , the @xmath3 ratio can be taken as a sensitive probe for investigating the density dependence of the symmetry potential of nuclear matter at high densities ( 1 - 4 times of normal baryon density ) .
this sensitivity of the symmetry potential to both the @xmath2 and @xmath3 ratios is found to depend strongly on the incident beam energy .
furthermore , the @xmath3 ratio is shown to carry the information about the isospin - dependent part of the @xmath1 hyperon single - particle potential . |
the interplay between superconductivity and ferromagnetism has been the subject of extensive research for many years .
the ferromagnetism , being usually much stronger than superconductivity , is supposed to destroy the latter .
this suppression is caused by two mechanisms .
one of them is related to the internal magnetic field which is created by the ordered magnetic moments .
the internal magnetic field , which is proportional to the magnetization @xmath9 , induces meissner currents and suppresses the superconducting order parameter @xmath10 .
this mechanism was first analyzed by ginsburg @xcite .
he concluded that the coexistence phase could not take place in an ordinary bulk sample .
the second mechanism is due to the direct action of an exchange field @xmath1 on spins of electrons@xcite .
the exchange field tends to align the spins of the electrons in one direction and thereby destroys the singlet cooper pairs .
the influence of the magnetization on the orbital condensate motion may be reduced drastically if the magnetization changes its direction on the scale of the superconducting correlation length in such a way that the averaged magnetic field is zero ( as in the case of a spiral magnetic structure ) .
the action of the exchange field may be also decreased if it is realized via the rkky interaction @xcite ( for details see the review @xcite ) .
the coexistence of superconductivity and ferromagnetism was observed in ternary rare earth compounds @xcite .
the layered superconductor / ferromagnet ( s / f ) structures open new possibilities to achieve the coexistence of ferromagnetism and superconductivity . in such systems
ferromagnetic and superconducting regions are spatially separated ; therefore if the magnetization is oriented parallel to the layers , it does not strongly affect the condensate in the s layers .
on the other hand the cooper pairs penetrates into the f layer over a length which in the dirty limit ( @xmath11 , @xmath12 is the elastic scattering time ) is of the order of @xmath13 , where @xmath3 is the diffusion coefficient in the f layer .
the s / f structure becomes superconducting if the f layers are thin enough .
layered structures can be created artificially or can be found in some high-@xmath5 cuprates , as in the compound @xmath14 ( @xmath15 : rare earth ) .
its structure is similar to the one of the high-@xmath5material @xmath16 .
magnetic ordering in @xmath17 layers ( it is assumed that mostly an antiferromagnetic order is realized ) occurs at @xmath18 and superconducting transition presumably in @xmath19 layers occurs at @xmath20 @xcite .
it is interesting to note that another material from the rutinate class , @xmath21 is an exotic triplet superconductor @xcite .
a few interesting effects were predicted and observed in layered s / f structures . since the experimental pioneering work performed by werthamer et al .
@xcite , it is known that the superconducting critical temperature @xmath22 is lowered in s / f multilayered structure with increasing thickness of the f layers .
this is due to the proximity effect , i.e. superconductivity in the s layer is suppressed to some extent if it is brought into contact with a non - superconducting ( specially magnetic ) layer .
another interesting phenomenon which can occur in s / f layered structures is an oscillatory dependence of the critical temperature @xmath22 and the critical current on the thickness of the f layer @xcite .
these oscillations are related to oscillations of the condensate in the f layers which can be formally obtained by the replacement @xmath23 @xmath24 ( here @xmath25 is the matsubara frequency which enters in the expression for the condensate green s functions ) .
an interesting behavior of the critical josephson current @xmath26 as a function of the exchange field @xmath1 was predicted first by bulaevskii et al .
@xcite ; they found that with increasing @xmath1 the current @xmath26 changes sign and the phase difference @xmath27 is established in a josephson s / f / s junction ( the so called @xmath28-junction ) .
this prediction was confirmed in a recent experiment @xcite .
despite of a large number of theoretical works on equilibrium and nonequilibrium effects in s / f / s structures , some important features of the effects remain unclear . in this report
we discuss the influence of different relative orientations of the magnetizations on the critical josephson current in a tunnel sf / i / fs junction .
its electrodes consist of sf bilayers .
the critical current @xmath26 is
not zero if the thickness of the ferromagnetic f layer is thin enough and depends on relative orientation and absolute values of magnetization or exchange energy @xmath1 .
surprisingly the current @xmath26 turns out to be not decreasing , but increasing function of @xmath1 in the case of antiferromagnetic configuration ( the magnetization vectors are aligned in opposite directions in the electrodes ) .
therefore the critical current at nonzero @xmath1 exceeds its value in the absence of @xmath1 .
the second problem which we discuss here is a possible mechanism of the enhanced conductance measured experimentally in mesoscopic s / f structures .
recent experiments on s / f structures showed that below the critical temperature @xmath22 the conductance of the ferromagnetic wire ( or film ) varies with the temperature in a nonmonotonic way and may exceed its value above @xmath22 @xcite .
the decrease of the conductance , which was also observed , has been explained in a few theoretical papers @xcite .
although it was assumed in some papers @xcite that an increase in the conductance may be due to scattering at the s / f interface , careful measurements demonstrated that the entire change of the conductance is due to an increase of the conductivity of the ferromagnet @xcite .
such an increase would not be a great surprise if instead of the ferromagnet one had a normal metal n. it is well known ( see for review @xcite ) that in s / n structures the proximity effect can lead to a considerable increase of the conductance of the n wire provided its length does not exceed the phase breaking length @xmath29 .
however , in a s / f structure if the superconducting pairing is singlet , the proximity effect is negligible at distances exceeding a much shorter length @xmath30 where @xmath1 is the exchange energy in the ferromagnet .
this reduction of the proximity effect due to the exchange field of the ferromagnet is clear from the picture of cooper pairs consisting of electrons with opposite spins .
the proximity effect is not considerably affected by the exchange energy only if the latter is small enough , _
i.e. _ @xmath31 . in the experiments of refs.@xcite strong ferromagnets , as @xmath32 or @xmath33 , were used .
their exchange energy @xmath1 is by several orders of magnitude larger than @xmath22 , and therefore a singlet pairing is impossible due to the strong difference in the energy dispersions of the two spin bands . at the same time , an arbitrary exchange field can not destroy a triplet superconducting pairing because the spins of the electrons forming cooper pairs are already parallel .
we suggest a new mechanism for the increase of the conductance in s / f structures .
this mechanism is based on the formation of the triplet component in the f wire , which is due to a local inhomogeneity of the magnetization @xmath9 in the vicinity of the s / f interface .
we show that the inhomogeneity generates a triplet component of the superconducting order parameter with an amplitude comparable with that of the singlet pairing .
the penetration length of the triplet component into the ferromagnet is equal to @xmath34 , where the energy @xmath35 is of the order of temperature @xmath36 or the thouless energy @xmath37 @xmath38 is the sample size .
the length @xmath39 is of the same order as that for the penetration of the superconducting pairs into a normal metal and therefore the increase of the conductance due to the proximity effect can be compared to that in an s / n structure .
the inhomogeneity may appear in a natural way as an edge effect or may be created intentionally .
the triplet component we considered differs from that in @xmath40 ; it is almost symmetrical in momentum @xmath41 ( the antisymmetrical part is small ) and is an odd function of the matsubara frequency @xmath42 .
this type of the odd triplet component was for the first time suggested by berezinskii @xcite as a possible condensate phase in superfluid he@xmath43 ( in fact a condensate function symmetrical in @xmath42 and antisymmetrical in momentum @xmath44 takes place in he@xmath43 ) .
later the so called odd superconductivity was discussed in refs.@xcite as a possible mechanism for high @xmath22 superconductivity .
this odd in @xmath42 and even in @xmath44 condensate component is not suppressed by the impurity scattering .
in this section we consider a layered system consisting of two f / s bilayers separated by an insulating layer ( see fig.1 ) . in this case
the josephson critical current is determined by the transparency of the insulating layer and depends on the relative orientation of magnetization in the f layers .
we assume that the f and the s layers @xmath45 are thin enough : @xmath46 , where @xmath47 and @xmath48 . with this assumption
one comes to effective values of the superconducting order parameter @xmath49 , of the coupling constant @xmath50 , and of the magnetic moment @xmath51 described by the following equations @xmath52 @xmath53 where @xmath54 and @xmath55 are the densities of states in the superconductor and ferromagnet , respectively .
first , we analyze the case of a high s / f interface transparency , i.e. @xmath56 . under these conditions
all the green s functions are nearly constant in space and continuous across the s / f interface . in order to find the green s functions we use the usadel equation @xcite , which in the presence of a nonhomogeneous exchange field has the general form @xmath57 + \left[\check{m}_{h},\check{{\bf g}}% \right]\!\ ! = \!\!0
\label{usadel}\ ] ] here @xmath58 is the quasiclassical green s function , which has the form @xmath59
@xmath60 , @xmath61 and the matrices @xmath62 and @xmath63 are the pauli matrices in the nambu and spin space respectively ; @xmath64 , where @xmath65 and @xmath66 are the corresponding unit matrices . the eq.([usadel ] ) is supplemented by the normalization condition @xmath67 the current density is determined by the usual expression @xmath68 in order to find the green s functions @xmath69 , we multiply the components ( 1,1 ) and ( 2,2 ) of the matrix equation ( [ usadel ] ) ( the usadel equations ) by the density of states @xmath70 in the f and the s layers respectively , and integrate over the thickness of the bilayers . neglecting the influence of one bilayer on the other
( this means that @xmath71 at the f / i interface ) , we obtain , in the matsubara representation , the following equation : @xmath72 + \left [ \hat{% \delta}_{eff}\otimes \hat{\sigma}_{3},\check { g}\right ] = 0\,\ , , \label{usa1}\ ] ] here @xmath73 denotes the matsubara green s function , and the matrix @xmath74 is given by @xmath75 where @xmath76 is the matsubara frequency . in what follows we will skip the indices @xmath77 .
we assume that the vector @xmath78 in the left layer is oriented along the z - axis and has the components @xmath79 in the right electrode .
one can simplify eq .
( [ usa1 ] ) in the right bilayer with the help of the following unitary transformation @xmath80 where @xmath81 . in this case one obtains for the both layers the same equation : @xmath82 + \left [ \hat{\delta}_{s}\otimes \hat{\sigma}_{3},% \check{g}\right ] = 0\;. \label{usadelrot1}\ ] ] we can solve eq.([usadelrot1 ] ) by making the ansatz @xmath83 from eq .
( [ usadelrot1 ] ) and the normalization condition ( [ normalization ] ) one can obtain the coefficients @xmath84 s and @xmath85 s . in the left bilayer @xmath73
is given by the expression ( [ ansatz1 ] ) while in the right bilayer it is given by @xmath86 , i.e. @xmath87 according to eq .
( [ current ] ) only the coefficients @xmath88 and @xmath89 will enter in the expression for the josephson current , and they are given by @xmath90 where @xmath91 , and @xmath92 . by writing @xmath93 in the right side one obtains the following expression for the critical current @xmath94 where @xmath95 and @xmath96 is the tunnel resistance of the i layer .
we note that the system under consideration is equivalent to a josephson junction consisting of two magnetic superconductors .
the same structure was also analyzed in ref.@xcite , where the critical current was calculated for different s / f interface transparencies .
the authors have found the conditions under which the system undergoes a transition to the @xmath97 state ; however they analyzed only the case of parallel magnetization .
here we consider two limiting cases : a ) a parallel relative orientation of the magnetizations , i.e. @xmath98 and b ) an antiparallel orientation : @xmath99 . in the case
@xmath98 according to eq .
( [ crit_current_sfisf ] ) , the critical current is given by the expression @xmath100 in writing eq .
( [ crit_curr_para ] ) we assumed that @xmath51 and @xmath101 are the same in both bilayers ( symmetric structure ) .
the dependence of the critical current on the exchange field @xmath51 is shown in fig.2 . at @xmath102
the current @xmath26 is constant up to the value @xmath103 where it drops to zero ; @xmath104 is the effective energy gap @xmath105 at zero temperature and zero exchange field .
this is a consequence of the fact that the order parameter @xmath106 is also constant .
we do not consider here a possible transition to the loff phase predicted by larkin and ovchinnikov ( lo ) @xcite and fulde and ferrell ( ff ) @xcite for the region @xmath107 .
we argue that since the homogeneous superconducting state in this region is a metastable state , its realization is possible .
nevertheless our result is definitely valid for the region of small @xmath51 , and a possible transition to the loff would manifest itself in a drop of the the critical current .
more interesting is the case when the relative orientation of the magnetizations is antiparallel , i.e. @xmath108 .
this case was considered in ref . @xcite .
then , the critical current is given by the expression @xmath109 in this case the dependence of @xmath26 on @xmath51 is completely different from that given by eq .
( [ crit_curr_para ] ) ( see fig.3 ) . the critical current determined by eq .
( [ crit_curr_anti ] ) increases with increasing @xmath51 ( i.e. with increasing either @xmath1 or @xmath110 ) and even diverges at zero temperature when @xmath111 .
of course , there is no real divergence of @xmath26 since , for example , finite temperatures smear out this divergency .
the critical current has a maximum at some value of @xmath112 close to @xmath104 . with decreasing @xmath36 the maximum value of @xmath26 increases and its position
is shifted towards @xmath104 .
for arbitrary relative orientations of magnetizations the expression for @xmath113 can be presented in the form @xmath114 therefore , the singular part is always present and its contribution reaches 100% at @xmath108 .
all the conclusions given above remain valid also for two magnetic superconductors with uniformly oriented magnetizations in each layer . as in the previous case of parallel orientations ,
the state with @xmath115 might be unreachable for the antiparallel orientation due to the appearance of the inhomogeneous loff state .
however the singular behavior of @xmath26 can be realized at smaller values of @xmath1 in a structure with large enough s / f interface resistance @xmath116 . in this case
the bulk properties of the s film are not changed by the proximity of the f film ( to be more precise the condition @xmath117 must be satisfied ; @xmath118 is the specific resistance of the f film ) .
then , as one can readily show @xcite , a subgap @xmath119 arises in the f layer .
the green s functions in the f layer have the same form as in eq .
( [ ansatz1 ] ) with @xmath105 replaced by @xmath120 .
the singularity in @xmath121 occurs at @xmath51 equals to @xmath120 , and the loff state does not arise because the subgap @xmath120 is not determined by the self - consistency equation .
another model of a sf / i / fs josephson junction was considered in a recent work @xcite .
the authors assumed that @xmath122 and that the conductance of the s layer in the normal state is much larger than the conductance of the f layer . according to this model
the critical current may either increase or decrease as a function of @xmath1 , and for certain values of @xmath1 may become negative ( @xmath123-junction ) .
it is worthwhile noting that real structures are much more complicated than the models presented above . in order to obtain a complete description of the systems one should take into account that electrons with up and down spins have different density - of - states , conductivities and transmission coefficient through the s / f interface .
although in almost all known superconductors cooper pairs are in spin singlet state , the triplet superconductivity have been studied in some works .
many years ago berezinskii studied a possible triplet phase in superfluid @xmath124 @xcite .
the triplet component of the condensate function proposed in ref.@xcite was odd in frequency and even in momentum .
however nowadays it is known that this hypothetical condensate function does not take place in @xmath124 .
the discovery of superconductivity in @xmath125 @xcite has arisen the general interest in triplet superconductivity . in the case of s / f structures , the possible role of a triplet component in transport properties
was studied in refs .
@xcite . in both cases
the triplet component arose as a result of mesoscopic fluctuations , and the correction to the conductance were very small . in this section
we suggest another mechanism of formation of triplet pairing in s / f structures .
this is due to a local inhomogeneity of the magnetization in the vicinity of the s / f interface .
we consider the system shown in fig.4 and assume that the magnetization orientation varies linearly from @xmath98 at @xmath126 to @xmath127 at @xmath128 . here
@xmath129 is the angle between @xmath9 and the @xmath130-axis ( the @xmath131-axis is parallel to the f wire ) .
thus in this region the magnetization is given by @xmath132 we also consider the diffusive limit corresponding to short mean free path and to the condition @xmath133 .
thus we may describe the system using eq .
[ usadel ] .
this equation contains the normal @xmath73 and anomalous @xmath134 green s functions , which are @xmath135 matrices in the nambu@xmath136spin space .
due to the strong mismatch of the fermi surfaces the transmission coefficient through the s / f interface is small and therefore we can assume that the anomalous condensate function @xmath134 is also small .
in the case of high transparency the order parameter in the superconductor might be suppressed , and therefore it is also possible to assume a weak proximity effect .
thus in both cases we can use the linearized usadel equation for the retarded matrix ( in spin space ) green s function @xmath137 , which has the form(the index @xmath15 is dropped ) @xmath138 here the matrix @xmath139 is defined as @xmath140 , where @xmath141 varies with @xmath131 as shown in fig.4 .
( [ a2 ] ) is supplemented by the boundary conditions at the interface that can also be linearized @xcite .
assuming that there are no spin - flip processes at the s / f interface , we have @xmath142 where @xmath143 is the resistivity of the ferromagnet , @xmath96 is the s / f interface resistance per unit area in the normal state , and @xmath144 . the solution of eq .
( [ a2 ] ) is trivial in the superconductor but needs some care in the ferromagnet . in the region
@xmath145 the solution @xmath146 can be sought in the form @xmath147 where @xmath148 is again an unitary transformation given by @xmath149 .
substituting eq .
( [ rotation2 ] ) into eq .
( [ a2 ] ) and assuming that the solution depends on the coordinate @xmath131 only we obtain the following equation for @xmath150 @xmath151 here @xmath152 is the anticommutator . in the region
@xmath153 , @xmath150 satisfies eq .
( [ a5 ] ) with @xmath154 .
we see from eq .
( [ a5 ] ) that the singlet and triplet components of the anomalous function @xmath150 inevitably coexist in the ferromagnet .
they are mixed by the rotating exchange field @xmath1 . in the region
@xmath153 these components decouple and their amplitudes should be found by matching the solutions at @xmath128 .
( [ a5 ] ) can be solved exactly .
the solution @xmath150 can be written in the form @xmath155 the function @xmath156 in eq . ( [ 2 ] ) is the amplitude of the triplet pairing , whereas the first and the second term describe the singlet one .
substituting eq .
( [ 2 ] ) into eq . ( [ a5 ] )
we obtain a system of three equations for the functions @xmath157 , @xmath158 and @xmath159 , which can be sought in the form @xmath160 the functions @xmath161 and @xmath162 can be written in a similar way . the eigenvalues @xmath163 obey the algebraic equations @xmath164 where @xmath165 and @xmath166 ( indices @xmath167 were dropped ) .
the eigenvalues @xmath168 are the values at which the determinant of eqs .
( [ 4 ] ) turns to zero . from the first equation of eqs .
( [ 4 ] ) we see that in the homogeneous case ( @xmath154 ) the triplet component has a characteristic penetration length @xmath169 , but we see from eq .
( [ bound - cond ] ) that its amplitude is zero . if @xmath170 , the triplet component @xmath171 is coupled to the singlet component ( @xmath172 , @xmath158 ) induced in the ferromagnet according to the boundary condition eq .
( [ bound - cond ] ) ( proximity effect ) .
if the width @xmath173 is small , the triplet component changes only a little in the region @xmath174 and spreads over a large distance of the order @xmath175 in the region @xmath176 . in the case of a strong exchange field @xmath1
, @xmath177 is very short ( @xmath178 ) , the singlet component decays very fast over the length @xmath177 , and its slowly varying part turns out to be small .
in this case the first two eigenvalues @xmath179 can be used everywhere in the ferromagnet @xmath180 , where @xmath38 is the length of the ferromagnet . as concerns the third eigenvalues , we obtain @xmath181 in the interval @xmath174 , and @xmath182 in the interval @xmath183 .
the amplitude @xmath184 of the slowly varying part of the singlet component is equal to @xmath185 .
all the amplitudes should be chosen to satisfy the boundary conditions at @xmath186 ( eq . ( [ bound - cond ] ) ) and zero boundary condition at @xmath187 . for the triplet component
we obtain ( we restore the indices r(a ) ) @xmath188 ^{-1}\!\right\ } ^{r(a)}\ ; , \label{5}\ ] ] where @xmath189 , @xmath190 is the amplitude of the singlet component at the s / f interface , @xmath191 , @xmath192 , and @xmath193 .
it is clear from eq .
( [ 5 ] ) , that the triplet component is of the same order of magnitude as the singlet one at the interface . indeed ,
for the case @xmath194 we obtain from eq . (
[ 5 ] ) @xmath195 , where @xmath196 is the angle characterizing the rotation of the magnetization . therefore if the angle @xmath197 and the s / f interface transparency is not too small , the singlet and triplet components are not small .
they are of the same order in the vicinity of the s / f interface , but while the singlet component decays abruptly over a short distance ( @xmath198 ) , the triplet one varies smoothly along the ferromagnet , turning to zero at the f reservoir . in fig.5
we plot the spatial dependence of the singlet @xmath199 and the triplet @xmath200 components for two different @xmath201 .
one can see that the singlet component decays abruptly undergoing the well known oscillations @xcite while the triplet one decays to zero slowly .
this decay in the region @xmath174 increases with increasing @xmath201 .
thus , we come to a remarkable conclusion : the penetration of the superconducting condensate into a ferromagnet may be similar to the penetration into a normal metal .
the only difference is that , instead of the singlet component in the case of the normal metal , the triplet one penetrates into the ferromagnet . of course
, in order to induce the triplet component one needs an inhomogeneity of the exchange field at the interface .
the presence of the condensate function ( triplet component ) in the ferromagnet can lead to interesting long - range effects .
one of them is a change of the conductance of a ferromagnetic wire in a s / f structure ( see inset in fig.4 ) when the temperature is lowered below @xmath22 .
this effect was observed first in s / n structures and later was successfully explained ( see , e.g. reviews @xcite ) .
now we consider the s / f structure shown in the inset of fig.1 .
the normalized conductance variation @xmath202 is given by the expression @xcite : @xmath203 ^{2}\right\rangle \;. \label{6}\ ] ] here @xmath204 is the conductance in the normal state , @xmath205 denotes the average over the length of the ferromagnetic wire between the f reservoirs , and @xmath206 is given by the expression @xmath207\ ; .\ ] ] the function @xmath146 is given by the third term of eq .
( [ 2 ] ) with @xmath208 ( we neglect the small singlet component ) .
substituting eqs .
( [ 2 ] , [ 5 ] ) into eq . ( [ 6 ] ) one can determine the temperature dependence @xmath209 .
fig.6 shows this dependence .
we see that @xmath210 increases with decreasing temperature and saturates at @xmath102 .
this monotonic behavior of @xmath210 contrasts with the so called reentrant behavior of @xmath210 in s / n structures @xcite and is a result of broken time - reversal symmetry of the system under consideration .
available experimental data are still controversial . it has been established in a recent experiment @xcite that the conductance of the ferromagnet does not change below @xmath22 and all changes in @xmath211 are due to changes of the s / f interface resistance @xmath212 . however ,
in other experiments @xmath212 was negligibly small @xcite .
the mechanism suggested in our work may explain the long - range effects observed in the experiments @xcite . at the same time , the result of the experiment @xcite is not necessarily at odds with our findings .
the inhomogeneity of the magnetic moment at the interface , which is the crucial ingredient of our theory , is not a phenomenon under control in these experiments .
one can easily imagine that such inhomogeneity existed in the structures studied in refs .
@xcite but was absent in those of ref . @xcite .
the magnetic inhomogeneity near the interface may have different origins .
anyway , a more careful study of the possibility of a rotating magnetic moment should be performed to clarify this question . in order to explain the reentrant behavior of @xmath213 observed in refs .
@xcite one should take into account other mechanisms , as those analyzed in refs .
@xcite . however , this question is beyond the scope of the present paper .
we note that at the energies @xmath214 of the order of thouless energy @xmath215 the triplet component spreads over the full length @xmath38 of the ferromagnetic wire ( see fig.2 ) .
this long - range effect differs completely from the proximity effect in a ferromagnet with a uniform magnetization considered recently in ref.@xcite . in the latter case
the characteristic wave vector is equal to @xmath216 ( cf .
( [ 4 ] ) ) .
it was noted in ref .
@xcite that if @xmath217 , then @xmath218 and the singlet component penetrates in the ferromagnet .
if the characteristic energies @xmath219 are much less than @xmath1 , the penetration length @xmath220 is of the order @xmath177 and is much shorter then @xmath221 or @xmath38 .
we analyzed two effects related to a nonhomogeneous magnetization in s / f structures .
first we have calculated the josephson critical current @xmath26 in a tunnel s / f - i - s / f junction .
we have shown that in contrast to a common view , in the case of an antiparallel configuration ( the magnetization vectors in the s / f electrodes are antiparallel to each other ) the critical current @xmath26 increases with the increasing exchange field @xmath1 .
this means that in a s / f - i - f / s junction the current @xmath26 can be even greater than that in a s - i - s junction with the same s electrodes .
secondly we have calculated the conductance of a mesoscopic f wire attached to a superconductor s. we have shown that in the presence of a local inhomogeneity near the s / f interface , both singlet and triplet components of the condensate are created in the ferromagnetic wire due to the proximity effect .
the singlet component penetrates into the ferromagnet over a short length @xmath177 , whereas the triplet component can spread over the full mesoscopic length of the ferromagnet .
this long - range penetration of the triplet component should lead to a significant variation of the ferromagnet conductance below @xmath22 .
we thank sfb 491 _ magnetische heterostrukturen _ for financial support .
l. bauernfeind _
et al . _ , physica c * 245*,151 ( 1995 ) ; c. bernhard _ et al .
b * 59 * , 14 ( 1999 ) ; j. l. tallon _ et al . _ , phys . rev .
b * 61 * , r6471 ( 2000 ) ; a. c. mclaughlin _
et al . _ ,
b * 60 * , 7512 ( 1999 ) ; o. chmaissem _
et al . _ ,
b * 61 * , 6401 ( 2000 ) ; ken - ichi kumagai _
et al . _ ,
b * 63 * , 180509(r ) ( 2001 ) .
v.t.petrashov _ et al _ , jetp lett . * 59 * , 551 ( 1994 ) ; v.t .
petrashov , i.a.sosnin , i.cox , a.parsons , and c.troadec , phys.rev.lett . * 83 * , 3281 ( 1999 ) ; v.t .
petrashov , i.a.sosnin , and c.troadec , cond - mat/0007278 .
fig.2:dependence of the normalized critical current on @xmath1 for different temperatures in the case of a parallel orientation .
here @xmath222 , @xmath51 is the effective exchange field , @xmath223 and @xmath224 is the superconducting order parameter at @xmath102 and @xmath225 .
fig.3 : dependence of the normalized critical current on @xmath1 for different temperatures in the case of an antiparallel orientation . here
@xmath222 , @xmath51 is the effective exchange field , @xmath223 and @xmath224 is the superconducting order parameter at @xmath102 and @xmath225 .
fig.4 : schematic view of the structure under consideration . in the inset
is shown the structure , for which we calculate the conductance variation : two ferromagnetic wires connected to two ferromagnetic and two superconducting reservoirs .
fig.5 : spatial dependence of the singlet ( dashed line ) and triplet ( solid line ) components of @xmath226 in the f wire for different values of @xmath227 .
here @xmath228 , @xmath229 and @xmath230 .
@xmath231 is the thouless energy . | we study two different superconductor - ferromagnet ( s / f ) structures . we consider first a josephson junction which consists of two s / f bilayers separated by an insulating layer .
we show that for an antiparallel alignment of the magnetization in the two f layers the josephson critical current @xmath0 increases with increasing exchange field @xmath1 .
the second system we consider is a s / f structure with a local inhomogeneity of the magnetization in the ferromagnet near the s / f interface . due to the proximity effect not only a singlet but also a triplet component of the superconducting condensate is induced in the ferromagnet .
the latter penetrates over the length @xmath2 ( @xmath3 is the diffusion coefficient and @xmath4 the energy ) . in the case of temperatures of the order of the thouless energy
this length is comparable to the length of the ferromagnet .
this long - range penetration leads to a significant increase of the ferromagnet conductance below the superconducting critical temperature @xmath5 .
contrary to the case of the singlet component , the contribution to the conductance due to the odd triplet component is not zero at @xmath6 and @xmath7 ( @xmath8 is the voltage ) and decays with increasing temperature t in a monotonic way .
pacs numbers : 74.80.dm , 74.25.fy , 74.50.+r .
keywords : proximity effect , josephson effect , triplet superconductivity , ferromagnetism . |
unification theories applied to cosmology suggest a possibility of variation of the fundamental constants in the expanding universe ( see e.g. review @xcite ) . there are hints of variation of @xmath6 and @xmath7 in quasar absorption spectra , big bang nucleosynthesis and oklo natural nuclear reactor data ( see @xcite and references therein ) . here
@xmath8 is the quantum chromodynamics ( qcd ) scale , and @xmath9 and @xmath10 are the quark and electron masses .
however , the majority of publications report only limits on possible variations ( see e.g. @xcite ) . a very sensitive method to study the variation in a laboratory
consists of the comparison of different optical and microwave atomic clocks ( see e.g. measurements in @xcite ) .
an enhancement of the relative effect of @xmath6 variation can be obtained in a transition between almost degenerate levels in dy atom @xcite .
these levels move in opposite directions if @xmath6 varies .
an experiment is currently underway to place limits on @xmath6 variation using this transition @xcite , but unfortunately one of the levels has quite a large linewidth and this limits the accuracy .
an enhancement of 1 - 5 orders exists in narrow microwave molecular transitions @xcite .
some atomic transitions with enhanced sensitivity are listed in ref .
@xcite . in ref .
@xcite it was suggested that there may be a five orders of magintude enhancement of the variation effects in the low - energy transition between the ground and the first excited states in the @xmath0th nucleus .
the exsitence of the enhancement was confirmed in @xcite .
this transition in @xmath0th was suggested as a possible nuclear clock in ref .
indeed , the transition is very narrow .
the width of the excited state is estimated to be about @xmath11 hz @xcite ( the experimental limits on the width are given in @xcite ) .
the latest measurement of the transition energy @xcite gives @xmath12 ev , compared to earlier values of @xmath13 ev @xcite and @xmath14 ev @xcite .
this makes @xmath0th a possible reference for an optical clock of very high accuracy , and opens a new possibility for a laboratory search for the variation of the fundamental constants .
large interest in experiments exploring this possibility was expressed , for example , in e. hudson and e. peik talks at special workshop ( 14 - 18 july 2008 , perimeter institute ) devoted to the variation of the fundamental constants and privite communications of e. peik , p. beiersdorfer , zheng - tian lu , d. demille , d. habs and j. torgerson .
this motivated us to perform an independent calculation of the sensitivity of the transition frequency @xmath15 to the variation of @xmath6 based on the following relation suggested in ref .
@xcite : @xmath16 where @xmath17 is the contribution of the coulomb interaction to the energy spacing between these two levels .
the calculation of the difference in total coulomb energy between ground and excited nuclear states can hardly be done within existing nuclear models with the required accuracy .
instead , the calculation of the correction to the coulomb energy when one proton or one neutron is added into a single particle orbital @xmath18 is a much simpler problem .
the ground state of @xmath0th nucleus is @xmath19=[633\,5/2^+]$ ] ; i.e. the deformed oscillator quantum numbers are @xmath20 , @xmath21 , the projection of the valence neutron orbital angular momentum on the nuclear symmetry axis ( internal z - axis ) is @xmath22 , the spin projection @xmath23 , and the total angular momentum and the total angular momentum projection are @xmath24 .
the excited state is @xmath19=[631\,3/2^+]$ ] ; i.e. it has the same @xmath20 and @xmath21 .
the values @xmath25 , @xmath26 and @xmath27 are different . with relation to the variation of alpha , we shall discuss below the contribution of the coulomb interaction to the energy spacing between these two levels .
we start from an even @xmath28th nucleus and calculate the shift of nuclear coulomb energy when an odd particle is added into single particle state @xmath18 . in an even nucleus the coulomb energy can be presented as @xmath29 where @xmath30 is the charge density . for uniformly charged sphere
we obtain the well known expression @xmath31 which gives for z=90 , and a=228 , the estimate , @xmath321 gev .
the above equation can also be used to estimate the change in the coulomb energy when one neutron is added .
an addition of the neutron changes a , a@xmath33a+1 , therefore @xmath34 that gives for @xmath17 the value of the order of -1 mev .
this estimate gives some average macroscopic value for @xmath17 because in reality it depends on the state @xmath18 where we put the neutron .
adding an odd particle to the state @xmath18 we obtain a change in the charge density @xmath35 where @xmath30 is the charge density of a neighbor even nucleus .
for the change in the coulomb energy eq.([eq1 ] ) we obtain @xmath36 where @xmath37 is the single particle coulomb potential @xmath38 although @xmath28th is a deformed nucleus , for the estimates we shall neglect a small quadrupole component in the charge density @xmath30 .
effects of deformation will be discussed below . in approximation of uniformly charged sphere @xmath39 where @xmath40 is a coulomb radius of @xmath28th , @xmath41fm .
we use this very simple form in order to obtain an estimate for the change in the coulomb energy when going from one nuclear state to another state , and not in the absolute value .
the ground state of the nucleus @xmath0th is polarized due to the addition of the nucleon to the @xmath28th core .
the added nucleon introduces an admixture of the monopole state ( which is a radial excitation ) @xcite and the nuclear state is now : @xmath42 and @xmath43 are the ground state and the isovector monopole in @xmath28th and @xmath44 is the wave function of the added nucleon . with @xmath45 where @xmath46 is a two - body nuclear interaction .
now we evaluate the coulomb energy of this new polarized state @xmath47 .
the correction is : @xmath48 @xmath49 is the coulomb potential .
this correction can be written in the form : @xmath50 @xmath51 is the transition density between the ground state and the monopole : @xmath52 the z / a factor takes into account the fact that only protons interact via the coulomb .
one form used for the monopole transitions is : @xmath56 where @xmath57 is a constant and @xmath58 is the ground state density .
this form for the transition density corresponds to a change obtained by uniformly expanding the nucleus , but also can be derived from a sum rule @xcite .
this transition density has a node . because of the node in @xmath59 the result for
the polarization correction will be sensitive to the spatial form of the wave function @xmath44 , and even may change sign for different single - particle states . to obtain a quantitative estimate
let us switch to the particle - hole basis @xmath60 instead of the single monopole resonance @xmath43 .
the correction to the charge density introduced in eq.([eq2 ] ) can be presented as follows : @xmath61 where @xmath62 is the single - particle wave function , and @xmath63 is the correction to the proton density matrix of a neighbohr even nucleus , due to the odd nucleon added into the state @xmath64 .
@xmath65 here @xmath66 is the proton density matrix of the neighbohr even nucleus .
the equation for @xmath63 can be found for example in @xcite .
it states @xmath69 here @xmath70 is the matrix element of effective residual interaction between quasi - particles .
the first term in r.h.s . of eq.([eq8 ] ) exists only for an odd proton .
if the odd particle is a neutron , this term is absent .
the second term is the polarization correction which exists both for an odd proton and an odd neutron . in the first order in proton neutron residual interaction we obtain for the correction to the proton density matrix @xmath71 where the state @xmath72 refers to a neutron .
the correction to coulomb energy becomes @xmath73 eq.([eq10 ] ) can be presented in the following way , @xmath74 here we introduced the correction to potential energy @xmath75 that describes the reaction of the proton core to the presence of an additional neutron @xmath76 it follows from eqs.([eq10],[111 ] ) that the correction @xmath75 does not depend on the neutron state .
however , its mean value given by eq.([eq11 ] ) does depend on the particular state @xmath77 .
the radial dependence of @xmath75 is shown in fig.1 .
this radial dependence shown is typical for a double charged layer . due to small compressibility of a nucleus
the change of a charge density can take place on nuclear surface only , and in good approximation it can be presented as a double charged layer .
the mean value of the correction @xmath75 over nuclear volume is negative and equal to -440 kev . due to oscillating behavior
, the expectation value of @xmath75 will obviously be state dependent . the calculations for @xmath0th were performed using nilsson wave functions @xcite for the valence neutron in order to account for deformation . in @xmath0th the ground state
has the quantum numbers [ 633 5/2 + ] while the excited state has [ 631 3/2 + ] .
both of them belong to n=6 shell where the possible angular momenta in the nilsson orbitals are l=2 , 4 , 6 .
the shift of coulomb energy calculated due to add of valence neutron into these states is shown in table 1 .
note , that the shift for @xmath78 and @xmath79 becomes positive .
the radial wave function for l=6 has a sharp peak near nuclear surface where @xmath75 is positive .
the state @xmath79 has smaller binding energy and its peak , compared to @xmath78 state , is shifted to larger r where @xmath75 becomes negative .
this leads to a smaller value of the shift for @xmath79 state .
the difference between the first excited state with quantum numbers [ 631 3/2 ] and the ground state [ 633 5/2 ] in @xmath0th is extremely small ( @xmath80 10 ev ) and can not be reproduced in nilsson model at reasonable deformation @xmath81 . with the standard parameters of the nilsson model @xcite these two levels are crossed at the deformation @xmath82 .
therefore , we calculated the difference in coulomb energies @xmath83 at different deformations , as a function of the parameter @xmath84 , where @xmath85 is the spin - orbit constant .
the result is shown in fig.2 for @xmath86 .
this interval corresponds to deformations @xmath87 .
note that the difference becomes negative at the deformations @xmath88 .
however , the position of this interval depends on the parameters of the nilsson model , while the fact that the difference changes sign at some deformation remains in all sets of the parameters . the value of @xmath2 as a function of deformation changes from 1.5 mev at zero deformation down to -0.5 mev at @xmath89 . in our approach we can hardly give a better estimate , due to simplified treatment of deformation .
however , very small value of the coulomb energy shift seems unprobable . note that near the nilsson level crossing the difference of the coulomb energies is about 1 mev .
the shift of the order from several tens kev to several hundreds kev provides an enhancement of sensitivity to changes of @xmath6 , as it was mentioned in @xcite .
indeed , the frequency shift from eq.([eq0 ] ) may be presented in the following form @xmath90 this shift is four to five orders of magnitude larger than the shift which was studied in optical atomic clocks @xcite ) .
an additional enhancement may be due to the very small width of this transition ( @xmath11 hz @xcite ) .
it is also instructive to present the relative enhancement : @xmath91 these estimates confirm conclusion of ref .
@xcite that experiments with @xmath0th have potential to improve sensitivity of laboratory measurements of the variation of the fundamental constants up to 8 orders of magnitude . n.a .
thanks byron jennings for the hospitality at triumf , v.f . is grateful to perimeter institute for the hospitality and support .
this work was supported in part by the us - israel binational science foundation and australian research council 99 j - p .
uzan , rev .
* 75 * , 403 ( 2003 ) ; v. v. flambaum , int . j. mod
a * 22 * , 4937 ( 2007 ) .
m.t.murphy , j.k .
webb , v.v .
mon . not .
r. astron .
345 , 609 - 638 ( 2003 ) .
dmitriev , v.v .
flambaum , j.k .
webb , phys .
d69 , 063506 ( 2004 ) .
s. k. lamoreaux and j. r. torgerson _ phys .
d _ * 69 * 121701(r ) ( 2004 ) .
e. reinhold , r. buning , u.hollenstein , a. ivanchik , p. petitjean , w. ubachs .
* 96 * , 151101 ( 2006 ) .
murphy , j.k .
webb , v.v .
flambaum , phys .
99 * , 239001 ( 2007 ) .
s. g. karshenboim , v. v. flambaum , e. peik .
atomic clocks and constraints on variations of fundamental constants . in `` handbook of atomic , molecular and optical physics ''
drake , springer 2005 ) , chapter 30 , pp.455 - 463 ; arxiv : physics/0410074 .
e. peik , b. lipphardt , h. schnatz , t. schneider , chr .
tamm , s. g. karshenboim , arxiv : physics/0504101 .
s. blatt , a.d .
ludlow , g.k .
campbell , j.w .
thomsen , t. zelevinsky , m.m .
boud , j. ye , x. baillard , m. fouche , r. le targat , a. brusch , p. lemonde , m. takamoto , f .-
hong , h. katory , v.v .
100 , 140801 ( 2008 ) .
t. rosenband _
et al . _ ,
science * 319 * , 1805 ( 2008 ) . v. a. dzuba , v. v. flambaum , j. k. webb , phys . rev .
a * 59 * , 230 ( 1999 ) .
a. t. nguyen , d. budker , s. k. lamoreaux and j. r. torgerson _ phys .
a. _ * 69 * , 022105 ( 2004 ) .
s. j. ferrell , a. cingz , a. lapierre , a .- t .
nguyen , n. leefer , d. budker , v. v. flambaum , s. k. lamoreaux , and j. r. torgerson , phys .
a. 76 , 062104 ( 2007 ) .
v. v. flambaum , phys .
a * 73 * , 034101 ( 2006 ) . v. v. flambaum , m. g. kozlov , phys .
rev . lett.*99 * , 150801 ( 2007 ) .
d. demille , s. sainis , j. sage , t. bergeman , s. kotochigova , e. tiesinga , phys .
rev . lett.*100 * , 043202 ( 2008 ) .
e. j. angstmann , v. a. dzuba , v. v. flambaum , s. g. karshenboim , a. yu .
nevsky , j. phys .
b. * 39 * , 1937 ( 2006 ) , physics/0511180 .
v. v. flambaum , phys .
lett . * 97 * , 092502 ( 2006 ) .
xiao - tao he and zhong - zhou ren , j. phys .
g : nucl . part .
phys . * 34 * , 1611 ( 2007 ) .
e. peik and chr .
tamm , europhys .
* 61 * , 181 ( 2003 ) .
e. v. tkalya , a. n. zherikhin , v. i. zhudov , phys .
c * 61 * , 064308 ( 2000 ) .
a. m. dykhne , e. v. tkalya , pisma zh.eks.teor.fiz . * 67 * , 233 ( 1998 ) [ jetp lett .
* 67 * , 251 ( 1998 ) ] .
i. d. moore , i. ahmad , k. bailey , d. l. bowers , z .-
lu , t. p. oconnor , z. yin .
argonne national laboratory report phy-10990-me-2004 .
b. r. beck , j. a. becker , p. beiersdorfer , g. v. brown , k. j. moody , j. b. wilhelmy , f. s. porter , c. a. kilbourne , and r. l. kelley , phys .
rev . lett . * 98 * , 142501 ( 2007 ) .
r. g. helmer and c. w. reich .
c * 49 * , 1845 ( 1994 ) .
guimaraes - filho , o. helene .
c * 71 * , 044303 ( 2005 ) .
g. m. irwin , ki - hyon kim , bull .
* 42 * , 1071 ( 1997 ) .
a. c. hayes and j. l. friar , phys .
b * 650 * , 229 ( 2007 ) .
n. auerbach , phys .
36b * , 293 ( 1971 ) .
n. auerbach , nucl .
a229 * , 447 ( 1974 ) ; n. auerbach , phys . rep .
* 98*,273 , ( 1983 ) .
migdal `` theory of finite fermi systems '' , ( wiley interscience , new york , 1967 ; nauka , moscou , 1983 ) .
benjamin e. chi , nucl . phys . * 83 * , 97 ( 1966 ) . | we calculated the contribution of coulomb energy to the spacing between the ground and first excited state of @xmath0th nucleus as a function of the deformation parameter @xmath1 .
we show that despite the fact that the odd particle is a neutron , the change in coulomb energy @xmath2 between these two states can reach several hundreds kev .
this means that the effect of the variation of the fine structure constant @xmath3 may be enhanced @xmath4 times in the @xmath57.6 ev `` nuclear clock '' transition between the ground and first excited states in the @xmath0th nucleus . |
the number of dark spots in the sun s surface has been counted in a systematic way since rudolf wolf introduced the concept , in the first half of the nineteenth century .
more than any other solar observable , the sunspot number is considered the strongest signature of the 22-year magnetic cycle . moreover , since the sunspot number is the longest time series from all solar observables @xcite , it makes it the preferred proxy to study the variability and irregularity of the solar magnetic cycle . in the sun
s interior the large scale magnetic field is generated by a magnetohydrodynamic dynamo that converts part of the kinetic energy of the plasma motions into magnetic energy .
polarity reversals occur every 11 years approximately , as it can be observed directly in the sun s dipolar field , and taking a full 22-years to complete a magnetic cycle .
in fact during each magnetic cycle , the sun experiences two periods of maximum magnetic activity , during which magnetic flux tubes created in the tachocline layer , rise to the sun s surface by the action of buoyancy , emerging as sunspots pairs @xcite .
the polarity switch is also observed in the change of polarity alignment of these bipolar active regions .
although we know that the solar dynamo resides within the convection zone , we still do nt have a complete picture where all the physical mechanisms operate @xcite .
there is a strong consensus that the physical mechanism behind the production of the large scale toroidal field component , the so called @xmath0-effect , is located in the tachocline , a shear layer created by differential rotation and located at the base of the convection zone .
the major source of uncertainty is the location of the @xmath1-effect , the physical mechanism responsible to convert toroidal into poloidal field and close the system . in truth
, this effect could be in fact a collection of several physical mechanisms that operate at different places and with different efficiencies .
some examples are the babcock - leighton mechanism that operates in the solar surface and converts the product of decaying active regions into poloidal field , or the action of the turbulent magnetic helicity that takes place in the bulk of the convection zone .
one of the main questions that is still being debated is the quantification of the importance and relative contribution of each component to the operation of the solar dynamo . because different authors choose to give the leading role to one or another @xmath1 source term , there is vast number of dynamo models .
most of these are two dimensional models ( usually referred as 2.5d because they include two spatial coordinates plus time ) and are constructed using the mean - field theory framework proposed by @xcite . despite some short - comes , fruit of the approximations and formulation used , this type of models running in the kinematic regime , i.e. with prescribed large scale flows , has been very popular within the solar community because they can explain many of the observable features of the solar cycle . a detailed discussion on solar dynamo models , stellar magnetism and corresponding references to the vast literature on this subject can be found in the reviews by @xcite and @xcite . another way of tackling the solar dynamo problem is by producing 3d magnetohydrodynamic ( mhd ) simulations of the solar convection zone . these computer intensive simulations
solve the full set of the mhd equations ( usually under the anelastic approximation ) and are fully dynamical in every resolved scale , i.e. they take into consideration the interactions between flow and field and vice - versa unlike the kinematic regime usually used in mean field models , where only the flow influences the field .
recently these simulations have started to show stable large scale dynamo behaviour and they are starting to emerge as virtual laboratories for understanding in detail some of the mechanisms behind the dynamo @xcite . on the other end of the modelling spectrum , we can find oscillator models , that use simplified parameterizations of the main physical mechanisms that participate in the dynamo process .
although in the sun s interior the magnetic field generated by the dynamo has a very rich and complex structure , as a consequence of the structure of the magnetohydrodynamic differential equations , some of its main properties can be understood by analyzing low order differential equations obtained by simplification and truncation of their original mhd counterparts
. then , several properties of the cycle that can be extracted by studying these non - linear oscillator models , as is usually done in nonlinear dynamics .
these models have a solid connection to dynamical systems and are , from the physics point of view the most simple .
this does not mean that they are the easiest to understand because the reduction in the number of dimensions can sometimes be difficult to interpret ( viz .
introduction section of @xcite ) . these low order dynamo models ( lodm ) ,
as they are some times called , allow for fast computation and long integration times ( thousands of years ) when compared to their 2.5d and 3d counterparts .
they can be thought as a first order approximation to study the impact of certain physical mechanisms in the dynamo solution , or some of the properties of the dynamo itself as a dynamical system .
the variability exhibited by the sunspot number time series , inspired researchers to look for chaotic regimes in the equations that describe the dynamo . for a complete review on this subject consult @xcite and references therein .
some of the first applications of lodm were done in this context ( e.g. @xcite ) .
these authors found solutions with cyclic behaviour and variable amplitude , including extended periods of low amplitude reminiscent of the grand minima behaviour we see in the sun .
the downside of these initial works was the fact that although the proposed model equations made sense from a mathematical point of view , the physics realism they attained was small .
these low order models , with higher or lower degrees of physical complexity , can be used in many areas and several of their results have been validated by 2.5d spatially distributed mean field models , which grants them a certain degree of robustness .
this happens specially in lodm whose formulation is directly derived from mhd or mean - field theory equations .
some examples of the results obtained with lodm that have been validated by more complex mean field models are : the study of the parameter space , variability and transitions to chaos in dynamo solutions @xcite ; the role of lorentz force feedback on the meridional flow @xcite ; and the influence of stochastic fluctuations in the meridional circulation and in the @xmath1-effect @xcite .
some models even include time delays that embody the spatial segregation and communication between the location of source layers of the @xmath1- and @xmath0-effects .
these have been applied to a more general stellar context by @xcite and recently , @xcite showed that one of this type of time delay lodm that incorporates two different @xmath1 source terms working in parallel , can explain how the sun can enter and exit in a self - consistently way from a grand minimum episode .
a couple of lodm even ventured in the `` dangerous '' field of predictions .
for example @xcite combined his lodm with an autoregressive model in order to forecast the amplitude of future solar cycles . in this article
we show how can one of these lodm be used as a tool to study the properties of the solar magnetic cycle . for this purpose
we use the international sunspot number time series during the past 23 solar magnetic cycles .
nevertheless , the main focus of this work is to present a strategy inspired by helioseismology , were an _ inversion methodology _
is used to infer variations of some of the lodm parameters over time .
since these parameters are related to the physical mechanisms that regulate the solar dynamo , this should in principle , allow for a _ first order _ reconstruction of the main dynamo parameters over the last centuries . in a similar manner to helioseismology , the comparison between model solutions and data can be done by means of a _ forward method _ in which solar observational data is directly compared with the theoretical predictions , or by means of a _
backward method _ in which the data is used to infer the behaviour of leading physical quantities of the theoretical model .
naturally , it is necessary to develop an inversion technique or methodology that allows to reconstruct the quantities that have changed during the evolution of solar dynamo .
this type of studies is well suited to explore several aspects of the solar and stellar dynamo theory .
this can be done by : _
( i ) _ building a tool to study the dynamo regimes operating in stars ; _ ( ii ) _ establishing an inversion methodology to infer the leading quantities responsible for the dynamics and variability of the solar cycle over time ; _ ( iii ) _ comparing the dynamo numerical simulations with the observational data ; _ ( iv ) _ use this tool as a toy model to test global properties of the solar dynamo . here
we particularly focus in discussing the three last items of this list , with special attention on the development of an inversion method applied here to the sunspot number time series .
this is used to infer some of the dynamics of the solar dynamo back - in - time . in principle
this should allow us to determine the variation profiles of the quantities that drive the evolution of the magnetic cycle during the last few centuries . in section [ sec : lodmlsmf ]
, we present a non - linear oscillator derived from the equations of a solar dynamo that is best suited to represent the sunspot number . in section [ sec : lodminversion ] , we discuss how the non - linear oscillator analogue can be used to invert some of the leading quantities related with solar dynamo . in section
[ sec : smc ] is discussed how solar observational data is use to infer properties of the solar magnetic cycle . in section [ sec : numericmodels ] we present a discussion about how the low order dynamo model can be used to test the basic properties of modern axisymmetric models and numerical simulations , as well to infer some leading properties of such dynamo models . in section [ sec :
outlook ] , we discuss the outlook for the sun and other stars .
the basic equations describing the dynamo action in the interior of a star are obtained from the magnetic - hydrodynamic induction , and the navier - stokes equations augmented by a lorentz force @xcite . under the usual kinematic approximation the dynamo problem consists in finding a flow field with a velocity @xmath2 that has the necessary properties capable of maintaining the magnetic field , @xmath3 against ohmic dissipation @xcite . for a star like the sun
such dynamo models should be able to reproduce well - known observational features such as : cyclic magnetic polarity reversals with a period of 22 years , equatorward migration of @xmath3 during the cycle ( dynamo wave ) , the @xmath4 phase lag between poloidal and toroidal components of @xmath3 , the antisymmetric parity across the equator , predominantly negative / positive magnetic helicity in the northern / southern hemisphere , as well as many of the empirical correlations found in the sunspot records , like the waldmeier rule anti - correlation between cycle duration and amplitude ; the gnevyshev - ohl rule alternation of higher - than - average and lower - than - average cycle amplitude and grand minima episodes ( like the maunder minimum ) epochs of very low surface magnetic activity that span over several cycles . given the amount of complex features that a solar dynamo model has to reproduce , the task at hand is far from simple .
the vast majority of dynamo models currently proposed to explain the evolution of the solar magnetic cycle ( kinematic mean - field models ) became very popular with the advance of helioseismology inversions and the inclusion of the differential rotation profile . in the kinematic regime approximation , the flow field @xmath2 is prescribed and only the magnetic induction equation is used to determine the evolution of @xmath3 . generally , the large scale magnetic field , the one responsible for most of the features observed in the sun is modelled as the interaction of field and flow where two source terms ( @xmath0 and @xmath1 ) naturally emerge from mean - field theory ( e.g. , * ? ? ? * ; * ? ? ?
* ; * ? ? ?
from the mean - field electrodynamics , the induction equation reads @xmath5 where @xmath2 is the large - scale mean flow , and @xmath6 is the total magnetic diffusivity ( including the turbulent diffusivity and the molecular diffusivity ) .
currently , as inferred from helioseismology , @xmath2 can be interpreted as a large - scale flow with at least two major flow components , the differential rotation throughout the solar interior , and the meridional circulation in the upper layers of the solar convection @xcite .
given all the points above , and based on their popularity among the community , we start our study by considering a reference model based in the kinematic mean - field flux transport framework .
although the results obtained here are based on this specific type of model , most of the analysis method used , as well the conclusions reached , can easily be extended to other models . as usual , under the simplification of axi - symmetry the large - scale magnetic field @xmath3 can be conveniently expressed as the sum of toroidal and poloidal components , that in spherical polar coordinates @xmath7 can be written as @xmath8 similarly , the large - scale flow field @xmath2 as probed by helioseismology can be expressed as the sum of an axisymmetric azimuthal ( differential rotation ) and poloidal ( meridional flow ) components : @xmath9 where @xmath10 , @xmath0 in the angular velocity and @xmath11 is the velocity of the meridional flow . accordingly
, such decomposition of @xmath3 ( that satisfy the induction equation [ eq : vecb ] ) and @xmath2 leads to the following set of equations : @xmath12\cdot \nabla \omega -\gamma ( b_\phi ) b_\phi \label{eq : bphi}\end{aligned}\ ] ] where @xmath6 is the magnetic diffusivity and @xmath1 is the source term of @xmath13 ( the mechanism to convert toroidal to poloidal field ) .
moreover , following the suggestions of @xcite we also considered that the toroidal field can be removed from the layers where it is produced by magnetic buoyancy and obeying @xmath14 , where @xmath15 is a constant related to the removal rate and @xmath16 is the plasma density .
as the sun s magnetic field changes sign from one solar cycle to the next it is a plausible idea to attribute alternating signs in odd / even cycles also to other solar activity indicators such as the sunspot number ( ssn ) .
the resulting time series displays cyclic variations around zero in the manner of an oscillator .
this suggests an oscillator as the simplest mathematical model of the observed ssn series . as , however
, the profile of sunspot cycles is known to be markedly asymmetric ( a steep rise in 34 years from minimum to maximum , followed by a more gradual decline to minimum in @xmath17 years ) , a simple linear oscillator would be clearly a very poor representation of the sunspot cycle .
a _ damped _ linear oscillator @xmath18 will , on the other hand , naturally result on asymmetric profiles similar to what is observed .
the obvious problem that the oscillation will ultimately decay due to the damping could be remedied somewhat artificially by applying a periodic forcing or by reinitializing the model at each minimum . a much more natural way to counteract the damping , however ,
is the introduction of nonlinearities into the equation indeed , such nonlinearities are naturally expected to be present in any physical system , see below .
as long as the nonlinearity is relatively weak , the parameters @xmath19 and @xmath20 can be expanded into taylor series according to @xmath21 . due to the requirement of symmetry ( i.e. the behaviour of the oscillators should be invariant to a sign change in @xmath21 ) only terms of even degree will arise in the taylor series . to leading order , then , we can substitute @xmath22 into equation ( [ eq : linosc ] ) resulting in @xmath23 in the particular case when @xmath24 ( i.e. the nonlinearity affects the damping only ) and the other parameters are positive , the system described by equation ( [ eq : nonlinosc ] ) is known as a van der pol oscillator .
the alternative case when nonlinearity affects the directional force / frequency only , i.e. @xmath25 , @xmath26 and @xmath27 , in turn , represents a _
duffing oscillator_. due to their simplicity and universal nature these two systems are among those most extensively studied in nonlinear dynamics .
it is straightforward to see that the oscillator is non - decaying , i.e. the origin is repeller , whenever @xmath28 ( negative damping ) in the case of a van der pol oscillator and/or @xmath29 and @xmath30 in a duffing oscillator.when a nonlinearity is present in both paramters ( i.e. @xmath31 and @xmath32 are both non - zero ) a combined _ van der pol - duffing oscillator _ results .
the van der pol
duffing oscillator , however , is more than just a good heuristic model of the solar cycle .
in fact , an oscillator equation of this general form can be derived by a truncation of the dynamo equations . as noted before
, we are especially interested in capturing the temporal dynamics associated with the large scale magnetic field . in order to construct a low order model aimed at capturing this dynamics
, we follow the procedures described in @xcite .
it has been suggested by @xcite and @xcite that a dimensional truncation of the dynamo equations ( [ eq : ap ] ) and ( [ eq : bphi ] ) is an effective method to reduce the system s dimensions and capture phenomena just on that scale . following that ansatz , gradient and laplacian operators are approximated by a typical length scale of the system @xmath33 ( e.g. convection zone length or width of the tachocline ) , leading to @xmath34 and @xmath35 .
analogously this can be interpreted as a collapse of all spatial dimensions , leaving only the temporal behaviour . in terms of dynamical systems
, we are projecting a higher dimensional space into a single temporal plane .
after grouping terms in @xmath36 and @xmath13 ( now functions only dependent of the time ) we get @xmath37 where we have defined the _ structural coefficients _ , @xmath38 , as @xmath39 we now concentrate in creating an expression for the time evolution of @xmath36 since it is the field component directly associated with the productions of sunspots .
we derive expression ( [ eq : dbdt ] ) in order to the time , and substitute ( [ eq : dadt ] ) in it to take away the @xmath13 dependence yielding @xmath40 where @xmath41 , @xmath42 , @xmath43 and @xmath44 are model parameters that depend directly on the structural coefficients .
the name used to describe @xmath38 comes from the fact that these coefficients contain all the background physical structure ( rotation , meridional circulation , diffusivity , etc . )
in which the magnetic field evolves .
this oscillator ( equation [ eq : vdp ] ) is a van der pol - duffing oscillator and it appears associated with many types of physical phenomena that imply auto - regulated systems .
this equation is a quite general result which @xmath3 should satisfy . in this case , unlike in the classical van der pol - duffing oscillator , the parameters are interconnected by a set of relations that link the present oscillation model with the original set of dynamo equations ( [ eq : ap]-[eq : bphi ] ) .
this interdependency between parameters will eventually constrain the solution s space . as in the classical case , @xmath45 controls the frequency of the oscillations or the period of the solar magnetic cycle , @xmath20 controls the asymmetry ( or non - linearity ) between the rising and falling parts of the cycle and @xmath32 affects directly the amplitude .
the @xmath31 parameter , related to buoyancy loss mechanism sets the overall amplitude peak amplitude of the solution .
figure ( [ fig:1 ] ) shows the solution of equation ( [ eq : vdp ] ) in a time vs. amplitude diagram ( left ) and in a @xmath46 , @xmath47 phase space . from this figure
we find that this dynamo solution , under suitable parametrization ( viz .
next section ) is a self - regulated system that rapidly relaxes to a stable 22-year oscillation . in the phase space
the solution tend to a limit cycle or attractor .
a complete ( clockwise ) turn in the phase space corresponds to a complete solar magnetic cycle .
obtained from equation ( [ eq : vdp ] ) with parameters @xmath48 , @xmath49 , @xmath50 , @xmath51 . in the right
we have a ( @xmath36 , @xmath52 ) phase space representation of the solution .
the blue arrows indicate the direction of increasing time and the red dot the initial value used .
also indicated in this panel are the regions corresponding to the maxima and minima of the cycle . _
adapted from _
@xcite . ]
in order to estimate values for the coefficients in equation ( [ eq : vdp ] ) , @xcite and @xcite fitted this oscillator model either to a long period of the solar activity ( several solar cycles ) or to each magnetic cycle individually .
we shall return to this point in subsequent sections .
a more general approach to the problem of finding the parameter combinations with which the classical van der pol - duffing oscillator returns solar - like solutions was taken by @xcite .
the authors mapped the parameter space of the oscillator by adding stochastic noise to its parameters using different methods .
the objective was to constrain the parameter regime where this nonlinear model shows the observed attributes of the sunspot cycle , the most important requirement being the presence of the waldmeier effect @xcite according to the definition of @xcite .
noise was introduced either as an ornstein uhlenbeck process @xcite or as a piecewise constant function keeping a constant value for the interval of the correlation time .
the effect of this noise was assumed to be either additive or multiplicative .
the amplitudes and correlation times of the noise defined the phase space .
the attributes of the oscillator model were first examined in the case of the van der pol oscillator ( no duffing cubic term ) with perturbation either in the nonlinearity parameter , @xmath32 or the damping parameter , @xmath20 , as shown in the equations below : @xmath53\dot{x}\\ \label{e : vdp - xi}\ddot x&=&-\omega_0 ^ 2x-\mu_0\left [ \xi(t ) x^2 - 1\right]\dot{x}. \end{aligned}\ ] ] the constant parameters , @xmath54 , @xmath55 and @xmath56 used were taken from the fitted values listed by @xcite .
note that in this simple case , variations in these parameters were assumed to be independent from each other , whereas in reality they are interrelated ( see equation [ eq : vdp ] ) .
the results show that the model presents solar - like solutions when a multiplicative noise is applied to the nonlinearity parameter , as in equation ( [ e : vdp - xi ] ) .
an example of a time series produced by this type of oscillator is shown in figure ( [ fig : nmpk2013 ] ) . ) .
ssn values were defined as @xmath57 .
the noise applied ( piecewise constant in this case ) is shown in the top panel . ] and
ssn values were here defined as @xmath58 , following @xcite .
the noise applied is shown in the top and middle panels .
the duffing parameter was here given a constant value @xmath59 .
] as a next step towards a fully general study , let us consider the case where both @xmath32 and @xmath20 are simultaneously perturbed and the duffing term @xmath60 is also kept in the oscillator equation ( [ eq : nonlinosc ] ) .
noise is applied to @xmath20 but it also affects @xmath32 as the values of @xmath32 and @xmath20 are assumed to be related as @xmath61 @xcite ; here , @xmath62 and @xmath63 are constants .
a mapping of the parameter space shows that in this case solar - like solutions are more readily reproduced compared to the case when only one parameter was assumed to vary ( see fig .
[ fig : nmpk201x ] ) .
this finding is in line with the information derived from the lodm developed in the previous section .
an ongoing study shows that the corresponding time dependence in the duffing parameter , as predicted by the lodm , has a significant effect on the character of the solution .
we note that additive noise was first applied to one parameter ( @xmath32 ) of a van der pol oscillator by @xcite but the focus of that work was on reproducing cycle to cycle fluctuations , withour considering the waldmeier effect .
this model was further analysed by @xcite ( and references therein ) who studied the behavior of the hurst exponent of this system and concluded that this type of fluctuations implies that the stochastic process which underlies the solar cycle is not simply brownian .
this means that long - range time correlations could probably exist , opening the way to the possibility of forecasts on time scales comparable to the cycle period .
an attempt to introduce the effects of such non - gaussian noise statistics into the lodm was made by @xcite who suggest that this may contribute to cyclic variations of solar activity on time scales shorter than 11 years .
in the previous example a perturbation method was studied in order to find solar like solutions for this non - linear oscillator .
another way of thinking is to pair the oscillator directly to some solar observable and try to constraint its parameters . as mentioned in the introduction we choose the international sunspot number , @xmath64 and we use it to build a proxy of the toroidal magnetic component .
since the @xmath64 is usually taken to be proportional to the toroidal field magnetic energy that erupts at the solar surface ( @xmath65 ) , @xcite , this makes it ideal for compare with solutions of the lodm . taking this in consideration ,
@xcite have built a toroidal field proxy based on the sunspot number by following the procedure proposed by , i.e. @xmath66 .
details about the construction of the toroidal proxy ( see fig . [ fig:2 ] ) can be found in @xcite and @xcite .
is obtained by calculating @xmath67 , changing the sign of alternate cycles ( represented in gray ) , and smoothing it down using an fft low pass filter of 6 months .
the vertical thin dotted lines represent solar cycle minima . ]
the solution obtained for equation ( [ eq : vdp ] ) presented in figure ( [ fig:1 ] ) shows that the solar cycle is a self - regulated system that tends to a stable solution defined by an attractor ( limit cycle ) .
if we allow for the different physical processes responsible for the solar dynamo and embedded in the structural coefficients ( [ eq : c1 ] , [ eq : c2 ] and [ eq : c3 ] ) , i.e. the differential rotation , the meridional circulation flow , the @xmath1 mechanism , and the magnetic diffusion , to change slowly from cycle to cycle then we start to observe deviations from the equilibrium state . if deviations from this sort of dynamical balance occur , such that if one of these processes changes due to an external cause , the other mechanisms also change to compensate this variation and ensure that the solar cycle finds a new equilibrium .
to test this idea of an equilibrium limit cycle , we fit the lodm parameters to the \{@xmath36,@xmath52 } phase space of the built toroidal proxy ( see fig .
[ fig:4 ] ) . :
the crosses correspond to local area averaged values found by dividing the data into 32 temporal intervals .
the red dashed curve is a fit to the crosses .
the continuous black curves correspond to a fit to all data points ( not grouped in intervals ) . for the red curve
we have that @xmath68 , @xmath69 , @xmath70 and @xmath71 .
figure _ adapted from _
@xcite . ]
if one of the parameter s variation is very large the system can be dramatically affected , leading to a quite distinct evolution path like the ones found during the solar grand minima .
we will develop this subject in a subsequent section .
solutions with fluctuation similar to those we see on the solar cycle are easily set by variations in the @xmath20 parameter ( and the physical processes associated with it ) . by definition the structural coefficient that regulates this parameter ( @xmath72 ) also has an important role in the other parameters ( @xmath45 , @xmath32 and @xmath31 ) . in the lodm equation ( [ eq : vdp ] ) the @xmath20 and @xmath73 quantities
regulate the strength and the non - linearity of the damping .
moreover , an occasional variation on @xmath20 , like a perturbation on the meridional flow amplitude , @xmath74 ( see structural coefficients [ eq : c1 ] ) will affect all sets of parameters leading to the solar dynamo ( equation [ eq : vdp ] ) to find a new equilibrium , which will translate into the solar magnetic cycle observable like the sunspots number , showing an irregular behaviour .
the well - known relation discover by max waldmeier @xcite , that the time that the sunspot number takes to rise from minimum to maximum is inversely proportional to the cycle amplitude in naturally captured by the lodm assuming discrete variations in @xmath20 .
notice that the waldmeier effect occurs as a consequence of the limit cycle becoming increasingly sharp as @xmath20 increases , i.e. , the sunspot number amplitude increases as the cycle s rising times gets shorter .
from the physical point of view , based on observations , we know that in the sun some of the physical background structures that are taken as constant in our standard dynamo solution are nt so . in order to test that specific changes of the background state lead to the observed changes in the amplitude of the solar cycle , the following strategy was devised . at a first approximation
we assume that the structural coefficients can change only discretely in time , more specifically from cycle to cycle while the magnetic field is allowed to evolve continuously .
the idea is that changing coefficients will generate theoretical solutions with different amplitudes , periods and eigen - shapes at different times and by comparing these different solution pieces with the observed variations in the solar magnetic field , we are able to infer information about the physical mechanisms associated with the coefficients . to do this we compare our theoretical solution with a proxy built from the international monthly averaged ssn since 1750 to the present .
as mentioned before we assume that @xmath75 .
the proxy data is separated into individual cycles and fitted using equation ( [ eq : vdp ] ) , considering that the buoyancy properties of the system are immutable , i.e. @xmath76 is constant throughout the time series .
this means that when we fit the lodm to solar cycle @xmath77 , we will retrieve the set of @xmath78 coefficients that best describe that cycle .
this allows to probe how these coefficients vary from cycle to cycle and consequently how the physical mechanisms associated with them evolve in time @xcite .
equation ( [ eq : vdp ] ) is afterwards solved by changing the parameters to their fit value , at every solar minimum using a stepwise function ( similar to that presented in figure ( [ fig : vp_c1_var ] ) for @xmath72 ) .
figures ( [ fig : vp_c1_var ] ) and ( [ fig : fit ] ) highlight this procedure . the fact that such a simplified dynamo model can get this degree of resemblance with the observed data just by controlling one or two parameters is an indication that it captures the most important physical processes occurring in the sun .
the simple procedure previously described allows to reconstruct the behavior of solar parameters back in time . using an improved fitting methodology
, @xcite obtained with this model the reconstruction of the variation levels of the solar meridional circulation for every solar ( sunspot ) cycle over the last 250 years .
one must notice that in this specific lodm the amplitude of the cycle depends directly on the amplitude of the meridional flow during the previous cycle .
it is completely possible to imagine that other models that consider a different theoretical setup might return a different behaviour .
looking at equation ( [ eq : c1 ] ) , we can see that the coefficient @xmath72 depends on two physical parameters , the magnetic diffusivity , @xmath6 , and the amplitude of the meridional circulation , @xmath74 .
the magnetic diffusivity of the system is a property tightly connected with turbulent convection and is generally believed to change only in time scales of the order of stellar evolution .
this leaves variations in @xmath74 as the only plausible explanation for the variation observed from cycle to cycle .
therefore , by looking at the evolution of @xmath72 we can can effectively assume that we are looking at the variation in the strength of meridional circulation .
the results obtained are presented in figure ( [ fig : vp_c1_var ] ) .
( black line ) compared to smoothed the sunspot number ( gray ) . _ adapted from _
@xcite ] although this result is in itself interesting , a more important concept came from this study .
when @xcite presented their results for the first time , they introduced the idea that coherent long term variations ( of the order of the cycle period ) in the strength of the meridional circulation could provide an explanation for the variability observed in the solar cycle ( see fig .
[ fig : lodm_surya_comp ] ) .
this result was also _ a posteriori _ numerically validated using a 2.5d flux transport model ( @xcite and @xcite ) .
only a couple of years later , @xcite presented meridional circulation measurements spanning over the last solar cycle .
their measurements confirmed that the amplitude of this plasma flow changes considerably from cycle to cycle .
recently two other groups have tested this idea with their 2.5d dynamo models finding additional features based in this effect , c.f .
@xcite , @xcite and @xcite .
for example it was found that the instant at which the change in the meridional flow takes place , has an influence in the duration of the following solar cycle .
this was used as an explanation for the abnormally long duration of the last minimum . just for reference ,
the numbering of solar cycles only started after 1750 with solar cycle 1 beginning in 1755 . at this moment
we are in the rising phase of solar cycle 24 . ) .
_ adapted from _
@xcite and @xcite . ]
solar grand minima correspond to extended periods ( a few decades ) where very low or no solar activity occurs . during theses
periods no sunspots ( or very few ) are observed in the solar photosphere and it is believed that other solar phenomena also exhibit low levels of activity .
the most famous grand minimum that has been registered is the maunder minimum which occurred between the years of 1645 and 1715 ( @xcite ) .
a possible explanation for the origin of these quiescent episodes was put forward by @xcite . using a lodm
, they showed that a steep decrease in the meridional flow amplitude can lead to grand minima episodes like the maunder minimum ( see figs .
[ fig : grandmin ] and [ fig : bgrandmin ] ) .
this effect presents the same visual characteristics as the observed data , namely a rapid decrease of magnetic intensity and a gradual recovery into normal activity ( see fig .
[ fig : grandmin ] ) after the meridional circulation amplitude returns to its normal values .
a similar result was later obtained by @xcite , again using a more complex 2.5d numerical flux transport model .
nevertheless the reasons that could lead to a decrease of the meridional flow amplitude were not explored .
this served as a motivation to study the behavior of this lodm in the non - kinematic regime , explained in section 5.2 .
some examples mentioned in the introduction , hint that fluctuations in the @xmath1 mechanism can also trigger grand minima .
we focus on a specific example now , the lodm developed by @xcite . in this work
the authors used a time - delay lodm similar to that presented in @xcite but expanded with the addition of a second @xmath1 effect .
this model incorporates two of these mechanisms , one that mimics the surface babcock - leighton mechanism ( bl ) , and another one analogous to the classical mean - field ( mf ) @xmath1-effect that operates in the bulk of the convection zone .
this set up captures the idea that the bl mechanism should only act on strong magnetic fields that reach the surface , and that weak magnetic fields that diffuse through the convection zone should feel the influence of the mf @xmath1 .
the authors subject these two effects to different levels of fluctuations and find that in certain parameter regimes , the solution of the system shows the same characteristics as a grand minimum .
these results were also validated by implementing a similar set up into a 2.5d mean - field flux transport dynamo model @xcite .
again this shows the usefulness of low order models to probe ideas before their implementation into more complex models . for the near future , perhaps one of most interesting applications of this
lodm is its use in the predictability of future solar cycles amplitudes .
the first step towards this objective is presented in @xcite .
the authors studied the correlations between the lodm fitted structural coefficients and cycle s characteristics ( amplitude , period and rising time ) .
they found very useful relationships between these quantities measured for cycle n and the amplitude of cycle n+1 .
these relationships were put to the test by predicting the amplitude of current solar cycle 24 ( see fig .
[ fig : predict24 ] ) . .
[ fig : predict24 ]
in recent years there have been strong developments of different types of dynamo models to compute the evolution of the solar magnetic activity and to explore some of the causes of magnetic variability .
two classes of models have been quite successful , the kinematic dynamo models and , more recently , the global magnetohydrodynamical models .
both types of dynamo models have a quite distinct approach to the dynamo theory , the first one resolves the induction magnetic equation for a prescribed velocity field ( which is consistent with helioseismology ) , and the second one obtains global magnetohydrodynamical simulations of the solar convection zone .
many of these models are able to reproduce some of the many observational features of the solar magnetic cycle .
nevertheless , it remains quite a difficult task to successfully identify which are the leading physical processes in current dynamo models that actually drive the dynamo in the solar interior .
the usual method to test these dynamo models is to compare their theoretical predictions with the different sets of data , including the sunspot numbers , however , in many cases the conclusions obtained are very limited , as different physical mechanisms lead to very identical predictions .
this problem also arises in the comparison between different dynamo models , including different types of numerical simulations .
a possible solution to this problem is to use inverted quantities ( obtained form observational data ) to test the quality of the different solar dynamo model , rather than making direct comparison of data . for those of you familiarized with helioseismology
, there is a good analogue : it is the equivalent to compare the inverted sound speed profile ( obtained from observational data ) with the sound speed profile predicted by solar models ( _ backward approach _ ) , rather than compare predicted frequencies with observational frequencies ( _ forward approach _ ) .
the former method to test physical models is more insightful than the latter one . at the present level of our understanding of the solar dynamo theory , as a community
we could gain a more profound understanding of the mechanisms behind the solar magnetic variability , if we start developing some backward methods to analyse solar observational data and test dynamo models .
is constant ) ; ( b ) a variable kinematic dynamo model in which the @xmath74 for each magnetic cycle corresponds to the value obtained from the sunspot number temporal series .
both simulations correspond to 130-year time series .
the small variability present in the left panel is due to the stabilization of the numerical solution .
figure _ adapted from _ @xcite . , title="fig : " ] is constant )
; ( b ) a variable kinematic dynamo model in which the @xmath74 for each magnetic cycle corresponds to the value obtained from the sunspot number temporal series .
both simulations correspond to 130-year time series .
the small variability present in the left panel is due to the stabilization of the numerical solution .
figure _ adapted from _ @xcite .
, title="fig : " ] using the meridional velocity inverted from the sunspot number time series ( see fig . [
fig : vp_c1_var ] ) , @xcite showed that most of the long term variability of the sunspot number could be explained as being driven by the meridional velocity decadal variations , assuming that the evolution of the solar magnetic field is well described by an axisymmetric kinematic dynamo model .
figure ( [ fig : lodm_surya_comp ] ) shows a reconstructed sunspot times series that has been obtained using the meridional velocity @xmath74 inverted from the sunspot observational time series , and figure ( [ fig : compsuryssn ] ) shows the phase space of a standard axisymmetric kinematic dynamo model ( with the same @xmath74 for all cycles ) ( ) and a solar dynamo model where the @xmath74 changes from cycle to cycle as inverted from the sunspot times series ( @xcite ) .
it is quite encouraging to find that such class of dynamo models for which the @xmath74 changes overtime successfully reproduced the main features found in the observational data .
moreover , in their article @xcite tested two different methods of implementing the velocity variation for each magnetic cycle , namely , by considering that amplitude variations in @xmath74 that take place at sunspot minima or at sunspot maxima .
all the time series show a few characteristics that are consistent with the observed sunspot records .
in particular , all the simulations show the existence of low amplitudes on the sunspot number time series between 1800 and 1840 and between 1870 and 1900 .
the simulation that best reproduces the solar data corresponds to the model ssnrec[3 ] ( see fig . [
fig : lodm_surya_comp ] ) , in which was implemented a smoothed @xmath74 variation profile between consecutive cycles and taking place at the solar maximum .
this clearly highlights the potential of such methodology . here
, we discuss the same methodology as the one used in the previous section , but instead of applying it to observational sunspot number records , it is used to reconstruct the sunspot time series .
the results obtained clearly show that the present kinematic dynamo models can reproduce in some detail the observed variability of the solar magnetic cycle .
the fact that for one of the sunspot models model ssnrec[3 ] , it presents a strong level of correlation with the observational time series , lead us to believe that the main idea behind this _ backward approach _ is correct and it is very likely that the inverted @xmath74 variation is probably very close to the @xmath74 variation that happens in the real sun .
clearly , under the assumed theoretical framework the meridional circulation is the leading quantity responsible for the magnetic variability found in the sunspot number time series and current solar dynamo models are able to reproduce such variability to a certain degree .
so far , the vast majority of the lodm applications presented here followed the traditional assumption that the solar dynamo can be correctly modeled in the kinematic regime , where only the plasma flows influence the production of magnetic field , and not the other way around .
this kinematic approximation is used in the vast majority of the present 2.5d spatially resolved dynamo models . in the last couple of years
though , evidence started to appear supporting the claim that this kinematic regime might be overlooking important physical mechanisms for the evolution of the dynamo .
the idea that the meridional flow strength can change over time and affect the solar cycle amplitude coupled with the measurements of @xcite and @xcite indicate that the observed variation in this flow is highly correlated with the levels of magnetic activity .
this leads to the fundamental question : _ `` is the flow driving the field or is the field driving the flow ? ''
_ the first clues are starting to appear from 3d mhd simulations of solar convection .
the recent analysis of the output of one of the large - eddy global mhd simulations of the solar convection zone done by @xcite shows interesting clues .
these simulations solve the full set of mhd equations in the anelastic regime , in a broad , thermally - forced stratified plasma spherical shell mimicking the scz and are fully dynamical on all spatiotemporally - resolved scales .
this means that a two way interaction between field and flow is always present during the simulation .
the analysis shows that the interaction between the toroidal magnetic field and the meridional flow in the base of the convection zone indicates that the magnetic field is indeed acting on the equatorward deep section of this flow , accelerating it .
this observed relationship runs contrary to the usually assumed kinematic approximation . in order to check
if this non - kinematic regime has any impact in the long term dynamics of the solar dynamo , @xcite implemented a term that accounts for the lorentz force feedback in a lodm similar to the one presented here .
this allows to fully isolate the global aspects of the dynamical interactions between the meridional flow and magnetic field in a simplified way .
they assumed that the large - scale meridional circulation , @xmath74 , is divided into a `` kinematic '' constant part , @xmath79 ( due to angular momentum distribution ) and a time dependent part , @xmath80 , that encompasses the lorentz feedback of the magnetic field .
therefore they redefine @xmath74 as @xmath81 where the time dependent part evolves according to @xmath82 the first term is a magnetic nonlinearity representing the lorentz force and the second is a `` newtonian drag '' that mimics the natural resistance of the flow to an outside kinematic perturbation . under these conditions the lorentz force associated with the cyclic large - scale magnetic field
acts as a perturbation on the otherwise dominant kinematic meridional flow .
this idea was not new and it was used before in the context of magnetically - mediated variations of differential rotation in mean - field dynamo models @xcite , @xcite and @xcite .
the modified lodm equation they end up defining are @xmath83 where @xmath72 , is defined as @xmath84 and takes the role of magnetic diffusivity , while the other coefficients remain the same .
while the values used for the structural coefficients , are mean values extracted from the works presented in the previous in sections , the parameters associated with the meridional flow evolution , @xmath85 , @xmath86 and @xmath79 deserved now the attention .
these parameters have an important role in the evolution of the solution space .
the behavior observed in the solutions range from fixed - amplitude oscillations closely resembling kinematic solutions , multiperiodic solutions , and even chaotic solutions .
this is easier to visualize in figure ( [ fig - bifurcation_maps ] ) where are presented analogs of classical bifurcation diagrams by plotting successive peak values of cycle amplitudes , for solutions with fixed ( @xmath85 , @xmath79 ) combinations but spanning through values of @xmath86 .
transitions to chaos through bifurcations are also observed when holding @xmath86 fixed and varying @xmath85 instead . between @xmath87 and @xmath88 for different @xmath85 and @xmath79 .
( a ) single period regime , @xmath89 , @xmath90 ; ( b ) appearance of period doubling , @xmath89 , @xmath91 and ( c ) shows signatures of chaotic regimes with multiple attractors and windows , obtained with @xmath92 , @xmath93 . _
adapted from _
@xcite.,title="fig:",width=143 ] between @xmath87 and @xmath88 for different @xmath85 and @xmath79 .
( a ) single period regime , @xmath89 , @xmath90 ; ( b ) appearance of period doubling , @xmath89 , @xmath91 and ( c ) shows signatures of chaotic regimes with multiple attractors and windows , obtained with @xmath92 , @xmath93 .
_ adapted from _
@xcite.,title="fig:",width=143 ] between @xmath87 and @xmath88 for different @xmath85 and @xmath79 .
( a ) single period regime , @xmath89 , @xmath90 ; ( b ) appearance of period doubling , @xmath89 , @xmath91 and ( c ) shows signatures of chaotic regimes with multiple attractors and windows , obtained with @xmath92 , @xmath93 . _
adapted from _
@xcite.,title="fig:",width=143 ] the authors expanded the methodology used and applied stochastic fluctuations to parameter @xmath85 , the one that controls the influence of the lorentz force . as a result , and depending on the range of fluctuations , they observed that the short term stochastic kicks in the lorentz force amplitude create long term modulations in the amplitude of the cycles ( hundreds of years ) and even episodes where the field decays to near zero values , analog to the previously mentioned grand minima .
the duration and frequency of these long quiescent phases , where the magnetic field decays to very low values , is determined by the level of fluctuations of @xmath85 and the value of @xmath86 .
the stronger this drag term @xmath86 is , the shorter the minima are and the higher the level of fluctuation of @xmath85 , the more common these intermittency episodes become .
figure ( [ fig - lodm_stochastic_a ] ) shows a section of a solution that spanned for 40000 years and that presents all the behaviors described before .
$ ] , @xmath94 and @xmath95 .
all other model parameters are the same as in the reference solution .
panel ( a ) shows a section of the simulation where the long term modulation can be seen . in black
is @xmath96 , red @xmath97 and blue a scaled version of the meridional flow , in this case 5@xmath98 . in panel
( b ) the same quantities but this time zooming in into a grand minimum ( off phase ) period . _
adapted from _
@xcite.,width=377 ] in this specific example they used 100% fluctuation in @xmath85 and maintaining all the other parameters constant . in the the parameter space used to produce this figure , the solution without stochastic forcing is well behaved in the sense that it presents a single period regime .
therefore , the fluctuations observed in this solution are a direct consequence of the stochastic forcing of the lorentz force and not from a chaotic regime of the solution s space . to understand how the grand minima episodes arise they resort to visualizing one of these episodes with phase space diagrams of \{@xmath36 , @xmath13 ,
@xmath74}. this allows to see how these quantities vary in relation to each other and try to understand the chain of events that trigger a grand minimum .
the standard solution for the lodm without stochastic forcing , i.e. with @xmath85 fixed at the mean value of the random number distribution used , is the limit cycle attractor , i.e. , a closed trajectory in the \{@xmath36 , @xmath13 } phase space .
this curve is represented as a black dashed trajectory in the panels of figure ( [ fig - lodm_phasespace ] ) .
the gray points in this figure are the stochastic forced solution values sampled at 1 year interval .
these points scatter around the attractor representing the variations in amplitude of the solution .
occasionally the trajectories defined by these points collapse to the center of the phase space ( the point \{0,0,@xmath79 } is also another natural attractor of the system ) indicating a decrease in amplitude of the cycle , i.e. a grand minimum .
the colored trajectory evolving in time from purple to red represents one of those grand minimum .
this happens when the solution is at a critical distance from the limit cycle attractor and gets a random kick further away from it .
this kick makes the field grow rapidly . in turn , since the amplitude of the field grows fast , the lorentz force will induce a similar growth in @xmath80 eventually making @xmath74 change sign .
when this occurs , @xmath74 behaves as a sink term quenching the field growth very efficiently .
this behavior is seen in the two bottom panels of figure ( [ fig - lodm_phasespace ] ) where @xmath74 decays to its imposed
value @xmath79 after the fields decay . after this collapse of @xmath74
to @xmath79 it starts behaving has a source term again and the cyclic activity proceeds . .
_ adapted from _
@xcite.,width=377 ] one clear advantage of low order models emerges from this example .
currently 3d mhd simulations of solar convection spanning a thousand years take a couple of months to run in high efficiency computational clusters or in supercomputers .
longer simulations are at the moment prohibitive not only for the amount of time they take but also for the huge amount of data they generate .
statistical studies on grand minima originated by the kind of magnetic back - reaction described here , require long integration times where many thousands of cycles need to be simulated .
the lodm calculations can be done in a few minutes or hours in any current desktop . the grand minima mechanism presented in this section
is now being studied by looking at the data available from 3d simulations .
some effects are easier to find when you know what to look for .
so far we have shown that low order dynamo models ( for which the approximation must be carefully chosen to keep the relevant physics within ) could lead the way to explore some features of the solar magnetic activity including the long - term variability .
the study of the phase diagram @xmath99 clearly shows that on a scale of a few centuries the solar magnetic cycle shows evidence for a van der pool attractor put in evidence by the mean solar magnetic cycle , although on a time - scale of a few solar magnetic cycles the phase space trajectory changes dramatically . in some cases the trajectory collapses completely for several magnetic cycles as in the periods of grand minima
this gives us an indication about the existence of a well defined self regulated system under all this observed magnetic variability , for which we still need to identify the leading physical mechanisms driving the solar dynamo to extreme activity scenarios like periods of grand minima .
actually , the fact that a well - defined averaged van der pool limit curve exists for all the sunspot records , can be used to test different solar dynamo models , including numerical simulations , against observational data or between different dynamo models .
moreover , the fact that such well - defined attractor exists in the phase space , and several dynamo models are able to qualitatively reproduce the solar variability ( as observed in the phase space @xmath99 gives us hope that in the near future we will be able to make quite reliable short term predictions of the solar magnetic cycle variability , at least within certain time intervals of solar magnetic activity .
a significant contribution can be done by the utilization of more accurate sunspot time series in which many of the historical inaccuracies were corrected @xcite . in the future ,
similar inversion techniques could be developed , namely to study the possible asymmetry between the north and south hemispheres using the sunspot areas , either by treating each of the sunspot areas as two distinct times series or by attempting two - dimension inversions of sunspot butterfly diagrams . in the former case ,
recently @xcite have analysed these long - term sunspot areas time series and found that turbulent convection and solar granulation are responsible by the stochastic nature of the sunspot area variations . in the last case , we could learn about the evolution of the solar magnetic cycle in the tachocline during the last two and a half centuries . moreover ,
most of the inversion methods used for the sunspot number can be easily extended to other solar magnetic cycles proxies such as tsi , h@xmath100 and magnetograms .
the oscillator models , as a first order dynamo model are particularly suitable to study the magnetic activity in other stars . a good proxy of magnetic activity in stars in the chromospheric variations of ca ii h and k emission lines .
@xcite have found many f2 and m2 stars which seem to have cyclic magnetic cycle activity , as observed in the sun ( see fig . [
fig : msstar ] ) . in some of these stars
the observational time series covers several cycles of activity .
in particular , it will be interesting to identify how the dynamo operating in these stars differs from the solar case .
more recently , the corot and kepler space missions have observed photometric variability associated with solar - like activity in a very large number of main sequence and sub - giant stars . while the time coverage is too short to derive cycle periods for stars very close to the sun , the overall _ level _ of activity and its dependence on various stellar parameters can be studied on a large statistical sample @xcite .
nevertheless , with so many stars with quite distinct masses and radius , it is reasonable to expect that we will find quite different type of dynamos and regimes of stellar magnetic cycle .
actually , we think it is likely to find a magnetic diversity identical to the one found in the acoustic oscillation spectra measured for the more than 500 sun - like stars already discovered @xcite , some of which have already shown evidence of a magnetic cycle activity .
@xcite have obtained a proxy of the starspots number for the star hd49933 from amplitudes and frequencies of the acoustic modes of vibration . as in the nonlinear oscillator models
the activity level is determined by the structural parameters which in turn depends on the dynamo model .
these studies potentially offer a simple theoretical scheme against which to test the observational findings .
the authors thank the anonymous referee for the suggestions made to improve the quality of the article .
thanks the convenors of the workshop and the international space science institute for the invitation and financial support . i.l . and d.p .
would also like to thank arnab choudhuri and his collaborators for making the surya code publicly available .
i.l . would like to thank his collaborators in this subject of research : ana brito , elisa cardoso , hugo silva , amaro rica da silva and sylvaine turck - chize .
the work of i.l . was supported by grants from `` fundao para a cincia e tecnologia '' and `` fundao calouste gulbenkian '' .
d.p . acknowledges the support from the fundao para a cincia e tecnologia ( fct ) grant sfrh / bpd/68409/2010 .
m. n. acknowledges support from the hungarian science research fund ( otka grant no . | this article reviews some of the leading results obtained in solar dynamo physics by using temporal oscillator models as a tool to interpret observational data and dynamo model predictions .
we discuss how solar observational data such as the sunspot number is used to infer the leading quantities responsible for the solar variability during the last few centuries .
moreover , we discuss the advantages and difficulties of using inversion methods ( or backward methods ) over forward methods to interpret the solar dynamo data .
we argue that this approach could help us to have a better insight about the leading physical processes responsible for solar dynamo , in a similar manner as helioseismology has helped to achieve a better insight on the thermodynamic structure and flow dynamics in the sun s interior . |
the double or extended main - sequence turnoffs ( emstos ) were discovered in the color - magnitude diagram ( cmd ) of intermediate - age massive star clusters in the magellanic clouds @xcite .
one interpretation of the emstos is that the clusters have experienced an extended star - formation histories ( esfh ) with a duration of @xmath2 myr @xcite , which disagrees with classical understanding of star clusters being simple stellar populations ( ssps ) .
the emstos were also discovered in young clusters ngc 1856 @xcite , ngc 1755 @xcite , ngc 1850 @xcite , and ngc 1866 @xcite .
moreover , ngc 1856 , ngc 1755 , and ngc 1866 are found to exhibit dual main sequences ( ms ) below their msto @xcite .
an alternative interpretation of the emstos is the effects of star rotation @xcite .
@xcite show that the extension of msto caused by star rotations is related to the rotation rate of stars , the efficiency of rotational mixing , and the age of star clusters .
a relatively high rotation rate and a high efficient rotational mixing are required to explain the emstos of young clusters [ see figure 8 in ( * ? ? ?
* but see niederhofer et al .
2015a and dantona et al . 2015 ) ] .
@xcite claimed that the emsto of ngc 1856 can be explained by a rotation of @xmath3 times the keplerian rotation rate ( @xmath4 ) .
but in order to explain the dual mss of clusters ngc 1856 and ngc 1755 , the rotation rate of @xmath5 @xmath4 is required @xcite . * a large number of rapid rotating stars have been found in ngc 1850 and ngc 1856 by @xcite . *
* however , neither stellar models with different ages only , nor rapid rotating models with different rotation rates , properly reproduce the observed split ms and emsto of ngc 1866 @xcite .
the populations with both different ages and different rotation rates are needed to explain ngc 1866 @xcite .
* moreover , @xcite stated that their rotating models fail to reproduce the stars after the end of the central h - burning phase " of ngc 1856 .
* however , these stars might be stars with decretion disks @xcite seen nearly edge on , so they suffer from extinction which pushes them into this region .
* another coeval interpretation of the emstos is interacting binaries @xcite .
@xcite showed that interacting binaries including merged binary systems and the binaries with mass transfer can lead to both the emstos and the dual red - clumps .
the effects of the interacting binaries on the cmds of some clusters should not be neglected , although the number of the interacting binaries in a cluster could be not enough to explain the emstos alone .
one of the important predictions of the esfh scenario is that the ongoing star - formation should be observed in young star clusters with an age of a few hundred myr .
however , up to now , the expected ongoing star - formation is not observed in young clusters with age beyond 10 myr @xcite . @xcite and @xcite argued that the emstos can occur only in clusters with masses larger than a threshold of about @xmath6 and with escape velocity greater than @xmath7 km s@xmath8 .
however , the emstos of ngc 1755 @xcite and ngc 411 @xcite would represent a challenge for this scenario . * furthermore , the observation that there exists a strong correlation between cluster age and the inferred age spread as found by @xcite also rules out an actual age spread being the origin of the emsto . *
@xcite analyzed the sub - giant branch ( sgb ) of ngc 1651 harbouring an emsto and found that the sgb is narrower and offsets from what would be inferred from the emsto region if the large age spreads would be present within the cluster .
similar results were found in ngc 1806 and ngc 1846 @xcite and ngc 411 @xcite .
hence , they concluded that age spreads are not likely to be the cause of the emsto phenomenon .
however , @xcite found that the cross - sgb profiles of ngc 1651 , ngc 1806 , and ngc 1846 are consistent with their cross - msto profiles when the latter are interpreted as age distributions .
conversely , their sgb morphologies are inconsistent with those of simulated ssps .
the origin of the emstos is still an open question .
the overshooting of the convective core ( ovcc ) can bring more hydrogen - rich material into h - burning core , which significantly prolongs the lifetime of the burning of core hydrogen and enhances the he - core mass left behind .
the distance of the overshooting of a convection is defined as @xmath9 , where is a free parameter and @xmath9 is the local pressure scale - height .
recently , @xcite developed a method to determine the size of the convective core including the overshooting region from observed oscillation frequencies of low - degree @xmath10-modes .
it was found that the value of is variable for stars with an approximatelly equal mass .
for example , the value of is @xmath11 for kic 9812850 with @xmath12 @xcite , @xmath13 for kic 2837475 with @xmath14 @xcite , @xmath15 for procyon with @xmath14 @xcite , @xmath16 for hd 49933 with @xmath17 @xcite , and @xmath18 for kic 11081729 with @xmath19 @xcite . * the typical errors of the value of are @xmath20 . *
if a variable overshooting exists in stars with masses larger than @xmath21 , the msto of young- and intermediate - age star clusters would be affected by the overshooting . in this work , we mainly focus on whether the emstos of young- and intermediate - age star clusters can be explained by the effects of the ovcc .
the paper is organized as follows : we show our calculation results in section 2 , and the results are compared with observations in section 3 , then we discuss and summarize the results in section 4 .
in order to study the effects of overshooting of the convective core on the msto of star clusters , we computed a grid of evolutionary models with the initial metallicity @xmath22 , in the range of @xmath23 with a resolution of @xmath11 , supplemented by @xmath24 , and masses between @xmath5 and @xmath25 .
the resolution of mass varies from @xmath26 to @xmath27 .
we used the yale rotation evolution code ( yrec ) @xcite to construct the models in its non - rotation configuration .
the opal equation - of - state table eos2005 @xcite and opal opacity table gn93 @xcite were adopted , supplemented by the @xcite opacity tables at low temperature .
the diffusion and settling of both helium and heavy elements are computed by using the diffusion coefficients of @xcite for models with a mass less than @xmath28 .
convection is treated according to the standard mixing - length theory .
the value of the mixing - length parameter is fixed at @xmath29 .
the full mixing of material is assumed in the overshooting region .
all models are evolved from zero - age ms to the red giant branch ( rgb ) .
figure [ fig1 ] shows the evolutionary tracks of models with @xmath30 but with different .
the evolutionary tracks are obviously affected by the ovcc .
the model with a large mainly exhibits to be bluer than the model with a smaller at a given age and the lifetime of the burning of core hydrogen is significantly prolonged , but the lifetime of sgb is shrunk . for example , the lifetime of the sgb of the model with @xmath30 and = 0 is @xmath31 myr , which is @xmath32 of its ms lifetime . but the lifetime of the sgb of the model with @xmath30 and = @xmath33 is only @xmath24 myr , which is about @xmath34 of its ms lifetime .
the hertzsprung gap of the model with a large is obviously narrower than that of the model without the ovcc .
the values of the central hydrogen @xmath35 at points b , e , and g in figure [ fig1 ] are about @xmath36 , @xmath37 , and @xmath38 , respectively , while the values of @xmath35 at points c , f , and h are around @xmath39 , @xmath40 , and @xmath41 , respectively .
compared with the whole ms lifetime of stars with a large , the timescale of their ms hook is very short .
thus we define the points b , e , and g as the end of central h - burning phase .
due to the fact that more hydrogen - rich material is brought into h - burning core by the ovcc , the stars with the ovcc produce more nuclear energy .
part of the energy makes the stars more luminous .
the other is transformed into internal energy , which leads to the fact that the stars expand obviously , i.e. , the radius of the stars increases significantly . as a consequence ,
the effective temperature of the stars decreases at the end of central h - burning phase . and the star with a larger is brighter and redder than the star with a small at the end of central h - burning phase .
thus , the ms width and msto of a star cluster must be extended by the presence of the ovcc with different .
moreover , due to the fact that the timescale of the ms hook of the model with a large is significantly shrunk , the structure of ms hook of a young star cluster including many stars with the large is hard to be exhibited in the observed cmd of the cluster . in order to understand the effects of an ovcc on the msto of young- and intermediate - age star clusters , we calculated isochrones with different . to obtain the cmds of the isochrones , the theoretical properties
( [ fe / h ] , , @xmath42 , @xmath43 ) of models are transformed into colors and magnitudes using the color transformation tables of @xcite .
a distance modulus of @xmath44 is adopted , and the value of @xmath45 is adopted for @xmath46 in the calculation .
figure [ fig2 ] shows the cmds of isochrones of models with different . comparing the isochrones of models with a large to those of models with a smaller
, one can find that the upper ms and the msto of young- and intermediate - age star clusters are extended by the ovcc when the different ovccs exist in clusters .
however , the lower ms of the clusters are almost not affected . for the isochrones with the age of @xmath47 myr , the separation between the isochrone of models with @xmath48 and that of models with @xmath49 is about 0.1 mag in color u - v at m@xmath50 , but it is only around @xmath51 mag at m@xmath52 . however , the separation is about @xmath27 mag in color v - i at m@xmath53 .
the emsto of young clusters is better visible in u - v than in v - i ( see the panel @xmath54 of figure [ fig2 ] and figure [ fig3 ] ) .
these characteristics are similar to those found in ngc 1856 and ngc 1755 @xcite * and are also expected in the rotational scenario . *
we assumed that a distribution of such as a bimodal of @xmath55 and @xmath33 exists in a star cluster .
the isochrones of models with @xmath56 are chosen as the standard one .
the isochrones of models with a large are compared to those of models with @xmath56 to obtain the equivalent age spread caused by the ovcc .
hereafter , the beginning and the end points of ms hook of an isochrone are labeled as b and c , respectively . the local bluest point
( local minimum u - v ) of ms is marked as point a , and the brightest point ( minimum @xmath57 ) after point c is labeled as point d ( see figure [ fig4 ] ) .
the values of the central hydrogen @xmath35 of models at points a , b , c , and d on the isochrone with age @xmath58 myr are about @xmath59 , @xmath36 , @xmath39 , and @xmath55 , respectively . according to the initial mass function , the more massive the stars , the less the number of the stars . moreover , the more massive the stars , the faster their evolutions .
the stars between point c and d ( @xmath35 between about @xmath39 and 0 for @xmath56 but between about @xmath40 and 0 for @xmath48 ) should be rare in a young star cluster .
thus we compare the non - overshooting isochrone between point a and b to the isochrone with an overshooting to obtain the equivalent age spread .
for example , inspected by eyes , the non - overshooting isochrone between point a and b in figure [ fig4 ] is considered to overlap with the overshooting isochrone with @xmath48 and age @xmath60 myr .
thus , the equivalent age spread caused by the overshooting with @xmath48 is @xmath61 myr when the age of the cluster is @xmath47 myr .
the equivalent age spreads for other isochrones with different and ages are obtained in the similar way .
the `` points c and d '' on the isochrone with age @xmath60 myr and @xmath48 can not be matched by those of any isochrone with @xmath56 .
thus , we can not obtain the equivalent age spread through comparing the points c or d of isochrones with different . due to the fact that the models with a large are obviously brighter and redder than the models with a smaller at the end of central h - burning phase , the end point of the central h - burning phase of an isochrone with @xmath48 can not be reproduced by that of the isochrones with @xmath56 ( see figure [ fig4 ] ) .
thus , when one uses the isochrones with a small to fit the cmd of a cluster including many stars with a larger , one could find that there are many stars that can not be reproduced after the end of the central h - burning phase . in order to understand whether the effects of the ovcc can explain the observed emstos of star clusters , the equivalent age spreads caused by the ovcc are extracted as mentioned above and compared to observed ones .
* we do a similar experiment as @xcite .
* figure [ fig5 ] shows the comparison between the equivalent age spreads and those observed by @xcite , @xcite , @xcite , @xcite , and @xcite .
the values of the age spreads of goudfrooij sample adopted here are the values of @xmath62 , which are considered to can not be explained by rotating models ( * ? ? ?
* but see * brandt & huang 2015 and * niederhofer et al .
the equivalent age spreads increase with an increase in age of star clusters and , and are in good agreement with the observed ones , which shows that the observed emstos of young- and intermediate - age star cluster can be interpreted by the effects of the ovcc . *
the inferred age spreads of a given shown in figure [ fig5 ] are relative to the same age isochrone with @xmath56 , i.e. the extension of msto caused by a at a given age is relative to the isochrone with the same age but with @xmath56 .
* keep in mind that a single value of can not produce any emsto in the cmd of a cluster . * and
the trend between the age of star clusters and the inferred age spread is also expected in the rotational scenario .
* figure [ fig5 ] shows that the equivalent age spreads caused by the ovcc increase with an increase in age of star clusters .
however , when the age of star clusters is larger than 2 gyr , the equivalent age spreads caused by = 0.2 almost not increase with an increase in age of clusters .
this is because the convective core of low mass stars shrink during their ms .
but for more massive stars , the convective core increases in size during the initial stage of ms of these stars before the core begins to shrink .
the larger the , the later the core begins to shrink . in the late stage of the ms of stars , when the core shrinks to a certain degree , the chemical composition of the core including the overshooting region is mainly helium . at this time , the effects of the overshooting on the evolution of stars are not significant .
the smaller the value of the , the earlier this scenario appears .
therefore , the equivalent age spreads caused by the ovcc can not increase continuously with an increase in age of star clusters . as a consequence ,
the equivalent age spreads caused by = 0.2 reach a maximum of about @xmath63 myr at the age of about @xmath64 gyr .
* @xcite have shown that observationally there is a peak in the inferred age spread at age @xmath65 gyr , which is consistent with that expected by rotation models of @xcite .
the equivalent age spread is still present at age @xmath66 gyr in ovcc scenario .
but the width of emsto caused by the ovcc reaches a maximum in color v - i of cmd of star clusters at age @xmath67 gyr , and then decreases with an increase or decrease in age ( see panels @xmath68 and @xmath69 of figure [ fig2 ] and figure [ fig6 ] ) .
the equivalent age spread caused by @xmath70 is about 300 myr at age @xmath71 gyr .
but the separation between the ms of models with = 0.2 and that of models with = 0 is only @xmath72 mag at @xmath73 mag , which is comparable with the observational error .
the separation is around @xmath74 mag at @xmath75 mag ( see panel @xmath68 of figure [ fig6 ] ) .
when the age of clusters is about @xmath76 gyr , the ms of models with @xmath56 is very close to that of models with @xmath48 in the cmd ; the separation between the @xmath48 isochrone and the @xmath77 isochrone is about @xmath78 mag at @xmath79 mag .
the maximum separation between the ms of the @xmath77 or @xmath33 isochrone and that of the @xmath56 isochrone in color v - i at age @xmath80 gyr is about @xmath28 times as large as that at age @xmath81 gyr , and is around @xmath82 times as large as that at age @xmath83 gyr .
the emsto caused by a large should be present in the cmd of star clusters with ages @xmath84 gyr unless the fraction of stars with the large is too small .
but it is not as evident as that in intermediate - age star clusters .
this trend is not reflected by the equivalent age spreads .
* ngc 1856 exhibits the extended mstos .
@xcite claimed that the emstos of ngc 1856 can be explained by the effects of rotation with initial rotation rate @xmath85 @xmath86 . in figure
[ fig7 ] , we compare the data of ngc 1856 to models with different , which shows that the stars at the end of the central h - burning phase and the subgiants can be reproduced by the models with @xmath48 and @xmath87 but can not be reproduced by the models with @xmath56 or @xmath11 . the age of a star cluster obtained from isochrones increases with an increase in .
for example , when one uses the isochrones of models with @xmath56 to fit the cmd of ngc 1856 , one could obtain an age of @xmath88 myr ; however , when one uses the isochrones of models with @xmath48 to fit the cmd , an age of @xmath89 myr would be obtained .
moreover , figure [ fig7 ] also shows that one would obtain a smaller age spread if one uses the isochrones with a small to fit the emstos of ngc 1856 .
panel @xmath68 of figure [ fig7 ] shows that the emstos of ngc 1856 can be explained by an age spread no more than @xmath90 myr of models with @xmath48 .
* ngc 1856 also exhibits the dual mss .
the dual mss of ngc 1856 are separated by @xmath91 mag in color u - v at @xmath57 @xmath92 and merge together @xmath93 magnitudes below @xcite . in order to explain the split ms ,
the rotation as high as 0.9 @xmath86 is required @xcite .
moreover , the simulated populations of @xcite can not reproduce the stars near the end of the central h - burning phase ( see their figure 4 ) .
but this might be due to stars with decretion disks seen nearly edge on @xcite .
* the separation between the isochrone with age @xmath94 myr and that with age @xmath60 myr is @xmath91 mag in color u - v at @xmath57 @xmath92 , which is in good agreement with observation of @xcite , and is only @xmath95 mag at @xmath57 @xmath96 . * recently , * @xcite analyzed the data of ngc 1856 and speculated that the rapid stellar rotation scenario is the favored explanation of the multiple sequences of ngc 1856 rather than esfh scenario .
we computed rapid rotation models with @xmath85 @xmath86 and with the rotational mixing efficiency for the sun @xcite . in figure
[ fig8 ] , we compare the evolutionary tracks of our models with those of geneva models @xcite , which shows that our rotating and non - rotating models with @xmath56 are comparable with geneva models . figure [ fig8 ] shows that the effects of the ovcc with @xmath48 on the end point of central h - burning phase can not be mimicked by the effects of the rapid rotation .
when both the ovcc and the rapid rotation exist in a star , the effects of the ovcc will play a dominant role because the ovcc can mix all material in the overshooting region with that in the convective core but the rotational mixing only brings a small part of material in the shell on the top of the overshooting region into the convective core . due to the centrifugal effect ,
rotating models are fainter and redder than non - rotating overshooting models in the early stage of ms . however , in the late stage of ms , due to the fact that slightly more h - rich material is brought into h - burning core by rotational mixing , rotating models produce slightly more nuclear energy .
thus , they exhibit slightly brighter than non - rotating overshooting models ( see the panel @xmath54 of figure [ fig8 ] ) .
however , the rotational mixing in the radiative region also leads to an increase in the mean density of the region , i.e. a decrease in radius of stars , which results in the fact that rotating models have a higher effective temperature at the end of central h - burning phase . as a consequence , rotating overshooting models are slightly brighter and bluer than non - rotating overshooting models at the end of the central h - burning phase . in panels
@xmath97 , @xmath54 , and @xmath68 of figure [ fig9 ] , we compare rotating and non - rotating models with different to the data of ngc 1856 , which show that the effects of the rotation of @xmath3 @xmath86 are not enough to explain the emstos of ngc 1856 in our calculations . due to the fact that the centrifugal effect plays a dominant role in the early stage of ms phase of stars whose envelope is radiative ( with masses @xmath98 ) and leads to the fact that rotating models are fainter and redder than non - rotating models ,
the ms of rotating models is almost parallel with that of non - rotating models between @xmath57 @xmath99 and @xmath57 @xmath100 .
the separation between the ms of rotating models and that of non - rotating models is about @xmath101 mag
. a higher rotation rate will lead to a larger separation .
but the characteristic of the parallel separation seems to be not in good agreement with that of ngc 1856 . *
if there exist a large and a rapid rotation in stars , the elements produced in the core can be easily brought to the surface of the stars .
for the model with @xmath102 and @xmath85 in young clusters , the surface nitrogen abundance at the end of ms is about @xmath103 times as large as the initial abundance .
but for the models with @xmath77 and @xmath85 and the models with @xmath104 without rotation , the surface nitrogen abundance is almost invariable in ms phase . if there is a nitrogen spread in msto stars that would reflect the coexistence of rapid rotation and large .
due to the mixing effect of deep convective envelope , the surface nitrogen abundance in rgb is about @xmath105 times as large as the initial abundance .
but for the cluster with an age of @xmath106 myr , the envelopes of rgb stars with @xmath107 and @xmath108 are radiative . their nitrogen abundance can not be enhanced . for rgb stars with an age @xmath109 gyr
, the surface nitrogen abundance of models with @xmath102 is about @xmath110 times as large as that of models with @xmath56 in non - rotation case , but is about @xmath111 times in rotation case . * in figure [ fig10 ] , we compare overshooting models with the same age but with different to the data of ngc 1856 .
panel @xmath97 of figure [ fig10 ] shows that the emstos of ngc 1856 can be explained well by the effects of between 0 and 0.7 .
the separation between the isochrone with @xmath107 and that with @xmath56 is about 0.1 mag at @xmath57 @xmath112 and is around 0.014 mag at @xmath57 @xmath113 . in order to understand whether or not overshooting models can produce the emstos and the split mss of ngc 1856
, we performed a stellar population synthesis by way of monte carlo simulations following the initial mass function of @xcite .
@xcite found that about @xmath114 and @xmath115 of stars belong to the blue- and red - ms in ngc 1856 , respectively .
we assumed @xmath116% of population with @xmath107 , @xmath117% of population with @xmath102 , @xmath118% of population with @xmath48 , @xmath117% of population with @xmath77 , and @xmath116% of population with @xmath56 . in the synthesized population ,
we included observational errors taken to be a gaussian distribution with a standard deviation of 0.03 in magnitude and color .
panel @xmath54 of figure [ fig10 ] shows that the emsto of ngc 1856 is reproduced well by the simulated coeval population .
moreover , panel @xmath97 of figure [ fig10 ] shows that the bottom of the rgb of our variable models is located at u - v @xmath119 and @xmath57 @xmath120 .
both the observational and the simulated results show that there are many stars
. these stars can not appear in the simulations of esfh models and rapid rotation models . *
the intermediate - age star clusters have a relatively tight sgb @xcite . *
the subgiants of ngc 1856 seem to be not reproduced well by the simulation .
the larger the , the shorter the timescale of subgiants . moreover , the number of stars with @xmath102 and @xmath24 are less than that of stars with @xmath48 . thus , it is hard for the subgiants with @xmath102 and @xmath24 to occur in the cmd .
the number of stars with @xmath56 are much less than that of stars with @xmath48 .
therefore , the number of subgiants obtained by the simulation are also rare . in order to understand why there are many stars with u - v between about 0.6 and 1.2 in the cmd of ngc 1856 , we calculated a binary - star stellar population following @xcite by using the hurley rapid binary evolution codes @xcite .
the calculation shows that the effects of binary stars are not enough to explain the emstos of ngc 1856 .
however , the calculation clearly shows that there are many interacting binary stars with core - helium burning ( cheb ) whose luminosity and the effective temperature are very close to that of subgiants ( see figure [ fig11 ] ) .
the subgiant branch of ngc 1856 could be polluted by the cheb interacting binary stars .
many of stars of ngc 1856 with u - v between about @xmath87 and @xmath121 and with @xmath57 around @xmath122 mag might be the cheb interacting binaries .
moreover , the cheb interacting binaries could make us misunderstand that the `` subgiants '' of ngc 1856 mainly concentrate towards the end of the blue - ms expected from the mstos of the cluster .
when both the effects of a rapid rotation and that of different ovccs exist in the star cluster ngc 1856 , figure [ fig12 ] clearly shows that the effects of the rapid rotation of @xmath3 @xmath86 on the emstos are negligible compared with the effects of the different .
the emstos of the isochrones mainly result from the effects of the ovcc .
the separation between the isochrone with @xmath107 and @xmath108 and that with @xmath77 and @xmath85 is about @xmath123 mag at @xmath57 @xmath112 and is around @xmath124 mag at @xmath57 @xmath113 . the centrifugal effect increases with an increase in rotation rate
. a higher rotation rate will lead to the more significant split at @xmath57 @xmath125 , which can not be mimicked by the effects of ovcc .
thus , the determinations of rotation rates of stars and the split of ms of young clusters at @xmath57 @xmath126 aid in determining the effects of rotation . * the split ms of ngc 1856 was not reproduced by ovcc models .
this could not rule out ovcc scenario because variable and rapid rotation could coexist in a star cluster .
moreover , the blue ms could be affected by interacting binaries ( see figure [ fig11 ] ) .
* * there is a pulsation constant @xmath127 for pulsating stars , where @xmath128 is a period of pulsating stars , or a large frequency separation @xmath129 for stars with solar - like oscillations . in a star cluster ,
the radii of msto stars with a large are much larger than those of msto stars with a small .
thus , the periods of the former could be about @xmath130 times as long as those of the latter .
for example , at the end of the central h - burning phase of a star with @xmath30 , the period of the model with @xmath102 is about @xmath131 times as long as that of the model with @xmath56 .
in other words , if the variable exists in a star cluster , long - periodic pulsating / variable stars ( in bright msto ) and short - periodic pulsating stars ( in faint msto ) would be observed in the star cluster .
this case can not occur in rotation scenario and esfh scenario .
for example , a difference of @xmath90 myr in age can only lead to a change of about @xmath132 , depending on the age , in frequencies or periods of pulsation of msto stars .
gravity darkening caused by the angle of view can not result in any change in the frequencies or periods .
thus , asteroseismical observation for the young- and intermediate - age star clusters would directly rule out or confirm ovcc scenario , and show what plays a dominant role in the origin of the emsto . *
figures [ fig5 ] and [ fig10 ] show that the ovcc can lead to the observed emstos .
the variable ovcc is a good potential mechanism for explaining the emstos of star clusters .
however , it is dependent on the fraction of stars with a large that whether or not the ovcc is responsible for the observed emstos of star clusters .
asteroseismology only tells us that the value of can be different for different stars with an approximatelly equal mass .
it has not revealed the fraction of stars with a large .
why there are different for stars is another open question for asteroseismology and stellar physicists .
the ovcc is more efficient than rotation at bringing h - rich material into the h - burning core .
the effects of the ovcc with @xmath48 on the msto of star clusters is hard to be mimicked by the effects of a rapid rotation . in order to reproduce the stars of ngc 1856 at the end of central h - burning phase , an overshooting of around 0.4 @xmath9 is required for our models .
when both the ovcc and a rapid ratation exist in stars , the msto is mainly affected by the effects of the ovcc . however , the centrifugal effect of the rapid rotation playing a role in the ms stars whose envelope is radiative leads to the fact that the split of ms is more significant .
the points c and d of the rotating isochrones with @xmath85 can be matched by those of non - rotating isochrones with the same but with different ages .
the equivalent age spread obtained by comparing points c and d of a rotating isochrone to those of non - rotating isochrone with the same is twice more than that obtained through comparing points a and b of the rotating isochrone to those of non - rotating isochrone . @xcite
might overestimate the equivalent age spreads caused by rotation .
moreover , the higher the efficiency of the rotational mixing , the more the h - rich material brought into the h - burning core , the larger the luminosity of the rotating models at the end of central h - burning phase , which can lead to an increase in the equivalent age spread caused by rotation .
these may be the reason why the equivalent age spread of @xcite rotating models can match that of ngc 1856 but that of our rotating models can not .
however , a high efficiently rotational mixing in the radiative region of stars can hinder stellar expansion and lead to the fact that the evolutionary tracks of the stars move towards the upper left of the hertzsprung - russell diagram .
thus , the effects of a large are different from that of rotation . the equivalent age spread between the isochrone with @xmath48 and that with @xmath56 in panel @xmath97 of figure [ fig10 ] is @xmath133 myr , and the spread between the isochrone with @xmath102 and that with @xmath77 is @xmath134 myr , which are close to the spread of 150 myr determined by @xcite . the equivalent age spread between the isochrone with @xmath102 and that with @xmath56 is @xmath135 myr , which is larger than that determined by @xcite .
however , figure [ fig10 ] shows that the width between the msto of the isochrone with @xmath102 and that with @xmath56 does not exceed that of ngc 1856 . due to the fact that the lifetimes of the sgb of the stars with a large are much shorter than those of the stars without overshooting or with a small
, the population of sgb is not only dependent on the number of stars but also dependent on .
moreover , the sgb of young star clusters could be polluted by the cheb interacting binaries .
thus , the sgb of young star clusters can not be used to study the origin of the emstos . * the effects of ovcc on the msto of young- and intermediate - age star clusters are similar to that of rotation . the equivalent age spread caused by rotation
has a peak at age @xmath136 gyr @xcite , which is dependent on the efficiency of rotational mixing .
the observationally inferred age spread also has a peak at age @xmath136 gyr @xcite .
this characteristic does not occur in ovcc models .
the emsto caused by the ovcc is still present in star clusters with age @xmath84 gyr .
the emsto is hard to be exactly described by the equivalent age spreads relative to the same age isochrone with @xmath56 , especially for star clusters with age @xmath137 gyr .
our models show that the width of the emsto caused by a of @xmath11 or @xmath33 at age @xmath138 gyr is about @xmath3 times as wide as that at age @xmath13 gyr .
the width of the emsto caused by the ovcc has a peak in color v - i of cmd of star clusters at age @xmath67 gyr , and then decreases with an increase or decrease in age , which can not be reflected by the equivalent age spreads . in rotation models of @xcite ,
the emsto can occurs in star clusters with age larger than @xmath139 gyr because rotational mixing exists in stars like the sun @xcite .
thus the emsto of star clusters with age @xmath140 gyr can not be used to differentiate ovcc models and rotation models . if emsto does not exist in star clusters with ages between @xmath139 and @xmath141 gyr that is hard to be explained by ovcc scenario .
perhaps , that is due to the fraction of stars with a large being small in the star clusters .
* * at the end of the central h - burning phase of rapid rotation models , the surface nitrogen abundance can be enriched by @xmath142 in young star clusters for a between 0.4 and 0.6 .
this does not happen in the rotating models with @xmath70 and the non - rotating models .
moreover , the pulsating periods of bright msto stars would be longer than those of faint msto stars in ovcc scenario because the former has a low mean density , which also does not occur in rotation scenario and esfh scenario .
these characteristics could aid us in differentiating ovcc scenario and rotation scenario . *
asteroseismology has shown that the value of overshooting parameter of the convective core is variable for stars with an approximatelly equal mass .
we calculated the effects of the ovcc on the msto of young- and intermediate - age star clusters , and found that the observed emstos of young- and intermediate - age star clusters can be explained well by the effects of the ovcc with different . the equivalent age spread caused by the ovcc increases with an increase in age of star clusters and an increase in . due to the fact that the convective core of stars can shrink during the late stage of ms
, the equivalent age spread can not increase continuously with an increase in age of star clusters .
the emstos of ngc 1856 can be reproduced well by the coeval populations with different .
the effects of a rotation with @xmath85 and the rotational mixing efficiency for the sun on the msto of a cluster with an age of @xmath106 myr are negligible compared with the effects of the ovcc . however , the presence of the rotation can result in the fact that the split of ms of the cluster is more significant .
the emstos of star clusters may be related to the effects of the ovcc with different . * but keep in mind that a single value of @xmath0 can not produce any emsto in the cmd of clusters .
asteroseismical observation would confirm or rule out ovcc scenario . *
we thank the anonymous referee for helpful comments that helped the authors improve this work , a. p. milone for providing the data of ngc 1856 , and f. niederhofer for the data of niederhofer sample , as well as the support from the nsfc 11273012 .
alexander , d. r. , & ferguson , j. w. 1994 , apj , 437 , 879 bastian , n. , cabrera - ziri , i. , davies , b. , & larsen , s. s. 2013 , mnras , 436 , 2852 bastian , n. , cabrera - ziri , i. , niederhofer , f. , et al .
2017 , , arxiv : 1611.06705 bastian , n. , & de mink , s. e. 2009 , mnras , 398 , l11 bastian , n. , & niederhofer , f. 2015 , , 448 , 1863 bastian , n. , niederhofer , f. , kozhurina - platais , v. , et al .
2016 , , 460 , l20 bond , h. e. , gilliland , r. l. , schaefer , g. h. et al .
2015 , apj , 813 , 106 brandt , t. d. , huang , c. x. 2015 , , 807 , 25 cabrera - ziri , i. , bastian , n. , davies , b. , magris , g. , bruzual , g. , schweizer , f. 2014 , , 441 , 2754 cabrera - ziri , i. , bastian , n. , hilker , m. , et al .
2016 , , 457 , 809 correnti , m. , goudfrooij , p. , kalirai , j. s. , et al .
2014 , , 793 , 121 correnti , m. , goudfrooij , p. , puzia , t. h. , de mink , s. e. 2015 , , 450 , 3054 georgy , c. , ekstrm , s. , granada , a. , et al .
2013 , a&a , 553 , a24 girardi , l. , rubele , s. , & kerber , l. 2009 , mnras , 394 , l74 glatt , k. , grebel , e. k. , & sabbi , e. et al .
2008 , aj , 136 , 1703 goudfrooij , p. , girardi , l. , kozhurina - platais , v. , et al .
2014 , , 797 , 35 goudfrooij , p. , girardi , l. , rosenfield , p. , et al .
2015 , , 450 , 1693 goudfrooij , p. , puzia , t. h. , chandar , r. , & kozhurina - platais , v. 2011 , , 737 , 4 goudfrooij , p. , puzia , t. h. , kozhurina - platais , v. , & chandar , r. 2009 , aj , 137 , 4988 guenther , d. b. , demarque , p. , gruberbauer , m. 2014 , , 787 , 164 lejeune , t. , cuisinier , f. , & buser , r. 1998 , , 130 , 65 li , c. , de grijs , r. , bastian , n. , et al . 2016a , mnras , 461 , 3212 li , c. , de grijs , r. , deng , l. 2014 , nature , 516 , 367 li , c. , de grijs , r. , deng , l. 2016b , raa , 16 , 179 li , c. , de grijs , r. , deng , l. , & milone , a. p. 2016c
, apj , arxiv : 1611.04659 li , z. , mao , c. , chen , l. 2015 , , 802 , 44 li , z. , mao , c. , chen , l. , zhang , q. 2012 , , 761 , 22 li , z. , mao , c. , zhang , l. , zhang , x. , chen , l. 2016 , , 255 , 7 liu , z. , yang , w. , bi , s. , et al .
2014 , , 780 , 152 mackey , a. d. , & broby nielsen , p. 2007
, mnras , 379 , 151 mackey , a. d. , broby nielsen , p. , ferguson , a. m. n. , & richardson , j. c. 2008 , apj , 681 , l17 milone , a. p. , bedin , l. r. , piotto , g. , & anderson , j. 2009 , a&a , 497 , 755 milone , a. p. , bedin , l. r. , piotto , g. , et al .
2015 , , 450 , 3750 milone , a. p. , marino , a. f. , dantona , f. , et al . 2016a , , 458 , 4368 milone , a. p. , marino , a. f. , dantona , f. , et al .
2016b , arxiv:1611.06725 niederhofer , f. , georgy , c. , bastian , n. , ekstrm , s. 2015a , , 453 , 2070 niederhofer , f. , hilker , m. , bastian , n. , silva - villa , e. 2015b , , 575 , 62 niederhofer , f. , bastian , n. , kozhurina - platais , v. , et al .
2016 , , 586 , a148 piatti , a. e. 2013 , mnras , 430 , 2358 piatti , a. e. , & bastian , n. 2016 , mnras , 463 , 1632 pinsonneault , m. h. , kawaler , s. d. , sofia , s. , & demarqure , p. 1989
, apj , 338 , 424 rogers , f. j. , & nayfonov , a. 2002 , apj , 576 , 1064 rubele , s. , kerber , l. , & girardi , l. 2010 , mnras , 403 , 1156 salpeter , e. e. 1955 , apj , 121 , 16 thoul , a. a. , bahcall , j. n. , loeb , a. 1994 , apj , 421 , 828 yang , w. 2015 , arxiv:1508.00955 yang , w. 2016a , apj , 821 , 108 yang , w. 2016b , apj , 829 , 68 yang , w. , & bi , s. 2007 , apjl , 658 , l67 yang , w. m. , bi , s. l. , meng , x. c. , liu , z. 2013 , , 776 , 112 yang , w. m. , meng , x. c. , bi , s. l. , et al .
2011 , , 731 , l37 yang , w. , tian , z. , bi , s. , ge , z. , wu , y. , & zhang , j. 2015 , mnras , 453 , 2094 | recent investigations have shown that the extended main - sequence turnoffs ( emstos ) are a common feature of intermediate - age star clusters in the magellanic clouds .
the emstos are also found in the color - magnitude diagram ( cmd ) of young - age star clusters .
the origin of the emstos is still an open question .
moreover , asteroseismology shows that the value of the overshooting parameter @xmath0 of the convective core is not fixed for the stars with an approximatelly equal mass .
thus the msto of star clusters may be affected by the overshooting of the convective core ( ovcc ) .
we calculated the effects of the ovcc with different @xmath0 on the msto of young- and intermediate - age star clusters . *
if @xmath0 varies between stars in a cluster , * the observed emstos of young- and intermediate - age star clusters can be explained well by the effects .
the equivalent age spreads of msto caused by the ovcc are related to the age of star clusters and are in good agreement with observed results of many clusters . moreover ,
the observed emstos of ngc 1856 are reproduced by the coeval populations with different @xmath0 .
the emstos of star clusters may be relevant to the effects of the ovcc .
the effects of the ovcc * are similar to that of rotation in some respects .
but the effects can not result in a significant split of main sequence of young star clusters at @xmath1 .
* the presence of a rapid rotation can make the split of main sequence of young star clusters more significant . |
the discovery of the cosmic acceleration @xcite and the search for dark energy responsible for its origin @xcite have stimulated the study of different field models driving the cosmological evolution .
such a study usually is called the potential reconstruction @xcite , because the most typical examples of these models are those with a scalar field , whose potential should be found to provide a given dynamics of the universe . in the flat friedmann models with a single scalar field ,
the form of the potential and the time dependence of the scalar field are uniquely determined by the evolution of the hubble variable ( up to a shift of the scalar field ) . during last years
the models with two scalar fields have also become very popular .
this is connected with the fact that some observations point out that the relation between the pressure and the energy density could be less than -1 @xcite .
such equation of state arises if the matter is represented by a scalar field with a negative kinetic term .
this field is called `` phantom '' @xcite . moreover , according to some observations @xcite the universe undergoes a transition between normal and phantom phase .
such an effect is dubbed `` phantom divide line crossing '' @xcite . in principle , the effect of phantom divide line crossing can be explained in the model with the only scalar field provided a special form of the potential and initial conditions is chosen @xcite or in models with a non - minimally coupled scalar field @xcite .
however , the models with two scalar fields , one standard and one phantom , look more `` natural '' for the description of the phantom divide line crossing @xcite . in our preceding paper @xcite
we have studied the procedure of reconstruction of the potential in two - field models .
it was shown that there exists a huge variety of potentials and time dependences of the fields realizing the same cosmological evolution .
some concrete examples were considered , corresponding to the evolution beginning with the standard big bang singularity and ending in the big rip singularity @xcite .
one can ask oneself : what is the sense of studying different potentials and scalar field dynamics if they imply the same cosmological evolution ?
the point is that the scalar and phantom field can interact with other fields and influence not only the global cosmological evolution but also other observable quantities .
one of the possible effects of the presence of normal and phantom fields could be their influence on the dynamics of cosmic magnetic fields .
the problem of the origin and of possible amplification of cosmic magnetic fields is widely discussed in the literature @xcite . in particular ,
the origin of such fields can be attributed to primordial quantum fluctuations @xcite and their further evolution can be influenced by hypothetic interaction with pseudoscalar fields breaking the conformal invariance of the electromagnetic field @xcite . in the present paper
we consider the evolution of magnetic fields created as a result of quantum fluctuations , undergoing the inflationary period with unbroken conformal invariance and beginning the interaction with pseudoscalar or pseudophantom fields after exiting the inflation and entering the big bang expansion stage , which is a part of the bang - to - rip scenario described in the preceding paper @xcite .
we shall use different field realizations of this scenario and shall see how the dynamics of the field with negative parity influences the dynamics of cosmic magnetic fields . to our knowledge the possible influence of the two - field dynamics , ( when one of two ( pseudo)scalar fields is a phantom one ) on the magnetic fields was not yet discussed in the literature .
speaking of cosmic magentic fields we mean the large - scale galactic , intergalactic or super - cluster magnetic fields of order from @xmath0 to @xmath1 with correlation from 100 kpc to several mpc to the extent that they are originated from scalar and , possibly gauge field fluctuations after exiting the inflation .
their seeds may well have @xmath2 or less ( see @xcite ) .
the structure of the paper is as follows : in sec . 2 we recall the bang - to - rip scenario and describe some examples of different dynamics of scalar and phantom fields ; in sec .
3 we introduce the interaction of the fields ( phantom or normal ) with an electromagnetic field and write down the corresponding equations of motion ; in sec .
4 we describe the numerical simulations of the evolution of magnetic fields and present the results of these simulations ; sec .
5 is devoted to concluding remarks .
we shall consider a spatially flat friedmann universe with the metric @xmath3 here the spatial distance element @xmath4 refers to the so called comoving coordinates .
the physical distance is obtained by multiplying @xmath4 by the cosmological radius @xmath5 .
we would like to consider the cosmological evolution characterized by the following time dependence of the hubble variable @xmath6 , where `` dot '' denotes the differentiation with respect to the cosmic time @xmath7 : @xmath8 this scenario could be called `` bang - to - rip '' @xcite because it is obvious that at small values of @xmath7 the universe expands according to power law : @xmath9 while at @xmath10 the hubble variable explodes and one encounters the typical big rip type singularity .
( the factor one third in ( [ btr ] ) was chosen for calculation simplicity ) . in our preceding paper @xcite ) we considered a class of two - field cosmological models , where one field was a standard scalar field @xmath11 , while the other was a phantom one @xmath12 .
the energy density of the system of these two interacting fields is @xmath13 analyzing the friedmann equation and @xmath14 . in this system
the planck mass @xmath15 , the planck length @xmath16 and the planck time @xmath17 are equal to @xmath18 .
then when we need to make the transition to the `` normal '' , say , cgs units , we should simply express the planck units in terms of the cgs units . in all that follows
we tacitly assume that all our units are normalized by the proper planck units .
thus , the scalar field entering as an argument into the dimensionless expressions should be divided by the factor @xmath19 . ]
@xmath20 we have shown that in contrast to models with one scalar field , there is huge variety of potentials @xmath21 realizing a given evolution . moreover , besides the freedom in the choice of the potential , one can choose different dynamics of the fields @xmath22 and @xmath23 realizing the given evolution .
we have studied in @xcite some particular exactly solvable examples of forms of potentials and time dependences of fields @xmath11 and @xmath12 providing the evolution ( [ btr ] ) . here
we shall present and apply some of them .
consider the potential . ]
@xmath24 and the fields @xmath25 @xmath26 ( here the expressions for the fields @xmath22 and @xmath23 should be multiplied by @xmath27 . for the relation between planck units and cgs ones
see e.g. @xcite if we would like to substitute one of these two fields by the pseudoscalar field , conserving the correct parity of the potential , we can choose only the field @xmath11 because the potential @xmath28 is even with respect to @xmath11 , but not with respect to @xmath12 . in what follows
we shall call the model with the potential ( [ v - i ] ) , the pseudoscalar field ( [ phi1 ] ) and the scalar phantom ( [ xi1 ] ) `` model @xmath29 '' . consider another potential @xmath30 with the fields @xmath31 @xmath32 this potential is even with respect to the field @xmath12 .
hence our model @xmath33 is based on the potential ( [ v - ii ] ) , the pseudophantom field ( [ xi - new3 ] ) and the scalar field ( [ phi - new3 ] ) .
they will be the fields with the negative parity which couple to the magnetic field .
the action of an electromagnetic field interacting with a pseudoscalar or pseudophantom field @xmath11 is @xmath34 where @xmath35 is an interaction constant and the dual electromagnetic tensor @xmath36 is defined as @xmath37 where @xmath38 with the standard levi - civita symbol @xmath39},\ \epsilon_{0123 } = + 1 .
\label{lc}\ ] ] variating the action ( [ action ] ) with respect to the field @xmath40 we obtain the field equations @xmath41 @xmath42 the klein - gordon equation for the pseudoscalar field is @xmath43 the klein - gordon equation for the pseudophantom field ( which is the one that couples with the magnetic field in the model @xmath33 ) differs from equation ( [ kg ] ) by change of sign in front of the kinetic term . in what follows
we shall neglect the influence of magnetic fields on the cosmological evolution , i.e. we will discard the electromagnetic coupling in equation ( [ kg ] ) .
if one wants to rewrite these formulae in terms of the three - dimensional quantities ( i.e. the electric and magnetic fields ) one can find the expression of the electromagnetic tensor in a generic curved background , starting from a locally flat reference frame in which it is well known the relation between electromagnetic fields and @xmath44 and using a coordinate transformation .
it is easy to see that we have , for the metric ( [ friedmann ] ) : @xmath45 the field equations ( [ maxwell ] ) , ( [ maxwell1 ] ) and ( [ kg ] ) rewritten in terms of @xmath46 and @xmath47 become @xmath48 @xmath49 , \label{maxwell3}\ ] ] @xmath50 @xmath51 for a spatially homogeneous pseudoscalar field equations ( [ maxwell2 ] ) and ( [ maxwell3 ] ) look like @xmath52 @xmath53taking the curl of ( [ maxwell7 ] ) and substituting into it the value of @xmath46 from ( [ maxwell4 ] ) we obtain @xmath54 where @xmath55 stands for the three - dimensional euclidean laplacian operator .
let us introduce @xmath56 and its fourier transform @xmath57 here the field @xmath47 is an observable magnetic field entering into the expression for the lorentz force .
the field equation for @xmath58 is @xmath59 f(\vec{k},t)=0 , \label{max - four}\ ] ] where `` dot '' means the time derivative .
this last equation can be further simplified : assuming @xmath60 and defining the functions @xmath61 one arrives to @xmath62f_{\pm}=0 , \label{max - four1}\ ] ] where we have omitted the arguments @xmath63 and @xmath7 .
assuming that the electromagnetic field has a quantum origin ( as all the fields in the cosmology of the early universe @xcite ) the modes of this field are represented by harmonic oscillators . considering their vacuum fluctuations responsible for their birth we can neglect the small breakdown of the conformal symmetry and treat them as free .
in conformal coordinates @xmath64 such that the friedmann metric has the form @xmath65 the electromagnetic potential @xmath66 with the gauge choice @xmath67 , @xmath68 satisfies the standard harmonic oscillator equation of motion @xmath69 hence the initial amplitude of the field @xmath66 behaves as @xmath70 , while the initial amplitude of the functions @xmath44 is @xmath71 .
the evolution of the field @xmath44 during the inflationary period was described in @xcite , where it was shown that the growing solution at the end of inflation is amplified by some factor depending on the intensity of the interaction between the pseudoscalar field and magnetic field .
here we are interested in the evolution of the magnetic field interacting with the pseudoscalar field after inflation , where our hypothetic bang - to - rip scenario takes place .
more precisely , we would like to see how different types of scalar - pseudoscalar potentials and field dynamics providing the same cosmological evolution could be distinguished by their influence on the evolution of the magnetic field .
we do not take into account the breaking of the conformal invariance during the inflationary stage and all the effects connected with this breakdown will be revealed only after the end of inflation and the beginning of the bang - to - rip evolution .
this beginning is such that the value of the hubble parameter , characterizing this evolution is equal to that of the inflation , i.e. @xmath72 in turn , this implies that we begin evolution at the time moment of the order of @xmath73 .
we shall consider both the components @xmath74 and @xmath75 and we shall dwell on the scenarios @xmath29 and @xmath33 described in the preceding section . anyway , our assumption regarding the initial conditions for equation ( [ max - four1 ] ) can be easily modified in order to account for the previous possible amplification of primordial magnetic fields as was discussed by @xcite .
thus , all estimates for the numerical values of the magnetic fields in today s universe should be multiplied by some factor corresponding to the amplification of the magnetic field during the inflationary stage .
hence , our results refer more to differences between various models of a post inflationary evolution than of the real present values of magnetic fields , whose amplification might be also combined effect of different mechanism @xcite .
in this section we present the results of numerical simulations , for the two models @xmath29 and @xmath33 introduced in sec . 2 . in our models , the equation of motion for the modes @xmath76 ( [ max - four1 ] ) reads , where today - time is taken to be near the crossing of the phantom divide line , i.e. at @xmath77 .
this implies in turn that at the beginning of the `` bang - to - rip '' evolution the cosmological radius is @xmath78 . ] : @xmath79f_{\pm}=0 , \label{edm}\ ] ] where @xmath80 stands for the scalar field @xmath11 in the model @xmath29 and for the phantom @xmath12 in the model @xmath33 , so that @xmath81 equation ( [ edm ] ) is solved for different values of the wave number @xmath63 and the coupling parameter @xmath35 .
( the parameter @xmath35 has the dimensionality inverse with respect to that of the scalar field ; the wave number @xmath63 has the dimensionality of inverse length ; the time @xmath82 ) .
qualitatively we remark that in ( [ max - four1 ] ) the coupling term influence becomes negligible after some critical period .
after that the magnetic fields in our different scenarios evolve as if the parameter @xmath35 in ( [ max - four1 ] ) had been put equal to zero .
indeed , it can be easily seen that the interaction term vanishes with the growth of the cosmological radius @xmath83 .
then the distinction between the two models is to be searched in the early time behavior of the field evolution . noting that in both our models @xmath29 and @xmath33 the time derivative @xmath84 is positive , by inspection of the linear term in equation ( [ max - four1 ] ) we expect the amplification to be mainly given for the mode @xmath75 provided the positive sign for @xmath35 is chosen
; so we will restrict our attention on @xmath75 .
we can also argue that the relative strength of the last two terms in the left - hand side of ( [ edm ] ) is crucial for determining the behavior of the solution : when the coupling term prevails ( we remark that we are talking about @xmath85 so this term is _ negative _ in our models ) then we expect an amplification , while when the first term dominates we expect an oscillatory behavior . for future reference
it is convenient to define @xmath86 which is just the ratio between the last two terms in the left - hand side of equation ( [ edm ] ) .
indeed our numerical simulations confirm these predictions .
let us consider the model @xmath29 with @xmath87 and @xmath88 , where @xmath16 is the planck length .
such a value of the wave number @xmath63 corresponds to the wave length of @xmath89 at the present moment .
we obtain an early - time amplification of about 2 orders of magnitude , with the subsequent oscillatory decay .
notice that the parameter @xmath90 in this model at the beginning is very small : this corresponds to the dominance of the term proportional to @xmath84 and , hence , to the amplification of the field @xmath75 . at the time scale of the order of @xmath91 , where @xmath17 is the planck time
, this regime turns to that with big values of @xmath90 where the influence of the term proportional to @xmath84 is negligible . for
the same choice of the parameters @xmath35 and @xmath63 in the model @xmath33 the amplification is absent . for the model @xmath29 , with the parameter choice @xmath92 , @xmath88 .
it can be easily seen that at a time scale of order @xmath91 the ratio becomes greater than 1 . ]
( given in planck units ) in model @xmath29 with the parameter choice @xmath92 , @xmath88 .
the behavior , as said above in the text , consists in an amplification till a time of order @xmath91 , after which the oscillations begin . ]
( given in planck units ) in model @xmath29 with the parameter choice @xmath93 , @xmath88 .
the behavior , consists in an amplification till a time of order @xmath94 , after which the oscillations begin . ] in figure [ mag1 ] we present the time dependence of the function @xmath90 for the model @xmath29 for the values of @xmath92 and @xmath88 chosen above .
the figure [ mag2 ] manifests the amplification of the magnetic field in the model @xmath29 .
naturally the effect of amplification of the magnetic field grows with the coupling constant @xmath35 and diminishes when the wave number @xmath63 increases . in figure [ mag3 ]
we display the results for the case of @xmath95 , which is admittedly extreme and possibly non realistic , but good for illustrative purposes . here
the amplification is more evident and extends for a longer time period .
let us try to make some estimates of the cosmic magnetic fields in the universe today , using the correlation functions .
the correlation function for the variable @xmath44 is defined as the quantum vacuum average @xmath96 and can be rewritten as @xmath97 integrating over the angles , we come to @xmath98 to estimate the integral ( [ corr2 ] ) we notice that the main contribution to it comes from the region where @xmath99 ( see e.g.@xcite ) and it is of order @xmath100 where @xmath101 . in this estimation
the amplification factor is @xmath102 where the subscript @xmath103 is not present since we have taken the trace over polarizations .
now we are in a position to give numerical values for the magnetic fields at different scales in the model @xmath29 for different values of the coupling parameter @xmath35 .
these values ( see table [ tab ] ) correspond to three values of the coupling parameter @xmath35 ( 1,10 and 100 @xmath104 ) less than 1 make the effect of the coupling of the magnetic field with the pseudoscalar field negligible . ] and to two spatial scales @xmath105 determined by the values of the wave number @xmath63 . we do not impose some physical restrictions on the value of @xmath35 .
it is easy to see that the increase of @xmath35 implies the growth of the value of the magnetic field @xmath106 .
we also shall consider a rather large value of the wave number @xmath63 corresponding to the physical wavelength @xmath107 pc at the present moment when @xmath108 . while this scale looks too small for the description of large - scale cosmic magentic fields , we use it for illustrative purposes to underline the strong dependence of the amplification of magnetic field on the corresponing wavelength values .
let us stress once again that we ignored the effects of amplification of the magnetic fields during inflation to focus on seizable effects during evolution .
@lccc & @xmath109&@xmath110 & @xmath111 + & & & + @xmath112&@xmath113&@xmath114&@xmath115 + & & & + @xmath116&@xmath117&@xmath118&@xmath119 + [ tab ] finally notice that our quantum `` initial '' conditions correspond to physical magnetic fields which for presented values of @xmath63 are @xmath120 is equal to @xmath121 for @xmath122 and @xmath123 for @xmath124 .
we have seen that the evolution of the cosmic magnetic fields interacting with a pseudoscalar ( pseudophantom ) field is quite sensitive to the concrete form of the dynamics of this field in two - field models where different scalar field dynamics and potentials realize the same cosmological evolution .
we confirm the sensitivity of the evolution of the magnetic field with respect to its helicity given the sign of the coupling constant @xmath35 and that the @xmath80 is a monotonic function of time ( as it is really so in our models ) .
we give also some numerical estimates of the actual magnetic fields up to the factor of amplification of such fields during the inflationary period .
the toy model of the bang - to - rip evolution studied in this paper , can not be regarded as the only responsible for the amplification of cosmic magnetic fields implying their present observable values .
it rather complements some other mechanisms acting before .
however , the difference between cosmic magnetic fields arising in various models ( giving the same expansion law after the inflation ) is essential . it may provide a discriminating test for such models
naturally , the influence of the interaction between a pseudoscalar ( phantom ) field and a cosmic magnetic field on the dynamics of the latter depends on the velocity of change with time of the former .
the larger is the time derivative of the pseudoscalar field , the more intensive is the growth of the magnetic field ( remember that the evolution of the scalar field is monotonic ) .
the results of numerical calculations illustrated in sec .
4 confirm these qualitative considerations .
moreover , one can see that there exists a certain range of values of the wave number @xmath63 ( and hence of the corresponding wavelengths of the magnetic field ) where the effect is stronger .
indeed , if @xmath63 is too small the interaction term is small as well and the evolution of the magnetic field is damped . on the other hand
if the wave number @xmath63 is too large the interaction term is small compared to the oscillatory term , proportional as usual to @xmath125 and the evolution has practically oscillatory character .
thus , the study of interplay between the dynamics of global scalar fields providing the cosmological evolution and the magnetic fields looks promising .
this interplay has another interesting aspect .
the pseudoscalar - electromagnetic field interaction can imply a conversion of the photons into axions .
such an effect can cause the observable dimming of supenovae .
while it was shown @xcite that this effect can not mimic the cosmic acceleration , it could nevertheless mimic the dark energy fluid with a phantom equation of state @xcite .
thus , the interrelation amongst an electromagnetic field , a scalar and a phantom field can reveal some surprises .
this work was partially supported by grants rfbr 08 - 02 - 00923 and lss-4899.2008.2 .
the work of a.a . was supported by grants fpa2007 - 66665 , 2005sgr00564 , 2007piv10046 , by the consolider - ingenio 2010 program cpan ( csd2007- 00042 ) and program rnp 2.1.1.1112 .
99 riess a et al .
1998 _ astron . j. _ * 116 * 1009 ; perlmutter s j et al .
1999 _ astroph .
j. _ * 517 * , 565 sahni v and starobinsky a a 2000 _ int .
_ d * 9 * 373 ; padmananbhan t 2003 _ phys . rep . _ * 380 * 235 ; peebles p j e and ratra b 2003 _ rev
phys . _ * 75 * 559 ; sahni v 2002 _ class .
quantum grav . _ * 19 * 3435 ; copeland e j , sami m and tsujikawa s 2006 _ int
_ d * 15 * 1753 ; sahni v and starobinsky a a 2006 _ int . j. mod .
_ d * 15 * 2105 ; frieman j a , turner m s and huterer d 2008 dark energy and the accelerating universe _ preprint _ 0803.0982 [ astro - ph ] starobinsky a a 1998 _ jetp lett . _
* 68 * 757 ; boisseau b , esposito - farese g , polarski d and starobinsky a a 2000 _ phys .
* 85 * 2236 ; burd a b and barrow j d 1988 _ nucl
_ b * 308 * 929 ; barrow j d 1990 _ phys .
_ b * 235 * 40 ; gorini v , kamenshchik a , moschella u and pasquier v 2004 _ phys .
_ d * 69 * 123512 ; zhuravlev v m , chervon s v and shchigolev v k 1998 _ jetp _ * 87 * 223 ; chervon s v and zhuravlev v m the cosmological model with an analytic exit from inflation _ preprint _ gr - qc/9907051 ; yurov a v phantom scalar fields result in inflation rather than big rip _
astro - ph/0305019 ; yurov a v and vereshchagin s d 2004 _ theor .
* 139 * 787 ; guo z k , ohta n and zhang y z 2007 _ mod .
_ a * 22 * 883 ; guo z k , ohta n and zhang y z 2005 _ phys . rev .
_ d * 72 * 023504 ; zhuk a 1996 _ class .
* 13 * 2163 ; szydlowski m and czaja w 2004 _ phys .
_ d * 69 * 083507 ; szydlowski m and czaja w 2004 _ phys .
_ d * 69 * 083518 ; szydlowski m 2005 _ int . j. mod .
_ a * 20 * 2443 ; szydlowski m , hrycyna o and krawiec a 2007 _ jcap _ * 0706 * 010 ; vernov s yu construction of exact solutions in two - fields models and the crossing of the cosmological constant barrier _
astro - ph/0612487 alam u , sahni v , saini t d and starobinsky a a 2004 _ mon . not .
soc . _ * 354 * 275 ; padmanabhan t and choudhury t r 2003 _ mon . not .
_ * 344 * 823 ; choudhury t r and padmanabhan t 2005 _ astron.astrophys . _
* 429 * 807 ; alam u , sahni v and starobinsky a a 2004 _ jcap _ * 0406 * 008 ; wang y and freese k 2006 _ phys . lett .
_ b * 632 * 449 ; upadhye a , ishak m and steinhardt p j 2005 _ phys .
_ d * 72 * 063501 ; dicus d a and repko w w 2004 _ phys .
_ d * 70 * 083527 ; nesseris s and perivolaropoulos l 2004 _ phys .
_ d * 70 * 043531 ; lazkoz r , nesseris s and perivolaropoulos l 2005 _ jcap _ * 0511 * 010 caldwell r r 2002 _ phys .
_ b * 545 * 23 alam u , sahni v and starobinsky a a 2007 _ jcap _ * 0702 * 011 ; nesseris s and perivolaropoulos l 2007 _ jcap _ * 0702 * 025 esposito - farese g and polarski d 2001 _ phys . rev .
_ d * 63 * 063504 ; vikman a 2005 _ phys .
_ d * 71 * 023515 ; perivolaropoulos
l 2005 _ phys .
_ d * 71 * 063503 ; mcinnes
b 2005 _ nucl .
_ b * 718 * 55 ; arefeva i ya , koshelev a s and vernov s yu 2005 _ phys .
_ d * 72 * 064017 ; perivolaroupoulos
l 2005 _ jcap _ * 0510 * 001 ; caldwell r r and doran m 2005 _ phys .
_ d * 72 * 043527 ; sahni v and shtanov yu 2003 _ jcap _ * 0311 * 014 ; sahni v and wang l 2000 _ phys .
_ d * 62 * 103517 andrianov a a , cannata f and kamenshchik a y 2005 _ phys .
_ d * 72 * 043531 ; cannata f and kamenshchik a yu 2007 _ int .
_ d * 16 * 1683 ; li m , feng b and zhang x 2005 _ jcap _ * 0512 * 002 gannouji r , polarski d , ranquet a and starobinsky a a 2006 _ jcap _ * 0609 * 016 zhang x f , li h , piao y s and zhang x m 2006 _ mod .
lett . _ a * 21 * 231 ; feng b , wang x and zhang x 2005 _ phys .
_ b * 607 * 35 ; guo z k , piao y s , zhang x and zhang y z 2005 _ phys .
_ b * 608 * 177 ; hu w 2005 _ phys .
_ d * 71 * 047301 ; perivolaropoulos
l 2005 _ phys . rev .
_ d * 71 * 063503 ; caldwell r r and doran m 2005 _ phys .
_ d * 72 * 043527 .
andrianov a a , cannata f and kamenshchik a yu 2006 _ int . j. mod . phys .
_ d * 15 * 1299 ; andrianov a a , cannata f and kamenshchik a yu 2006 _ j. phys . _
a * 39 * 9975 andrianov a a , cannata f , kamenshchik a yu and regoli d 2008 _ jcap _ * 0802 * 015 starobinsky a a 2000 _ grav .
* 6 * 157 ; caldwell r r , kamionkowski m and weinberg n n 2003 _ phys .
lett . _ * 91 * 071301 mukhanov v f and chibisov g v 1981 _ jetp lett . _ * 33 * 532 grasso d and rubinstein h r 2001 _ phys . rep .
_ * 348 * 163 ; giovannini m 2004 _ int . j. mod .
_ d * 13 * 391 ; giovannini m 2008 _ lect.notes phys .
_ * 737 * 863 turner m s and widrow l m 1988 _ phys .
_ d * 77 * 2743 ; bamba k , ohta n and tsujikawa s , _ generic estimates for magnetic fields generated during inflation including dirac - born - infeld theories _ , 2008 _ preprint _ 0805.3862 [ astro - ph ] ; diaz - gil a , garcia - bellido j , garcia perez m and gonzalez - arroyo a , _ primordial magnetic fields from preheating at the electromagnetic scale _ , 2008
_ preprint _ 0805.4159 [ hep - ph ] ratra b 1992 _ astrophys .
j _ * 391 * l1 ; garretson w d , field g b and carroll s m 1992 _ phys . rev . _
d * 46 * 5346 ; dolgov a d 1993 _ phys
_ d * 48 * 2499 ; field g b and carroll s m 2000 _ phys .
_ d * 62 * 103008 ; finelli f and gruppuso a 2001 _ phys .
_ b * 502 * 216 ; bamba k and sasaki m 2007 _ jcap _ * 0702 * 030 anber m m , sorbo l _ jcap _ 2006 * 0610 * 018 mukhanov v 2005 _ physical foundations of cosmology _ ( cambridge university press ) mirizzi a , raffelt g g and serpico p d 2008 _ lect .
notes phys . _
* 741 * 115 csaki c , kaloper n and terning j 2005 _ annals of phys _ * 317 * 410 | we consider two different toy cosmological models based on two fields ( one normal scalar and one phantom ) realizing the same evolution of the bang - to - rip type .
one of the fields ( pseudoscalar ) interacts with the magnetic field breaking the conformal invariance of the latter .
the effects of the amplification of cosmic magnetic fields are studied and it is shown that the presence of such effects can discriminate between different cosmological models realizing the same global evolution of the universe . _ keywords _ : phantom dark energy , scalar fields , magnetic fields + |
the vast majority of the @xmath4 catalogued quasars have very similar optical spectral energy distributions : a very blue continuum with prominent broad emission lines .
several studies provide evidence that a population of _ red _ quasars also exists , overlooked by surveys optimized to find quasars by their uv excess and strong emission lines .
the size of any such `` hidden '' population of quasars is important for a complete understanding of the quasar phenomenon .
low et al . ( 1989 ) found that the majority of quasars in a small ir - selected sample were significantly redder than optically selected objects .
sprayberry & foltz ( 1992 ) argue that the larger dust extinction and very weak emission features in low ionization broad absorption line ( bal ) quasars means that they are underrepresented in the known quasar population . using a much larger sample , webster et al .
( 1995 ) have argued that the large range of @xmath5 colors ( 1 to 8) of _ radio - selected _ quasars implies that up to 80% of quasars are undetected because of several magnitudes of dust obscuration .
adding fuel to the fire are the discoveries of striking examples of individual qso s with extremely red spectral energy distributions and/or lacking prominent emission lines .
the continuum of 3cr 68.1 ( @xmath6 ) is highly reddened ( brotherton et al .
1998 ) and polarized . and the original radio - loud bal , first j155633.8 + 351758 ( @xmath7 ; becker et al . 1997 ) , is also quite red with @xmath8 ( hall et al .
1997 ) and very weak or absent broad emission lines . further evidence of red quasars comes from x - ray surveys .
dust reddened quasars are faint or invisible in soft x - rays because of the high columns of associated absorbing gas ( risaliti et al .
2001 ) but are relatively bright in hard x - ray surveys now being done with chandra ( barger et al .
2001 ) . countering the dusty quasar argument
are the studies of benn et al .
( 1998 ) and whiting , webster , & francis ( 2001 ) showing that radio - selected quasars are often red because of an excess of flux at long wavelengths from host galaxy contributions or a red synchrotron component rather than from extinction at shorter wavelengths .
determining the fraction of very red quasars in the total population , regardless of the cause of their redness , is essential for a precise determination of the quasar luminosity function , and bears directly on the evolution of quasars and the nature of the quasar - galaxy connection . also important is whether the red quasar population may harbor a significant number of optically faint , gravitationally lensed objects , where extinction in an intervening dusty lens reddens the quasar light and cancels the magnification bias which accompanies dust - free lensing ( bartelmann & loeb 1999 ) .
a sizable population of red lensed quasars would have significant ramifications for any conclusions based on the statistics of gravitational lensing .
the 2-micron all - sky survey ( 2mass ; kleinmann et al .
1994 ) provides an opportunity for gauging the population of very red quasars , and determining the fraction missed by optically - selected quasar surveys . with a resolution of 4 and @xmath9 limiting @xmath10 magnitude of 14.3
, 2mass will catalog @xmath11 sources with high - precision coordinates and ir fluxes . the overwhelming majority of unresolved 2mass sources are ordinary galactic stars , so to use 2mass effectively in selecting extragalactic point sources requires additional information .
we have matched 2mass with the vla first survey 20 cm catalog ( becker et al .
1995 ; white et al .
1997 ) , obtaining a sample dominated by extragalactic sources .
particularly interesting for the red quasar question is the subset of first/2mass matches that are too faint in the optical to appear in extant optical catalogs such as the apm ( mcmahon & irwin 1992 ) .
we are in the early stages of a program using imaging and spectroscopy to investigate the nature of these objects .
we report here the discovery of two extremely red quasars from our sample , first j013435.7@xmath0093102 ( also known as pks 0132@xmath0097 ) and first j073820.1 + 275045 ( hereafter j0134@xmath00931 and j0738 + 2750 , respectively ) .
we list their 2mass magnitudes and first positions and 20 cm fluxes in table 1 .
neither is bright enough optically to appear in the apm catalog , though both are faintly visible on the digitized second generation sky survey .
both appear to be intrinsically red .
one object , j0134 - 0931 , is extremely bright in the ir , which we argue below is due to gravitational lensing .
the gravitationally lensed nature of j0134 - 0931 has been independently discovered by winn et al.(2001a ) , who investigate its radio properties in detail .
we discuss whether the redness of these two quasars arises in the host galaxy or an intervening galaxy along the line of sight , and compare their properties to other very red quasars in the literature , in particular mg 0414 + 0534 ( hewitt et al .
1992 ) and pks 1830@xmath0211 ( jauncey et al.1991 ; courbin et al .
1998 ) which have similarly extreme optical near - infrared colors ( table 1 ) .
if even one or two of the lensed objects with extreme spectral energy distributions is intrinsically red , then the statistics imply that a significant _ unlensed _ population of red quasars probably exists , of which j0738 + 2750 is an example , and the discovery of the two ir - bright , optically faint quasars in the 2mass - first sample is further evidence that optically selected quasars do not provide a complete sampling of the quasar population .
the intrinsically red interpretation leads us to speculate that many , perhaps even most , quasars may undergo a red phase early in their lives when the active galactic nucleus ( agn ) first turns on . if , on the other hand , the lensed quasars are reddened by dusty lensing galaxies , then there must exist numerous undiscovered , optically faint , gravitationally lensed quasars with small ( @xmath12 ) image separations .
j0738 + 2750 has @xmath13 and a first 20 cm flux density of 2.6 mjy . in 1999 november
, we obtained an optical spectrum of j0738 + 2750 using the low resolution imaging spectrograph ( lris , oke at al .
1995 ) at keck observatory .
the spectrum reveals a very red continuum and identifiable , but relatively weak , quasar emission features ( figure 1 ) at @xmath14 .
the emission is particularly weak and both and show narrow line doublet absorption at or near the emission line reference frame .
the absorption comprises three discrete systems at redshifts of 1.97496 , 1.98535 , and 1.99127 , while the appears to be single and at the emission line redshift .
the absorption features appear marginally resolved ; higher resolution data may reveal multiple components in the absorption , but at much lower velocity dispersion than in .
the velocity dispersion of the three systems is @xmath15 in the quasar rest frame .
a broad absorption trough extends from -13000 to -17500 , reaching a depth of @xmath16 of the continuum level ; we identify this as due to , making j0738 + 2750 a weak bal quasar .
absorption systems ( @xmath17 ) corresponding to each of the three absorbers are also present , as is ] emission , which is unusually strong relative to the other emission lines .
there is also an intervening + absorption system at @xmath18 . in 2000 january
, we obtained 36 minutes of spectropolarimetry of j0738 + 2750 , again using lris .
these data show that this quasar highly polarized in the optical , ranging from 5% at to more than 9% at ( figure 1 ) . in 2000 april
, we obtained @xmath19 and @xmath20 band images of j0738 + 2750 ( figure 2 ) in photometric conditions using the near infrared imaging camera ( nirc , matthews & soifer , 1994 ) at keck observatory .
the 9 point dither pattern totaled 270s of integration time .
although the seeing was @xmath21 , instrument rotator problems limited the resolution to roughly twice that . at this level , j0738 + 2750 is unresolved , but 18 away is a faint galaxy which is perhaps the source of the intervening absorption lines .
j0134 - 0931 is distinguished by being both a strong radio source ( 900 mjy at 20 cm in the first survey ) and a very bright infrared object , with 2mass @xmath22 .
it is identified in ned as a quasar , but no redshift is given .
the original veron - cetty & veron catalog listed objects as quasars if they were stellar optical identifications of radio sources , without requiring spectral confirmation .
the optical `` identification '' of j0134 - 0931 is found in bolton , shimmens , & wall ( 1975 ) , the seventh part of the parkes catalog , where finding charts for optical counterpart candidates are presented for 122 radio sources . for 121 of these ,
the photographic finding charts show an obvious optical source , but for pks0132 - 097 , the tick marks enclose blank sky .
thus , this quasar , `` discovered ''
nearly thirty years ago , not only had no redshift , but also had no optical identification . in 2000
august , we obtained an infrared spectrum of j0134 - 0931 at lick observatory using the gemini ir imaging spectrograph on the shane 3 m telescope .
the spectrum ( figure 3 ) covers windows in @xmath23 , and @xmath24 , with resolution varying from 30 in @xmath19 to 120 in @xmath24 .
a single broad emission feature is seen at a wavelength of @xmath25 , which we ascribed to h@xmath26 at @xmath27 .
the observed fwhm of this line is 280 , more than twice the instrumental resolution of 120 .
this identification was confirmed by an optical spectrum obtained with the andalucia faint object spectrograph camera ( alfosc ) on the nordic optical telescope ( not ) at la palma on 2000 october 30 and 31 .
the observations consisted of two 1800s exposures covering the 4000 - 9000 region , and a single 1800s exposure with a different grism to extend the wavelength coverage redwards to @xmath28 .
both sets of observations were made with a 12 slit at a pa of 125at an airmass of @xmath29 .
the seeing was 09 on both nights and the resolution was 15 for both spectra .
the optical spectrum clearly shows emission lines of ] and at a redshift of @xmath30 ( figure 3 ) ; is weakly detected .
images of j0134 - 0931 were obtained with nirc in @xmath19 and @xmath20 in 2000 september , totaling 270s of integration each .
conditions were photometric , the seeing at @xmath31 was 037 , and the instrument rotator problem ( which compromised the nirc observation of j0738 + 2750 ) had been fixed .
the data were reduced using the dimsum package in iraf up to the point of sky subtraction , after which the dithered exposures were subsampled with a spline interpolation by a factor of 3.14 , then shifted and combined .
the @xmath20 image ( figure 4 ) clearly resolves two components , the brighter component ( a ) being extended along one axis in both @xmath19 and @xmath20 bands .
daophot / iraf was used to construct a point spread function ( psf ) from a star near the edge of the image .
fitting and subtracting the psf shows that component d is consistent with a point source while a leaves large residuals .
( for consistency , we have retroactively adopted the nomenclature of winn et al .
2001a where the components are ranked by radio flux ; see also below . ) the centroids of the two objects are 0635 apart .
a 110s observation of j0134 - 0931 made on 1992 december 31 with the very large array ( vla ) in the a - configuration at 8.4 ghz ( 3.6 cm ) was recovered from the nrao archive .
the data were calibrated using observations of 0141 - 094 and 3c286 with the aips software package . the resulting map ( figure 4 ) has a resolution of 03 and is very similar to the nirc image , showing two components , the brighter being resolved in one axis at the same position angle as in the infrared .
using aips imfit to fit elliptical gaussians to components a and d yields total fluxes of 518 and 69.0 mjy , respectively , separated by 0660 .
see the accompanying paper by winn et al .
( 2001a ) for an in - depth discussion of the radio properties of j0134 - 0931 .
the apparent extremely large luminosity of j0134 - 0931 initially led us to suspect that gravitational lensing was at work . adopting h@xmath32 = 70 and @xmath33
, the redshift of j0134 - 0931 implies that @xmath34 even before dereddening . in the absence of lensing
, it would be one of the most luminous objects in the universe .
the morphological similarity of the nirc k-band image and the vla 3.6 cm map virtually confirms that gravitational lensing is at work .
early in our analysis of j0134 - 0931 , we became aware of the independent work of winn et al .
( 2001a ) showing that j0134 - 0931 is a 5-component radio source , which they also interpreted as due to lensing . to explore the lensing nature of this object further and to compare the infrared and radio morphologies , we deconvolved our nirc @xmath20-band image of j0134 - 0931 using a variety of techniques .
all produced qualitatively similar results ; here we discuss those from the maximum entropy task mem in iraf .
the deconvolved image is shown in figure 5 as greyscale , smoothed to reduce the pixelation , and has an effective resolution of @xmath35 ( 2 pixels ) .
the deconvolution breaks component a into two peaks , a and b ( marked by crosses in figure 5 ) , perhaps with some associated extended structure at lower surface brightness .
component d is consistent with a point source in our @xmath20 image .
a fainter and much less certain source , e , lies to the northeast , plus a hint of excess flux remaining between components a and d. we next used the difmap package ( shepherd 1997 ) to map and fit models to the archival vla data .
a check of the nasa extragalactic database ( ned ) shows that j0134 - 0931 is a gigahertz - peaked spectrum ( gps ) radio source , and therefore we expect the dominant emission regions to have a small size compared to the vla beam ( odea , baum & stanghellini 1991 ) , so we fit models consisting only of point sources to the data .
initially we used four point sources : three to account for the resolved brighter component , plus a fourth for component d. a fifth source , c , just to the se of the brightest component ( a ) was required to reduce the otherwise large residuals and to produce a reasonable fit to the high s / n data .
these five sources account for 96% of the flux and there is no evidence for further discrete components in the residual map , to a limit of @xmath36mjy ( @xmath37 ) .
the resulting model is shown in figure 5 , convolved with a @xmath35 restoring beam to match approximately the resolution of the deconvolved nirc image ; the position of c is indicated by a black dot .
the four components a , b , c , and e have fluxes of 346 , 84 , 79 , and 20 mjy respectively , while d is 65 mjy .
these ratios are roughly 100 : 24 : 23 : 19 : 5.8 ( a through e ) .
the radio positions and flux ratios are consistent with the more detailed , higher resolution results of winn et al .
( 2001a ) .
the radio and ir images in figure 5 have been placed on the same scale and registered using component d. there is good positional agreement for components a and b ; the nirc data suggest an object consistent with e , and at still lower significance there is a hint of another component roughly midway between a and d. better signal - to - noise and resolution are needed to test the reality of these fainter components .
fitting psfs at the locations of the five radio components in the original @xmath20 image with daophot / iraf greatly reduces the residuals over the single psf a - component case , and significantly improves the fit over those with just two or even three components .
the relative fluxes a : b : c : d : e are , roughly , 100 : 90 : 10 : 25 : 5 . using the cplucy deconvolution task iraf , we arrive at very similar flux ratios for the 5 component case , 100 : 90 : 14 : 26 : 6 . given the limited spatial resolution and depth of the @xmath20 data , the results of fitting five components is evidence only for a qualitative consistency between the radio and infrared images .
the overall similarity of the @xmath20-band and 3.6 cm images leads us to conclude that j0134 - 0931 is gravitationally lensed , but the exact nature of the lensing is unclear .
one possibility is a quadruple image , the 4 components being a , b , d , and e , though the latter is not unambiguously detected in the nirc images .
component c , clearly detected in the radio and perhaps marginally needed to account for the 2@xmath38 flux , is difficult to interpret , but is perhaps structure in the quasar radio source and aligned optical emission .
the observed mismatch between the radio and infrared images can be ascribed to insufficient resolution and signal - to - noise of the @xmath24-band data , real optical / radio morphology differences among the components , or even variability of the radio source .
one obvious discrepancy between the radio and infrared is the relative brightness of a and b ; though comparable at 2@xmath38 , there is a factor of 4 difference at 3.6 cm . in support of the @xmath20 result ,
similar analysis with cplucy and daophot of the somewhat lower resolution ( 05 ) nirc @xmath19-band image finds that b is at least 80% as bright as a. at the same time , the flux ratio of the sum of the close components ( a , b , c , e ) to component d is remarkably similar at 3.6 cm ( 8.1 ) and @xmath31 ( 8.0 ) especially given the brightness differences seen among the four closest components . in the @xmath19-band image
, we find the a / d brightness ratio to be much greater , @xmath39 . at the redshift of j0134 - 0931 ,
the observed @xmath19-band is rest frame near - uv , so the change in color between @xmath19 and @xmath20 of the two components implies significant differential reddening , or possibly microlensing effects , or both .
the reader is referred to winn et al .
( 2001a ) for an in - depth multi - frequency , high resolution radio analysis of the gravitational lensing of j0134 - 0931 , including a more in - depth discussion of the lensing model possibilities . from the flux calibrated lris spectrum of j0738 + 2750
, we obtain @xmath40 ; combining with the 2mass data yields a relatively red @xmath41 . from our composite optical / ir spectrum of j0134 - 0931 , we measure @xmath42 and a _ lower limit _ @xmath43 there are no detected counts in the @xmath44 passband . the extreme red color of j0134 - 0931 is confirmed by the broad band photometry of winn et al .
( 2001a ) .
the foreground galactic extinctions ( schlegel , finkbeiner , & davis 1998 ) are small in both cases ( table 1 ) .
these colors are approximate given the uncertainties of the spectrophotometric calibration , but the extremely red observed spectral energy distributions of these objects are beyond question . a critical issue is whether the redness of these unusual objects is due to environmental effects local to the quasar or to intervening extinction somewhere along our line of sight . the strong absorption features in the spectrum of j0738 + 2750 at the same redshift as the emission lines argue strongly that in this particular case , the reddening arises locally to the quasar .
the intervening absorption features are not particularly strong ; many quasars with similar intervening systems are not noticeably reddened ( white et al .
the strong polarization ( figure 1 ) , indicative of large amounts of local scattering , is also strong evidence that reddening by dust is responsible for the atypical spectral energy distribution of this object .
sprayberry & foltz ( 1992 ) have found that bal quasars are intrinsically redder than non - bal quasars , so even the modest bal ( figure 1 ) is consistent with local reddening in j0738 + 2750 .
determining the origin of the redness of j0134 - 0931 is more difficult .
there are no prominent absorption lines in the optical spectrum and the gravitational lens interpretation raises the real possibility of intervening reddening and extinction .
dusty lenses have been debated as the source of the redness in lensed quasars ( malhotra , rhoads , & turner 1997 ; kochanek et al .
2000 ) .
we have attempted to constrain the redshift of the dust using simple least squares minimization and various popular extinction laws to redden the composite quasar template spectrum of brotherton et al.(2001 ) to match the observed spectra of j0738 + 2750 and j0134 - 0931 . the standard milky way reddening curve fails because of the 2175 silicate `` bump , '' which does not appear in either source . the starburst reddening law of calzetti et al .
( 1994 ) can produce a red quasar spectrum that matched the targets , but could not simultaneously fit the 2mass photometry points , regardless of the assumed redshift for the dust .
only the smc extinction curve ( pei 1992 ) could produce results approximating the overall spectral energy distributions of j0738 + 2750 and j0134 - 0931 , though still not with complete success .
figure 6 compares j0738 + 2750 with the result of reddening the composite spectrum by @xmath45 , assuming the reddening occurs in the quasar frame .
the 2mass photometry points were not used to constrain the fit because of the uncertainty of the scaling between the slit spectroscopy flux calibration and the ir aperture photometry , which we estimate to be 10 - 30% . given the uncertainties of applying the smc reddening in this environment , the fit is reasonably good over the range covered by the keck spectrum , and fits the 2mass points within the overall scaling uncertainty .
there is a continuum slope mismatch at longer wavelengths , perhaps because of a real difference between the intrinsic unreddened continuum of j0738 + 2750 and the quasar composite .
the observed turndown is a general characteristic of a reddened power law spectral energy distribution ( francis , whiting , & webster 2000 ) . adjusting the redshift of the extinction to the intervening absorber system at 1.424 leads to a slight improvement in the fit over the region covered by the lris spectrum , but the fit is the ir is significantly worse .
there is evidence for excess light in j0738 + 2750 compared to the reddened quasar spectrum at short wavelengths ( @xmath46 ) , which is probably scattered quasar light , though a contribution from young stars in the host galaxy can not be ruled out .
based on the high polarization and the overall consistency of the reddening fits , we conclude that j0738 + 2750 is reddened by dust local to the quasar .
figure 7 shows the observed optical - ir spectrum of j0134 - 0931 , with the 2mass and @xmath47 band photometry ( winn et al .
2001a ) overplotted .
we have scaled the photometry points by a factor of 0.7 by eye to bring them into agreement with the spectrum ; the overall agreement confirms that the two spectral regions are scaled correctly .
the @xmath24-band point drops to the continuum level of the spectrum if corrected for the presence of the strong h@xmath26 line .
two reddened versions of the composite spectrum are also overplotted .
the solid line is the composite reddened by @xmath48 with dust local to the quasar .
while the overall shape of the spectrum is reproduced , the fit does not account for the flux in the @xmath19-band , and the slope of the continuum is not reproduced for wavelengths redward of @xmath49 .
the dotted line spectrum shows the result of placing the dust at @xmath50 , where the best fit using the smc law yields @xmath51 . moving the dust to lower redshift results in a slightly improved fit in the ir and perhaps marginally better fit in the optical , but
this is not compelling , given that the same is true for j0738 + 2750 where the polarization points conclusively to dust in the quasar environment . lacking additional evidence , such as polarization measurements or absorption features in the spectrum ,
we conclude that our present data are consistent with dust reddening of j0134 - 0931 , but the redshift of the dust is unconstrained by the modeling .
the existing circumstantial evidence is equally ambiguous as to the location of the reddening of j0134 - 0931 .
gps sources like j0134 - 0931 frequently occur in systems obscured by large amounts of gas and dust ( odea et al .
1996 ) and where the hosts are often morphologically `` disturbed '' , perhaps mergers or interacting galaxies ( see review by odea 1998 ) .
a significant amount of dust associated with the interaction could account for the extreme red energy distribution of j0134 - 0931 .
if j0134 - 0931 is a merger or interacting system with large amounts of dust , then the difficulty in consistently interpreting its ir and radio morphology ( see also winn et al .
2001a ) may not be surprising .
the differential reddening that we find in j0134 - 0931 , however , is highly suggestive of a dusty lens , though microlensing , which may be needed to explain at least some of the details , can not be ruled out .
this object is slated for imaging during cycle 10 of the hubble space telescope ( falco et al .
go9133 ) ; the higher resolution optical data may resolve the situation .
spectropolarimetry would also help determine the whereabouts of the dust .
it is , of course , possible that both the host and the lens are dusty .
the differences between the reddened composite quasar and the observed red quasars can be ascribed to a multitude of factors .
strong iron emission can possibly account for much of the difference in the @xmath19 band .
the turnover and slope mismatch is perhaps due to intrinsic differences between the continuum slopes .
there may be host galaxy contributions with different levels of reddening than the quasars , and the extinction laws are almost certainly different from the smc .
an infrared spectrum of j0738 + 2750 and a higher s / n infrared spectrum for j0134 - 0931 documenting h@xmath52 will permit detailed modeling , as was done by lawrence et al .
( 1995a ) for the red lensed quasar mg 0414 + 0534 , and may help clarify these issues . determining the optical colors of the individual components , which will be possible with the hst images , will also aid in unraveling the nature of j0134 - 0931 . to conclude , our data for j0738 + 2750 place the source of extinction firmly in the host galaxy .
our results are somewhat less certain in the case of j0134 - 0931 .
the lack of strong absorption features in its spectrum suggests that if the lens is dusty , then it must be at a low enough redshift to place telltale absorption features far enough into the blue that they are undetected in our spectrum . with the presently available information
, we can not pin down the redshift of the dust for j0134 - 0931 .
how numerous are dust - reddened objects such as j0134 - 0931 and j0738 + 2750 ?
such unusually red quasars suffer so much extinction in their restframe ultraviolet that they are rendered practically invisible even at moderate redshifts and are potentially severely underestimated in optically - selected samples . establishing
the size of this red population is of paramount importance to a full understanding of the general quasar phenomenon and to the interpretation of quasar gravitational lensing statistics .
although the very red spectral energy distributions of j0134 - 0931 and j0738 + 2750 are unusual compared to the large majority of optically - selected quasars , such red objects are much more common among radio - selected samples of both unlensed ( webster et al .
1995 ; francis et al .
2000 ) and lensed quasars ( malhotra et al .
using matched populations of radio- and optically - selected quasars , francis et al .
( 2000 ) demonstrate that most optically - selected quasars fall in a narrow color range , @xmath53 ; roughly 50% of radio - selected quasars have similar colors , but the other half are distributed over the range @xmath54 .
malhotra et al .
( 1997 ) have shown that a similar color dichotomy exists for the known lensed quasars .
optically - discovered lensed quasars have broad band colors consistent with the general quasar population , typically @xmath55 , while radio - discovered lensed quasars span a much wider and redder color range , @xmath56 having @xmath57 , including one of the reddest extragalactic objects known , mg 0414 + 0534 with @xmath58 ( lawrence et al .
1995a ) .
a color of @xmath59 corresponds to an effective optical - infrared spectral index @xmath60 ( defining @xmath26 in the sense that flux density @xmath61 ) ; we hereafter adopt this our working definition of a `` red quasar '' . to estimate the fraction of dusty red quasars in the general population requires an understanding of why the color distribution of unlensed , radio - selected quasars differs from that of the optically - selected counterparts .
benn et al .
( 1998 ) , francis et al.(2000 ) , and whiting et al .
( 2001 ) argue that many of the red quasars in the sample of webster et al .
( 1995 ) have intrinsically red spectra due to optical / near - infrared synchrotron emission from the radio jets in these relatively bright , flat - spectrum radio sources , but francis et al .
( 2000 ) conclude that dust is necessary to account for a significant fraction of red quasars , amounting to @xmath62 of the total population .
the 3crr sample of 173 quasars and galaxies ( laing , riley & longair 1983 ) contains a number of quasars which have colors significantly redder than the general population .
as 3crr is selected at low radio frequencies , the beamed emission from the radio jet which can produce strong optical synchrotron emission is much reduced compared to the sample of webster et al .
( 1995 ) , and thus these quasar spectra are unlikely to have significant red optical synchrotron components .
the fraction of quasars significantly reddened by dust in 3crr has been estimated by simpson et al .
( 1999 ) as @xmath63% , larger than the fraction deduced by francis et al.(2000 ) , but with small number statistics producing the large error bars .
most optical surveys for quasars rely on some form of blue color selection criterion , so it is not surprising that red quasars suffering large extinction should be missing from optically - selected samples .
it is conceivable , however , that the radio - quiet quasar population simply has , for some reason , no red objects ( @xmath64 ) .
j0738 + 2750 , although not radio - quiet , is radio - intermediate ( if the optical emission is dereddened ) , suggesting that this is probably not the case .
evidence that such objects may be relatively common comes from identifications of chandra hard x - ray sources ( barger et al .
2000 ) which find significant numbers of lightly - reddened quasars , and one example of a `` quasar-2 '' , in which the broad line region is completely hidden by dust ( norman et al .
similarly , using an objective prism survey and rosat data , risaliti et al .
( 2001 ) have identified quasars with anomalously low soft x - ray fluxes , indicating a significant absorbing column , and find that these quasars are significantly redder in the optical than the general population .
all of these results suggest that there is a substantial population of radio - quieter red quasars , and that the intrinsic color distributions of radio - loud and radio - quiet quasars may well be similar .
the different color distributions of optical- and radio - selected lensed quasars documented by malhotra et al .
( 1997 ) can be explained largely as a natural consequence of the above discussion red quasars are missing from the optically - selected samples to begin with .
two other factors enhance the numbers of red , radio - selected lensed quasars , explaining why malhotra et al .
( 1997 ) find that the colors of radio - loud lensed quasars are even redder than those of the general radio - loud unlensed population .
first , a very dusty lensing galaxy can significantly redden the quasar light .
this appears to be the explanation for the very red quasar pks1830 - 211 .
the high differential reddening ( courbin et al . 1998 ) between the components of pks 1830 - 211 and the strong differential molecular absorption at an intervening redshift ( frye , welch , & broadhurst 1997 ; menten et al .
1999 ) is near - conclusive evidence that the responsible dust is in the lensing galaxy and that the color differences are not intrinsic to the source or due to microlensing effects ( courbin et al .
1998 ) .
the second factor contributing to the high numbers of red radio - selected lensed objects is that lensing can magnify the host galaxy contributions , invariably redder than the quasar light .
host galaxy starlight probably explains the very red colors of mg 1131 + 0456 and b 1938 + 666 ( kochanek et al .
but hubble space telescope nicmos images and near - infrared spectra show that host galaxy contributions are negligible for the extremely red lensed quasar mg 0414 + 0534 ( lawrence et al .
1995a ) , which has colors comparable to j0134 - 0931 ( see table 1 ) .
although lawrence et al . ( 1995a ) conclude from multicomponent spectral modeling that a dusty lens is responsible for the red color of mg 0414 + 0534 , the circumstantial evidence argues that the redness arises from dust local to the quasar .
first , it is a gps source .
there are also very strong low - ionization absorption features of and at the host redshift ( lawrence et al.1995b ) , much like the case of j0738 + 2750 discussed above , as well as 21 cm hi absorption with a column consistent with the observed reddening ( moore , carilli , & menten 1998 ) .
in addition , the lensing galaxy seems to be a typical elliptical with a low dust content ( tonry & kochanek 1999 ) .
the size of the quasar population which is reddened by dust local to the source can be estimated from the statistics of radio - discovered gravitationally lensed quasars .
there are now roughly 18 to 20 gravitationally lensed quasars from various radio searches , including the jodrell bank - vla astrometric survey ( jvas , patnaik & narasimha 1993 ) , the cosmic lens all - sky survey ( class , phillips et al.2001 ) and winn et al .
( 2000 ; 2001b , c ) . among these
are the three extremely red objects pks 1830 - 211 ( courbin et al .
1998 ) , mg0414 + 0534 ( hewitt et al .
1992 ) and j0134 - 0931 ( winn et al . 2001a and
this work ) . assuming that both j0134 - 0931 and mg0414 + 0534 are reddened by dust local to the quasar ,
then @xmath62 of the radio - selected objects are intrinsically red quasars .
if , however , j0134 - 0931 and mg0414 + 0534 are , along with pks 1830 - 211 , reddened by their lensing galaxies , the implication is that a potentially large number of gravitationally lensed quasars exists . as only @xmath65 of quasars are radio - loud , such a population will be @xmath66 times as large as the radio - loud lensed quasars . as
dusty lenses are likely to be lower mass , late - type galaxies , the image separations will be small and extinction in the lensing galaxy can more than counteract any amplification of the quasar light ( bartelmann & loeb 1998 ) .
such radio - quiet lensed objects are extremely difficult to discover at any wavelength .
two possible explanations for the existence of intrinsically dust - reddened quasars are youth and orientation .
both the gps and bal phenomena have been conjectured to occur early in the lives of quasars ( odea et al .
1996 ; becker et al .
the reddest quasars , exemplified by j0134 - 0931 and mg 0414 + 0534 , both gps sources , and j0738 + 2750 , with a bal feature , may thus be early evolutionary phases in the lives of quasars , when the agn is first activated or is refueled . furthermore , baker & hunstead ( 1997 ) , in a sample of quasars selected from the molongolo radio source catalogue found that the compact steep spectrum ( css ) quasars in their sample typically had redder than average optical spectra and higher balmer decrements . like gps sources , css sources
are thought to be young , with source evolution proceeding from the gps through the css to the fully - developed frii stage ( de vries et al .
1998 ) . alternatively , within the context of `` unified schemes '' ( antonucci 1993 )
, red quasars may represent objects whose quasar light is just grazing the dusty torus which hides the quasar nucleus of narrow - line agn .
there are well - studied examples from the 3crr sample which suggest that youth and orientation are each partly responsible for the red quasar population : 3c190 is a css radio source in a spectacular merging host galaxy and so possibly a young quasar ( stockton & ridgway 2001 ) , whereas 3c22 and 3c68.1 have all the hallmarks of evolved radio sources at an angle to the line of sight roughly corresponding to the hypothesised torus opening angle of @xmath67deg .
if j0134 - 0931 and mg0414 + 0534 are intrinsically red , then the statistics of lensed and unlensed radio - selected red quasars imply that @xmath68% of radio - loud quasars are significantly reddened by dust in the host galaxy .
the case of j0738 + 2750 proves that there are at least some dust reddened quasars with radio luminosities less than the very high values typical of 3crr and jvas / class sources .
if such intrinsically red objects exist in similar proportion to the radio - loud red quasars , then a substantial population of red , radio - quiet objects awaits discovery .
the relative numbers of red quasars and their redshift distribution can set limits on the relative lengths of the red phase and its possible relation to the bal and gps phases
. if j0134 - 0931 and mg0414 + 0534 are reddened by dust in their lensing galaxies , however , then no large population of intrinsically red quasars exists .
these objects , along with pks 1830 - 211 , have been discovered because they are @xmath69 jy radio sources .
if they are red because of intervening material along the line of sight , then a substantial population of radio - quiet , red , gravitationally lensed systems remains undiscovered , perhaps in numbers 5 - 10 times as great as the radio - loud objects .
their actual numbers are critical to the interpretation of the statistics of lensed quasars , bearing on fundamental issues of cosmology and the masses and evolution of intervening galaxies .
quasar surveys utilizing 2mass and other ir catalogs will eventually build the statistics of the red quasar population , allowing these issues to be addressed definitively .
it will be necessary to carry out much of the exploratory spectroscopy at infrared wavelengths to accurately gauge the numbers of the most extreme objects .
we thank j. winn for sharing his radio and optical results for j0134 - 0931 ahead of publication and for allowing us to use his broad band photometry in figure 7 and table 1 .
he is also thanked for helpful comments on our manuscript .
some of the data presented here 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 keck observatory was made possible by the financial support of the w.m .
keck foundation .
the first survey is supported by grants from the national science foundation ( grant ast-98 - 02791 ) , nato , the national geographic society , sun microsystems , and columbia university .
part of the work reported here was done at the institute of geophysics and planetary physics , under the auspices of the u.s .
department of energy by lawrence livermore national laboratory under contract no . w-7405-eng-48 .
the vla is a facility of the national radio astronomy observatory , operated by associated universities inc .
, under contract with the national science foundation .
the not is operated on the island of la palma jointly by denmark , finland , iceland , norway , and sweden , in the spanish observatorio del roque de los muchachos of the instituto de astrofisica de canarias .
alfosc is owned by the instituto de astrofisica de andalucia ( iaa ) and operated at the not under agreement between iaa and the nbifafg of the astronomical observatory of copenhagen .
k. brand assisted with the not observations .
the two micron all sky survey ( 2mass ) is a joint project of the university of massachusetts and the infrared processing and analysis center / california institute of technology , funded by the national aeronautics and space administration and the national science foundation ( nsf ) .
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 .
lllcllrrrrrrl first 0134@xmath00931 & @xmath70 & @xmath71 & 0.10 & 2.216 & @xmath72 & 22.57 & 20.61 & 18.77 & 16.17 & 14.75 & 13.55 & 900 .
+ first 0738@xmath732750 & @xmath74 & + 275046 & 0.14 & 1.985 & 23.8 & 22.1 & 20.6 & 19.2 & 17.1 & 16.2 & 15.3 & 2.6 + mg 0414@xmath730534 & @xmath75 & @xmath76 & 0.10 & 2.639 & 24.1 & 23.8 & 21.8 & 19.5 & 15.7 & 14.2 & 13.7 & @xmath77 + pks 1830@xmath02111 & @xmath78 & @xmath79 & 1.54 & 2.507 & & & & 21.4 & 18.7 & & 15.0 & @xmath80 + fig .
1. flux - calibrated keck lris spectrum of j0738 + 2750 showing the very red continuum .
the 2800 and 1549 regions have narrow line absorption at the emission line redshift and the absorption is triple with a velocity dispersion of @xmath15 .
absorption corresponding to each of these is also present , as is a broad absorption feature extending from @xmath81 to @xmath82 in the quasar restframe .
there is also an intervening absorber contributing narrow and lines at @xmath83 ( in italics ) .
the polarization data , also from keck lris , show an increase from 4% to more than 9% between and , supporting the interpretation that the unusual spectral energy distribution is due to reddening by dust .
the polarization error bars show @xmath84 errors ; wavelength error bars indicate the binning used to measure the polarization signal . | in a survey of quasar candidates selected by matching the first and 2mass catalogs , we have found two extraordinarily red quasars .
first j013435.7@xmath0093102 is a 1 jy source at @xmath1 and has @xmath2 , while first j073820.1 + 275045 is a 2.5 mjy source at z=1.985 with @xmath3 .
first j073820.1 + 275045 has strong absorption lines of and in the rest frame of the quasar and is highly polarized in the rest frame ultraviolet , strongly favoring the interpretation that its red spectral energy distribution is caused by dust reddening local to the quasar .
first j073820.1 + 275045 is thus one of the few low radio - luminosity , highly dust - reddened quasars known .
the available observational evidence for first j013435.7@xmath0093102 leads us to conclude that it too is reddened by dust .
we show that first j013435.7@xmath0093102 is gravitationally lensed , increasing the number of known lensed , extremely dust - reddened quasars to at least three , including mg0414@xmath00534 and pks1830@xmath0211 .
we discuss the implications of whether these objects are reddened by dust in the host or lensing galaxies .
if reddened by their local environment , then we estimate that between 10 and 20% of the radio - loud quasar population is reddened by dust in the host galaxy .
the discovery of first j073820.1 + 275045 and objects now emerging from x - ray surveys suggests the existence of an analogous radio - quiet red quasar population .
such objects will be entirely missed by standard radio or optical quasar surveys . if dust in the lensing galaxies is primarily responsible for the extreme redness of the lensed quasars , then an untold number of gravitationally lensed quasars are being overlooked . |
zwicky showed that there is an inconsistency between the dynamical mass and the luminous mass for the coma cluster .
he postulated that there must exist dark matter at the scale of clusters of galaxies @xcite .
the observations of the rotation curves of spiral galaxies by rubin also indicated that galaxies need missing mass in the context of newtonian gravity @xcite .
the main paradigm to interpret missing mass in the universe is the theory of dark matter , supposing that the unseen matter accounts for more than @xmath7 of the universe s matter content .
this hypothetical matter is made of particles ( e.g.,wimps ) that may only interact weakly with ordinary baryonic matter .
ongoing experiments have been conducted to directly detect dark matter particles by sensitive particle physics detectors .
so far , no positive signal has been detected @xcite .
an alternative approach for solving the problem of missing mass is to modify the law of gravity .
one of the first phenomenological efforts to modify newtonian gravity was modified newtonian dynamics ( mond ) @xcite .
this model is based on a non - linear poisson equation for the gravitational potential with a characteristic acceleration @xmath8 ; newtonian gravity is modified for particles with acceleration @xmath9 .
the model can successfully describe the rotation curves of galaxies @xcite , but it fails to be consistent with the temperature and gas density profiles of clusters of galaxies without invoking dark matter @xcite .
there are other modified gravity models such as the relativistic extension of mond by bekenstein ( 2004 ) , conformal gravity @xcite and non - local gravity @xcite .
we attempt in the following to test observationally a covariant modified gravity ( mog ) theory , also called scalar
tensor vector gravity ( stvg ) at the scale of clusters of galaxies @xcite . in this theory , the dynamics of a test particle are given by a modified equation of motion where , in addition to the conventional geodesic motion of a particle , a massive vector field couples to the fifth force charge of a particle .
the fifth force charge of a particle is proportional to the inertial mass of the particle .
we have shown that in the weak field approximation the potential for a matter distribution of an extended object behaves like the newtonian potential with an enhanced gravitational constant and an additional yukawa potential @xcite .
we applied the effective potential in the weak field approximation to the study of the dynamics of galaxies and demonstrated that mog can successfully explain the rotation curves of galaxies without the need for dark matter .
we fixed the two free parameters of the potential and chose the stellar mass - to - light ratio as the only free parameter in calculating the dynamics of the galaxies .
we extend in the following the observational tests in moffat & rahvar ( 2013 ) to one order of magnitude larger system scale .
we use the observational data of relaxed clusters of galaxies in the chandra x - ray sample for this purpose . for a set of clusters of galaxies
, we have the temperature profile as well as the plasma gas and galaxy distribution .
these observables are sufficient to calculate the dynamical mass and compare it to the baryonic mass of clusters . in section ( [ formalism ] ) , we briefly introduce the stvg theory and review the weak field approximation limit . in section ( [ sectionboltzmann ] ) , we obtain a full relativistic version of the boltzmann equation in mog .
furthermore , we derive the virial equation in the weak field approximation . in section ( [ data ] ) , the observational data from the chandra x - ray sample of clusters are examined in conjunction with the mog and finally in ( [ conc ] ) , we provide a summary and conclusions .
the stvg theory is a covariant theory of gravity .
the action of stvg is composed of scalar , tensor and vector fields @xcite : @xmath10 the components of the action are ( i ) the einstein gravity action : @xmath11 where @xmath12 is the cosmological constant , ( ii ) the massive vector field @xmath13 action : @xmath14\sqrt{-g}~d^4x,\end{aligned}\ ] ] where @xmath15 is the vector field and @xmath16 , ( iii ) the action for the scalar @xmath17 and @xmath1 fields is : @xmath18\sqrt{-g}~d^4x , \label{scalar}\end{aligned}\ ] ] where @xmath19 denotes the covariant derivative with respect to the metric @xmath20 .
@xmath21 denotes a dimensionless coupling constant and @xmath22 and @xmath23 denote self - interaction potentials of @xmath1 and @xmath17 fields . while in the generic formalism of mog , @xmath17 and @xmath1
are taken into account as scalar fields , for simplicity we keep these two fields as constant parameters . finally , ( iv ) @xmath24 is the matter action .
the action for pressureless dust coupled to the vector field @xmath13 is given by @xmath25 where the density of the fifth force charge @xmath26 is related to the density of matter by @xmath27 where @xmath28 is a constant . for a test particle with the density @xmath29 and a fifth force charge @xmath30 , a variation of the action in eq .
( [ sm ] ) with respect to the comoving time @xmath31 , yields the equation of motion : @xmath32 where the right - hand side is a lorentz - type force .
since the fifth force charge is proportional to the inertial mass , the equation of motion is independent of the mass .
the main difference of the lorentz - type force with standard electromagnetism is the same sign for the charge of particles , unlike the normal electromagnetic force where there are opposite sign particles ( results in a neutral medium for an ensemble of particles in a plasma ) .
a physical consequence of the same sign charged particles is a repulsive force produced by an ensemble of particles . since the vector field in mog is massive , after a characteristic range of the field given by @xmath33 , the effect of the repulsive force vanishes and eq .
( [ geo ] ) reduces to the pure metric field geodesic equation of motion . for the astrophysical application of mog
, we need the weak field approximation of the theory .
we briefly review in this section the formalism of the weak field approximation .
a detailed derivation of this limit for extended distributions of matter was presented in moffat & rahvar ( 2013 ) . for a slow moving particle
, we expand the metric around minkowski space : @xmath34 the left - hand side of eq .
( [ geo ] ) can be written in terms of @xmath35 , while on the right - hand side we can ignore the magnetic - like force compared to the electric - like force due to the slow motion of the particle .
( [ geo ] ) can now be written as @xmath36 where @xmath37 and we define the effective potential as @xmath38 we substitute in the right - hand side of ( [ geo2 ] ) the solutions for @xmath35 and @xmath39 from the field equations . for the tensor component , @xmath40 where @xmath41 , and the zero component of the vector field @xmath39 satisfies the yukawa equation : @xmath42 where @xmath43 is the zero component of four dimensional current as the source of vector field and relates to density of matter as @xmath27 . substituting the solution of the poisson equation and the solution of @xmath39 from the yukawa equation into ( [ geo2 ] ) and taking into account the asymptotic solutions of
effective potential at smaller distances ( i.e. @xmath44 ) to the newtonian gravity @xcite , we obtain the following effective potential in the weak field approximation : @xmath45 where @xmath17 has been replaced by @xmath46 , @xmath0 is a dimensionless parameter , @xmath47 is the newtonian gravitational constant and @xmath1 is the mass of the vector field .
we have used this effective potential in moffat & rahvar ( 2013 ) and successfully fitted the rotation curves of galaxies , yielding the best values of the parameters : @xmath48 and @xmath49 kpc@xmath4 . the first term on the right - hand side of eq .
( [ potential22 ] ) is the potential of the newtonian attractive force with the effective gravitational constant @xmath50 and the second term results from the repulsive vector force . for scales much larger than @xmath51 ,
the yukawa term is suppressed and the gravitational potential is enhanced by factor of @xmath52 compared to the newtonian potential . on the other hand , for smaller scales
the repulsive force cancels the enhanced attractive force and the result is the standard newtonian potential at solar system scales .
in this section we derive the full relativistic version of the boltzmann equation in mog and for the astrophysical applications , we calculate the boltzmann equation in the weak field approximation .
the result will be used in studying the dynamics of x - ray clusters of galaxies . for an ensemble of collisionless particles moving due to their self - gravity ,
the number of particles inside a differential element in the phase - space does not change and @xmath53 .
this equation can be written as a sum of partial derivatives in terms of time , space and momentum . since in momentum space , we have the constraint @xmath54 , there are three degrees of freedom for @xmath55 and we keep only the spatial components of the momentum space in our calculation @xcite . the boltzmann equation can be written as @xmath56 we can write ( [ geo ] ) in terms of the momentum of a particle : @xmath57 by substituting ( [ 3 ] ) into ( [ 1 ] ) , we obtain the boltzmann equation in curved space with the extra vector field as follows : @xmath58f = 0.\ ] ] in order to have a relativistic euler equation , we multiply the boltzmann equation by the spatial component of four - momentum , @xmath59 , and integrate over momentum space : @xmath60 where @xmath61 is the differential volume in momentum space . in the first term , the space - time partial derivative can be moved outside the integral , and in the second term the christoffel symbol and the vector field strength can also be moved outside the integral .
integrating over momentum space gives the result : @xmath62 where the bar sign denotes the average value over momentum space .
for simplicity we rewrite the relativistic euler equation in terms of the four velocity @xmath63 : @xmath64 now we multiply this equation by @xmath65 and integrate over the 3-space volume and calculate this expression term by term . using the definition of the inertia tensor : @xmath66 the first term of ( [ b11 ] ) can be written in terms of partial time derivatives of @xmath67 and for the static and stationary systems it is zero .
the second term represents the energy- momentum term which can be written as @xcite : @xmath68 for simplicity , we can split the kinetic tensor into the energy - momentum tensor from the bulk velocity and the internal energy from the dispersion velocity , yielding the result : @xmath69 where @xmath70 and where @xmath71 is the dispersion tensor with @xmath72 .
the integrations of the third , fourth and fifth terms in eq .
( [ b11 ] ) contain kinetic energy contributions .
the newtonian gravitational part of the integrations can be interpreted as the gravitational potential of the system .
we assign the last three terms of eq .
( [ b11 ] ) to the potential energy tensor : @xmath73 combining all the integration terms , the conservation of the energy - momentum fluid can be written as @xmath74 for astrophysical systems such as globular clusters , elliptical galaxies and clusters of galaxies , the morphology of the systems do not change in time , in other words , these systems have been virialized and do not interact with other systems .
we set the left - hand side of ( [ v1 ] ) to zero for this case . by choosing the center - of - mass of the system as the center of the coordinate system , the global velocity cancels and @xmath75 , whereby the energy - momentum equation simplifies to the so - called virial equation : @xmath76 in the next section
, we will use the boltzmann equation in mog for studying clusters of galaxies .
while here we derived a generic euler equation and virial condition in mog , we will use just the equilibrium condition for studying the clusters . in this section
, we study the boltzmann equation for spherically symmetric systems in the weak field approximation . in order to calculate the euler equation for a spherically symmetric system , we start with eq .
( [ b11 ] ) .
let us choose @xmath77 and assume that @xmath78 is a function of radial distance @xmath79 , radial particle velocity @xmath80 and tangential particle velocity , @xmath81 .
for a virialized system , the shape of the system does not change in time and terms with partial derivatives with respect to time vanish .
then the jeans equation simplifies to @xmath82 by assuming a spherically symmetric metric for this structure , @xmath83 we can choose @xmath84 and assuming a perturbation @xmath85 and the definition @xmath86 , the jeans equation in the weak field approximation can be written as @xmath87 = -\rho\partial_r\phi_{eff}(r ) .
\label{jeans2}\ ] ] for an isotropic distribution @xmath88 due to rotational symmetry there is no preferred transverse direction i.e. , @xmath89 .
we can still have a degree of anisotropy in the velocity distribution at each point given by @xmath90 , and then the jeans equation can be written as @xmath91= g(r ) , \label{vir}\ ] ] where the acceleration at the right hand side , using @xmath92 , is given by @xmath93d^3{\bf r'},\ ] ] where for the scales @xmath94 , the exponential term approaches to one and by canceling the last two terms in the integrand , the gravitational acceleration reduces to newtonian gravity . on the other hand , for the case that @xmath95 , the exponential term approaches to zero and we get newtonian gravity with an enhanced gravitational constant @xmath96 . in what follows
, we can set @xmath97 in eq .
( [ vir ] ) for the isotropic distribution of the velocity . [ cols="^,^,^,^,^,^,^,^,^,^,^ " , ] here , for a spherical symmetric structure with scales of the order of @xmath98 mpc , we can simplify our numerical calculation , noting that the length corresponding to the mass of the vector field , @xmath99 kpc , is much smaller than the size of a cluster .
for these structures , the gradient of the density inside the cluster within the length scale of @xmath99 kpc is negligible and we can ignore the gravitational acceleration exerted on a test particle by the fluid surrounding it within this radius .
hence , the second term of eq .
( [ potential22 ] ) for a cluster - size structure is negligible .
we have numerically checked this simplification by comparing the overall effective gravitational acceleration with the simplified relation of keeping only the first term of equation ( [ potential22 ] ) .
figure ( [ acc ] ) compares the relative acceleration on a test mass particle with and without considering the yukawa term .
since the measurements of central parts of clusters in the chandra observations are not accurate enough and the data is scattered @xcite , using the simplified formula for the acceleration at the outer radii of the clusters ( i.e. @xmath100 kpc ) and ignoring the yukawa term is considered to be valid .
using gauss s theorem for a spherically symmetric system with a @xmath101 potential , the dynamical mass from eq .
( [ halo ] ) can be written as @xmath102 where @xmath103 is given in terms of solar mass , @xmath79 is the distance from the center of the cluster in kpc , @xmath104 is the temperature of the gas in kev .
we use the value @xmath2 which has been universally fixed by the fits to the rotation curve data of galaxies @xcite . in figure ( [ fig2 ] )
, we compare the profile of dynamical mass @xmath105 as a function of distance from the center of the cluster with the baryonic mass of clusters composed of the mass of gas plus the mass contribution from stars , @xmath106 .
the overall dynamical and baryonic masses at the distance @xmath107 are given in table ( [ tab3 ] ) .
the error bar associated to the mass of baryonic component is adapted from vikhlinin et al ( 2006 ) where they reported the fraction of baryonic mass to the overall mass in the dark matter scenario .
assuming the error bar corresponds to this fractional mass as a constant value throughout the cluster , we calculate the error bar of the baryonic mass being proportional to the mass .
we plot the error bars for the baryonic matter of clusters as a function of distance from the center of the cluster and with the cadence of @xmath108 mpc separation .
for some of clusters as cga262 and cga1991 there is an offset between the observation and theoretical mass profiles with about @xmath109 deviation . for the rest of the clusters
there is consistency between the theoretical and observational mass profiles , except for the central regions of clusters . we note that observational data at the central parts of clusters is poor .
moreover , this part might not completely virialized due to cooling flow . for most of the clusters at scales larger than @xmath110 kpc ,
the theoretical profiles are in good agreement with the observational data . from the mass profiles in figure ( [ fig2 ] ) ,
the outer regions of clusters at @xmath111 kpc contain almost @xmath112 of the mass of the clusters and for these regions the mog theory prediction follows the observational data .
for the overall mass of the clusters at @xmath107 , we compare the mass profiles for the data and theory .
the error bar associated with the observed mass is obtained from the uncertainty in the observational data , while for the theoretical profile , we obtain the upper and lower values for the mass resulting from the uncertainties in @xmath1 and @xmath0 , reported in @xcite .
figure ( [ fig3 ] ) compares the overall theoretical baryonic mass of clusters with the observational data .
following our previous paper @xcite , in this work we have investigated further the weak field approximation of stvg to test the dynamics of clusters of galaxies .
the theory of gravity in the weak field regime is composed of an enhanced newtonian gravity term with the gravitational constant @xmath113 and a yukawa repulsive term with the mass @xmath114 kpc . at the smaller scales compared to @xmath51 , e.g. , solar system scales , standard newtonian gravity can be recovered . on the other hand , at larger scales
we get newtonian gravity with a stronger effective gravitational constant .
our analysis of a sample of galaxies in the previous work showed that with the fixed universal parameters @xmath0 and @xmath1 in the effective gravitational potential @xcite , we can successfully describe the dynamics of a large set of galaxies with only their baryonic matter content . in order to test this universality ,
we have analyzed a sample of nearby chandra x - ray data in which the gas density profiles and the temperatures have been measured @xcite .
our analysis shows that , using the universal values for the parameters @xmath0 and @xmath1 in the effective mog potential , we get consistent mass profiles and overall values for the dynamical and baryonic mass of clusters , except for the central parts of clusters where the observational data is scattered . these results along with the analyses of the rotation curves of galaxies demonstrate that with a unique weak field mog effective potential , we can describe the dynamics of astrophysical systems from the solar system to @xmath115mpc scale systems without exotic dark matter . finally , we outline further questions in mog yet to be answered , ( i ) large scale structure formation in the universe .
the structure formation growth has to be described by mog without using dark a matter component and using cmb boundary conditions , ( ii ) interpretation of observational data of gravitational lensing at large scales , especially the case of the asymmetric bullet cluster without invoking dark matter for the lensing system .
a previous published investigation of the bullet cluster using a version of mog @xcite , showed compatibility between the predictions of mog and the available data at the time .
however , this work was based on a spherically symmetric point - like solution of the field equations , which has to be extended to continuous matter distributions using the methods of the present paper .
the john templeton foundation is thanked for its generous support of this research .
the research was also supported by the perimeter institute for theoretical physics .
the perimeter institute was supported by the government of canada through industry canada and by the province of ontario through the ministry of economic development and innovation .
we thank niayesh afshordi , bahram mashhoon , martin green and viktor toth for helpful discussions and comments .
finally , we would like to thank david weinberg `` as referee of this work '' for useful comments and suggestions for improving this work . | we apply the weak field approximation limit of the covariant scalar - tensor - vector gravity ( stvg ) theory , so - called modified gravity ( mog ) , to the dynamics of clusters of galaxies by using only baryonic matter .
the mog effective gravitational potential in the weak field approximation is composed of an attractive newtonian term and a repulsive yukawa term with two parameters @xmath0 and @xmath1 .
the numerical values of these parameters have been obtained by fitting the predicted rotation curves of galaxies to observational data , yielding the best fit result : @xmath2 and @xmath3 kpc@xmath4 @xcite .
we extend the observational test of this theory to clusters of galaxies , using data for the ionized gas and the temperature profile of nearby clusters obtained by the chandra x - ray telescope . using the mog virial theorem for clusters ,
we compare the mass profiles of clusters from observation and theory for eleven clusters .
the theoretical mass profiles for the inner parts of clusters exceed the observational data . however , the observational data for the inner parts of clusters ( i.e. , @xmath5 ) is scattered , but at distances larger than @xmath6 kpc , the observed and predicted mass profiles converge .
our results indicate that mog as a theory of modified gravity is compatible with the observational data from the the solar system to mega parsec scales without invoking dark matter . |
in refs .
@xcite , a unified model of dark matter has been presented , which posits a ds with a u(1 ) gauge symmetry , spontaneously broken in order to avoid long range interactions .
the ds is modelled by an ah model @xmath5 , where the dark scalar field , @xmath6 , couples to the gsw theory through a higgs portal coupling @xcite and the dark gauge field @xmath7 through a gauge kinetic mixing term @xcite .
we consider the following semilocal model coupled to a ds defined by the lagrangian . ] : @xmath8 where @xmath9 , @xmath10 , @xmath11 , @xmath12 , @xmath13 .
the potential , @xmath14 is a slight generalisation of that of the witten model @xcite , @xmath15 the parameters @xmath16 , @xmath17 , @xmath18 , @xmath19 are restricted by demanding that @xmath20 for @xmath21 , resulting in : @xmath22 , @xmath23 , and @xmath24 . for a description of the vacua of @xmath14 we refer to refs .
the parameters , @xmath18 and @xmath25 , correspond to the higgs portal and gauge kinetic mixing @xcite , respectively . the above model can be viewed as the @xmath26 limit of the gsw theory coupled to a ds , therefore we shall refer to the fields @xmath27 and @xmath28 as the `` visible sector '' ; and to @xmath6 and @xmath7 as the ds .
apart from the local u(1)@xmath3u(1 ) it has a global su(2 ) symmetry acting on the ( complex ) higgs doublet , @xmath27 , and we shall refer to to as `` semilocal - ds '' model . in the 2vev case , for @xmath29 ,
the gauge boson masses are given as @xmath30 and @xmath31 , where the vevs @xmath32 and @xmath33 expressed in terms of the parameters of the potential are listed in appendix a , eq . .
the scalar particles @xmath34 and @xmath6 mix , the analysis thereof is presented in appendix [ app : masses ] . the field @xmath35 remains massless ( in the gsw theory , it is the would - be goldstone boson corresponding to the longitudinal component of @xmath36 ) . for a detailed analysis of the effects of the gauge kinetic mixing we refer to refs . @xcite . unless the mass of the ds scalar @xmath6 is large ( @xmath37 tev ) compared to sm masses , @xmath38 @xcite . in the 2vev case ,
the dark sector higgs and gauge bosons do not directly make up dark matter @xcite . as a result , there are much less stringent experimental bounds on the model parameters , e.g. , if the mixing of the visible sector and the dark sector higgs particles is small enough , and the dark sector particles are heavy enough , the model is viable . by setting @xmath39 we obtain a semilocal model coupled through the higgs field to a dark scalar field ( portal model ) . assuming that there is an unbroken @xmath40 symmetry in the ds , the dark scalar can not take on a vev ( 1vev case ) .
the main interest of such portal models is their minimality in that the dark scalar field itself can be considered as a primary constituent of the dark matter . in the 1vev case ,
the gauge boson mass is @xmath41 , and the scalar masses are @xmath42 , @xmath43 . due to the global su(2 ) symmetry
the field @xmath35 stays massless .
experimental limits on the couplings can be found in refs .
we note that higgs decays into the dark sector pose rather strong constraints on the coupling @xmath18 and dark matter density on @xmath44 .
straight string solutions in a two - component extended abelian higgs model with both fields having a non - zero vev have been considered for two charged fields in refs .
@xcite and for one charged and one neutral in refs . @xcite . in the case of two electrically charged fields ,
unless the windings of the two scalar fields agree , the energy per unit length of such strings diverges logarithmically in eq .
( [ eq : pot ] ) . ] , and their flux is fractional . in the absence of the gauge kinetic mixing term ( @xmath29 ) the 2vev vortices of refs .
@xcite can be embedded in the model given by eq . , by setting @xmath45 .
for @xmath46 the angular component of the ds gauge field also becomes non - zero .
the ( straight ) string solutions we consider are translationally symmetric in the @xmath47 direction , and rotationally symmetric in @xmath48 plane , corresponding to the ansatz @xmath49 where @xmath50 are polar coordinates in the plane and the other field components ( @xmath35 , @xmath51 , @xmath52 , @xmath53 , @xmath54 ) vanish . using the field equations ( [ eq : radeq ] ) one obtains easily that the energy ( [ eq : edens ] ) is a monotonously increasing function of the dark charge @xmath55 [ see eq .
( [ eq : ergq2 ] ) ] .
the derivative w.r.t .
the gauge kinetic mixing is given by @xmath56 which vanishes at @xmath29 , since in that case the field equation for @xmath57 in eq .
( [ eq : radeq ] ) becomes homogeneous and a standard maximum principle argument implies @xmath58 . expanding the fields in a power series of @xmath25
[ see eq .
( [ eq : epsexpan ] ) ] , the energy of the vortex can be written as @xmath59 at @xmath60 , @xmath61 , @xmath62 , @xmath63 , and @xmath64 eq .
( [ eq : ergeps ] ) yields an excellent approximation up to @xmath65 .
moreover @xmath66 and the correction is @xmath67 .
a further approximation is to consider the @xmath68 limit [ see appendix [ app : radeq ] , esp .
( [ eq : c1 ] ) ] , in which case @xmath69 , simplifying the expression for @xmath70 : @xmath71 remarkably , @xmath70 in eq .
( [ eq : ergepsq ] ) is proportional to the magnetic energy of the unperturbed vortex . at @xmath60 , @xmath61 , @xmath62 , @xmath63 , and @xmath72 , eq .
( [ eq : ergepsq ] ) yields @xmath73 . for these parameter values , eq .
( [ eq : ergeps ] ) gives @xmath74 , which compares quite favourably .
next we summarise the main results of our stability analysis of string ( or vortex in the plane ) solutions corresponding to ansatz .
the perturbation equations around the straight string solutions are given in appendix [ app : linear ] . crucially the fluctuation equations for @xmath75 resp .
@xmath76 decouple from the other components ( and also from each other ) .
this decoupling is related to @xmath77 for the background solution and to the coupling structure of the ds .
the ( only ) known instabilities of the semilocal model ( without a ds ) have been found in the @xmath75 sector .
we argue that in the semilocal - ds model the only potential instabilities are expected to appear in the fluctuation equations for @xmath75 ( and for @xmath76 ) , at least for `` not too large '' values of @xmath18 , simplifying considerably the stability analysis . due to the translation symmetry in the @xmath78 resp .
@xmath47 variables the linearised equations for the corresponding vector - field components @xmath79 , @xmath80 , resp .
@xmath81 , @xmath82 , decouple from each other and from the other components . exploiting the symmetries of the background string solution
the linearised equations for the components @xmath83(@xmath84 , @xmath85 , @xmath86 , @xmath87 , @xmath88 , @xmath89 ) can be reduced to a coupled system of the form @xmath90 constituting a system of 8 second order radial ode s for a given value of the angular momentum @xmath91 . for more details of the small fluctuation equations we refer to appendix [ app : linear ] , and refs .
@xcite .
the coupled perturbation system is not expected to give rise to instabilities at least for not too large values of @xmath18 , @xmath25 .
when @xmath92 the string solution reduces to an abrikosov - nielsen - olesen ( ano ) one @xcite in the visible sector , embedded into the semilocal - ds model with @xmath77 and @xmath93 .
therefore eqs .
decouple into the perturbations of the ano vortex in the visible , and that of the vacuum in the dark sector . in the visible sector
the lowest eigenvalues are well known to be positive @xcite [ e.g. , for @xmath60 , the lowest bound state eigenvalue is @xmath94 , the lowest continuum state is at @xmath95 , while in the ds positivity is rather obvious as we are perturbing around a true vacuum state [ continuum above @xmath96 .
simple perturbation theoretic arguments show that for @xmath97 , @xmath98 the spectrum remains positive . therefore in this paper
we shall investigate only the decoupled fluctuation equations for @xmath75 , which can be written as : @xmath99 for @xmath100 , the potential in eq .
( [ eq : pertdeltaphi22vev ] ) has a negative valley close to the origin ( the core of the vortex ) , while for @xmath101 it is given as @xmath102 .
the existence of negative eigenvalues depends on the depth of the attractive valley .
the stabilizing effect of the scalar condensate comes from making this attractive potential valley shallower .
more quantitatively , for a given value of @xmath103 by increasing @xmath19 ( remember @xmath104 ) the negative eigenvalue approaches zero and for some value @xmath105 , it actually reaches zero . for @xmath106 ,
dc vortices are then stable . quite importantly , a large value of the coupling @xmath19 is also compatible with the experimental bounds on the model , which is quite promising for electroweak - dark strings .
numerical data are presented in table [ tab : stab2 ] .
an unstable vortex and the potential in its perturbation equation , eq .
( [ eq : pertdeltaphi22vev ] ) , is shown in fig .
[ fig:2vev ] , and a stable one in fig .
[ fig:2vevs ] .
as the parameters are tuned , the valley in the potential around the origin becomes shallow , and the bound mode disappears .
importantly , both the higgs portal coupling and the gauge kinetic mixing act to stabilise semilocal vortices . in table [
tab : stab2 ] , some numerical data of dc vortices are given for @xmath107 , i.e. , at the value of @xmath19 when the change of stability sets in .
note that to larger values of @xmath25 correspond lower values of @xmath108 ( i.e. , a larger domain of stability ) .
one may note that the values of @xmath108 decrease of the order of @xmath109 while @xmath25 increases from @xmath110 to @xmath111 .
therefore it may appear surprising that the change in the energy is rather small and positive although @xmath112 while @xmath25 changes considerably more than @xmath19 .
this effect can be accounted for by observing that the energy is rather more sensitive to a change in @xmath19 than to one in @xmath25 , e.g. at @xmath60 , @xmath113 , @xmath62 , @xmath114 , @xmath115 and @xmath116 .
the relative smallness of @xmath117 as compared to @xmath118 can be understood from eq . .
since @xmath119 is exponentially suppressed for large values of @xmath120 the main contribution to the integral is expected to come from the region of @xmath121 , however @xmath122 for @xmath123 , accounting for the relative smallness of @xmath70 . on the other hand , @xmath124
[ see eq .
( [ eq : ergalpha ] ) in appendix [ app : radeq ] ] .
.stabilisation of 2vev vortices : the value of @xmath19 where the vortex becomes stable , and additionally the energy of the vortex at that value of @xmath19 is displayed .
the hidden sector charge is @xmath64 . [
cols="^,^,^,^,^,^,^,^,^",options="header " , ]
we have investigated the effect of a dark scalar field with higgs portal coupling and a u(1 ) gauge field with a gauge - kinetic mixing term on semilocal strings with a local u(1 ) and global su(2 ) symmetry in the visible sector .
the strings considered in this paper have unit winding number with respect to the visible u(1 ) and zero winding number with respect to the dark u(1 ) .
we have found that in a minimal higgs portal model ( with a single dark scalar field ) , semilocal strings get stabilized by a dark scalar condensate at the core of the string . considering also a dark u(1 ) gauge field with a gauge - kinetic mixing term an additional stabilising effect is found .
these observations open up the possibility of the existence of classically stable dark core electroweak strings .
this work has been supported by the grant otka k101709 .
to obtain scalar masses in the 2vev case , we linearise the potential ( [ eq : pot ] ) about the vacuum @xmath125 , @xmath45 and @xmath126 , with @xmath127 we also introduce new variables , @xmath128 , @xmath129 .
the would - be goldstone bosons , that are later gauged into the longitudinal components of the gauge fields are then @xmath130 and @xmath131 .
the propagating scalar particles are mixed out of @xmath132 and @xmath133 , where their mixing matrix is @xmath134 the higgs particle @xmath135 and the dark higgs particle @xmath136 are related to these as @xcite @xmath137 where @xmath138 the resulting scalar masses are @xmath139 the second semilocal component , @xmath35 remains massless .
inserting the ansatz ( [ eq:2vevans ] ) into the field equations corresponding to the lagrangian ( [ eq : lagphi ] ) yields the radial equations @xmath140f\ , , \\
\frac{1}{r}(rg ' ) ' & = \left [ \frac{q^2c^2}{r^2 } + \beta_2g^2 -\alpha + \beta ' f^2\right]g\ , .
\\ \end{aligned}\ ] ] the boundary conditions at the origin are demanded by regularity of the fields in the plane , for @xmath141 , @xmath142 , @xmath143 , @xmath144 and @xmath145 .
for @xmath146 we impose @xmath147 , @xmath148 , @xmath149 and @xmath150 in the 2vev case , and @xmath151 and @xmath152 in the 1vev one .
the energy density of a field configuration in the ansatz ( [ eq:2vevans ] ) is @xmath153 + ( f')^2 + ( g')^2 + \frac{n^2 ( 1-a)^2}{r^2}f^2+\frac{q^2 c^2}{r^2}g^2 + v(f , g)\,,\ ] ] where @xmath154 in eq .
( [ eq : potrad ] ) , @xmath155 is the term subtracted in the 2vev case to set the potential to zero at its minimum . in the 1vev case ,
no such term is necessary . as the fields satisfy the euler - lagrange equations , in the derivatives of the energy w.r.t .
the parameters of the model , terms proportional to the implicit derivatives of the fields vanish , and only explicit terms remain , e.g. , @xmath156 we spell out explicitly the derivative used in sec .
[ sec:2vev ] in the 2vev case , @xmath157 where the second term is due to the subtraction of @xmath155 in eq .
( [ eq : potrad ] ) . in the 1vev case , the derivative is the same as in eq .
( [ eq : ergalpha ] ) without the subtraction of @xmath158 .
a series expansion of the solutions in @xmath25 is as follows , @xmath159 the resulting equations of motion are obtained for @xmath160 by setting @xmath29 and @xmath161 in the radial equations ( [ eq : radeq ] ) .
the leading order correction , @xmath162 satisfies @xmath163 which can be approximated in the limit @xmath68 : in this limit , the right hand side of eq .
( [ eq : c1 ] ) becomes the same as that of the equation of @xmath164 [ see eq .
( [ eq : radeq ] ) ] , and therefore @xmath69 .
although the @xmath68 limit is not uniform in @xmath120 , the dominant contribution in the energy correction eq .
( [ eq : ergepsq ] ) is expected to come from the core , which is numerically verified .
for the linearised field equations we use the formalism of ref .
@xcite ( see also refs . @xcite for applications to multi - component vortices ) . for the 2vev case . in the linearised field equations we introduce a set of new variables for the gauge fields , @xmath165 which diagonalise the gauge kinetic terms at the cost of introducing couplings between both gauge fields and both scalars , @xmath166 the linearised equations assume a particularly simple form in the background field gauge @xcite , @xmath167 the components @xmath168 and @xmath169 decouple from the rest of the variables due to the @xmath170 independence of the background , satisfying @xmath171 where @xmath172 infinitesimal gauge transformations act on the fields as @xmath173 due to the residual gauge freedom allowed by the gauge fixing ( [ eq : gauge ] ) , there are ghost modes , satisfying the equations @xmath174 which agree with those of the @xmath175 gauge field components , eq .
( [ eq : k03l03 ] ) , and cancel part of the spectrum , including all modes in the @xmath168-@xmath169 sector , therefore , in what follows , we omit these components .
the following ansatz is compatible with the field equations , due to the cylindrical symmetry of the background , @xmath176 where @xmath177 and similarly for @xmath178 ( note that @xmath179 ) . with the ansatz ( [ eq : pertansatz ] ) , the perturbation equation assume the form @xmath180 where @xmath181 .
note that the lowest eigenvalue corresponds to @xmath182 , therefore in what follows , we shall only consider such perturbations .
we write the @xmath183 matrix operator @xmath184 in eq .
( [ eq : perteqapp ] ) as ( suppressing all zero entries ) @xmath185 in eq .
( [ eq : opmell ] ) , the following notation is used @xmath186 with @xmath187 and @xmath188 for @xmath29 , the ghost mode equations ( [ eq : ghost ] ) decouple .
the visible sector one has been solved numerically ; it has positive eigenvalues , which change slowly with the parameters .
the ds one has a positive potential .
therefore , no modes corresponding to instabilities are cancelled by ghosts .
lukcs , in : a. zichichi ( ed ) : _ what is known and unexpected at the lhc _ ,
proceedings of the international school of subnuclear physics , erice - sicily , italy , 29 august 7 september 2010 , world scientific , 2013 . | semilocal and electroweak strings are well - known to be unstable against unwinding by the condensation of the second higgs component in their cores .
a large class of current models of dark matter contains dark scalar fields coupled to the higgs sector of the standard model ( higgs portal ) and/or dark u(1 ) gauge fields .
it is shown , that higgs - portal - type couplings and a gauge kinetic mixing term of the dark u(1 ) gauge field have a significant stabilising effect on semilocal strings in the `` visible '' sector .
cosmic strings and their observational signatures have been studied since a long time as they are expected to form in the early universe @xcite .
even if by now it seems unlikely that cosmic strings could have significantly contributed to structure formation in the universe , string - like excitations in the standard model ( sm ) continue to be of great interest not only from a theoretical point of view , but such objects may eventually leave observable signatures , e.g. , in the large hadron collider @xcite .
remarkable string solutions have been uncovered in the bosonic sector of the glashow - salam - weinberg ( gsw ) theory ( in this paper we shall refer to a generalisation of the electroweak sector of the sm allowing its parameters to take on non - physical values as the gsw theory . ) , for a review see ref .
@xcite .
a rather interesting class of models emerges by taking the @xmath0 limit of the gsw theory , where @xmath1 denotes the electroweak mixing angle .
one obtains this way an abelian higgs model with an extended scalar sector having an @xmath2 symmetry acting on the higgs doublet , this a a prototype of semilocal models .
its strings solutions are referred to as semilocal strings @xcite and these are quite instructive to study as being potentially important object in the gsw theory .
an important criterion for the physical relevance of string - type objects is their classical stability .
semilocal strings turned out to be stable only when the mass of the scalar particle is smaller than that of the ( single ) gauge boson , as shown in refs .
@xcite .
the stability of electroweak strings ( whose progenitors are the semilocal ones ) has been considered in refs .
@xcite ; it was found that for physically realistic values of @xmath1 , electroweak strings are unstable .
moreover there are good reasons to consider extended versions of the gsw theory by a dark sector ( ds ) , motivated by the mystery of dark matter . in such extended models
the question of the possible rle of strings appears naturally .
a minimalistic extension of the gsw theory is to couple a ( dark ) scalar field to the by now firmly established higgs sector of the gsw theory ( higgs portal ) @xcite , but there are also well motivated extensions of the gsw theory containing u(1 ) gauge fields in the ds @xcite . in refs .
@xcite physical properties and possible observational signatures of cosmic strings in the ds ( dark strings ) have been considered .
a more detailed investigation of string solutions in abelian higgs theories modelling a `` visible '' and a `` dark '' u(1 ) gauge sector was presented in ref .
@xcite . in subsequent works
@xcite semilocal - type models with a `` visible '' and a `` dark '' u(1 ) gauge field spontaneously broken in both sectors have been investigated .
it has been observed that the stability region of semilocal string solutions with a non - zero winding number in the ds can be extended in function of the couplings between the visible and the ds .
higher winding vortices in the u(1)@xmath3u(1 ) model and its supersymmetric generalisation have been considered in refs .
@xcite . an earlier work on string solutions in a portal - type theory is ref .
@xcite . in all these works only strings with non - zero winding in the ds have been considered , because of the known instabilities of `` visible '' semilocal strings .
the main goal of the present paper is to complement these studies on dark strings by concentrating on the influence of the ds on `` visible '' semilocal - type string solutions ( i.e. with zero winding in the ds ) .
we consider a u(1)@xmath3u(1 ) abelian higgs ( ah ) model , whose scalar sector consist of a complex higgs doublet with ( global ) su(2 ) symmetry coupled to a dark scalar field with an u(1)@xmath3u(1 ) symmetric potential , which is a simple generalisation of the model of witten @xcite .
we use this simplified model to study the effect of the ds on semilocal strings .
it is convenient to distinguish between two symmetry breaking patterns ; either both the visible higgs and the dark scalar field have non - zero vacuum expectation values ( 2vev ) , or there is no symmetry breaking in the ds ( 1vev ) .
the case with 1vev is directly relevant to the higgs portal ( scalar phantom ) model of refs .
@xcite , whereas when the ds contains gauge fields to model interaction among the dark matter particles the symmetry must be broken in both the visible and the dark sector ( 2vev case ) @xcite .
generically , semilocal strings are unstable with respect to condensation of the dark scalar field at their core [ we shall refer to such strings as dark core , ( dc ) ones ] . in the absence of the gauge kinetic mixing
, the dc strings investigated in the present paper , correspond to embeddings of the solutions previously found in refs .
@xcite resp .
@xcite into the su(2)@xmath3u(1 ) symmetric semilocal model coupled to a ds . when the gauge kinetic mixing is different from zero the string solutions we consider here differ from those of ref .
@xcite in that our strings have nontrivial winding only in the visible sector .
our main result is the stability of dc strings with respect to small perturbations for a rather large parameter domain .
it has to be pointed out that a number of mechanisms to stabilise semilocal strings have already been investigated . in ref .
@xcite , a stabilising effect due to a bound state of an additional scalar field on semilocal and electroweak strings has been found . in ref .
@xcite , it has been shown that a special ( dilatonic - type ) coupling between the gauge and the scalar field also has a stabilising effect on semilocal strings . in the complementary limit of the electroweak theory , @xmath4
, it has been demonstrated that quantum fluctuations of a heavy fermion doublet coupled to the string can also lead to stabilisation in refs .
@xcite .
stabilisation of electroweak strings due to the interaction with thermal photons has been demonstrated in ref .
@xcite .
the plan of the paper is as follows : in sec .
[ sec : model ] we introduce the models considered , which is followed by the discussion of visible straight string solutions in the 2vev case and their stability properties in sec .
[ sec:2vev ] .
next we analyse the 1vev case in sec .
[ sec:1vev ] .
we conclude in sec .
[ sec : conclusions ] .
some details have been relegated to various appendices : scalar masses in the 2vev case to appendix [ app : masses ] , radial equations of vortices to appendix [ app : radeq ] and the linearisation of the field equations about the vortices to appendix [ app : linear ] . |
the aim of combinatorial optimization is to find good approximations to the solution(s ) of minimization problems .
many of the most famous algorithms currently used in this field @xcite were inspired by analogies with physical systems . among them
the most celebrated is _ thermal simulated annealing _ @xcite proposed in 1983 by kirkpatrick _
et al . _ : the space of all admissible solutions is endowed with a potential profile dependent on the cost function associated to the optimization problem .
the exploration of this space is represented by a temperature dependent random walk .
an opportunely scheduled temperature lowering ( annealing ) stabilizes then the walk around a , hopefully global , minimum of the potential profile .
+ the _ quantum annealing _ approach to combinatorial optimization @xcite , instead , was originally suggested by the behaviour the stochastic process @xmath0 associated @xcite to the ground state @xmath1 of a hamiltonian of the form : @xmath2 where the potential function @xmath3 encodes the cost function to be minimized . the behavior of @xmath4 is characterized by long sojourns around the stable configurations , i.e. minima of @xmath5 , interrupted by rare large fluctuations which carry @xmath4 from one minimum to another : @xmath4 in thus allowed to `` tunnel '' away from local minima to the global minimum of @xmath5 .
the deep analysis of the semiclassical limit performed in @xcite shows , indeed , that as @xmath6 _ `` the process will behave much like a markov chain whose state space is discrete and given by the stable configurations''_. + however , the ground state of @xmath7 is seldom exactly known and approximations are required .
one of the earliest proposals in this direction , advanced in @xcite and applied in @xcite , was to construct an unnormalized approximation of @xmath8 by acting on a suitably chosen initial condition @xmath9 with the hamiltonian semigroup @xmath10 , namely by solving , with the initial condition @xmath11 , the _ imaginary time _
schrdinger equation .
similar ideas appear in the chemical physics literature @xcite and , with more specific reference to the optimization problems considered here , in @xcite .
yet , the inability to autonomously construct the ground state process , without recourse to the unphysical step of imaginary time evolution , substantially detracts from what is otherwise a physical route to optimization by dynamical evolution toward the ground state .
+ in this note , encouraged by the progress in quantum annealing in the last twenty years , as reviewed for instance in @xcite , by its close relationship with adiabatic quantum computation @xcite and by proposals of its hardware implementation @xcite , we try to eliminate the above unphysical step : we try to implement , instead of imaginary time evolution , the idea of reaching the ground state with the help of viscous friction @xcite .
+ we first introduce the nonlinear schrdinger - langevin - kostin ( slk ) equation and illustrate , by means of examples on two toy models , how a frictional force acts in the continuous case . then we turn to quantum combinatorial optimization and show how dissipation can , in the discrete case , balance genuinely quantum effects , such as bloch oscillations and anderson localization , which can hinder the search of optimal solutions .
the slk equation is the analogue of the heisenberg - langevin equation and represents a quantum analogue of classical motion with frictional force proportional to velocity @xcite ; it can be seen as a rough analogue of the classical drude - lorentz model of ohmic friction , i.e. an approximate description of the motion of a quantum particle through matter with inelastic scattering . + a solution @xmath12 of the equation : @xmath13 satisfies the inequality for , @xmath7 being the hamiltonian ( [ eq : schr1 ] ) .
what we will show below is how the _ norm preserving _ dissipative evolution described by ( [ eq : slk ] ) can dynamically drive a suitable initial condition @xmath14 toward the ground state @xmath8 of @xmath7 . + * toy model 1 : * require @xmath15 ( the parameters @xmath16 being chosen so that @xmath17 ) to be the ground state of @xmath7 and to belong to the eigenvalue @xmath18 .
the above two requirements determine the potential @xmath19 .
+ figure [ fig : tela2008g2 ] follows the evolution @xmath20 of the initial condition @xmath21 under ( [ eq : slk ] ) for a time interval @xmath22 .
it shows that , as @xmath23 decreases with time , the `` vacuum overlap '' @xmath24 approaches the value 1 . while the state @xmath25 approaches the ground state , some of the probability mass `` tunnels '' from the leftmost ( local ) to the rightmost ( global ) minimum .
+ we point out that this class of examples , where both the ground state wave function @xmath8 and the ground state energy @xmath26 are known , allows also for the calibration of the numerical method . in our case we have used the built - in ` ndsolve ` resource of _ wolfram mathematica 6_. + * toy model 2 * : for the sake of comparison with classical literature on quantum annealing @xcite , we consider here a double - well potential of the form @xmath27 as shown in figure [ fig : tela2008g3a ] , for the parameters used there , the local minimum of the potential @xmath5 is wider than the global one .
we refer the reader to section 2.2 of @xcite and to @xcite for a discussion of the meaning of the parameters and for the presentation of numerical experiments comparable with ours . here
, we use this well known toy model as an example in which the ground state is _ unknown _ and the dissipative dynamics of slk type provides a method to find it . + while the initial condition evolves ( figure [ fig : tela2008g3a ] ) , the mean value of @xmath7 decreases as in figure [ fig : tela2008g3cd ] . that @xmath28 is a good approximation of the ground state
is shown by comparing , in the inset of figure [ fig : tela2008g3cd ] , @xmath5 with the potential @xmath29 evaluated at @xmath30 , of which @xmath28 is the ground state belonging to the eigenvalue 0 .
comparison of the two curves in the inset of figure [ fig : tela2008g3cd ] is , furthermore , suggestive of a real - time version of piela s _ method of deformation of the potential energy hypersurface _ @xcite .
+ in the continuous case , then , the slk hamiltonian enriches quantum annealing of what can be seen as a _ probability percolation _
: while the state of the system converges toward the ground state , some probability mass tunnels from one minimum to another and , instead of tunneling back , as would happen in a reversible dynamics , remains there .
this autonomous stabilization would represent an alternative to _
adiabatic quantum computation _ @xcite .
most of the effort in ref .
@xcite went into making the intuition developed so far available in a context of _ combinatorial _ optimization .
in such a context the domain of the function @xmath3 to be minimized is a finite set @xmath31 and the search of the minimum of @xmath3 is modeled on a graph @xmath32 , where the edges @xmath33 describe the moves allowed in the search .
for instance , in @xcite , @xmath31 was taken to be the boolean hypercube @xmath34 , for some positive integer @xmath35 , and an edge was placed between any two points in @xmath36 separated by a unit hamming distance .
+ in this note , we consider the much simpler instance in which @xmath31 is , for some positive integer @xmath37 , the finite set @xmath38 equipped with the set of edges according to the general approach outlined in @xcite , this amounts to a search of the minimum of the function @xmath3 defined on @xmath39 by means of an interacting continuous time quantum walk @xcite on @xmath39 governed by a hamiltonian of the form @xmath40 we wish to show here that the _ quantum _ search outlined above can suffer from two , typically _ quantum _ , problems , namely bloch oscillations @xcite and anderson localization @xcite and that a certain amount of `` viscous '' friction can provide some relief to both these problems .
+ on a finite box @xmath39 , we consider the evolution of an initial condition of the form @xmath41 .
+ + we refer the reader to section 5 of @xcite for a motivation of this choice : suffice here to say that it describes a spatially well located wave packet that in _ the absence of any potential _ moves back and forth , with speed close to 1 , inside the box @xmath39 , as in figure [ fig : tela2008g4a ] .
the effect on this _ ballistic _ evolution of a linear potential @xmath42 is shown in figure [ fig : tela2008g4b ] . for @xmath43 ,
the peculiar energy - momentum relation @xmath44 , holding on a discrete lattice , determines bloch oscillations that prevent the wave packet from approaching the point @xmath45 at which the minimum of the cost function @xmath5 is located .
figure [ fig : tela2008g4b ] is therefore a reminder of the fact that a _ greedy _ quantum optimization driven by the cost function itself acting as a potential can be hindered by the fact that , on a lattice , increasing momentum @xmath46 can mean decreasing velocity @xmath47 . + we propose here to introduce a certain amount of viscous friction in the discrete schrdinger equation , as a `` kostin potential '' @xmath48 : @xmath49 the idea is that friction can prevent the momentum @xmath46 from crossing the first brillouin zone and thus can prevent the velocity @xmath50 from being inverted before the wave packet reaches the boundary of @xmath39 . this unwanted inversion is illustrated in figure [ fig : tela2008g4b ] , the effect on it of a suitable kostin potential is shown in figure [ fig : tela2008g4c ] . + as it is easy to check that , for @xmath25 evolving @xmath51 the actual form of @xmath52 that we adopt in order to achieve decrease of @xmath53 is , with @xmath54 .
+ + + figure [ fig : tela2008g5a ] shows , instead , the effect , in the form of anderson localization , of a random gaussian potential of mean 0 and variance @xmath55 , acting independently on each site of @xmath39 .
the order of magnitude @xmath56 of the noise parameter @xmath57 is suggested by a scaling + whereas the fact that friction can wipe out bloch oscillations is well known @xcite , the less well known fact that we show here is that the pseudo - ballistic motion shown in figure [ fig : tela2008g4c ] is _ much more stable _ than the truly inertial motion represented in figure [ fig : tela2008g4a ] with respect to the onset of anderson localization .
as a final remark , we observe that the same framework developed in this note for the combinatorial optimization metaphor can be used , with minor changes , to describe an excitation travelling along a spin chain or a light pulse propagating through a waveguide lattice @xcite .
we conjecture , therefore , that slk dynamics can be exploited also in those fields .
for example , we can , maybe , increase the fidelity of state transmission , in presence of imperfections , along a spin chain , by applying a `` tension '' at both ends of it @xcite ( see figure [ fig : tela2008g5d ] ) .
the sole convergence toward the ground state could , instead , find applications in all - optical switching of light in waveguide arrays @xcite : the injected light pulse can be steered toward a given position by a suitable tuning of the thermal gradient which determines the potential profile of the lattice .
future work should be devoted to further investigation of this open research problems .
b. apolloni , n. cesa - bianchi , and d. de falco . a numerical implementation of quantum annealing .
( eds . ) , stochastic processes , physics and geometry , proceedings of the ascona / locarno conference , 4 - 9 july 1988 _ , pages 97111 .
world scientific , 1990 . | we show , in the context of quantum combinatorial optimization , or quantum annealing , how the nonlinear schrdinger - langevin - kostin equation can dynamically drive the system toward its ground state .
we illustrate , moreover , how a frictional force of kostin type can prevent the appearance of genuinely quantum problems such as bloch oscillations and anderson localization which would hinder an exhaustive search . |
what is irreversibility of a process ?
this question , in this form , does not make much sense .
we first have to specify `` irreversibility with respect to what '' .
it means we first need to decide a set of rules
e. a set of allowed transformations together with some free resource to which one has to conform when trying to revert the process .
we can then say that irreversibility basically measures the deterioration of some resource that does not come for free , within the rules we specified .
when studying quantum error correction , one usually considers an extremely strict scenario , where legitimate corrections only amount to a fixed quantum channel applied after the action of the noise .
this scenario corresponds to the task of trying to restore the entanglement initially shared by the input system ( undergoing the noise ) with an inaccessible reference , only by using local actions on the output system , being any kind of communication between the two systems impossible .
being quantum error correction a basic task in quantum information theory , the literature on the subject grew rapidly in the last 15 years @xcite .
it is however possible to devise two main sectors of research : the first one is devoted to the design of good quantum error correcting codes , and directly stems from an algebraic approach to _ perfect _ quantum error correction ; the second one tries to understand conditions under which _ approximate _ quantum error correction is possible . usually , while the former is more practically oriented , the latter is able to give information theoretical bounds on the performance of the optimum correction strategy , even when perfect correction is not possible , while leaving unspecified the optimum correction scheme itself .
our contribution follows the second approach : we will derive some bounds relating the loss of entanglement due to the local action of a noisy channel on a bipartite state with the possibility of undoing such a noise . the original point in our analysis
is that we will consider many inequivalent ways to measure entanglement in bipartite mixed states , hence obtaining many inequivalent measures of irreversibility . after reviewing the main results of ref .
@xcite , we will show how we can relate such entropic quantities with different norm - induced measures of irreversibility , like those exploiting the cb - norm distance @xcite or the channel fidelity @xcite , therefore providing measures of the overall i .
e. state independent irreversibility of a quantum channel .
in the following , quantum systems will be often identified with the ( finite dimensional ) hilbert spaces supporting them , that is , the roman letter @xmath0 [ resp .
@xmath1 , rigorously denoting the system only , will also serve as a shorthand notation instead of the more explicit @xmath2 [ resp . @xmath3 .
the ( complex ) dimension of @xmath2 [ resp .
@xmath3 will be denoted as @xmath4 [ resp . @xmath5 .
the set of possible states of the system @xmath0 [ resp .
@xmath1 , that is , the set of positive semi - definite operators with unit trace acting on @xmath2 [ resp .
@xmath3 , will be equivalently denoted with @xmath6 [ resp .
@xmath7 or @xmath8 [ resp .
@xmath9 . a general quantum noise @xmath10 is described as a completely positive trace - preserving map i .
channel_. if the input system @xmath0 is initially described by the state @xmath11 , we will write @xmath12 to denote @xmath13 . the aim of this section is to understand how one can measure the coherence of the evolution @xmath14 induced by @xmath15 on @xmath11 .
( we will see in the following how to get rid of the explicit dependence on the input state and obtain a quantity measuring the overall invertibility of a given channel , as a function the channel only . ) before continuing the discussion , we should clarify what we mean with the term `` coherence '' .
imagine that the input system @xmath0 is actually the subsystem of a larger bipartite system @xmath16 , where the letter @xmath17 stands for _ reference _ , initially described by a pure state @xmath18 , such that @xmath19=\rho^a.\ ] ] the situation is depicted in fig .
[ fig:3 ] . is purified with respect to a reference system @xmath17 into the state @xmath18 .
the noise @xmath20 acts on the system @xmath0 only , in such a way that @xmath18 is mapped into @xmath21.,width=302 ] notice that the input state @xmath11 is mixed if and only if the pure state @xmath18 is entangled .
then , the coherence of the evolution ( [ eq : evolution ] ) can be understood as the amount of residual entanglement survived in the bipartite output ( generally mixed ) state @xmath21 after the noise locally acted on @xmath0 only .
however , any naive attempt to formalise such an intuitive idea is soon frustrated by the fact that there exist many different and generally inequivalent ways to measure the entanglement of a mixed bipartite system @xcite .
this well - known phenomenon turns out in the existence of many different and generally inequivalent , but all in principle valid , ways to measure the coherence of an evolution .
one possibility to overcome such a problem was considered already in ref .
@xcite . there
, schumacher introduced the quantity called _ entanglement fidelity _ of a channel @xmath22 with respect to an input state @xmath11 , defined as @xmath23 such a quantity ( which does not depend on the particular purification @xmath18 considered ) accurately describes how close the channel @xmath15 is to the noiseless channel @xmath24 on the support of @xmath11 @xcite .
however , it was noticed that , as defined in eq .
( [ eq : ent_fid ] ) , @xmath25 is _ not _ related to the coherence of the evolution @xmath26 , in that it is easy to see that a unitary channel
e. completely coherent can result in a null entanglement fidelity .
we then have to consider a more general situation , like the one depicted in fig . [ fig:1 ] .
, here , after the noise @xmath15 , we apply a subsequent correction via a local restoring channel @xmath27 .
the corrected bipartite output state @xmath28 is denoted by @xmath29.,width=377 ] after the local noise produced the bipartite state @xmath30 , we apply a local restoring channel @xmath27 to obtain @xmath31 notice that in general the restoring channel can explicitly depend on the input state @xmath11 and on the noise @xmath15 .
however , for sake of clarity of notation , we will leave such dependence understood , and make it explicit again , by writing @xmath32 , only when needed .
we now compute the _ corrected _ entanglement fidelity @xmath33 and take the supremum over all possible corrections @xmath34 this is now a good measure of the coherence of the noisy evolution @xmath26 : by construction it is directly related to the degree of invertibility of the noise @xmath15 on the support of @xmath11 .
the maximisation over all possible correcting channels in eq .
( [ eq : corr_ent_fid ] ) can be extremely hard to compute . moreover , we are still interested in understanding how the coherence of a transformation is related to the theory of bipartite entanglement .
the idea is that of finding some quantity ( typically an entropic - like function ) which is able to capture at one time both the amount of coherence preserved by the channel as well as the invertibility of the channel itself , possibly bypassing the explicit evaluation of @xmath35 , for which accurate upper and lower bounds would suffice .
a key - concept in the theory of approximate quantum error correction is that of _ coherent information _
@xcite , which , for a bipartite state @xmath36 , is defined as @xmath37 where @xmath38 $ ] is the von neumann entropy of the state @xmath39 .
notice that , in the definition of coherent information , system @xmath0 and system @xmath40 play apparently different roles : such asymmetry acknowledges that the flow of quantum information is considered as being from @xmath0 to @xmath40 .
accordingly , _ channel coherent information _ is defined as @xmath41 where @xmath42 stand for reference , input , and output system , respectively . in our picture , the input state @xmath18 is pure , so that @xmath43 .
we then compute the coherent information loss due to the action of the noise @xmath15 on subsystem @xmath0 as @xmath44 where the non - negativity follows from the data - processing inequality @xcite . the following theorem (
whose first part is in ref .
@xcite and second part in ref .
@xcite ) is exactly what we were searching for let @xmath11 be the input state for a channel @xmath20 .
let @xmath45 be the corresponding loss of coherent information .
then , there exists a recovering channel @xmath46 such that @xmath47 conversely , for every channel @xmath27 , it holds @xmath48 where @xmath49 is an appropriate positive , continuous , concave , monotonically increasing function such that @xmath50 . in particular , for @xmath51 , we can take @xmath52 .
@xmath53 notice that , in particular , we have @xmath54 and @xmath55 where @xmath35 was given in eq .
( [ eq : corr_ent_fid ] ) .
the above theorem can be summarised by stating that the loss of coherent information of an input pure state @xmath18 due to a channel @xmath15 acting on @xmath0 is small ( that means @xmath45 close to zero ) if and only if the channel @xmath15 can be approximately corrected on the support of @xmath11 ( that means @xmath35 close to one ) .
this has been a very important generalisation of the previous theorem appeared ref .
@xcite concerning _ exact _ channel correction , namely @xmath56 on the support of the input state @xmath11 , which turns out to be possible , as a corollary , if and only if @xmath57 .
such a generalisation lies at the core of some recent coding theorems for quantum channel capacity see for example ref .
@xcite .
coherent information loss is now an extremely handy quantity to deal with , easy to compute and providing sufficiently tight bounds on the invertibility of the noise .
however , coherent information still lacks of some requirements we asked for in our original program .
indeed , we would like to relate the degree of invertibility of a general quantum noise to some function quantifying the loss of entanglement .
in fact , it is known that coherent information is not a satisfactory measure of entanglement , and it is not straightforward to generalise theorem 1 to other entanglement measures loss . to find
a relation between noise invertibility and various entanglement measures will be the aim of the next section .
in the following we will focus on a widely studied family of possible entanglement measures , namely those which stem from von neumann entropy and analogous quantities . among these measures , that often gain an operational interpretation as the optimum asymptotic rate at which a particular entanglement transformation can be done , we find , e. g. the _ distillable entanglement _ @xmath58 , the _ distillable key _
@xmath59 , the _ squashed entanglement _ @xmath60 , the _ relative entropy of entanglement _ @xmath61 , the _ entanglement cost _ @xmath62 , and the _ entanglement of formation _ @xmath63 , just to mention some of them ( for an accurate review of definitions and properties of a large class of entanglement measures see ref .
@xcite and references therein ) .
in particular , in the following we will explicitly call for the entanglement of formation , which is defined as @xcite @xmath64 where the minimum is taken over all possible ensemble decompositions @xmath65 , for pure @xmath66 s , and where @xmath67)$ ] , is the so - called _ entropy of pure - state entanglement_. here we refrain from provide even a short review of the other entropic - like entanglement measures we mentioned , which would be far beyond the scope of the present contribution .
the interested reader is directed to refs .
@xcite and @xcite . for our purposes , we are content with recalling that , given a bipartite state @xmath36 , the following inequalities hold @xmath68 where @xmath69 is the _ quantum mutual information_. moreover @xmath70 notice that it is commonly found that @xmath71 and , as dimensions of subsystems @xmath0 and @xmath40 increase , a mixed state picked up at random in the convex set of mixed bipartite states almost certainly ( that is , with probability approaching one exponentially fast in the dimension ) displays an even more dramatic separation @xcite @xmath72 our motivation is to work out a result analogous to theorem 1 , where , instead of the coherent information loss @xmath45 introduced in eq .
( [ eq : coherent - info - loss ] ) , we would like to use some other entanglement measure loss @xmath73 where the letter `` @xmath74 '' could stand , for example , for `` @xmath75 '' ( squashed entanglement loss ) or `` @xmath76 '' ( entanglement of formation loss ) .
already at a first glance , we can already say that , thanks to eqs .
( [ eq : bounds]-[eq : hashing ] ) , the second part of theorem 1 can be extended to other entanglement loss measures , that is @xmath77 for every channel @xmath27 . instead , the generalisation of the first part of theorem 1 is not straightforward : because of the typical entanglement behaviour summarised in eq .
( [ eq : extreme - bounds ] ) , we could easily have , for example , a channel causing a _ vanishingly small _ entanglement of formation loss with , at the same time , a relatively _ severe _ coherent information loss .
still , the following argument suggests that _ there must be _ an analogous of eq .
( [ eq : miao ] ) for alternative entanglement losses : in fact , when evaluated on pure states , all mentioned entanglement measures coincide with the entropy of pure - state entanglement .
moreover , many of these entanglement measures are known to be continuous in the neighbourhood of pure states .
this is equivalent to the fact that , in the neighbourhood of pure states , they have to be reciprocally boundable .
therefore , if the action of the noise @xmath15 is `` sufficiently gentle '' and the output state @xmath30 exhibits an entanglement structure which is `` sufficiently close '' to pure - state entanglement itself being pure .
a trivial example of a mixed state with pure - state entanglement structure is given by @xmath78 , where @xmath79 and @xmath80 are two subsystems of @xmath40 .
] , then it should be possible to write the analogous of eq .
( [ eq : miao ] ) in terms of @xmath81 or @xmath82 , for example , as well .
the problem is to explicitly write down such analogous formula . in ref .
@xcite , the interested reader can find the proof of the following theorem let @xmath11 be the input state for a channel @xmath20 .
let @xmath81 and @xmath83 be the corresponding losses of squashed entanglement and entanglement of formation , respectively . then @xmath84 and @xmath85 notice the large numerical factor , depending on the dimensions of the underlying subsystems , in front of the entanglement of formation loss : this feature is reminiscent of the previously mentioned irreversibility gap between distillable entanglement and entanglement of formation , and makes it possible the situation where the noise causes a vanishingly ( in the dimensions ) small entanglement of formation loss , even though its action is extremely dissipative with respect to the loss of coherent information . on the contrary
, the loss of squashed entanglement seems to be an efficient indicator of irreversibility , almost as good as the coherent information loss in fact , only an extra constant factor of @xmath86 appears in eq .
( [ eq : direct1 ] ) with respect to eq .
( [ eq : miao]) ; on the other hand , it is symmetric under the exchange of the input system with the output system , a property that does not hold for the coherent information loss .
summarising this section , the important thing is that there always exist a threshold ( which is strictly positive for finite dimensional systems ) below which all entanglement losses become equivalent , in the sense that they can be reciprocally bounded ( it is noteworthy that , in the case of squashed entanglement loss and coherent information loss , we can have dimension - independent bounds , which is a desirable property when dealing with quantum channels alone , see section 5 below ) . it is interesting now to forget for a moment about the channel @xmath15 itself , and see what eqs .
( [ eq : miao ] ) , ( [ eq : direct1 ] ) , and ( [ eq : direct2 ] ) mean in terms of a given bipartite mixed state only .
first of all , notice that , for every mixed state @xmath36 , there exist two pure states , @xmath87 and @xmath88 , and two channels , @xmath89 and @xmath90 , such that @xmath91 and @xmath92 .
now , for a given state @xmath36 , let us define @xmath93 and @xmath94 where the letter @xmath74 is used as before are defined in the same way , by simply exchanging subsystems labels , as @xmath95 . ] .
then , theorems 1 and 2 tell us that there exist channels @xmath96 and @xmath97 , and two pure states , @xmath98 and @xmath99 , with @xmath100=\tau^a$ ] and @xmath101=\tau^b$ ] , such that @xmath102 where @xmath103 , @xmath104 , and @xmath105 . in a sense , either @xmath106 or @xmath107 being small . ] , it means that the entanglement present in the state @xmath36 is basically pure - state entanglement , even if @xmath36 is itself a mixed state .
this is the reason for which we can establish a quantitative relation between typically inequivalent entanglement measures , as the following corollary of theorems 1 and 2 clearly states @xcite for an arbitrary bipartite mixed state @xmath36 , with @xmath108 , the following inequality holds @xmath109 where @xmath110 is a function as in eq .
( [ eq : converse ] ) in theorem 1 .
@xmath53 this corollary is in a sense the quantitative version of the intuitive argument given before theorem 2 , and it represents a first attempt in complementing the findings of ref .
@xcite , summarised in eq .
( [ eq : extreme - bounds ] ) . ) shows the behaviour , for a bipartite system of two qutrits , of the lower bound in eq .
( [ eq : fgprime ] ) for coherent information as a function of entanglement of formation .
coherent information , and hence distillable entanglement , are bounded from below by the thick curve .
notice that @xmath63 has to be extraordinarily close to its maximum value in order to have a non trivial bound from eq .
( [ eq : fgprime ] ) .
this fact suggests that the bound itself could be improved.,width=377 ] it is also possible to invert eq .
( [ eq : gap ] ) and obtain a function @xmath111 such that @xmath112 for all bipartite state @xmath113 .
the plot of @xmath111 is given in fig .
[ fig:2 ] for @xmath114 ( for qubits every entangled state is also distillable ) , for a state for which @xmath115 . the plotted curve displays the typical behaviour of the bound ( [ eq : fgprime ] ) .
notice from fig .
[ fig:2 ] that entanglement of formation has to be extremely close to its maximum attainable value in order to obtain a non trivial bound from eq .
( [ eq : fgprime ] ) .
this is a strong evidence that the bound itself could probably be improved . nonetheless , we believe that such an improvement , if possible , would only make smaller some ( unimportant ) constants which are independent of the dimension , while leaving the leading order of dependence on @xmath116 in the right hand side of eq .
( [ eq : gap ] ) untouched .
the previous analysis , following ref .
@xcite , was done in order to quantify the invertibility of a noisy evolution with respect to _ a given _ input state @xmath11 . in this section ,
we want to derive quantities characterising the `` overall '' invertibility of a given channel . in other words
, we would like to get rid of the explicit dependence on the input state and obtain the analogous of eqs .
( [ eq : miao ] ) , ( [ eq : miao2 ] ) , ( [ eq : direct1 ] ) , and ( [ eq : direct2 ] ) as functions of the channel @xmath15 only .
intuitively , to do this , we should quantify how close the corrected channel @xmath117 can be to the noiseless channel @xmath24 , for all possible corrections @xmath118 .
however , in doing this , we have to be very careful about which channel distance function we adopt in order to measure `` closeness '' . a safe choice consists in using the distance induced by the so - called _ norm of complete boundedness _ , for short _ cb - norm _ , defined as @xmath119 where @xmath120 is the identity channel on @xmath121 density matrices , and @xmath122\le 1}\tr\left[\,|\mn(\rho)|\,\right].\ ] ] ( we put the absolute value inside the trace because in literature one often deals also with non completely positive maps , so that the extension @xmath123 can be non positive . ) notice , that , in general , @xmath124 , and the two norms can be inequivalent @xcite . a part of the rather technical definition of cb - norm ( the extension in eq .
( [ eq : def - cb ] ) is necessary , basically for the same reasons for which we usually consider complete positivity instead of the simple positivity ) , we will be content with knowing that , for channels , @xmath125 and @xmath126 , and that the following theorem holds @xcite let @xmath22 be a channel , with @xmath127
. then @xmath128 where the infimum of the entanglement fidelity is done over all normalised states @xmath129 .
@xmath53 it is then natural to define a cb - norm based measure of the overall invertibility of a given channel @xmath20 as @xmath130 with the infimum taken over all possible correcting channels @xmath131 . for a moment , let us now go back to the other functions we introduced before
. we will be able to relate them , in some cases with dimension independent bounds , to the cb - norm based invertibility @xmath132 . given the loss function @xmath133 , where @xmath134 is used to denote the coherent information loss , the squashed entanglement loss , and the entanglement of formation loss , respectively , we define the following quantity @xmath135 where the supremum is taken over all possible input states @xmath11 .
analogously , from eq .
( [ eq : corr_ent_fid ] ) , let us define @xmath136 such quantities are now functions of the channel only , and we want to understand how well @xmath137 and @xmath138 capture the `` overall '' invertibility of a channel .
first of all , let us understand how they are related .
let @xmath139 be the state for which @xmath137 is achieved .
then , @xmath140 on the other hand , let @xmath138 be achieved with @xmath141 . then , @xmath142 where , as usual , @xmath103 , @xmath104 , and @xmath105 .
we are now in position , thanks to theorem 3 , to show how @xmath132 , @xmath137 , and @xmath138 are related to each other .
let the value @xmath138 be achieved by the couple @xmath143 .
then , @xmath144{2k_x\delta_x(\mn ) } , \end{split}\ ] ] where in the second line we used theorem 3 , since the channel @xmath145 has equal input and output spaces .
conversely , let @xmath137 be achieved by @xmath139 .
then , thanks to eq .
( [ eq : converse2 ] ) @xmath146 for all channels @xmath27 .
let @xmath147 be the channel achieving the infimum in eq .
( [ eq : cb - invert ] ) .
then , @xmath148 summarising , we found that @xmath149{2k_x\delta_x(\mn)}. \end{split}\ ] ] in the function @xmath110 , the dependence on the dimension @xmath150 is present ( see theorem 1 ) , however only inside a logarithm : this is not bad , in view of coding theorems .
the dependence on @xmath150 can instead be dramatic in @xmath151 ; on the contrary , both @xmath152 and @xmath153 are independent on the dimension .
this work is funded by japan science and technology agency , through the erato - sorst project on quantum computation and information .
the author would like to thank in particular m hayashi , and k matsumoto for interesting discussions and illuminating suggestions .
m gregoratti and r f werner , j. mod .
opt . * 50 * , 915 ( 2003 ) ; f buscemi , g chiribella , and g m dariano , phys .
lett . * 95 * , 090501 ( 2005 ) ; j a smolin , f verstraete , and a winter , phys . rev . a * 72 * , 052317 ( 2005 ) ; f buscemi , phys . rev . lett . *
99 * , 180501 ( 2007 ) .
m a nielsen and i l chuang , _ quantum computation and quantum information _
( cambridge university press , cambridge , 200 ) ; j kempe , in _ quantum decoherence , poincar seminar 2005 _ , progress in mathematical physics series ( birkhauser verlag , berlin , 2006 ) . | the action of a channel on a quantum system , when non trivial , always causes deterioration of initial quantum resources , understood as the entanglement initially shared by the input system with some reference purifying it .
one effective way to measure such a deterioration is by measuring the loss of coherent information , namely the difference between the initial coherent information and the final one : such a difference is `` small '' , if and only if the action of the channel can be `` almost perfectly '' corrected with probability one . in this work ,
we generalise this result to different entanglement loss functions , notably including the entanglement of formation loss , and prove that many inequivalent entanglement measures lead to equivalent conditions for approximate quantum error correction . in doing this , we show how different measures of bipartite entanglement give rise to corresponding distance - like functions between quantum channels , and we investigate how these induced distances are related to the cb - norm . |
we study scaling limits of _ internal diffusion limited aggregation _ ( `` internal dla '' ) , a growth model introduced in @xcite . in internal dla ,
one inductively constructs an * occupied set * @xmath8 for each time @xmath9 as follows : begin with @xmath10 and @xmath11 , and let @xmath12 be the union of @xmath13 and the first place a random walk from the origin hits @xmath14 . the purpose of this paper is to study the growing family of sets @xmath13 .
following the pioneering work of @xcite , it is by now well known that , for large @xmath1 , the set @xmath13 approximates an origin - centered euclidean lattice ball @xmath15 ( where @xmath16 is such that @xmath17 has volume @xmath1 ) .
the authors recently showed that this is true in a fairly strong sense @xcite : the maximal distance from a point where @xmath18 is non - zero to @xmath19 is a.s .
@xmath2 if @xmath3 and @xmath4 if @xmath5 .
in fact , if @xmath20 is large enough , the probability that this maximal distance exceeds @xmath21 ( or @xmath22 when @xmath5 ) decays faster than any fixed ( negative ) power of @xmath1 .
some of these results are obtained by different methods in @xcite .
this paper will ask what happens if , instead of considering the maximal distance from @xmath19 at time @xmath1 , we consider the `` average error '' at time @xmath1 ( allowing inner and outer errors to cancel each other out ) .
it turns out that in a distributional `` average fluctuation '' sense , the set @xmath13 deviates from @xmath17 by only a constant number of lattice spaces when @xmath23 and by an even smaller amount when @xmath5 .
appropriately normalized , the fluctuations of @xmath13 , taken over time and space , define a distribution on @xmath24 that converges in law to a variant of the gaussian free field ( gff ) : a random distribution on @xmath24 that we will call the * augmented gaussian free field*. ( it can be constructed by defining the gff in spherical coordinates and replacing variances associated to spherical harmonics of degree @xmath25 by variances associated to spherical harmonics of degree @xmath26 ; see [ ss.augmentedgff ] . )
the `` augmentation '' appears to be related to a damping effect produced by the mean curvature of the sphere ( as discussed below ) .
, with particles started uniformly on @xmath27 .
though we do not prove this here , we expect the cluster boundaries to be approximately flat cross - sections of the cylinder , and we expect the fluctuations to scale to the _ ordinary _ gff on the half cylinder as @xmath28 . ] to our knowledge , no central limit theorem of this kind has been previously conjectured in either the physics or the mathematics literature .
the appearance of the gff and its `` augmented '' variants is a particular surprise .
( it implies that internal dla fluctuations although very small have long - range correlations and that , up to the curvature - related augmentation , the fluctuations in the direction transverse to the boundary of the cluster are of a similar nature to those in the tangential directions . ) nonetheless , the heuristic idea is easy to explain . before we state the central limit theorems precisely ( [ ss.twostatement ] and [ ss.generalstatement ] ) ,
let us explain the intuition behind them .
write a point @xmath29 in polar coordinates as @xmath30 for @xmath31 and @xmath32 on the unit sphere
. suppose that at each time @xmath1 the boundary of @xmath13 is approximately parameterized by @xmath33 for a function @xmath34 defined on the unit sphere .
write @xmath35 where @xmath36 is the volume of the unit ball in @xmath24 .
the @xmath37 term measures the deviation from circularity of the cluster @xmath13 in the direction @xmath32 .
how do we expect @xmath38 to evolve in time ? to a first approximation , the angle at which a random walk exits @xmath13 is a uniform point on the unit sphere .
if we run many such random walks , we obtain a sort of poisson point process on the sphere , which has a scaling limit given by space - time white noise on the sphere .
however , there is a smoothing effect ( familiar to those who have studied the continuum analog of internal dla : the famous hele - shaw model for fluid insertion , see the reference text @xcite ) coming from the fact that places where @xmath38 is small are more likely to be hit by the random walks , hence more likely to grow in time .
there is also secondary damping effect coming from the mean curvature of the sphere , which implies that even if ( after a certain time ) particles began to hit all angles with equal probability , the magnitude of @xmath38 would shrink as @xmath1 increased and the existing fluctuations were averaged over larger spheres .
the white noise should correspond to adding independent brownian noise terms to the spherical fourier modes of @xmath38 .
the rate of smoothing / damping in time should be approximately given by @xmath39 for some linear operator @xmath40 mapping the space of functions on the unit sphere to itself .
since the random walks approximate brownian motion ( which is rotationally invariant ) , we would expect @xmath40 to commute with orthogonal rotations , and hence have spherical harmonics as eigenfunctions . with the right normalization and parameterization , it is therefore natural to expect the spherical fourier modes of @xmath38 to evolve as independent brownian motions subject to linear `` restoration forces '' ( a.k.a . ornstein - uhlenbeck processes ) where the magnitude of the restoration force depends on the degree of the corresponding spherical harmonic .
it turns out that the restriction of the ( ordinary or augmented ) gff on @xmath24 to a centered volume @xmath1 sphere evolves in time @xmath1 in a similar way . of course
, as stated above , the `` spherical fourier modes of @xmath38 '' have not really been defined ( since the boundary of @xmath13 is complicated and generally _ can not _ be parameterized by @xmath41 ) . in the coming sections
, we will define related quantities that ( in some sense ) encode these spherical fourier modes and are easy to work with .
these quantities are the martingales obtained by summing discrete harmonic polynomials over the cluster @xmath13 .
the heuristic just described provides intuitive interpretations of the results given below .
theorem [ t.fluctuations ] , for instance , identifies the weak limit as @xmath42 of the internal dla fluctuations from circularity at a fixed time @xmath1 : the limit is the two - dimensional augmented gaussian free field restricted to the unit circle @xmath43 , which can be interpreted in a distributional sense as the random fourier series @xmath44\ ] ] where @xmath45 for @xmath46 and @xmath47 for @xmath48 are independent standard gaussians .
the ordinary two - dimensional gff restricted to the unit circle is similar , except that @xmath49 is replaced by @xmath50 .
the series unlike its counterpart for the one - dimensional gaussian free field , which is a variant of brownian bridge is a.s .
divergent , which is why we use the dual formulation explained in [ ss.generalstatement ] . the dual formulation of amounts to a central limit theorem , saying that for each @xmath48 the real and imaginary parts of @xmath51 converge in law as @xmath52 to normal random variables with variance @xmath53 ( and that @xmath54 and @xmath55 are asymptotically uncorrelated for @xmath56 ) .
see @xcite for numerical data on the moments @xmath55 in large simulations . before we set about formulating our central limit theorems precisely , we mention a previously overlooked fact
. suppose that we run internal dla in continuous time by adding particles at poisson random times instead of at integer times : this process we will denote by @xmath57 ( or often just @xmath58 ) where @xmath59 is the counting function for a poisson point process in the interval @xmath60 $ ] ( so @xmath59 is poisson distributed with mean @xmath1 ) .
we then view the entire history of the idla growth process as a ( random ) function on @xmath61 , which takes the value @xmath62 or @xmath63 on the pair @xmath64 accordingly as @xmath65 or @xmath66 .
write @xmath67 for the set of functions @xmath68 such that @xmath69 whenever @xmath70 , endowed with the coordinate - wise partial ordering .
let @xmath71 be the distribution of @xmath72 , viewed as a probability measure on @xmath67 .
[ t.fkg ] _
( fkg inequality ) _ for any two increasing functions @xmath73 , the random variables @xmath74 and @xmath75 are nonnegatively correlated .
one example of an increasing function is the total number @xmath76 of occupied sites in a fixed subset @xmath77 at a fixed time @xmath1 .
one example of a decreasing function is the smallest @xmath1 for which all of the points in @xmath78 are occupied .
intuitively , theorem [ t.fkg ] means that on the event that one point is absorbed at a late time , it is conditionally more likely for all other points to be absorbed late .
the fkg inequality is an important feature of the discrete and continuous gaussian free fields @xcite , so it is interesting ( and reassuring ) that it appears in internal dla at the discrete level .
note that sampling a continuous time internal dla cluster at time @xmath1 is equivalent to first sampling a poisson random variable @xmath79 with expectation @xmath1 and then sampling an ordinary internal dla cluster of size @xmath79 .
( by the central limit theorem , @xmath80 has order @xmath81 with high probability . )
although using continuous time amounts to only a modest time reparameterization ( chosen independently of everything else ) it is aesthetically natural .
our use of `` white noise '' in the heuristic of the previous section implicitly assumed continuous time .
( otherwise the total integral of @xmath38 would be deterministic , so the noise would have to be conditioned to have mean zero at each time . ) [ cols="^,^ " , ] the restriction to harmonic @xmath82 ( as opposed to a more general test function @xmath83 ) seems to be necessary in large dimensions because otherwise the derivative of the test function along @xmath43 appears to have a non - trivial effect on ( see [ ss.fixedtimeproof ] ) .
this is because has a lot of positive mass just outside of the unit sphere and a lot of negative mass just inside the unit sphere . it may be possible to formulate a version of theorem [ t.highdconvergence ] ( involving some modification of the `` mean shape '' described by @xmath84 ) that uses test functions that depend only on @xmath32 in a neighborhood of the sphere ( instead of using only harmonic test functions ) , but we will not address this point here . deciding whether theorem [ t.lateness ] as stated extends to higher dimensions requires some number theoretic understanding of the extent to which the discrepancies between @xmath84 and @xmath85 ( as well as the errors that come from replacing a @xmath86 with a smooth test function @xmath83 ) average out when one integrates over a range of times .
we will not address these points here either .
we may write a general vector in @xmath24 as @xmath30 where @xmath87 and @xmath88 .
we write the laplacian in spherical coordinates as @xmath89 a polynomial @xmath90 $ ] is called _ harmonic _ if @xmath91 is the zero polynomial .
let @xmath92 denote the space of all homogenous harmonic polynomials in @xmath93 $ ] of degree @xmath25 , and let @xmath94 denote the space of functions on @xmath95 obtained by restriction from @xmath92 . if @xmath96 , then we can write @xmath97 for a function @xmath98 , and setting to zero at @xmath99 yields @xmath100 i.e. , @xmath101 is an eigenfunction of @xmath102 with eigenvalue @xmath103 .
note that continues to be zero if we replace @xmath25 with the negative number @xmath104 , since the expression @xmath105 is unchanged by replacing @xmath25 with @xmath106 .
thus , @xmath107 is also harmonic on @xmath108 . now , suppose that @xmath101 is normalized so that @xmath109 by scaling , the integral of @xmath110 over @xmath111 is thus given by @xmath112 . the @xmath113 norm on all of @xmath114 is then given by @xmath115 a standard identity states that the dirichlet energy of @xmath101 , as a function on @xmath95 , is given by the @xmath113 inner product @xmath116 .
the square of @xmath117 is given by the square of its component along @xmath95 plus the square of its radial component .
we thus find that the dirichlet energy of @xmath118 on @xmath114 is given by @xmath119 now suppose that we fix the value of @xmath118 on @xmath111 as above but harmonically extend it outside of @xmath114 by writing @xmath120 for @xmath121 .
then the dirichlet energy of @xmath118 outside of @xmath114 can be computed as @xmath122 which simplifies to @xmath123 combining the inside and outside computations in the case @xmath124 , we find that the harmonic extension @xmath125 of the function given by @xmath101 on @xmath95 has dirichlet energy @xmath126 .
if we decompose the gff into an orthonormal basis that includes this @xmath125 , we find that the component of @xmath125 is a centered gaussian with variance @xmath127 .
if we replace @xmath125 with the harmonic extension of @xmath128 ( defined on @xmath111 ) , then by scaling the corresponding variance becomes @xmath129 .
now in the augmented gff the variance is instead given by , which amounts to replacing @xmath127 with @xmath130 .
considering the component of @xmath128 in a basis expansion the space of functions on @xmath111 requires us to divide by @xmath131 ( to account for the scaling of @xmath118 ) and by @xmath132 ( to account for the larger integration area ) , so that we again obtain a variance of @xmath133 for the augmented gff , versus @xmath129 for the gff . in some ways ,
the augmented gff is very similar to the ordinary gff : when we restrict attention to an origin - centered annulus , it is possible to construct independent gaussian random distributions @xmath134 , @xmath135 , and @xmath136 such that @xmath134 has the law of a constant multiple of the gff , @xmath137 has the law of the augmented gff , and @xmath138 has the law of the ordinary gff . in light of theorem [ t.fluctuations ] , the following implies that ( up to absolute continuity ) the scaling limit of fixed - time @xmath13 fluctuations can be described by the gff itself . [ p.abscont ]
when @xmath3 , the law @xmath139 of the restriction of the gff to the unit circle ( modulo additive constant ) is absolutely continuous w.r.t .
the law @xmath140 of the restriction of the augmented gff restricted to the unit circle .
the relative entropy of a gaussian of density @xmath141 with respect to a gaussian of density @xmath142 is given by @xmath143 note that @xmath144 , and in particular @xmath145 .
thus the relative entropy of a centered gaussian of variance @xmath62 with respect to a centered gaussian of variance @xmath146 is @xmath147 .
this implies that the relative entropy of @xmath140 with respect to @xmath139 restricted to the @xmath148th component @xmath149 is @xmath150 .
the same holds for the relative entropy of @xmath139 with respect to @xmath140 . because the @xmath149 are independent in both @xmath140 and @xmath139 , the relative entropy of one of @xmath140 and @xmath139 with respect to the other
is the sum of the relative entropies of the individual components , and this sum is finite .
we recall that increasing functions of a poisson point process are non - negatively correlated @xcite .
( this is easily derived from the more well known statement @xcite that increasing functions of independent bernoulli random variables are non - negatively correlated . )
let @xmath140 be the simple random walk probability measure on the space @xmath151 of walks @xmath152 beginning at the origin .
then the randomness for internal dla is given by a rate - one poisson point process on @xmath153 where @xmath139 is lebesgue measure on @xmath154 .
a realization of this process is a random collection of points in @xmath155 .
it is easy to see ( for example , using the abelian property of internal dla discovered by diaconis and fulton @xcite ) that adding an additional point @xmath156 increases the value of @xmath57 for all times @xmath1 .
the @xmath57 are hence increasing functions of the poisson point process , and are non - negatively correlated . since @xmath157 and @xmath158 are increasing functions of the @xmath57 , they are also increasing functions of the point process and are thus non - negatively correlated .
let @xmath159 be a polynomial that is harmonic on @xmath24 , that is @xmath160 in this section we give a recipe for constructing a polynomial @xmath161 that closely approximates @xmath82 and is discrete harmonic on @xmath0 , that is , @xmath162 where @xmath163 is the symmetric second difference in direction @xmath164 .
the construction described below is nearly the same as the one given by lovsz in @xcite , except that we have tweaked it in order to obtain a smaller error term : if @xmath82 has degree @xmath165 , then @xmath166 has degree at most @xmath167 instead of @xmath168 .
discrete harmonic polynomials have been studied classically , primarily in two variables : see for example duffin @xcite , who gives a construction based on discrete contour integration .
consider the linear map @xmath169 \to \r[x_1,\ldots , x_d]\ ] ] defined on monomials by @xmath170 where we define @xmath171 [ discretepolynomial ] if @xmath172 $ ] is a polynomial of degree @xmath165 that is harmonic on @xmath24 , then the polynomial @xmath173 is discrete harmonic on @xmath0 , and @xmath166 is a polynomial of degree at most @xmath167 .
an easy calculation shows that @xmath174 from which we see that @xmath175 = \xi [ \frac{\partial^2}{\partial x_i^2 } \psi ] .\ ] ] if @xmath82 is harmonic , then the right side vanishes when summed over @xmath176 , which shows that @xmath177 $ ] is discrete harmonic .
note that @xmath178 is even for @xmath165 even and odd for @xmath165 odd .
in particular , @xmath179 has degree at most @xmath167 , which implies that @xmath180 has degree at most @xmath167 . to obtain a discrete harmonic polynomial @xmath86 on the lattice
@xmath181 , we set @xmath182 where @xmath165 is the degree of @xmath82 . for each fixed @xmath82 with @xmath183 the process @xmath184 is a martingale in @xmath1 .
each time a new particle is added , we can imagine that it performs brownian motion on the grid ( instead of a simple random walk ) , which turns @xmath185 into a continuous martingale , as in @xcite . by the martingale representation theorem ( see ( * ? ? ?
* theorem v.1.6 ) ) , we can write @xmath186 , where @xmath187 is a standard brownian motion and @xmath188 is the quadratic variation of @xmath185 on the interval @xmath60 $ ] . to show that @xmath189 converges in law as @xmath28 to a gaussian with variance @xmath190 , it suffices to show that for fixed @xmath1 the random variable @xmath191 converges in law to @xmath192 . by standard riemann integration and the @xmath13 fluctuation bounds in @xcite ( the weaker bounds of @xcite
would also suffice here ) we know that @xmath193 converges in law to @xmath194 as @xmath28 .
thus it suffices to show that @xmath195 converges in law to zero .
this expression is actually a martingale in @xmath1 .
its expected square is the sum of the expectations of the squares of its increments , each of which is @xmath196 .
the overall expected square of is thus @xmath197 , which indeed tends to zero .
recall and note that if we replace @xmath1 with @xmath59 , this does not change the convergence in law of @xmath191 when @xmath183 .
however , when @xmath198 , it introduces an asymptotically independent source or randomness which scales to a gaussian of variance @xmath199 ( simply by the central limit theorem for the poisson point process ) , and hence remains correct in this case .
similarly , suppose we are given @xmath200 and functions @xmath201 .
the same argument as above , using the martingale in @xmath1 , @xmath202 implies that @xmath203 converges in law to a gaussian with variance @xmath204 the theorem now follows from a standard fact about gaussian random variables on a finite dimensional vector spaces ( proved using characteristic functions ) : namely , a sequence of random variables on a vector space converges in law to a multivariate gaussian if and only if all of the one - dimensional projections converge .
the law of @xmath205 is determined by the fact that it is a centered gaussian with covariance given by .
recall that @xmath13 for @xmath206 denotes the discrete - time idla cluster with exactly @xmath1 sites , and @xmath207 for @xmath208 denotes the continuous - time cluster whose cardinality is poisson - distributed with mean @xmath1 .
define @xmath209 and @xmath210 fix @xmath211 , and consider a test function of the form @xmath212 where the @xmath213 are smooth functions supported in an interval @xmath214 .
we will assume , furthermore , that @xmath215 is real - valued .
that is , the complex numbers @xmath213 satisfy @xmath216 it follows from lemma [ dual ] below ( with @xmath220 , @xmath221 ) , that theorem [ gffdisk0 ] can be interpreted as saying that @xmath222 tends weakly to the gaussian random distribution associated to the hilbert space @xmath223 with norm @xmath224\frac{dr}{r}\ ] ] where @xmath225 and @xmath226 .
( the subscript @xmath227 means nonradial : @xmath223 is the orthogonal complement of radial functions in the sobolev space @xmath228 . ) [ dual ] let @xmath229 and let @xmath82 be a real valued function on @xmath230 .
denote @xmath231 where the supremum is over real - valued @xmath118 , compactly supported and subject to the constraint @xmath232 then @xmath233 in the case @xmath234 , replace @xmath118 in @xmath235 with @xmath236 change order of integration and apply the cauchy - schwarz inequality .
for the case @xmath237 , multiply by the appropriate factors @xmath238 and @xmath239 to deduce this from the case @xmath234 .
theorem [ gffdisk1 ] is a restatement of theorem [ t.lateness ] .
( as in the proof of theorem [ t.highdconvergence ] , the convergence in law of all one - dimensional projections to the appropriate normal random variables implies the corresponding result for the joint distribution of any finite collection of such projections . ) theorem [ gffdisk1 ] says that @xmath245 tends weakly to a gaussian distribution for the hilbert space @xmath228 with the norm @xmath246\frac{dr}{r}.\ ] ] by way of comparison , the usual gaussian free field is the one associated to the dirichlet norm @xmath247\frac{dr}{r}.\ ] ] comparing these two norms , we see that the second term in @xmath248 has an additional @xmath249 , hence our choice of the term `` augmented gaussian free field . ''
as derived in [ ss.augmentedgff ] , this @xmath249 results in a smaller variance @xmath250 in each spherical mode of degree @xmath25 of the augmented gff , as compared to @xmath251 for the usual gff .
the surface area of the sphere is implicit in the normalization , and is accounted for here in the factors @xmath252 above .
let @xmath254 , and for @xmath48 let @xmath255 , where @xmath256\ ] ] is the discrete harmonic polynomial associated to @xmath257 as described in [ s.polynomials ] .
the sequence @xmath258 begins @xmath259 for instance , to compute @xmath260 , we expand @xmath261\ ] ] and apply @xmath262 to each monomial , obtaining @xmath263\ ] ] which simplifies to @xmath264 .
one readily checks that this defines a discrete harmonic function on @xmath265 .
( in fact , @xmath266 is itself discrete harmonic , but @xmath267 is not for @xmath268 . ) to define @xmath258 for negative @xmath165 , we set @xmath269 . part ( a ) of this lemma was proved by van der corput in the 1920s ( see @xcite , theorem 87 p. 484 ) .
part ( b ) follows from the same method , as proved below .
part ( c ) follows from part ( b ) and the stronger estimate of lemma [ discretepolynomial ] , @xmath280 for @xmath281 ( and @xmath282 ) .
we prove part ( b ) in all dimensions .
let @xmath283 be a harmonic polynomial on @xmath24 of homogeneous of degree @xmath165 .
normalize so that @xmath284 where @xmath285 is the unit ball .
in this discussion @xmath165 will be fixed and the constants are allowed to depend on @xmath165 .
consider @xmath294 a smooth , radial function on @xmath24 with integral @xmath62 supported in the unit ball . then define @xmath295 characteristic function of the unit ball .
denote @xmath296 then @xmath297 this is because @xmath298 is nonzero only in the annulus of width @xmath299 around @xmath300 in which ( by the van der corput bound ) there are @xmath301 lattice points .
the poisson summation formula implies @xmath302*\hat p_k(\xi)/r^k\ ] ] in the sense of distributions .
the fourier transform of a polynomial is a derivative of the delta function , @xmath303 .
because @xmath286 and @xmath178 is harmonic , its average with repect to any radial function is zero .
this is expressed in the dual variable as the fact that when @xmath304 , @xmath305 ) = 0\ ] ] so we our sum equals @xmath306*\hat p_k ( \xi)/r^k\ ] ] next look at @xmath307 @xmath308 all the terms in which fewer derivatives fall on @xmath309 and more fall on @xmath310 give much smaller expressions : the factor @xmath311 corresponding to each such differentiation is replaced by an @xmath312 .
the asymptotics of this oscillatory integral above are well known . for any fixed polynomial @xmath313 they are of the same order of magnitude as for @xmath314 ,
namely @xmath315 this is proved by the method of stationary phase and can also be derived from well known asymptotics of bessel functions .
denote @xmath320 applying the formula above for @xmath321 , @xmath322 to estimate the error term @xmath323 , note first that the coefficients @xmath213 are supported in a fixed annulus , the integrand above is supported in the range @xmath324 .
furthermore , by @xcite , there is an absolute constant @xmath20 such that for all sufficiently large @xmath311 and all @xmath1 in this range , the difference @xmath325 is supported on the set of @xmath326 such that @xmath327 .
thus @xmath328 moreover , lemma [ discrete - error ] applies and @xmath329 next , lemma [ vandercorput](a ) says ( since @xmath330 ) @xmath331 thus replacing @xmath82 by @xmath332 gives an additional error of size at most @xmath333 in all , @xmath334 for @xmath335 , consider the process @xmath336 note that @xmath337 as @xmath338 .
note also that lemma [ vandercorput](c ) implies @xmath339 because @xmath258 are discrete harmonic and @xmath340 for all @xmath341 , @xmath342 is a martingale .
it remains to show that @xmath343 in law .
as outlined below , this will follow from the martingale central limit theorem ( see , e.g. , @xcite or @xcite ) . for sufficiently large @xmath311 , the difference @xmath344 is nonzero only for @xmath345 in the range @xmath346 ; and @xmath347 .
we now show that this implies @xmath348 so that the martingale central limit theorem applies . to prove , observe that @xmath349 where @xmath350 is the @xmath351th point of @xmath13 .
then @xmath352 implies @xmath353 , and hence @xmath354 recalling that @xmath355 unless @xmath356 , we have @xmath357 which confirms .
because @xmath13 fills the lattice @xmath265 as @xmath358 , we have @xmath359 we prove in three steps : replace @xmath360 by @xmath267 ( or @xmath361 if @xmath362 ) ; replace the lower limit @xmath363 by @xmath364 ; replace the sum of @xmath350 over lattice sites with the integral with respect to lebesgue measure in the complex @xmath350-plane .
we begin the proof of by noting that the error term introduced by replacing @xmath258 with @xmath267 is @xmath365 in the integral this is majorized by @xmath366 since there are @xmath367 such terms , this change contributes order @xmath368 to the sum .
next , we change the lower limit from @xmath363 to @xmath369 . since @xmath370 ,
the integral inside @xmath371 is changed by @xmath372 thus the change in the whole expression is majorized by the order of the cross term @xmath373 again there are @xmath374 terms in the sum over @xmath350 , so the sum of the errors is @xmath375 .
lastly , we replace the value at each site @xmath376 by the integral @xmath377 where @xmath378 is the unit square centered at @xmath376 and @xmath379 . because the square has area @xmath62 , the term in the lattice sum is the same as this integral with @xmath380 replaced by @xmath376 at each occurrence . since @xmath381 , @xmath382
after we divide by @xmath383 , the order of error is @xmath384 .
adding all the errors contributes at most order @xmath384 to the sum .
we must also take into account the change in the lower limit of the integral , @xmath385 is replaced by @xmath386 . since @xmath381 , @xmath387 recall that in the previous step we previously changed the lower limit by @xmath388 .
thus by the same argument , this smaller change gives rise to an error of order @xmath384 in the sum over @xmath376 .
the proof of is now reduced to evaluating @xmath389 integrating in @xmath32 and changing variables from @xmath7 to @xmath390 , @xmath391 then change variables from @xmath1 to to @xmath392 to obtain @xmath393 this ends the proof of theorem [ gffdisk0 ] .
the proof of theorem [ gffdisk1 ] follows the same idea .
we replace @xmath13 by the poisson time region @xmath394 ( for @xmath395 ) , and we need to find the limit as @xmath217 of @xmath396 the error terms in the estimation showing this quantity is within @xmath397 of @xmath398 are nearly the same as in the previous proof .
we describe briefly the differences .
the difference between poisson time and ordinary counting is @xmath399 .
it follows that for @xmath400 , @xmath401 as in the previous proof for @xmath363 .
further errors are also controlled since we then have the estimate analogous to the one above for @xmath13 , namely @xmath402 we consider the continuous time martingale @xmath403 instead of using the martingale central limit theorem , we use the martingale representation theorem .
this says that the martingale @xmath404 when reparameterized by its quadratic variation has the same law as brownian motion .
we must show that almost surely the quadratic variation of @xmath185 on @xmath405 is @xmath406 . @xmath407
integrating with respect to @xmath345 gives the quadratic variation @xmath406 after a suitable change of variable as in the previous proof .
theorem [ t.fluctuations ] follows almost immediately from the @xmath3 case of theorem [ t.highdconvergence ] and the estimates above .
consider @xmath408 where @xmath409 is as in .
what happens if we replace @xmath83 with a function @xmath410 that is discrete harmonic on the rescaled mesh @xmath411 within a @xmath412 neighborhood of @xmath413 ? clearly ,
if @xmath83 is smooth , we will have @xmath414 .
since there are at most @xmath415 non - zero terms in , the discrepancy in @xmath416 which tends to zero as long as @xmath417 . the fact that replacing @xmath418 with @xmath409 has a negligible effect follows from the above estimates when @xmath3 .
this may also hold when @xmath419 , but we will not prove it here .
instead we remark that theorem [ t.fluctuations ] holds in three dimensions provided that we replace with , and that the theorem as stated probably fails in higher dimensions even if we make a such a replacement .
the reason is that is positive at points slightly outside of @xmath420 ( or outside of the support of @xmath84 ) and negative at points slightly inside .
if we replace a discrete harmonic polynomial @xmath82 with a function that agrees with @xmath82 on @xmath413 but has a different derivative along portions of @xmath43 , this may produce a non - trivial effect ( by the discussion above ) when @xmath421 .
r. durrett , _ probability : theory and examples _ , 2nd ed . ,
1995 . c. m. fortuin , p. w. kasteleyn and j. ginibre , correlation inequalities on some partially ordered sets , _ comm .
* 22 * : 89103 , 1971 .
a. ivi , e. krtzel , m. khleitner , and w.g .
nowak , lattice points in large regions and related arithmetic functions : recent developments in a very classic topic , _ elementare und analytische zahlentheorie _ , schr .
wiss . ges .
johann wolfgang goethe univ .
frankfurt am main , 20 , 89128 , 2006 .
d. jerison , l. levine and s. sheffield , internal dla : slides and audio .
_ midrasha on probability and geometry : the mathematics of oded schramm . _ http://iasmac31.as.huji.ac.il:8080/groups/midrasha_14/weblog/855d7/images/bfd65.mov , 2009 .
d. jerison , l. levine and s. sheffield , logarithmic fluctuations for internal dla .
d. jerison , l. levine and s. sheffield , internal dla in higher dimensions .
http://arxiv.org/abs/1012.3453[arxiv:1012.3453 ] g. lawler , _ intersections of random walks _ , birkhuser , 1996 .
l. levine and y. peres , strong spherical asymptotics for rotor - router aggregation and the divisible sandpile , _ potential anal . _ * 30 * ( 2009 ) , 127 .
http://arxiv.org/abs/0704.0688[arxiv:0704.0688 ] l. lovsz , discrete analytic functions : an exposition , _ surveys in differential geometry _ * ix*:241273 , 2004 . | in previous works , we showed that the internal dla cluster on @xmath0 with @xmath1 particles is almost surely spherical up to a maximal error of @xmath2 if @xmath3 and @xmath4 if @xmath5 .
this paper addresses `` average error '' : in a certain sense , the average deviation of internal dla from its mean shape is of _ constant _ order when @xmath3 and of order @xmath6 ( for a radius @xmath7 cluster ) in general .
appropriately normalized , the fluctuations ( taken over time and space ) scale to a variant of the gaussian free field . |
Subsets and Splits
No saved queries yet
Save your SQL queries to embed, download, and access them later. Queries will appear here once saved.