article
stringlengths 0
745k
| abstract
stringlengths 0
92.7k
|
|---|---|
the possibility of a fault - tolerant topological quantum computation @xcite based upon topological quasiparticles that obey non - abelian statistics ( non - abelian anyons ) @xcite motivated much recent interest in the new systems that support such quasiparticles .
the idea behind the topological quantum computation is that non - abelian anyons could be used to encode and manipulate information in a way that is resistant to error .
namely , if a quantum system has topological degrees of freedom , like non - abelian anyons , then the information contained in those degrees of freedom will be protected against errors caused by local interactions with the environment .
this provides the possibility of using such systems to perform fault - tolerant quantum computation without decoherence .
non - abelian states of matter also present the fundamental intellectual challenge of principle and of experimental realization @xcite .
the understanding of the origin and properties of non - abelian phases is far from complete and is at the frontier of current theoretical research .
the fundamental objectives are the understanding of the interplay between topology and quantum mechanics that leads to the formation of non - abelian phases and the investigation of new models that have non - abelian quasiparticles @xcite .
non - abelian anyons first appeared in the context of the fractional quantum hall ( fqh ) effect @xcite , since fqh systems are believed to have a series of exotic non - abelian states .
such states , like the pfaffian state @xcite , which is the exact ground state of quantum hall hamiltonians with three - body contact interactions , have elementary excitations that are non - abelian anyons .
similar states have also been predicted to occur in cold atoms @xcite , superconductors with @xmath0-wave pairing symmetry @xcite , hybrid systems of superconductors with topological insulators and/or semiconductors @xcite , and non - abelian lattice spin models @xcite .
although non - abelian states are associated with two - dimensional ( 2d ) systems , analogous states can be found in certain one - dimensional ( 1d ) models @xcite .
ultimately , such 1d non - abelian states must be braided in order to compute .
this could be achieved by combining 1d systems into a 2d network as previously proposed in the case of majorana fermions @xcite .
understanding how to build states that support non - abelian defects is an important building block towards topological computation . in this paper
we refine a previous proposal for a pfaffian - like state that was proposed as an ansatz for the ground state of bosonic atoms subject to three - body infinite repulsive interactions and in a 1d optical lattice @xcite .
although such three - body interactions are rare in nature , several experimentally realizable methods have been proposed to realize dominant three - body interactions between bosonic atoms in optical lattices @xcite . in particular
, three - body interactions can be efficiently simulated by mixtures of bosonic atoms and molecules under conditions that are achievable with current technology in systems of atoms and molecules in optical lattices @xcite .
therefore , the physical system that we consider is a collection of bosonic atoms and diatomic feshbach molecules trapped in a 1d optical lattice . under certain experimentally achievable conditions ,
the system can be described by an effective hamiltonian for bosonic atoms with two - body and three - body contact interactions @xcite .
we study the ground states and elementary excitations of the system in the limit of infinite repulsive three - body interactions and for a range of values of the two - body interaction strength .
the pfaffian - like ansatz was originally proposed as an ansatz for the ground - state wave function of the system in the absence of two - body interactions @xcite .
however , our results show that the pfaffian - like ansatz wave function most closely corresponds to the exact ground - state wave function of the system at some finite value of the two - body interaction strength .
the results also indicate that in the thermodynamic limit this value of the interaction strength might be close to the value where the system undergoes a quantum phase transition from the superfluid state to the mott insulating state , as previously found within the bosonization approach @xcite .
non - abelian states of matter order their constituent particles following a hidden global pattern that is not associated with the breaking of any symmetry @xcite . this leads to a degeneracy that is not based upon simple symmetry considerations and is robust against perturbations and interactions with the environment .
topological quasiparticles of such systems exhibit an exotic statistical behavior .
namely , the interchange of two identical quasiparticles takes one ground state into another .
if two different exchanges are performed consecutively among the quasiparticles , the final state of the system will depend upon the order in which these exchanges were carried out .
this ordering dependence is the reason why such states and their quasiparticles are called non - abelian or noncommutative .
in addition , the quasiparticles of a non - abelian system are neither fermions nor bosons , which motivated the name anyons .
the pfaffian - like states that we consider in this paper can not be characterized by any local order parameter and exhibit a global hidden order that is associated with the organization of bosons in identical indistinguishable clusters .
indistinguishability between the clusters is achieved by symmetrization over the subsets of coordinates of each cluster .
this symmetrization introduces the possibility of topological degeneracy in the space of quasiparticles and makes these states potential carriers of non - abelian excitations @xcite .
our explicit calculations use variationally optimized entangled - plaquette states for systems of up to 60 sites .
these are benchmarked with exact diagonalization ( ed ) studies of systems of up to 14 sites .
using the ed method , we first study ground - state properties and elementary excitations of the system for small system sizes and with periodic boundary conditions .
the pfaffian - like ansatz was originally proposed as an ansatz for the ground - state wave function of the system in the absence of two - body interactions ( vanishing two - body interaction strength ) @xcite .
however , our ed results clearly demonstrate that the pfaffian - like ansatz better approximates the ground state of the system at some finite value of the two - body interaction strength .
this interaction strength increases with increasing system size .
also , the overlap of the exact ground - state wave function at such a value of the two - body interaction strength and the pfaffian - like ansatz wave function , decreases more gradually with increasing system size in the presence of two - body repulsion than it does in the absence of the two - body interactions .
the ed results thus indicate that in the thermodynamic limit the pfaffian - like ansatz wave function most closely corresponds to the exact ground - state wave function of the system at some finite value of the two - body interaction strength . this might be close to the value where the system undergoes a quantum phase transition from the superfluid state to the mott insulating state , as previously found within the bosonization approach @xcite .
we also present preliminary evidence that these states support non - abelian excitations required for topological quantum computation .
we further study the ground - state properties of the system for larger system sizes .
motivated by the recent success of tensor network methods @xcite to numerically simulate a variety of strongly correlated models , we use the entangled - plaquette - state ( eps ) ansatz , also called the correlator - product - state ( cps ) ansatz , and the variational monte carlo ( vmc ) method @xcite .
in the eps approach , the lattice is covered with overlapping plaquettes and the ground - state wave function is written in terms of the plaquette coefficients .
configurational weights are then optimized using a vmc algorithm . here
, the plaquette coefficients that minimize the energy are found using the stochastic minimization method @xcite .
for small system sizes we find that the eps and vmc calculation gives quite accurate estimates of the ground - state energy and the one - body and two - body correlation functions . to examine the proximity of the ground state to the pfaffian - like ansatz for larger system sizes , we calculate the one - body and two - body correlation functions for the exact ground - state and the pfaffian - like ansatz wave functions and compare their asymptotic behavior .
since the eps wave function gives quite accurate estimates of the correlations within any plaquette , we estimate the asymptotic behavior of the correlation functions from the values of the correlation functions for the lattice sites within a plaquette .
we study the ground - state properties of the system for the system sizes @xmath1 and 60 sites .
the results obtained within the eps and vmc approach are consistent with the ed results for smaller system sizes and with the results for vanishing two - body interaction strength obtained previously using variational matrix product states ( mpss ) .
for the system size @xmath2 sites , the maximum system size that we have considered , the results indicate that at some finite value of the two - body interaction strength @xmath3 the exact ground - state wave function is still very close to the pfaffian - like ansatz wave function .
the paper is organized as follows . in sec .
ii we introduce the effective three - body interacting atomic hamiltonian for a system of bosonic atoms and diatomic feshbach molecules trapped in a 1d optical lattice . in sec .
iii we review the theory of the pfaffian - like states in 1d . in sec .
iv we present ed results for small system sizes and with periodic boundary conditions .
the results for larger system sizes obtained within the eps and vmc approach are presented in sec .
v. in the final section , sec .
vi , we draw our conclusions and discuss possible directions for future research .
we consider a systems of bosonic atoms and diatomic feshbach molecules trapped in a 1d optical lattice .
the system can be described by the hamiltonian @xcite @xmath4 where @xmath5 @xmath6 , \nonumber\ ] ] and @xmath7 the bosonic operators for atoms and molecules are denoted @xmath8 and @xmath9 , respectively .
the term @xmath10 describes the tunneling processes of atoms and molecules .
the term @xmath11 is the feshbach resonance term and the term @xmath12 describes the on - site atom - molecule and molecule - molecule interactions .
here @xmath13 , @xmath14 , @xmath15 , @xmath16 , and @xmath17 are hopping matrix elements for atoms and molecules and the on - site atom - atom , atom - molecule , and molecule - molecule interaction strengths , respectively .
the energy offset between open and closed channels in the feshbach resonance model is denoted @xmath18 , and @xmath19 is the coupling strength to the closed channel .
we further assume @xmath15 , @xmath16 , @xmath20 , and @xmath21 . in the limit @xmath22 ,
the formation of molecules is highly suppressed and the effective hamiltonian for the system can be obtained , to first order in @xmath23 , by projection of the hamiltonian , ( [ eq : hamiltonian1 ] ) , onto the subspace with no molecules .
the resulting effective hamiltonian is @xcite @xmath24 \\ & + & ( u_{aa}-g^2/\delta)\sum_i(a_i^\dagger)^2(a_i)^2 . \nonumber\end{aligned}\ ] ] in the limit @xmath25 , valid in typical experiments with @xmath26rb @xcite , the effective hamiltonian , ( [ eq : hamiltonian3 ] ) , further reduces to @xmath27 with @xmath28 , @xmath29 , and @xmath30 .
the effective hamiltonian , ( [ eq : hamiltonian4 ] ) , is the hamiltonian for a system of bosonic atoms in a 1d optical lattice with repulsive two- and three - body on - site interactions .
we further assume that @xmath31 : the limit accessible in typical setups with @xmath26rb atoms @xcite . in the limit @xmath32 ,
the hilbert space is projected onto the subspace of states with occupation numbers @xmath33 .
the bosonic operators subject to this condition , that is , the condition @xmath34 , are referred to as three - hard - core bosonic operators and satisfy the commutation relations @xmath35=\delta_{i , j}(1-\frac{3}{2}(a^\dagger _ { 3,i})^2(a_{3,i})^2)$ ] .
since @xmath36 , these operators can be represented by @xmath37 matrices of the form @xmath38 in terms of these three - hard - core bosons the projected effective hamiltonian is @xmath39 with @xmath40 .
we also note that within the experimental situation that we study , three - body losses are strongly suppressed ( the binding energy , and therefore the released energy , is not larger than the lattice depth ) @xcite . in this paper
we study the ground states of the hamiltonian , ( [ eq : hamiltonian5 ] ) , for a range of values of @xmath41 and at the fixed average filling factor of one boson per lattice site .
we compare the ground states to the pfaffian - like ground - state ansatz wave function @xcite .
the properties of the pfaffian - like state are reviewed in the following section .
the idea of using symmetrized indistinguishable cluster states as ansatz wave functions for non - abelian 1d bosonic liquids was proposed by paredes , kielmann , and cirac @xcite .
the pfaffian - like ansatz , inspired by the form of the ground state for fractional quantum hall bosons subject to a three - body interaction @xcite , was originally proposed as an ansatz for the ground - state wave function of the hamiltonian , ( [ eq : hamiltonian5 ] ) , at @xmath42 @xcite . for bosons in the lowest landau level subject to the three - body interaction potential @xmath43 , with @xmath44 being the complex coordinate in the 2d plane ,
the exact ground state in the limit @xmath32 is the pfaffian state @xcite @xmath45 this state is a symmetrized product of two identical laughlin states @xcite , @xmath46 with @xmath47 .
since the laughlin state of each cluster is a zero - energy eigenstate of the two - body interaction potential @xmath48 , three particles can never coincide in a state of the form of ( [ eq : pfaffian2d ] ) .
the operator that symmetrizes over the two virtual subsets of coordinates @xmath49 and @xmath50 is denoted @xmath51 .
schematic of the local projector @xmath52 at a lattice site @xmath53 .
the operator @xmath52 projects the two identical local degrees of freedom onto a new degree of freedom that is symmetric under exchange of the two components . here
the operator @xmath52 maps the single - site four - dimensional hilbert space of two species of hard - core bosons , @xmath54 and @xmath55 ( red and blue spheres ) to the single - site three - dimensional hilbert space of three - hard - core - bosons ( green spheres ) . ] an ansatz for the ground state of the hamiltonian , ( [ eq : hamiltonian5 ] ) , at @xmath42 was proposed @xcite in direct analogy with the wave function , ( [ eq : pfaffian2d ] ) : @xmath56 this ansatz has the same form as the pfaffian state , ( [ eq : pfaffian2d ] ) , with the laughlin state replaced by a tonks - girardeau state @xcite : @xmath57 the tonks - girardeau state , ( [ eq : tgstate ] ) , is the ground state of 1d lattice hard - core bosons described by the hamiltonian @xmath58 and with periodic boundary conditions @xcite .
the hard - core bosonic operators @xmath59 obey @xmath60 , allowing only occupation numbers of @xmath61 or 1 boson per site . here
@xmath62 and @xmath63 , with @xmath64 being the number of lattice sites . to write the ansatz wave function in second quantized form
, we define a projection operator @xmath65 such that latexmath:[\[\label{eq : pfaffian1d_2 } the projection operator @xmath65 is a local operator of the form @xmath67 where @xmath64 is the number of lattice sites and @xmath68 is the local projector at a lattice site @xmath53 , @xmath69 @xmath52 maps the single site four - dimensional hilbert space of two species of hard - core bosons , @xmath54 and @xmath55 , to the single - site three - dimensional hilbert space of three - hard - core bosons as illustrated in fig . 1 . it can be further shown that the one- and two - body correlation functions for the ansatz wave function , ( [ eq : pfaffian1d_2 ] ) , have the asymptotic behavior @xcite @xmath70 for large @xmath71 and for a large system size @xmath64 .
the two - body correlation function corresponds to the one - particle correlation function for on - site pairs @xcite , @xmath72 where @xmath73 , for large @xmath71 , is the well known result for a tonks - girardeau gas @xcite . in other words , although the system of atoms has some kind of coherence , with slowly decaying spatial correlations @xmath74 , the underlying system of on - site pairs is in a much more disordered state with a fast decay of spatial correlations @xmath75 .
overlap of the exact ground - state wave function of the hamiltonian , ( [ eq : hamiltonian5 ] ) , and the ansatz wave function , ( [ eq : pfaffian1d_2 ] ) , as a function of the two - body interaction strength @xmath41 for system sizes of @xmath76 sites . here
the filling factor @xmath77 , with @xmath78 being the number of particles . ]
we first examine the ground - state properties and elementary excitations of the system for small system sizes and with periodic boundary conditions using the ed method .
the ground - state properties are calculated for a range of values of the two - body interaction strength @xmath41 and at the filling factor of one particle per site .
we calculate the overlap of the exact ground - state wave function of the hamiltonian , ( [ eq : hamiltonian5 ] ) , and the ansatz wave function , ( [ eq : pfaffian1d_2 ] ) , the average occupation of sites with one and two particles , and the one- and two - body correlation functions .
+ overlap of the exact ground - state wave function of the hamiltonian , ( [ eq : hamiltonian5 ] ) , and the ansatz wave function , ( [ eq : pfaffian1d_2 ] ) , at the filling factor @xmath79 for the values of the two - body interaction strength @xmath80 ( red symbols ) and @xmath3 , where the overlap @xmath81 is maximal ( blue symbols ) .
dashed gray lines are included as guides for the eye and were obtained by the extrapolation of the ed results for @xmath76 . ] the value of the two - body interaction strength @xmath41 where the overlap @xmath81 is maximal ( @xmath82 ) for system sizes of @xmath76 sites and at filling factor @xmath79 .
dashed gray lines are included as guides for the eye and were obtained by the extrapolation of the ed results for @xmath76 . ]
average number of sites with one particle @xmath83 ( blue and red symbols ) and two particles @xmath84 ( green and purple symbols ) for the exact ground - state wave function of the hamiltonian , ( [ eq : hamiltonian5 ] ) , at @xmath3 ( blue and green symbols ) and the ansatz wave function , ( [ eq : pfaffian1d_2 ] ) , ( red and purple symbols ) . here
the filling factor @xmath79 and @xmath85 denotes the value of the two - body interaction strength where the overlap @xmath81 is maximal .
dashed gray lines are included as guides for the eye and were obtained by the extrapolation of the ed results for @xmath76 . ] the overlap of the exact ground - state wave function and the pfaffian - like ansatz wave function for the system sizes @xmath76 and as a function of the two - body interaction strength @xmath86 is shown in fig .
[ fig : overlap ] .
the results clearly demonstrate that the pfaffian - like ansatz wave function , ( [ eq : pfaffian1d_2 ] ) , is a better ansatz for the exact ground - state wave function of the hamiltonian , ( [ eq : hamiltonian5 ] ) , at some finite value of the two - body interaction strength @xmath85 than it is for the exact ground - state wave function at @xmath42 as suggested previously @xcite . also , the overlap decreases more gradually with increasing system size @xmath64 at @xmath85 than it decreases at @xmath42 ( fig .
[ fig : overlap0uc ] ) .
this indicates that in the thermodynamic limit ( @xmath87 ) , the pfaffian - like ansatz wave function most closely corresponds to the exact ground - state wave function of the hamiltonian , ( [ eq : hamiltonian5 ] ) , at some finite value of the two - body interaction strength @xmath85 . the one - body ( blue and red symbols ) and two - body ( green and purple symbols ) correlation functions for the exact ground - state wave function of the hamiltonian , ( [ eq : hamiltonian5 ] ) , at @xmath3 ( blue and green symbols ) and the ansatz wave function , ( [ eq : pfaffian1d_2 ] ) ( red and purple symbols ) , for the system size @xmath88 . here
the filling factor @xmath79 and @xmath85 denotes the value of the two - body interaction strength where the overlap @xmath81 is maximal .
the long - distance scaling of the correlation functions @xmath89 and @xmath90 is @xmath91 and @xmath92 for the exact ground state and @xmath93 and @xmath94 for the ansatz wave function . ] the ed results also indicate that the value of @xmath82 where the overlap is maximal increases with increasing system size ( fig .
[ fig : uc ] ) .
it can also be shown that the value of @xmath95 where the system undergoes a quantum phase transition from the superfluid state to the mott insulating state , @xmath96 , decreases with increasing system size .
the ed results thus suggest that the value of @xmath82 approaches the value of @xmath96 with an increase in the system size @xmath64 and that the pfaffian - like state might be the state at the superfluid - to - mott insulator boundary as previously found within the bosonization approach @xcite .
we have further calculated the average number of sites with one particle , @xmath97 , and with two particles , @xmath98 , at the filling factor @xmath79 and for a range of values of the two - body interaction strength @xmath86 .
the average number of sites with one particle and with two particles is @xmath99 @xmath100 where @xmath101 and @xmath102 .
the values of @xmath97 and @xmath98 for the pfaffian - like ansatz wave function , ( [ eq : pfaffian1d_2 ] ) , and for the exact ground - state wave function of the hamiltonian , ( [ eq : hamiltonian5 ] ) , at @xmath3 ( where the overlap @xmath81 is maximal ) are shown in fig .
[ fig : n1n2 ] .
the values of @xmath97 and @xmath98 for the exact ground - state wave function are very close to the values of @xmath97 and @xmath98 for the pfaffian - like ansatz wave function , as can be clearly seen in fig .
[ fig : n1n2 ] .
we have also calculated the one - body and two - body correlation functions , @xmath103 for the exact ground - state wave function at @xmath3 and for the ansatz wave function for a system of @xmath88 sites .
the results in fig .
[ fig : lnc ] show that the correlation functions for the exact and ansatz wave functions show very similar asymptotic behavior .
overlap @xmath104 , ( [ eq : ovl_es ] ) , for two degenerate first excited states of the hamiltonian , ( [ eq : hamiltonian5 ] ) , at the filling factor @xmath79 for values of the two - body interaction strength @xmath80 ( red symbols ) and @xmath3 , where the overlap @xmath81 for the ground state of the hamiltonian , ( [ eq : hamiltonian5 ] ) , is maximal ( blue symbols ) . dashed gray lines are included as guides for the eye and were obtained by the extrapolation of the ed results for @xmath76 . ]
elementary excitations further reveal the topological nature of the pfaffian - like state . by construction ,
the pfaffian - like ansatz , ( [ eq : pfaffian1d_2 ] ) , has a hidden global order associated with the organization of particles in two identical indistinguishable copies of the same state .
consequently , the elementary excitations above the pfaffian - like ground - state ansatz exhibit non - abelian statistics @xcite . the argument for this proceeds as follows .
the elementary excitations can be constructed by creating a quasihole in each of the copies and symmetrizing @xcite .
symmetrization leads to a topological degeneracy in the subspace of elementary excitations and non - abelian algebra of exchanges of elementary excitations ( quasiholes ) @xcite .
if the ground state of the system is close to the pfaffian - like ansatz , elementary excitations above it also exhibit non - abelian statistics . to confirm this statement we compare excited states of the hamiltonian , ( [ eq : hamiltonian5 ] ) , with the ansatz wave functions for the excited states @xcite @xmath105 are eigenstates for each of two copies .
in particular , we calculate the overlap of the first excited state of the hamiltonian , ( [ eq : hamiltonian5 ] ) , and corresponding ansatz wave function for the first excited state , ( [ eq : ansatz_es ] ) . for a fixed number of particles @xmath78 ( with @xmath78 either even or odd ) an excited state with @xmath106 elementary excitations ( quasiholes ) that are su(2)@xmath107 ( ising ) anyons ( similar to majorana fermions ) is expected to have topological degeneracy @xmath108 @xcite .
we find that the first excited state of the hamiltonian , ( [ eq : hamiltonian5 ] ) , is twofold degenerate and corresponds to a state with four quasiholes .
the overlap for each of the two degenerate first excited states is calculated by considering the total overlap with the manifold of degenerate ansatz states , ( [ eq : ansatz_es ] ) , that correspond to the first excited states .
we find four degenerate , linearly independent , ansatz states that correspond to two degenerate first excited states of the hamiltonian , ( [ eq : hamiltonian5 ] ) .
the degeneracy of the exact first excited state is two , and not four , since the hamiltonian , ( [ eq : hamiltonian5 ] ) , does not have particle - hole symmetry .
the overlap is then given by @xmath109 where @xmath110 , with @xmath111 , are two degenerate first excited states of the hamiltonian , ( [ eq : hamiltonian5 ] ) , and @xmath112 , with @xmath113 , are corresponding degenerate ansatz states .
we note that the states @xmath112 form an orthonormal basis within the degenerate manifold , which leads to expression ( [ eq : ovl_es ] ) for the total overlap .
we find that @xmath114 for all values of the two - body interaction strength @xmath41 that we have considered .
the overlap of the first excited states and corresponding ansatz wave functions is shown in fig .
[ fig : overlap0uc_es ] . in agreement with the results for the ground - state wave function ,
the overlap decreases more gradually with increasing system size @xmath64 at @xmath85 than it decreases at @xmath42 ( fig .
[ fig : overlap0uc ] ) .
the results also demonstrate that the overlap with the ansatz wave function for four quasiholes is very close to 1 , indicating that the elementary excitations of the system are non - abelian .
illustration of the six - site 1d plaquettes with a five - site overlap between the plaquettes used in the calculation of the ground - state properties of the system for larger system sizes . ] studying properties of the system for larger system sizes with the ed method is not possible due to the rapid increase in the hilbert - space size with increasing system size .
motivated by successes of the tensor network methods @xcite to numerically simulate a variety of strongly correlated models , we further study the properties of the system for larger system sizes using the entangled - plaquette - state ansatz optimized using the variational monte carlo method @xcite . within the eps approach , also called the correlator - product - state approach , the lattice is covered with overlapping plaquettes and the ground - state wave function is written in terms of the plaquette coefficients .
configurational weights can then be optimized using a vmc algorithm . for a lattice with @xmath64 sites
, an arbitrary quantum many - body wave function can be written as @xmath115 where @xmath116 denotes the vector of occupancies and @xmath117 is the amplitude or weight of a given configuration @xmath118 . for the system described by the hamiltonian ( [ eq : hamiltonian5 ] ) @xmath119 for the lattice sites @xmath63 .
+ in the eps ( cps ) description of a bosonic system , the weight @xmath117 is expressed as a product of the plaquette coefficients over the lattice @xmath120 where @xmath121 is the occupancy vector of the @xmath122-site plaquette @xmath0 . in many cases , the qualitative behavior of large systems
can be described even by plaquettes with a small number of sites .
the eps wave function corresponding to the ground - state wave function of the system gives reasonable estimates of the ground - state energy and short - range correlations .
the estimates improve with an increase in plaquette size and greater overlap between the plaquettes . here
we choose six - site 1d plaquettes with a five - site overlap between the plaquettes as illustrated in fig .
[ fig : plaquettes ] .
the weights @xmath123 for the ground - state wave function of hamiltonian ( [ eq : hamiltonian5 ] ) with periodic boundary conditions can then be written as @xmath124 .ground - state energy per site ( in units of @xmath125 ) for the system size @xmath88 and with periodic boundary conditions . here
@xmath126 stands for the relative error with respect to the exact ground - state energy and is defined as @xmath127 . [
cols="^,^,^,^,^,^,^",options="header " , ] within the variational mps approach@xcite this plaquette choice would correspond to matrices of dimension @xmath128 , with @xmath129 being the number of overlapping lattice sites . in a vmc algorithm
the energy @xmath130 is written as @xmath131 where it is assumed that the wave function @xmath132 is not normalized , and the local energy @xmath133 and the probability @xmath134 are given by @xmath135 the expectation value of any operator @xmath136 can be expressed in the same form by replacing the hamiltonian @xmath137 with the operator @xmath136 .
the probability @xmath138 is never explicitly calculated from eq .
( [ eq : e_vmc_2 ] ) . instead , for a given set of plaquette coefficients ,
the energy can be efficiently computed using the metropolis algorithm @xcite .
the one - body ( red and blue symbols ) and two - body ( green and purple symbols ) correlation functions obtained using the ed ( red and green symbols ) and the eps and vmc ( blue and purple symbols ) methods for the system size @xmath88 and at the value of the two - body interaction strength @xmath139 . ] within the metropolis algorithm , used to sample the probability distribution , the overall energy can be efficiently computed as an average of the sampled local energies . in our calculation the total number of atoms @xmath78 is fixed .
we start from a randomly chosen initial configuration @xmath140 , with @xmath141 and @xmath142 for @xmath63 , and then generate via the metropolis algorithm a large set of new configurations by replacing @xmath143 with @xmath144 ( if @xmath145 ) and @xmath146 with @xmath147 ( if @xmath148 ) at two neighboring sites @xmath53 and @xmath149 . starting from configuration @xmath150 , the acceptance probability of a new configuration @xmath151 is given by @xmath152.\ ] ] according to the variational principle , minimization of expression ( [ eq : e_vmc ] ) with respect to the weights gives an upper bound of the ground - state energy .
the plaquette coefficients that minimize the energy can be found by using the stochastic minimization method @xcite , which requires only the first derivative of the energy with respect to the plaquette coefficients , which is given by @xmath153\right\},\ ] ] where the wave function @xmath154 in eq .
( [ eq : e_vmc ] ) is approximated by the eps ( cps ) wave function and @xmath155 with @xmath117 given by eq .
( [ eq : wn_6 ] ) . here
@xmath156 denotes the number of times the plaquette coefficient @xmath157 appears in the product , ( [ eq : wn_6 ] ) , for the amplitude @xmath117 for configuration @xmath150 .
if the same plaquette coefficient is used for multiple sites ( e.g. , for a translationally invariant choice of plaquettes ) , @xmath158 .
if each plaquette coefficient is used once , @xmath159 .
equivalently to the overall energy the first derivative can be efficiently calculated using the metropolis algorithm from the same sample used to compute the overall energy . the one - body ( @xmath160 ) and two - body ( @xmath161 ) correlation functions for the system sizes @xmath1 ( red symbols ) , and @xmath2 ( blue symbols ) sites and at the value of the two - body interaction strength @xmath80 .
here the asymptotic behavior of the correlation functions is @xmath162 and @xmath163 .
] the steps of the vmc algorithm used to calculate the ground - state properties of the system are as follows : ( i ) start from the randomly chosen complex values for the plaquette coefficients , ( ii ) evaluate the energy and its gradient vector , ( iii ) update all plaquette coefficients @xmath157 according to @xmath164 and ( iv ) iterate from ( ii ) until convergence of the energy is reached .
here @xmath165 is a random number between @xmath166 and @xmath167 , and @xmath168 is the step size for a given iteration @xmath169 . the one - body ( @xmath160 ) and two - body ( @xmath161 ) correlation functions for the system size @xmath2 sites and at the values of the two - body interaction strength @xmath80 ( red symbols ) , @xmath170 ( purple symbols ) , and @xmath171 ( blue symbols ) .
here the asymptotic behavior of the correlation functions is @xmath162 and @xmath163 . ] in each iteration @xmath169 , the energy and its derivative are estimated from @xmath172 values , where @xmath64 is the number of lattice sites and @xmath173 is called the number of sweeps per sample . in a given sweep each lattice site is visited sequentially and a move , @xmath174 , to a new configuration is proposed by changing the occupancy numbers @xmath175 ( for @xmath145 ) and @xmath176 ( for @xmath148 ) at two neighboring lattice sites , @xmath53 and @xmath149 . also , to achieve convergence and reach the optimal energy value it is important to carefully tune the gradient step @xmath168 . for each iteration @xmath169 ,
the number of sweeps @xmath177 is increased linearly , @xmath178 , and the procedure of evaluating the energy and updating the coefficients is repeated @xmath179 times .
the step size is gradually reduced per iteration . here
we use a geometric form , @xmath180 , with @xmath181 .
the number of sweeps per iteration is increased because the derivatives become smaller as the energy minimum is approached and require more sampling in order not to be dominated by noise .
an increasing @xmath182 effectively corresponds to a slower cooling rate .
here we take @xmath183 , @xmath184 , and @xmath181 .
the initial minimization routine is performed with @xmath185 for 50 iterations .
the resulting plaquette coefficients are then used as a starting point for a new run of 50 iterations with @xmath186 .
after the minimization is complete the expectation values are calculated by repeating the procedure for a single iteration with 0 step size and large @xmath177 and @xmath182 to obtain more accurate estimates of the expectation values .
it is also important to note that it is more difficult to obtain good estimates of the ground - state energies and the correlation functions for small system sizes due to the presence of the statistical error in the stochastic algorithm .
having a larger number of parameters allows the optimization method more freedom in finding the minimum energy state and the statistical error can be controlled by increasing the system size .
the results for the ground - state energy for the system size @xmath88 and for several values of the two - body interaction strength @xmath41 are listed in tab . [
table : e0 ] and compared to the ed results ( exact ground - state energy values ) . as can be seen in table [ table : e0 ] , the relative error , defined as @xmath187 does not exceed @xmath188 for any value of the two - body interaction strength @xmath41 .
the eps and vmc calculation also gives quite accurate estimates of the correlation functions as demonstrated in fig .
[ fig : cf_12_l14 ] . to examine the proximity of the ground - state wave function for larger system sizes to the pfaffian - like ansatz wave function
, we further calculate correlation functions for system sizes @xmath1 and 60 sites . since the eps wave function gives quite accurate estimates of the correlations within any plaquette @xmath0 , to estimate the asymptotic behavior of the correlation functions we calculate the one - body and two - body correlation functions , ( [ eq : c1c2 ] ) , for the lattice sites within a plaquette @xmath0 ( @xmath189 ) .
the results for the one - body and two - body correlation functions at @xmath42 and for the system sizes @xmath1 and 60 sites are shown in fig . [
fig : cf_u0 ] .
the values of @xmath190 and @xmath191 that describe the asymptotic behavior of the correlation functions , @xmath192 move away from the values expected for the pfaffian - like state ansatz ( @xmath193 and @xmath194 ) with increasing system size .
the results thus suggest that the overlap between the exact ground - state wave function at @xmath42 and the pfaffian - like ansatz decreases with increasing system size .
this is consistent with the ed results and with the previously obtained results for the system sizes @xmath195 obtained using the variational mps @xcite .
the ed results presented in the previous section also suggest that the pfaffian - like ansatz wave function , ( [ eq : pfaffian1d_2 ] ) , better approximates the exact ground state of the hamiltonian , ( [ eq : hamiltonian5 ] ) , at some finite value of the two - body interaction strength @xmath3 than it approximates the exact ground - state wave function at @xmath42 . the results for the one - body and two - body correlation functions for the system size @xmath2 sites ( fig .
[ fig : cf_l60 ] ) show that the values of @xmath190 and @xmath191 increase with an increase in the value of the two - body interaction strength @xmath196 .
in other words , at some value @xmath85 the values of @xmath190 and @xmath191 will be the closest to the values expected for the pfaffian - like ansatz wave function ( @xmath193 and @xmath194 ) .
this indicates that the overlap between the exact ground - state wave function and the pfaffian - like ansatz wave function is maximal at @xmath85 , in agreement with the ed results for smaller system sizes .
previous calculations with variational mps found the values of @xmath190 and @xmath191 for the system size @xmath197 sites to be @xmath198 and @xmath199 for the exact ground - state wave function at @xmath42 and @xmath200 and @xmath201 for the pfaffian - like ansatz wave function @xcite .
the corresponding overlap between the exact ground - state and the pfaffian - like state wave functions was found to be @xmath202 @xcite .
also , for the system size @xmath1 sites at @xmath42 the overlap is @xmath203 @xcite , which corresponds to
@xmath204 and @xmath205 obtained within our eps and vmc calculation for the exact ground - state wave function ( fig .
[ fig : cf_u0 ] ) and @xmath206 and @xmath207 for the pfaffian - like ansatz wave function .
it is difficult to determine the exact values of @xmath85 from our eps and vmc calculation .
however , for the system size @xmath2 sites , the maximum system size that we have considered , we find that @xmath208 and @xmath209 at @xmath3 . therefore , based on the results mentioned in the previous paragraph , we estimate that the overlap between the exact ground - state and the pfaffian - like - state wave functions is still good for the system size @xmath2 and at @xmath3 .
we have studied ground states and elementary excitations of a system of bosonic atoms and diatomic feshbach molecules trapped in a 1d optical lattice . under certain conditions , which are experimentally achievable with current technology in systems of cold atoms and molecules in optical lattices ,
the system can be described by an effective hamiltonian for bosonic atoms with two- and three - body interactions .
we have considered the limit of infinitely strong three - body interactions for a range of values of the two - body interaction strength .
the ground - state properties of the system were calculated using the ed method for small system sizes , and the eps and vmc method for larger system sizes .
the pfaffian - like ansatz was originally proposed as an ansatz for the ground - state wave function of the effective hamiltonian in the absence of two - body interactions .
however , our results clearly demonstrated that the pfaffian - like ansatz wave function is a better ansatz for the ground - state wave function of the effective hamiltonian at some finite value of the two - body interaction strength .
this value of the two - body interaction strength might be close to the value where the system undergoes a quantum phase transition from the superfluid state to the mott insulating state , as previously found within the bosonization approach .
we also demonstrate that these states support non - abelian excitations required for quantum computation .
further work is necessary to find an experimentally realizable model with a ground - state wave function that can be even better approximated by the pfaffian - like ansatz wave function .
this can possibly be achieved in a system with long - range interactions .
an additional direction for future research is to consider similar 2d non - abelian models , for example , anisotropic systems consisting of coupled interacting 1d wires .
such anisotropic 2d lattice models can have interesting non - abelian chern insulating phases and fractional topological insulating phases . in order to use non - abelian states for quantum computation
, one must be able to braid them .
although this is not possible for a strictly 1d system , creating a network of such 1d systems connected by t - junctions , as suggested previously in the context of majorana quantum wires @xcite , potentially allows this . in the case of majorana quantum wires
a t - junction @xcite allows for adiabatic exchange of two majorana fermions .
such a t - junction has topological and non - topological regions that can be controlled by individually tunable gates . in principle ,
similar t - junction networks can be created for the bosonic system that we have considered , where topological and nontopological regions of the network can be controlled by tuning the two - body interaction in different regions of the t - junction network ( for example , by changing the depth of the optical lattice in certain regions of the t - junction ) . in a similar manner to fractional quantum hall states ,
non - abelian anyons can be created by creating pairs of quasiholes , with one quasihole in each cluster @xcite .
these non - abelian anyons are ising anyons [ su(2)@xmath107 anyons ] , similar to majorana fermions , and excitations of the @xmath210 fractional quantum hall state ( pfaffian state ) .
we also note that braiding of su(2)@xmath107 anyons alone does not permit universal quantum computation @xcite .
however , to obtain a universal set of gates , braiding of su(2)@xmath107 anyons needs to be strengthened only by a single - qubit @xmath211 phase gate and a two - qubit measurement @xcite .
also , pfaffian - like states obtained by symmetrization of two identical copies can be generalized to states obtained by symmetrization of @xmath169 identical copies that support su(2)@xmath212 anyons @xcite and can be used for universal quantum computation .
for example su(2)@xmath213 anyons
( like fibonacci anyons ) can be used for universal quantum computation @xcite .
the results presented here thus constitute an important step towards understanding generalized states that support su(2)@xmath212 anyons .
we thank jakub zakrzewski for very helpful suggestions and discussions .
we acknowledge support from the epsrc through grants nos .
ep / k02163x/1 , and ep / i004831/2 , and topnes program grant no .
ep / i031014/1 .
n. chancellor was funded by lockheed martin corporation at the time this work was carried out . t. uri also acknowledges support from the eu grant quic ( h2020-fetproact-2014 , grant no .
641122 ) .
a. y. kitaev , ann .
( n.y . ) * 303 * , 2 ( 2003 ) .
m. h. freedman , m. larsen , and z. wang , commun .
phys . * 227 * , 605 ( 2002 ) .
s. das sarma , m. freedman , and c. nayak , phys .
today * 59 * , 32 ( 2006 ) . c. nayak , s. h. simon , a. stern , m. freedman , and s. das sarma , rev .
phys . * 80 * , 1083 ( 2008 ) .
j. k. pachos , _ introduction to topological quantum computation _
( cambridge university press , cambridge , uk , 2012 ) .
j. k. pachos , and s. h. simon , new j. phys .
* 16 * , 065003 ( 2014 ) .
g. moore and n. read , nucl .
b , * 360 * , 362 ( 1991 ) .
a. stern , ann .
* 323 * , 204 ( 2008 ) .
a. stern , nature * 464 * , 187 ( 2010 ) .
s. das sarma , m. freedman , and c. nayak , phys .
lett . * 94 * , 166802 ( 2005 ) .
a. stern , and b. i. halperin , phys .
rev . lett . * 96 * , 016802 ( 2006 ) .
p. bonderson , a. kitaev , and k. shtengel , phys .
lett . * 96 * , 016803 ( 2006 ) .
a. y. kitaev , ann . phys .
* 321 * , 2 ( 2006 ) .
j. c. y. teo , and c. l. kane , phys .
b * 89 * , 085101 ( 2014 ) .
m. greiter , x.g .
wen , and f. wilczek , nucl .
b * 374 * , 567 ( 1992 ) . c. nayak and f. wilczek , nucl . phys .
b * 479 * , 529 ( 1996 ) .
n. read , e. rezayi , phys .
b * 54 * , 16864 ( 1996 ) .
n. read , e. rezayi , phys .
b * 59 * , 8084 ( 1999 ) . n. k. wilkin and j.m.f .
gunn , phys .
* 84 * , 6 ( 2000 ) .
a. cappelli , l. s. georgiev , and i. t. todorov , nucl .
b * 599 * , 499 ( 2001 ) .
n. r. cooper , n. k. wilkin , and j. m. f. gunn , phys .
rev . lett . * 87 * , 120405 ( 2001 ) .
v. gurarie , and l. radzihovsky , ann .
* 322 * , 2 ( 2007 ) .
g. mller and n. r. cooper , phys .
* 103 * , 105303 ( 2009 ) .
t. uri , and a. lazarides , phys . rev .
b * 86 * , 115135 ( 2012 ) .
h . wu and j. k. jain , phys .
b * 87 * , 245123 ( 2013 ) .
a. sterdyniak , b. a. bernevig , n. r. cooper , and n. regnault , phys .
b * 91 * , 035115 ( 2015 ) .
l. fu , and c. l. kane , phys .
lett . * 100 * , 096407 ( 2008 ) .
l. fu , and c. l. kane , phys .
lett . * 102 * , 216403 ( 2009 ) .
j. nilsson , a. r. akhmerov , and c. w. beenakker , phys .
* 101 * , 120403 ( 2008 ) .
p. nikoli , t. uri , and z. teanovi , phys .
lett . * 110 * , 176804 ( 2013 ) .
j. d. sau , r. m. lutchyn , s. tewari , and s. das sarma , phys .
104 * , 040502 ( 2010 ) .
m. a. levin , and x. g. wen , phys .
b * 71 * , 045110 ( 2005 ) .
a. m. tsvelik , sov .
phys . * 47 * ( 1988 ) 172 - 175 , yad . fiz . * 47 * ( 1988 ) 272 - 278 .
a. m. tsvelik , phys .
lett . * 113 * , 066401 ( 2014 ) .
p. fendley , and h. saleur , phys .
d * 65 * , 025001 ( 2001 ) .
tu , phys .
b * 87 * , 041103(r ) ( 2013 ) . b. paredes , t. keilmann , and j. i. cirac , phys . rev . a * 75 * , 053611 ( 2007 ) .
b. paredes , phys .
b * 85 * , 195150 ( 2012 ) .
a. e. b. nielsen , j. i. cirac , and g. sierra , j. stat .
2011 , p11014 .
m. greiter , _ mapping of parent hamiltonians : from abelian and non - abelian quantum hall states to exact models of critical spin chains .
springer tracts in modern physics , vol .
244 _ ( springer - verlag , berlin , 2011 ) .
r. thomale , s. rachel , p. schmitteckert , and m. greiter , phys .
b * 85 * , 195149 ( 2012 ) . j. alicea , y. oreg , g. refael , f. von oppen , and m. p. a. fisher , nature phys .
* 7 * , 412 ( 2011 ) .
cooper , phys .
lett . * 92 * , 220405 ( 2004 ) . s. will , t. best , u. schneider , l. hackermuller , d .- s .
luhmann , and i. bloch , nature * 465 * , 197 ( 2010 ) .
a. j. daley , and j. simon , phys .
rev . a * 89 * , 053619 ( 2014 ) .
d. s. petrov , phys .
rev . a * 90 * , 021601(r ) ( 2014 ) .
d. s. petrov , phys .
lett . * 112 * , 103201 ( 2014 ) .
lee , y .- l .
lee , and m .- f .
yang , phys .
b * 76 * , 075117 ( 2007 ) .
wen , _ quantum field theory of many - body systems _
( oxford university press , oxford , uk , 2004 ) .
b. paredes , arxiv:1402.3567 .
r. ors , ann . phys . * 349 * , 117 ( 2014 ) . h. j. changlani , j. m. kinder , c. j. umrigar , and g. k .- l .
chan , phys .
b * 80 * , 245116 ( 2009 ) .
f. mezzacapo and j. i. cirac , new j. phys .
* 12 * , 103039 ( 2010 ) .
e. neuscamman , h. changlani , j. kinder , and g. k .- l .
chan , phys .
b * 84 * , 205132 ( 2011 ) .
f. mezzacapo , phys .
b * 86 * , 045115 ( 2012 ) .
f. mezzacapo , n. schuch , m. boninsegni , and j. i. cirac , new j. phys .
* 11 * , 083026 ( 2009 ) .
f. mezzacapo , phys .
b * 83 * , 115111 ( 2011 ) .
f. mezzacapo and m. boninsegni , phys .
b * 85 * , 060402(r ) ( 2012 ) .
e. neuscamman and g. k .- l .
chan , phys .
b * 86 * , 064402 ( 2012 ) .
s. al - assam , s. r. clark , c. j. foot , and d. jaksch , phys .
b * 84 * , 205108 ( 2011 ) .
t. uri , n. chancellor , and i. f. herbut , phys .
b * 89 * , 165123 ( 2014 ) .
a. w. sandvik and g. vidal , phys .
lett . * 99 * , 220602 ( 2007 ) .
j. lou and a. w. sandvik , phys .
b * 76 * , 104432 ( 2007 ) .
e. timmermans , p. tommasini , r. ct , m. hussein , and a. kerman phys .
lett . * 83 * , 2691 ( 1999 ) .
m. holland , j. park , and r. walser , phys .
* 86 * , 1915 ( 2001 ) .
r. b. laughlin , phys .
lett . * 50 * , 1395 ( 1983 ) .
m. girardeau , j. math . phys .
* 1 * , 516 ( 1960 ) . b. paredes , a. widera , v. murg , o. mandel , s. flling , i. cirac , g.v .
shlyapnikov , t. w. hnsch and i. bloch , nature * 429 * , 277 ( 2004 ) . n. metropolis , a. w. rosenbluth , m. n. rosenbluth , a. h. teller , and e. teller , j. chem . phys .
* 21 * , 1087 ( 1953 ) .
w. m. c. foulkes , l. mitas , r. j. needs , and g. rajagopal , rev .
phys . * 73 * , 33 ( 2001 ) . | we study ground states and elementary excitations of a system of bosonic atoms and diatomic feshbach molecules trapped in a one - dimensional optical lattice using exact diagonalization and variational monte carlo methods .
we primarily study the case of an average filling of one boson per site . in agreement with bosonization theory ,
we show that the ground state of the system in the thermodynamic limit corresponds to the pfaffian - like state when the system is tuned towards the superfluid - to - mott insulator quantum phase transition .
our study clarifies the possibility of the creation of exotic pfaffian - like states in realistic one - dimensional systems .
we also present preliminary evidence that such states support non - abelian anyonic excitations that have potential application for fault - tolerant topological quantum computation . |
spacetimes with more than three spatial dimensions have been attracting interest for long time .
first proposals go back to kaluza and klein @xcite , in their attempt to unify electromagnetism with einstein gravity by proposing a theory with a compact fifth dimension of a size of the order of the planck length . later on
the emphasis shifted to a
brane world " picture @xcite with our world being confined to a subspace of higher - dimensional spacetime .
more recently it was put forward , inspired also in @xcite , that at energy scales of the standard model matter can not propagate into extra dimensions whilst gravity on the other hand can permeate all over . within brane worlds extra dimensions may be compact but large @xcite , or infinite and warped @xcite .
they have opened a rich and interesting phenomenological model - building aiming at solving long standing problems like the hierarchy and cosmological constant problem within high - energy physics @xcite , as well as testing models with observational data at the cosmological level @xcite .
we should stress at this point that to implement the brane scenario one looks for physical mechanisms which localize higher dimensional fields to a lower dimensional brane .
localization of gauge fields has turned out to be more difficult than that of scalar and spinors , however , some mechanisms have been developed in @xcite .
the one in @xcite is a particularly simple scenario that was probed to localize electromagnetic fields .
it consists of a single ( 3+@xmath0)-brane embedded in a ( 3+@xmath0 + 1)-dimensional space containing @xmath0 compact extra dimensions and one infinite and warped .
this is a ( 5+@xmath0)-dimensional brane world that we refer to as a randall - sundrum ii model modified by @xmath0 extra compact dimensions ( rsiin ) . as for fermions some mechanisms
have been developed in @xcite .
we shall combine some of these ideas for our purposes below .
it is generally assumed within brane models that relevant effects appear at high energies @xcite whereas low - energy effects have no chance to become relevant .
this is not necessarily the case since some high - precision tests may be sensible to such tiny effects or else , at worst , it may be possible to set bounds for the different parameters of the models @xcite . in this paper
we investigate two low - energy effects for rsiin .
the first one involves determining the electromagnetic potentials of point charge in uniform motion in a 3-dimensional subspace of the single ( 3+@xmath0)-brane . considering that experiments in atomic physics are sensitive to small frecuency shifts down to 1 mhz @xcite , one can in principle use the atom as an instrument to measure the anti - de sitter radius @xmath1 .
the second problem we investigate is a hydrogen atom , localized on the brane , where the binding electromagnetic field is just the one obtained in the first part .
we have calculated the @xmath2 splitting coming from the presence of the extra dimensions that corresponds to the lamb shift associated to qed .
this has allowed us to set bounds for the parameter of the scenario , namely the anti - de sitter radius or , equivalently , the tension of the brane . in the present work we provide details and extend some of the results of a previous letter @xcite .
the paper is organized as follows . in section
ii we describe the rsiin brane world and describe the localization mechanism for the electromagnetic and spinor fields .
section iii is devoted to calculate the effective electromagnetic potentials for a test point charge in uniform motion thus providing details and generalizing the static results of @xcite . in section
iv the @xmath2 shift of a localized hydrogen atom is calculated completely . finally in section
v we focus on the discussion of our results as well as some perspectives .
now we describe the brane world we use in the present work . looking for a localization mechanism for gauge fields as due to the gravity associated with the brane it was proved in @xcite that the following setting would accomplish such task : a single ( 3+@xmath0)-brane embedded in a ( 3+@xmath0 + 1)-dimensional space so that space - time is ( @xmath3)dimensional with @xmath0 compact dimensions and one infinite and warped .
we call such a scenario rsiin .
much as in the case of the randall - sundrum models to solve einstein equations the brane tension and the negative bulk cosmological constant must be fine - tuned .
the resulting metric is @xcite @xmath4 where @xmath5 is the four - dimensional minkowski metric , @xmath6 @xmath7 @xmath8 $ ] are compact coordinates and @xmath9 are the sizes of the compact dimensions .
@xmath1 is the anti - de sitter radius .
since our purpose is to study low energy - physics effects and hence the situation where electromagnetic fields and spinors are localized to a 3-dimensional subspace of the ( 3+@xmath0)-brane we next describe how this works in rsiin .
there has been extensive discussion about the localization of gauge fields in brane - worlds @xcite .
here we just reproduce the proposal in @xcite for the rsiin scenario .
consider a gauge field @xmath10 , with action @xmath11 for low - energies , hence @xmath9 small , and in the background ( [ ecu : fon ] ) , we truncate this action to the zero kaluza - klein modes of the compact @xmath0 dimensions .
the effective 4-dimensional electromagnetic action turns out to be @xmath12 hence @xmath13 is normalizable , up to a normalization factor proportional to @xmath14 . if @xmath15 there is no localized gauge field , but for @xmath16 the gauge field localizes on the brane and electrodynamics on it becomes 4-dimensional at large distances @xmath17 .
we proceed next to discuss the spinor localization in rsiin . as in the case of the electromagnetic field in rsiin ,
the observed 4-dimensional fields are the zero modes of higher dimensional ones . for explicit calculations we shall be considering @xmath0=1,2 , namely 6- and 7-dimensional spacetime unless we specify otherwise , and hence 8-component spinors .
the analysis should hold with the adequate generalizations for arbitrary @xmath0 .
we combine the localization mechanism of @xcite designed to work in 5-dimensional spacetime for massive fermions with the projection analysis of @xcite that allows one to identify 4-component spinors out of the higher dimensional ones . to begin with a domain wall
is considered to be produced by some scalar field @xmath18 .
the domain wall separates two regions in the non - compact extra dimension @xmath19 .
explicitly the region @xmath20 , @xmath21 separates from @xmath22 , @xmath23 .
one introduces a yukawa coupling between the spinor and the scalar @xmath24 . for a given sign of the yukawa coupling @xmath25 the zero modes have chirality defined .
in general there is no dependence on the details of the profile of the scalar field across the wall .
it turns out to be convenient to organize the chiral spinors into one field @xmath26 .
the mass term is given by @xmath27 . at low - energies , and
hence small @xmath9 , we truncate this action to the zero kaluza - klein modes of the compact @xmath0 dimensions .
the resulting fermion action reads , after an adequate internal rotation @xcite , @xmath28+g\chi \tau_1+\mu\tau_2\right)\psi , \\
\psi & = & \left(\begin{array}{c } { \psi_1 } \\ { \psi_2 }
\end{array}\right ) \,.\end{aligned}\ ] ] here @xmath29 and @xmath30 are spinors in ( @xmath3)-dimensions and @xmath31}\sigma^{ab}$ ] is the spin conection , with generators @xmath32 $ ] . in what follows
we restrict to the cases @xmath33 . in this way @xmath34 are six @xmath35 dirac matrices . the dirac equation which follows from this action in the background ( [ ecu : fon ] )
has the form @xmath36\psi=0 \,.\ ] ] since the details of the profile of the scalar filed across the wall play no role for the localization of fields , we use an infinitely thin wall approximation . in this limit
one has @xmath37sign@xmath38 .
the components @xmath39 are given in terms of the frames components through @xmath40 . where the only non zero frames components are @xmath41 with @xmath42 , the indices in the coordinate @xmath43 refer to @xmath33 and underlined indices refer to the orthonormal frame .
the non - vanishing ones are @xmath44 when @xmath45 and @xmath46 there are no true localized modes but a metastable massive state @xcite .
we assume such conditions hold for our @xmath18 and ( [ ecu : dir1 ] ) becomes @xmath47\psi=0 \,.\ ] ] we proceed now by choosing dirac matrices in the following form : @xmath48 , @xmath49 and @xmath50 . a convenient rearrangement for @xmath26 is @xcite : @xmath51 because of the presence of @xmath52 , from the point of view of 3-dimensional space , the spinor represents a quantum point - like particle .
now , the fundamental representation for @xmath26 corresponds to the eigenvalues @xmath53 and @xmath54 for @xmath55 , thus we use @xmath56 as in @xcite . in terms of @xmath57
the equation([ecu : dir2 ] ) translates into a set of coupled equations , @xmath58 by taking @xmath59 and after eliminating @xmath60 in the system ( [ ecu:1])-([ecu:2 ] ) one obtains a second order equation for @xmath61 that is given by @xmath62 > from here on the same argument of @xcite holds and there is a metastable state .
namely , for @xmath63 one has @xmath64 with @xmath65 ^ 2}\ ] ] with @xmath66 being the gamma function . in the opposite regime , @xmath67 ,
one has @xmath68 overall , for a given scalar field fulfilling the conditions above , that is to say @xmath69 , it is possible to envisage a long life - time for the spinor to be localized on the brane .
we assume this is the case and hence the role of the extra dimensions in rsiin at low - energies for a localized hydrogen atom we are interested in is reduced to an effective electromagnetic potential for a point - like proton which we describe in the following section .
as a first step in the calculation of the energy levels corrections for hydrogen atoms one has to find the effective electrostatic potential that replaces the coulomb potential . in @xcite we have presented the electrostatic potential produced by a point charge localized in the single brane .
a uniformly moving 4d point charge source will be considered as follows @xmath70 here @xmath71 is the charge density and @xmath72 is the velocity of the source and @xmath73 is a parameter that turns out to be useful in regularizing a delta product later on so we are interested at the end in the limit @xmath74 .
for completeness we briefly recall the calculation leading to the photon green s function as given in the reference @xcite . for the background ( [ ecu : fon ] ) the maxwell equations on the right to the brane yield @xmath75 here we use the gauge @xmath76 , @xmath77 .
@xmath78 and @xmath79 is the 4d dalambertian .
the term containing @xmath80 is pure gauge on the brane so we just drop it from now on .
one now proceeds to solve ( [ eq : maxrsn ] ) using the green s function approach , which satisfies @xmath81 we solve this equation using the eigenfunction expansion for the green s function @xmath82.\ ] ] thus the corresponding eigenvalue equation for @xmath83 can be rewritten as @xmath84\phi(y)=0.\label{ecsl}\ ] ] notice that this equation is invariant under a change of sign in the coordinate @xmath19 and therefore it is enough to solve the equation in the region @xmath85 ( on the right to the brane ) .
performing the change of variable @xmath86 and redefining the function as @xmath87 , one gets the bessel equation @xmath88 when @xmath89 the solution has the form @xmath90 the weight factor @xmath91 comes from the sturm - liuville form of eq .
( [ ecsl ] ) . in the case
@xmath92 the normalized modes become @xmath93 fulfilling the boundary condition @xmath94 as well as the normalization @xmath95 performing the integration over @xmath96 in eq.([ec : green ] ) yields the green s function on the brane @xmath97 here the first term is the contribution of the zero mode , @xmath98 , the second term comes from the continuum of massive modes , @xmath99 .
the induced electromagnetic potential is obtained as usual by integrating the green s function with the source which in this case becomes @xmath100}{4\pi\lambda}mj_1(m\lambda ) \phi_m ( 0)\phi_m ( y')r^{-n}\sqrt{g}\,\,j^{\mu}(x',y')\ , . \label{eq : ajmu}\ ] ] we have used equations ( [ eq : green])-([eq : dm ] ) , @xmath101 and the completeness relation @xmath102 notice that the eq .
( [ eq : ajmu ] ) is obtained by regularization of the product of @xmath103 s as follows : assume @xmath104 , with @xmath73 a regulating parameter going to zero in the limit .
this yields zero because of the delta product .
now we focus on the evaluation ( [ eq : ajmu ] ) .
for low energies only small masses @xmath105 ( @xmath106 ) are relevant , and we obtain for the electromagnetic potential @xmath107 } { 4\pi \gamma(\nu-1 ) } \left(\frac{\epsilon}{2}\right)^{\nu-1 } \sigma^{(5+n)}m^{\nu } j_{\nu}\left(\epsilon m\right)\frac{j_1\left(m \lambda\right)}{\lambda}\,.\ ] ] for the evaluation of equation ( [ eq : amur ] ) , there are two cases , whether @xmath108 is even or odd .
we evaluate first the equation ( [ eq : amur ] ) for @xmath0 odd ( @xmath109 half - integer ) .
we use the relation @xmath110 and introduce ( [ ecu : re2 ] ) in ( [ eq : amur ] ) to integrate over @xmath105 .
so we easily find the result @xmath111}{4\pi r^n \gamma(\nu-1)\lambda } \sigma^{(5+n)}v^{\mu}\frac{\epsilon^{2\nu-1}}{2^{\nu-1}}\left(\frac{d}{\epsilon d\epsilon}\right)^{\nu-1/2 } \frac{\theta[\lambda-\epsilon]}{\lambda\sqrt{\lambda^2-\epsilon^2}},\ ] ] where @xmath112 satisfies @xmath113 .
it is possible to calculate the potential when the particle moves with uniform speed .
we consider , for simplicity , the case when the motion takes place along the @xmath114 direction in the @xmath115 plane , the solution is @xmath116 for a static particle we have @xmath117,\quad r=|\vec{x}|.\ ] ] for the case of @xmath0 even ( @xmath109 integer ) , we can use the relation @xmath118 and integrating over @xmath105 the electromagnetic potential can be rewritten as @xmath119}{4\pi r^n\gamma(\nu-1 ) } \frac{\sigma^{(5+n)}v^{\mu}(-1)^{\nu}}{\lambda}\frac{\epsilon^{2\nu-1}}{2^{\nu-1 } } \left(\frac{d}{\epsilon d\epsilon}\right)^{\nu-1}\frac{\delta[\lambda-\epsilon ] } { \epsilon\lambda}.\ ] ] the electromagnetic potential for a point particle that moves with uniform speed is given by @xmath120 where @xmath121/[|\vec{x}-\vec{\mathrm { x}}(t_*)|]$ ] . for the static particle we have the following potential @xmath122 let us present the above results for @xmath123 for @xmath124 and @xmath125 .
@xmath126 : + liennard - wiechert + @xmath127 static case + @xmath128 @xmath129 : + liennard - wiechert @xmath130 } + \frac{\left(1-|\vec v|^2\right)}{2\left[\sqrt{1+r(t_*)^2}- \hat n\cdot \vec{v}r(t_*)\right]^3 } \right\},\label{ecua : lw}\\ r(t_*)&=&\frac{|\vec{x}-\vec{\mathrm { x}}(t_*)|}{\epsilon}\end{aligned}\ ] ] static case @xmath131 our electrostatic potentials for @xmath33 are consistent with the coulomb potential at long distances and they preserve spherical symmetry .
remarkably these electrostatic potentials are finite at the position of the charge .
( [ coulomb ] ) .
the finite character of our effective electrostatics potentials allows to maintain the idea of an effective 3d point particle .
( [ lwp ] ) shows the effective electromagnetic and standard 4d liennard - wiechert potentials . in polar
coordinates with the origin at the particle s position at the present time and @xmath132 given by the solution of @xmath133 so @xmath134 can take on the values @xmath135 in the electromagnetic potential eq .
( [ ecua : lw ] ) @xmath136 for @xmath137 $ ] and @xmath138 for @xmath139 $ ] .
in this section we calculate the possible corrections to the energy levels for a hydrogen atom caused by the existence of extra dimensions in rsiin .
the effective four - dimensional dirac equation is given by @xmath140 here @xmath141 is given by ( [ ecu : po1 ] ) in the case of one compact extra dimension and ( [ ecu : po2 ] ) for two compact extra dimensions and @xmath142 stands for the electron charge as coming from @xmath3 dimensional spacetime . given the complexity of the effective electrostatic potential we calculate the energy shifts by means of perturbation theory to first order .
we consider first the @xmath126 compact extra dimension case . expanding the electrostatic potential ( [ ecu : po1 ] ) around the coulomb potential , _
ie _ @xmath143 , gives @xmath144 > from here on we shall assume @xmath145 in order to compare with experimental data . within the present approximation the stationary part of dirac equation ( [ ecu : dira ] ) can be rewritten as @xmath146 where the eigenvalue @xmath147 is the energy and @xmath148 is the dirac hamiltonian for the unperturbed hydrogen atom and @xmath149 is the perturbation .
also @xmath150 are dirac s matrices . for the solutions to the dirac equation in a coulomb potential with energy @xmath147 and angular momentum quantum numbers ( @xmath151 )
we take the following form @xcite @xmath152 here @xmath153 is the principal quantum number . in the coulomb case the functions @xmath154 and @xmath155 are related to the confluent hypergeometric functions and @xmath156 are the spherical spinors .
the latter are eigenfunctions of @xmath157 with eigenvalues @xmath158 for @xmath159 .
they satisfy the orthogonality relations @xmath160 the corresponding energy eigenvalues are @xmath161 when calculating the energy shifts to lowest order in @xmath162 the contribution of @xmath154 can be neglected .
the energy shifts are determined by the matrix elements of the perturbation @xmath163 we want to calculate the energy split between the @xmath164 and @xmath165 states , @xmath166 , due to the presence of the @xmath0 compact extra dimensions and compare it with the experimental value for the lamb shift @xmath167 associated to qed . for the @xmath164 state , we have @xmath168 and @xmath169 , and @xmath155 is given by @xmath170,\end{aligned}\ ] ] where @xmath171 the energy shift for @xmath126 determined through eq.([ecu : deltae ] ) is @xmath172.\nonumber\end{aligned}\ ] ] for the @xmath165 state @xmath173 is given by @xmath174\,,\end{aligned}\ ] ] where @xmath175 and now the energy shift is given by @xmath176\gamma(2\gamma-2)+ [ \sqrt{2(1+\gamma)}+2]^2\gamma(2\gamma-3)\right].\nonumber\end{aligned}\ ] ] next we proceed to consider the case of two compact extra dimensions , @xmath129 .
let us take the electrostatic potential eq .
( [ ecu : po2 ] ) and expand it around coulomb s , _ ie _ @xmath143 , thus producing @xmath177 notice that in this case @xmath178 .
we have now the perturbation @xmath179 the energy shift for the @xmath164 state one arrives at is @xmath180.\nonumber\end{aligned}\ ] ] for the @xmath165 state the corresponding energy shift is @xmath181 \gamma(2\gamma-3)+[\sqrt{2(1+\gamma)}+2]^2\gamma(2\gamma-4)\right ] .\nonumber\end{aligned}\ ] ] now we are ready to compare the @xmath164-@xmath165 energy shift for hydrogen in the presence of @xmath0 compact and one warped extra dimensions with the qed lamb shift .
namely , by assuming a null result from the experiment that could be directly associated to the extra dimensions we have @xcite @xmath182 and with the use of the explicit formulae obtained above we can estimate bounds for @xmath1 .
notice this has the advantage with respect the use for comparison of the uncertainty in the measurement that one does not need to consider other important effects as the charge radius of the proton , recoil effect , among others , which are indeed less than @xmath183 @xcite . using the value of @xmath184 @xcite
the following results follow : @xmath185 for @xmath33 , respectively .
in this work we have investigated low - energy effects associated to the presence of extra dimensions .
the brane world considered here is a randall - sundrum like containing a @xmath186-brane embedded in a @xmath187 space with @xmath0 compact and one warped infinite extra dimensions or rsiin . apart from its hybrid character having both compact and non - compact extra dimensions this model is interesting because it localizes the zero mode gauge field due to the presence of the compact dimensions @xcite .
the effects we have considered are i ) the effective electromagnetic potentials of a point charge moving uniformly along the non compact directions of the @xmath186-brane and ii ) the lamb shift for a hydrogen atom localized in the same subspace .
we took a test point charge as seen from the 4d spacetime perspective , moving uniformly along a timelike straight geodesic spatially contained along the non compact directions of the @xmath186-brane .
some comments regarding the stability of such geodesic are in order . although unstable such geodesics are interesting whenever the tunneling time into the extra dimensions is large enough .
as it is clear from our discussion above for the spinor localization based on @xcite for rsiin this can be the case quantum mechanically .
it has also been proved for other similar scenarios in @xcite .
thus it is reasonable to think in terms of a semiclassical approximation leading to such a tunneling effect with long life - time for our uniformly moving particle .
we assume then that our description of the classical test charge is the leading order of the corresponding semiclassical approximation for the quantum situation .
moreover , at high enough energies , a particle gets excited along the compact extra dimensions and in this case the instability of the geodesic is removed .
our results summarize as follows . at low energies our point charge in uniform motion produces electromagnetic potentials given by the zero mode regarding the compact extra dimensions but include light massive modes associated to the non - compact dimension .
they reduce to the standard liennard - wiechert potentials in 4d minkowski spacetime at long distances , as determined by the characteristic scale of the scenario @xmath1 . remarkably below this scale
they differ from the standard ones and become finite at the position of the point charge .
they can easily be seen to produce finite self energies for the point particles as calculated in 4d namely @xmath188 and @xmath189 .
the details of the potentials for short distances depend on the number of extra dimensions @xmath0 .
it is tempting to explain the finiteness of the electromagnetic potentials and self energy due to the fact that our charge looks point - like in 3d but it is actually effectively smeared along the compact extra dimensions at low energies due to the zero mode approximation in the compact dimensions for the electromagnetic fields .
indeed as @xmath190 in ( [ ecu : po1 ] ) and ( [ ecu : po2 ] ) such finiteness is lost .
however it must be observed that for finite @xmath134 actually the standard divergences for the coulomb potential and self - energy are regained in the limit @xmath191 .
so the finiteness is better described as a combined effect of the presence of the compact and warped extra dimensions . indeed not having compact dimensions
amounts to an impossibility of localizing the particle s electromagnetic field to the brane . as for the hydrogen atom model
we take a dirac equation in rsiin for which we have adopted the localization mechanism of the spinor as given in @xcite . here
we pick the zero mode in both compact and non - compact extra dimensions for the localized spinor and couple it with the static form of the time component of the effective liennard - wiechert potential we have determined in the present work , eqs .
( [ ecu : po1 ] ) and ( [ ecu : po2 ] ) . for the static case
the effective potential can be seen as differing from the standard 4d coulomb potential by an additive term containing a different power of the spatial separation between the point like charges .
the power depends upon the number @xmath0 of compact extra dimensions . by making use of a perturbation analysis in terms of this deviation from the coulomb potential it was possible to calculate a @xmath2 shift which we compared with experimental lamb shift data @xmath184 @xcite . assuming the effect of the extra dimensions @xmath192 is smaller than the lambs s @xmath183 yields the bounds to the length scale of the scenario : @xmath193 for @xmath33 , respectively .
notice we estimated such bounds by comparing it with the lamb shift itself and not with the experimental uncertainty in its measurement .
so no other effects like the charge radius of the proton or the recoil effect , among others , are relevant since they are indeed less than @xmath183 .
of course one could insist on using such uncertainty but no significant improvement is obtained because although the bounds improve one has then to incorporate the afore mentioned extra effects that essentially cancel the improvement .
furthermore our upper bounds for @xmath1 lie in the fermi range so no conflict with macroscopic electromagnetism tests is expected . to go beyond such scale
say scattering experiments one would actually require a quantum field approach or qed on rsiin which remains an interesting open problem .
overall both of our results add evidence on how known low - energy phenomena can be used to probe models with extra dimensions that were originally thought to be testable only at high energies while being automatically compatible with standard low - energy physics .
remarkably , finiteness of the electromagnetic potentials and self - energy has led in the past to new ideas as for instance the proposal that gravity regulates the self - energy of a point particle @xcite or a non linear electrodynamics of born and infeld @xcite implementing the finiteness of the electric field @xcite . in light of our results it would be of interest to study other singular problems in standard electrodynamics as the radiation reaction problem . in this regard
higher dimensional flat spacetimes have been considered @xcite but curved hybrid scenarios like rsiin have not to the best of our knowledge . as for the quantum extensions it would be interesting to study further the finite character we have found here for the potentials and self - energy of a point charge quantum mechanically and its possible bearing on qed in particular high precision tests @xcite or possibly other gauge fields of the standard model @xcite to see whether they may add something to realistic brane world models @xcite .
this work was partially supported by mexico s national council of science and technology ( conacyt ) , under grants conacyt 51132-f and conacyt - nsf e-120 - 0837 .
o.p . acknowledges the support of the conacyt fellowship 162767 .
he would also like to thank the young collaborator programme of abdus salam ictp trieste , italy , for supporting a visit where part of this work was done .
r. bluhm , lorentz and cpt tests in atomic systems .
presented at 3rd international symposium on symmetries in subatomic physics , adelaide , australia , 13 - 17 mar 2000 . published in aip conf.proc.539:109-118,2000 also in * adelaide 2000 , symmetries in subatomic physics * 109 - 118 e - print archive : hep - ph/0006033 w. r. johnson , k. t. cheng , and m. h. chen , accurate relativistic calculations including qed contributions for few - electron systems in relativistic electronic structure theory : part 2 .
applications ( theoretical and computational chemistry # 14 ) ed .
peter schwerdtfeger , chap .
3 , isbn 0444512993 ( elsevier , 2004 ) .
m. spiropulu , collider experiment : strings , branes and extra dimensions . to appear in the proceedings of theoretical advanced study institute in elementary particle physics ( tasi 2001 ) : strings , branes and extra dimensions , boulder , colorado , 3 - 29 jun 2001 . published in * boulder 2001 , strings , branes and extra dimensions * 519 - 558 [ hep - ex/0305019 ] | testing extra dimensions at low - energies may lead to interesting effects . in this work a test point charge is taken to move uniformly in the 3-dimensional subspace of a ( 3+@xmath0)-brane embedded in a ( 3+@xmath0 + 1)-space with @xmath0 compact and one warped infinite spatial extra dimensions .
we found that the electromagnetic potentials of the point charge match standard liennard - wiechert s at large distances but differ from them close to it .
these are finite at the position of the charge and produce finite self - energies .
we also studied a localized hydrogen atom and take the deviation from the standard coulomb potential as a perturbation .
this produces a lamb shift that is compared with known experimental data to set bounds for the parameter of the model .
this work provides details and extends results reported in a previous letter .
+ _ keywords : brane worlds , liennard - wiechert , hydrogen atom , point charges . _ |
in braneworld scenarios it is assumed that our observable spacetime corresponds to a submanifold of a higher - dimensional ambient space .
basically this picture comes from the idea that matter and non - gravitational fields might be confined to the ordinary four - dimensional spacetime ( the brane ) by some trapping mechanism which would keep extra dimensions hidden from direct observations at an energy level below tev scale @xcite . of course
, for the sake of consistency , the existence of extra dimensions could not be revealed by gravitational probe either at lower energy regime .
usually this is conceptually achieved by admitting the compactification of extra dimensions .
but , unlike kaluza - klein models , extra dimensions could be much larger than planck length in braneworld scenarios@xcite .
the randall - sundrum type ii ( rsii ) model @xcite , however , surpassing the compactification paradigm , shows that a zero mode perturbation of gravity can also be localized in the brane if it is embedded in a non - compact ambient space such as @xmath0 ( five - dimensional anti - de sitter ) space with @xmath1 symmetry . as a consequence , in the first order approximation , gravity in the brane
is effectively four - dimensional at large distance . in the rsii model
the brane is infinitely thin and possesses a tension which is responsible for a discontinuity in the first derivative of the metric .
there are solutions involving a real scalar field interacting minimally with five - dimensional gravity which can be considered as smooth versions of rsii model @xcite .
the so - called thick branes are regular domain wall solutions that have a finite thickness and give rise to well - behaved differentiable metrics . in the infinitely thin limit
, these thick brane solutions reduce to the rsii model @xcite .
thus , in this sense , the rsii model might be understood as a kind of idealization of thick branes scenarios .
the confinement of gravity is also valid in thick brane versions of rsii @xcite . in this paper
we are interested in studying the motion of test particles in thick branes focusing on the problem of confinement and stability of the motion of test particles .
if a test particle ( a meteorite , in a solar system , for example ) somehow gains a small momentum in the fifth dimension direction , what happens ? does the particle leave the brane or does it stay moving around it ?
the motion of test particle in higher - dimensional spaces has been investigated under different approaches @xcite . in ref .
@xcite necessary general geometric conditions for the geodesic motion to be stable around a hypersurface were established .
it turns out , however , that the rsii model does not match these conditions .
indeed , it has been shown that particles , whose geodesic motion is perturbed , will escape from the brane to infinity in the rsii model @xcite .
as we shall see there is no stable confinement for the thick brane case either .
then , we can conclude that the gravitational force itself is not strong enough to confine test particles in this kind of brane .
therefore , confinement of test particles must be provided by some additional mechanism . the possibility of localization of matter fields by domain walls was originally realized by rubakov and shaposhnikov @xcite .
it was shown that confinement of fermions can be provided by a yukawa - like interaction between the spinor dirac field and a scalar field@xcite .
of course , confinement of test particles must be a consequence of confinement of matter at the quantum level .
here we propose a realization of this confinement mechanism in the non - quantum picture .
we shall see that a stable confinement of the motion of test particle in the thick brane is possible if we consider a direct interaction between test particles and the scalar field .
due to this interaction the particles do not follow a geodesic motion . the interaction with the scalar field
is implemented by a modification of the lagrangian of test particle which is inspired by the properties of the yukawa - type interaction .
consider a real scalar field @xmath2 defined in the bulk , which is minimally coupled with the five - dimensional gravity described by the metric tensor @xmath3(here capital latin indices run from @xmath4 to @xmath5 ) .
the total action of this interaction is@xmath6 \label{action}%\ ] ] where @xmath7 is the determinant of the metric whose signature is admitted to be @xmath8 , @xmath9 represents the scalar curvature related to @xmath10 , @xmath11 corresponds to the potential of the scalar field and @xmath12 is the gravitational coupling constant in five dimensions . from that action
, we obtain the einstein equations @xmath13 , where @xmath14 is the energy - momentum tensor of the scalar field@xmath15 \label{t}%\ ] ] the dynamic equation for the scalar field is@xmath16 where @xmath17 is the covariant derivative compatible with @xmath18 let us consider a potential @xmath11 which has two degenerate minima at @xmath19 , at least . in the simplest domain wall solutions , without any further structures ,
the scalar field depends only on the extra dimension coordinate , @xmath20 .
asymptotically the field tends to the minima values , i.e. , @xmath21 for @xmath22 . in order to make contact with the rsii set up@xcite ,
let us assume that the five - dimensional ambient space is a warped product space whose metric can be written as@xmath23 where greek indices @xmath24 and @xmath25 represents the induced metric on the hypersurface @xmath26 , which is supposed to be the ordinary four - dimensional spacetime . with no loss of generality we are assuming here that @xmath27 .
considering the foliation of the ambient space adapted to the gaussian coordinate system given above , the einstein equations can be rewritten in terms of the intrinsic and the extrinsic curvatures of hypersurfaces @xmath28 @xmath29 in the following equivalent form@xmath30 where @xmath31 and @xmath32 are , respectively , the ricci tensor and the covariant derivative associated to the metric @xmath33 induced on the hypersurface @xmath34 and @xmath35 is the extrinsic curvature of @xmath36 .
@xmath37 is the trace of the energy - momentum tensor and @xmath38 is the dimension of the ambient space , which , in our case , is @xmath39 .
the metrics @xmath40 and @xmath41 ( the metric of the ordinary spacetime ) are related by a conformal factor which depends only on the extra dimension @xmath42 .
it follows , then , that @xmath43 and @xmath44 . under the assumption that @xmath45 , the field equations reduces to a much simpler form .
if we admit that the spacetime metric @xmath41 is ricci - flat @xmath46 , then we have the following equations@xmath47 where prime denotes derivative with respect to @xmath48 of course the solution of these equations shall depend on the explicit form of the function @xmath11 .
however , some general aspects of the behavior of the solutions can be anticipated .
a general feature of a domain wall solution is the existence of a scale parameter @xmath49 which is related to the width of the thick brane @xcite .
as we have mentioned , in order to have a domain wall solution , the scalar field @xmath50 must be a smooth function interpolating between the two real vacua .
thus , we should have @xmath51 @xmath52 for @xmath53 . as
a consequence , the equations ( [ a2 ] ) and ( [ a1 ] ) asymptotically ( @xmath53 ) yield@xmath54 assuming that @xmath55 , the above equations admit a solution in the form : @xmath56 which corresponds to the warping factor of the rsii model if we identify @xmath57 ( the bulk cosmological constant )
. the proper energy density of the domain wall which exceed the vacuum energy @xmath58 is @xmath59 where @xmath60 is the four - velocity of observers which are at rest with respect to the domain wall . using ( [ t ] ) ,
( [ a2 ] ) and ( [ a1 ] ) and integrating the energy density in the interval @xmath61 we find@xmath62 dy=-\frac{3}{\kappa^{2}}\left ( a^{\prime}\left ( \eta_{2}\right ) -a^{\prime } \left ( \eta_{1}\right ) \right ) -\int_{\eta_{1}}^{\eta_{2}}\left [ \frac { 6}{\kappa^{2}}a^{\prime2}+v\left ( \varphi_{0}\right ) \right ] dy\label{tension}%\ ] ] in the thin brane limit @xmath63 , the energy of the domain wall is supposed to be concentrated around the brane giving rise to the brane tension @xmath64@xmath65 dy\ ] ] if the thick brane has a well - defined limit to the rsii model , then the metric must satisfy some conditions .
let @xmath66 denote the warping function related to the parameter @xmath49 .
in order to guarantee that the thin brane limit gives the rsii brane , it is necessary that @xmath67 be locally bounded and @xmath68 locally square - integrable for any value of @xmath49 @xcite . besides , the sequence of functions @xmath67 and @xmath69 should converge locally in square integral to @xmath70 and @xmath71 respectively , where @xmath72 is the step function .
it turns out then that last integral in equation ( [ tension ] ) goes to zero in the thin brane limit .
thus , we have @xmath73 dy=\left\ { \begin{array } [ c]{c}% \sqrt{-\frac{6}{\kappa^{2}}v\left ( \varphi_{0}\right ) } , \qquad\eta _ { 1}<0\quad and\quad\eta_{2}>0\\ 0,\qquad\eta_{1}<\eta_{2}<0\quad or\quad0<\eta_{1}<\eta_{2}% \end{array } \right.\ ] ] therefore , the energy distribution of the thin domain wall is@xmath74 with @xmath75 .
if we had considered only the contribution due to the potential we would have found@xmath76 let us mention that an example of a regularized version of rsii brane is the thick brane considered in reference @xcite .
let us consider this version here as it leads to the following analytical solution of the equations ( [ a2 ] ) and ( [ a1]):@xmath77 where @xmath78 and @xmath79 are positive parameters which are related respectively to the vacuum value of the scalar field @xmath80 and to the width of the domain wall , respectively .
in the limit @xmath22 , the warping factor converges to @xmath81 .
thus , in the particular case in which @xmath82 , i.e. , the ricci - flat metric @xmath41 is the minkowski metric @xmath83 , then ( [ metric ] ) corresponds to the metric of a @xmath1 symmetric @xmath0 space .
therefore , the thick brane solution coincides asymptotically with the rsii solution provided the identification @xmath84 is made , where @xmath85 is the bulk cosmological constant . in the thin brane limit , @xmath86 , with @xmath87 the rsii solution
is recovered @xcite .
by definition test particles carry a very small amount of energy in such way that they are affected by external fields but the effects they would cause in the ambient are negligible .
as is well known it is assumed that test particles follow geodesics when they are under the influence of the gravitational field only .
the geodesic motion of a particle with rest mass @xmath88 can be derived from the extremization of the action@xmath89 where the parameter @xmath90 is the particle s proper time and @xmath91 means @xmath92 .
let us consider the motion of a massive test particle in the five - dimensional warped product space ( [ metric ] ) .
the equation of motion along the extra dimension @xmath42 is easily obtained@xmath93 where @xmath94 . using the constraint @xmath95 , we can write@xmath96 this shows that the motion of the particle along the extra dimension decouples from the motion in other directions .
multiplying equation([ygeo ] ) by the integration factor @xmath97 , the following fist integral can be readily obtained:@xmath98 taking @xmath26 , we get from the above equation that @xmath99 , where @xmath100 might be interpreted as the amount of initial kinetic energy related to the motion along the extra dimension per rest mass .
thus , we find@xmath101 therefore , given @xmath102 the particle can move only in the region where @xmath103 . in this sense
, we can say that ( [ motion ] ) is equivalent to an equation of motion of a particle under the influence of the potential @xmath104 .
the maximum geodesic distance the particle can reach , @xmath105 , is then given by the condition @xmath106 .
let us now investigate the motion of a test particle in the thick brane .
we shall see that the motion is not bound to the hypersurface @xmath107 .
indeed , the first derivative of the potential is : @xmath108 .
a necessary condition for the existence of confinement of test particles in @xmath107 is that @xmath26 be a critical point of @xmath109 , which , of course , demands that @xmath110 .
thus the function @xmath111 must satisfy the following ` initial conditions ' at @xmath107 : @xmath112 and @xmath113 . according to equation ( [ a2 ] ) , @xmath114 for any @xmath42 , it follows that @xmath111 has an absolute maximum at @xmath26 . to check this
let us suppose that there is a point @xmath79 where @xmath115 .
first , let us admit that @xmath116 . in this case
@xmath117 for some @xmath118 with @xmath119 then , as @xmath110 , @xmath120 would be positive for some @xmath42 , @xmath121 .
if @xmath122 , then @xmath123 for some @xmath124 again , @xmath120 would be positive for some @xmath4@xmath125@xmath126 . both cases would be in contradiction with equation ( [ a2 ] ) .
therefore , the function @xmath111 is non - positive and , as a consequence , @xmath127 for any @xmath42 .
so , we can conclude that any free test particle which acquires a non - null transversal velocity @xmath128 will not stay localized in the vicinity of @xmath107 no matter how small the velocity is , since its motion has no turning point .
this means that geodesic motion in the brane is unstable .
this result is in accordance with prior analysis which has shown that a particle is expelled from the thin brane if its motion is perturbed toward the extra dimension @xcite .
we can illustrate this point by considering the particular thick brane described by the solutions ( [ a ] ) and ( [ phi ] ) . in this case
, we have explicitly @xmath129 ^{-b\kappa^{2}}-1)\leq0 $ ] .
it is thus clear that gravity by itself can not localize test particles in thick branes of rsii - type .
then , there must exist some additional mechanism which explains the confinement of them to the brane .
in quantum regime , some mechanisms which play this role are known .
for instance , the dirac spinor field can be trapped in a domain wall by a yukawa - type interaction with the scalar field @xcite . the action for the five - dimensional massless fermion @xmath130 which describes this interaction is@xmath131 where @xmath132 is the coupling constant . when the classical domain wall solution @xmath133 known as kink , is considered
then the field equation for @xmath130@xmath134 presents some interesting features .
first , the scalar field generates mass for the five - dimensional fermion . at the scalar field vacua
, the fermion acquires a mass @xmath135 .
second , there exists a zero mode solution @xmath136 which is localized at @xmath137 i.e. decays exponentially for large @xmath138 and is proportional to the solution of the usual four dimensional dirac field .
the question we face now is how this confinement mechanism can emerge in a classical picture of test particles . as we shall see confinement is possible if we admit a direct interaction between test particles and the scalar field .
this interaction is carried out by a redefinition of the action ( [ s ] ) .
as we have mentioned , in the quantum picture , the mass of the dirac field is modified by the scalar field . indeed ,
if we admit that the five - dimensional dirac spinor field possesses a rest mass @xmath118 then , due to the yukawa interaction we get the relation@xmath139 where @xmath140 represents the @xmath141-momentum of the fermion .
our intention is to define a new lagrangian in such way that the energy relation ( [ energy ] ) could be derived .
one way to do this is changing the mass term @xmath88 by @xmath142 , then the new action for test particles would read@xmath143 calculating the @xmath141-momentum , @xmath144 , where @xmath145 is the lagrangian @xmath146 we find@xmath147 for massive test particles @xmath148 .
it is easily checked that the condition ( [ energy ] ) is satisfied .
of course , the usual action is obtained again turning off the interaction , i.e. , taking @xmath149 .
it is worthy of mention that a similar kind of lagrangian was also employed , in a different context , to describe the interaction between test particles and dilatonic fields @xcite .
let us now investigate the equation of motion of test particles in the warped product space ( [ metric ] ) .
first , we consider the motion in the extra dimension . from the euler - lagrange equation we get@xmath150 where we have used the constraint @xmath151 . multiplying the equation by the integration factor @xmath152 the first integral
can be immediately obtained and can be put in the following form@xmath153 \label{motion - scalar}%\ ] ] the effective potential is now @xmath154 $ ] , which reduces to the previous one when @xmath149 .
now let us study the motion of the particle around @xmath107 .
as we shall see there are thick branes of rsii - type in which test particles can be trapped in the domain wall due to the interaction with the scalar field . the first derivative of the potential is@xmath155\ ] ] for the sake of simplicity let us restrict our analysis to the case in which @xmath156 , i.e. , the case in which the scalar field is found in the false vacuum state at @xmath107 . then , @xmath26 will be a critical point of @xmath157 only if @xmath110 .
calculating the second derivative at @xmath107 , we find@xmath158 using the field equations ( [ a2 ] ) and ( [ a1 ] ) evaluated at @xmath107 , the above expression can be rewritten as@xmath159 where @xmath160 is the value of the potential of the scalar field at the false vacuum , which is necessarily non - negative according to equation ( [ a1 ] ) .
if @xmath161 then for the particles that obey the condition@xmath162 the point @xmath26 will represent a stable equilibrium point . therefore , these particles will be stably confined to @xmath107 if the energy of perturbation @xmath100 is small enough . in this case , the motion will be given by the equation @xmath163 which corresponds to an oscillation of small amplitude @xmath164 and angular frequency @xmath165 . moreover ,
if @xmath166 is high enough and @xmath167 is sufficiently small , then the period of oscillation will be so short and the amplitude so small that the movement in the extra dimension could not be detected by @xmath168observers .
it is clear that the confinement is stronger in cases of large values of @xmath160 .
the width of the brane also influence the intensity of the binding of the particle to @xmath107 . in the previous section
, we have seen that in the thin limit @xmath169 tends to a delta function whose support is on @xmath26 .
thus , @xmath160 can not be bounded in the limit @xmath170 .
indeed , consider the inequality@xmath171 } \left ( v\left ( y\right ) -v\left ( \varphi_{0}\right ) \right ) \right ] \left ( b - a\right)\ ] ] assuming that @xmath11 has an absolute maximum at @xmath172 in the range @xmath173 $ ] , then@xmath174 consider now the limit @xmath170 keeping @xmath85 constant .
for @xmath175 and @xmath176 , according to equation ( [ pot - delta ] ) , the integral has a finite value which is independent of @xmath78 and @xmath177@xmath178 but as @xmath78 and @xmath177 are arbitrary , the difference @xmath179 can be made arbitrarily small .
therefore , in the thin limit @xmath180 must increases with no bounds .
it is worthy of mention that for a particle whose rest mass is greater than the critical value @xmath181 , the interaction is not strong enough to keep it bound to the hypersurface @xmath107 . once more let us illustrate this point by considering the particular thick brane given by solutions ( [ a ] ) and ( [ phi ] ) . in this case , the effective potential is@xmath182 ^{2}}{\left [ \cosh\left ( \frac{2cy}{\kappa^{2}}\right ) \right ] ^{b\kappa^{2}}}-1\label{potential}%\ ] ] it can be directly checked that @xmath183 and that @xmath184 . as we have anticipated in the general case , @xmath26 is a stable equilibrium point for particles that obey the condition @xmath185 ( see fig.1 ) .
it can be easily verified also that the confinement is stronger in cases of small domain wall width @xmath186 .
another interesting characteristic is that the intensity of the binding is greater in the case of large values of @xmath85 @xmath187recall that @xmath188 .
it is important to stress that the latter feature is valid to this particular thick brane , but we can not assure its general validity . in this particular thick brane , as the potential is explicitly known ,
general motions , i.e. , those not restricted to small perturbations , can be studied .
the potential is an even function which vanishes at @xmath26 and tends to @xmath189 for @xmath190 .
if the condition @xmath191 holds , @xmath192 has a maximum value @xmath193 at two points @xmath194 .
therefore the potential is a bounded function : @xmath195 a rough estimate of an upper limit of @xmath196 can be readily obtained by taking the maximum value of numerator and the minimum value of the denominator simultaneously in ( [ potential ] ) .
this gives @xmath197 .
then , the initial kinetic energy @xmath100 of the light particles ( @xmath198 can not exceed this amount of energy in order that we have confinement , otherwise they certainly will escape from the brane and will never return .
in the case of confinement , @xmath199 the particle will oscillate around the hypersurface @xmath107 and will reach a maximum distance @xmath105 which is given by @xmath200 ^{2}}{\left [ \cosh\left ( \frac{2cy_{\max}}{\kappa^{2}}\right ) \right ] ^{b\kappa^{2}}% } -1=\dot{y}_{0}^{2}%\ ] ] let us define @xmath201
. in terms of this new variable the above equation does not depend on the parameter @xmath79 . then
, it is clear that @xmath202 it is interesting to note that the particle will stay closer to the hypersurface @xmath107 , or in other words , confinement will be stronger in case of thinner domains wall @xmath203 . now let us consider the general case again .
let us now study the motion of massless spinor particles by admitting that their motion in the fifth direction is given by eq.([motion - scalar ] ) in the limit @xmath204 .
multiplying that equation by @xmath88 and taking the mentioned limit , we get@xmath205 the unique solution for this equation is @xmath26 . as consequence the energy relation ( [ energy ] ) reduces to@xmath206 this means that the particle , which seems to be massless with respect to @xmath207-observers , stays strictly confined to the hypersurface @xmath107 .
this result can be confirmed by analyzing the second derivative of the effective potential , which gives the strength of the binding force between the particle and the domain wall .
as we have seen , it increases for particles with small mass .
then , it is expected that in the massless limit the binding force becomes so huge that the particles can not escape from @xmath107 .
let us now consider the issue we have already raised concerning the motion of a small meteorite ( considered as a test particle ) in the solar system in the context of thick branes . for the sake of simplicity , let us admit that the geometry around the sun is approximately given by the schwarzschild metric .
thus , the five - dimensional metric is @xcite@xmath208 + dy^{2}\label{schawrzbrane}%\ ] ] where @xmath209 . since the schwarzschild metric is ricci - flat ,
the field equations reduce to the same equations ( [ a2 ] ) and ( [ a1 ] ) .
therefore the general analysis of the previous section concerning the motion of test particles in the fifth dimension , which is described by equation ( [ motion - scalar ] ) , is also applicable to the case of the meteorite in the schwarzschild thick brane ( [ schawrzbrane ] ) .
then we can conclude that , for a small perturbation , the meteorite stays bound to the hypersurface @xmath107 and its motion in the fifth dimension might not be directly observed .
we now study the motion in the other directions .
first , let us obtain the euler - lagrange equation for the coordinate @xmath210@xmath211 as usual we consider the particular solution @xmath212 , i.e. , the motion in the equatorial plane . the equations for @xmath213 and @xmath214 are@xmath215 where @xmath216 and @xmath145 are respectively the energy and the angular momentum measured by an asymptotic observer @xmath217 in the brane @xmath218 .
the euler - lagrange equation related to the coordinate @xmath219 can be readily integrated and , using ( [ motion - scalar ] ) , ( [ t ] ) and ( [ angle ] ) , gives@xmath220 note that if there is no perturbation , i.e. , @xmath221 , then @xmath26 is a solution of ( [ motion - scalar ] ) and the equations ( [ t ] ) , ( [ angle ] ) and ( [ r ] ) give the usual four - dimensional motion of a test particle in the schwarzschild geometry .
now let us consider the equation of motion parameterized by the time measured by an asymptotic observer@xmath222 it is interesting to note that compared to the usual corresponding equation in the four - dimensional schwarzschild spacetime the eq.([rt ] ) differs only by the additional term @xmath223 .
therefore , if the perturbation is truly small , i.e. , @xmath224 , then , despite the oscillatory motion in the fifth dimension , the asymptotic observer can not detect the existence of a fifth dimension by examining the time evolution @xmath225 of the meteorite .
the same conclusion is valid for the trajectory , i.e. @xmath226 . substituting @xmath227 in the equation ( [ rt ] ) we have@xmath228 it is remarkable that , as in the previous case , the only modification of the trajectory equation , due to the fifth dimension , is again the term @xmath229 .
thus , the influence of the extra dimension in the trajectory of the meteorite might be neglected if the energy of perturbation is small enough .
we have studied the motion of test particles in thick branes of rsii - type .
we have seen that a particle following a geodesic motion does not stay bound to the four - dimensional spacetime if its motion is transversally perturbed .
this means that gravity alone can not be responsible for the confinement of test particles to this kind of brane .
based on the mechanism of localization of fermions in a domain wall by means of a yukawa - like interaction between fermions and the scalar field , we have proposed a direct interaction between test particles and the scalar field by changing the lagrangian of test particles .
we have shown that due to this interaction light test particles , i.e. particles which obey the condition @xmath185 will be bound to the brane for small transversal perturbations .
heavier particles if transversally perturbed will acquire so large an inertia that the interaction with scalar field will not be strong enough to stop their motion into the extra dimension .
we also analyze the motion of test particles in the schwarzschild thick brane . as expected for the consistency reasons of the brane world scenario , in the lower energy regime the motion of test particles does not reveal the existence of an extra dimension , neither by direct observation of the oscillatory motion in the fifth dimension nor by deviation of the motion in the four - dimensional ordinary spacetime .
the authors thank cnpq - fapesq ( pronex ) for financial support .
thanks also go to the referee for useful comments and suggestions . n. arkani - hamed , s. dimopoulos , and g. dvali , phys .
b 429 , 263 ( 1998 ) ; i. antoniadis , n. arkani - hamed , s. dimopoulos , and g. dvali , phys . lett .
b 436 , 257 ( 1998 ) .
l. randall and r. sundrum , phys .
83 , 3370 ( 1999 ) | we consider the motion of test particles in thick branes of ` randall - sundrum ii'-type .
it is known that gravity alone can not explain the confinement of test particles in this kind of brane . in this paper
we show that a stable confinement in a domain wall is possible by admitting a direct interaction between test particles and a scalar field .
this interaction is implemented by a modification of the lagrangian of the particle which is inspired by a yukawa - type interaction between fermions and scalar fields .
pacs : 04.20.jb , 11.10.kk |
there are many well known probabilistic proofs that objects with some properties exist . such a proof estimates the probability for a random object to violate the requirements and shows that it is small ( or at least strictly less than @xmath0 )
let us look at a toy example . consider a @xmath1 boolean matrix and its @xmath2 minor ( the intersection of @xmath3 rows and @xmath3 columns chosen arbitrarily ) .
we say that the minor is _ monochromatic _ if all its elements are equal ( either all zeros or all ones ) . for large enough @xmath4 and for @xmath5 , there exists a @xmath6-matrix that does not contain a monochromatic @xmath7-minor .
we repeat the same simple proof three times , in three different languages .
( probabilistic language ) let us choose matrix elements using independent tosses of a fair coin .
for a given @xmath3 colums and @xmath3 rows , the probability of getting a monochromatic minor at their intersection is @xmath8 .
( both zero - minor and one - minor have probability @xmath9 . )
there are at most @xmath10 choices for columns and the same number for rows , so by the union bound the probability of getting at least one monochromatic minor is bounded by @xmath11 and the last expression is less then @xmath0 if , say , @xmath12 and @xmath4 is suffuciently large .
( combinatorial language ) let us count the number of bad matrices .
for a given choice of columns and rows we have @xmath13 possibilities for the minor and @xmath14 possibilities for the rest , and there is at most @xmath10 choices for raws and columns , so the total number of matrices with monochromatic minor is @xmath15 and this is less than @xmath16 , the total number of boolean @xmath6-matrices .
( compression language ) to specify the matrix that has a monochromatic minor , it is enough to specify @xmath17 numbers between @xmath0 and @xmath4 ( rows and column numbers ) , the color of the monochromatic minor ( @xmath18 or @xmath0 ) and the remaining @xmath19 bits in the matrix ( their positions are already known ) .
so we save @xmath20 bits ( compared to the straightforward list of all @xmath21 bits ) using @xmath22 bits instead ( each number in the range @xmath23 requires @xmath24 bits ; to be exact , we may use @xmath25 ) , so we can compress the matrix with a monochromatic minor if @xmath26 , and not all matrices are compressible .
of course , these three arguments are the same : in the second one we multiply probabilities by @xmath16 , and in the third one we take logarithms .
however , the compression language provides some new viewpoint that may help our intuition .
in this example we want to put bits ( zeros and ones ) around the circle in a `` essentially asymmetric '' way : each rotation of the circle should change at least a fixed percentage of bits .
more precisely , we are interested in the following statement : there exists @xmath27 such for every suffuciently large @xmath4 there exists a sequence @xmath28 of bits such that for every @xmath29 the cyclic shift by @xmath3 positions produces a sequence @xmath30 that differs from @xmath31 in at least @xmath32 positions ( the hamming distance between @xmath31 and @xmath33 is at least @xmath32 ) . is bad if most of the dotted lines connect equal bits ]
assume that some rotation ( cyclic shift by @xmath3 positions ) transforms @xmath31 into a string @xmath33 that coincides almost everywhere with @xmath31 .
we may assume that @xmath34 : the cyclic shift by @xmath3 positions changes as many bits as the cyclic shift by @xmath35 ( the inverse one ) .
imagine that we dictate the string @xmath31 from left to right .
first @xmath3 bits we dictate normally .
but then the bits start to repeat ( mostly ) the previous ones ( @xmath3 positions before ) , so we can just say `` the same '' or `` not the same '' , and if @xmath36 is small , we know that most of the time we say `` the same '' .
technically , we have @xmath32 different bits , and at least @xmath37 bits to dictate after the first @xmath3 , so the fraction of `` not the same '' signals is at most @xmath38 .
it is well known that strings of symbols where some symbols appear more often than others can be encoded efficiently .
shannon tells us that a string with two symbols with frequencies @xmath39 and @xmath40 ( so @xmath41 ) can be encoded using @xmath42 bits per symbol and that @xmath43 only when @xmath44 . in our case , for small @xmath36 , one of the frequencies is close to @xmath18 ( at most @xmath38 ) , and the other one is close to @xmath0 , so @xmath45 is significantly less than @xmath0 .
so we get a significant compression for every string that is bad for the theorem , therefore most string are good ( so good string do exist ) .
more precisely , every string @xmath46 that does not satisfy the requirements , can be described by * @xmath3 * @xmath47 * @xmath48 for @xmath49 and for large enough @xmath4 the economy in the third part ( compared to @xmath35 ) is more important than @xmath24 in the first part .
of course , this is essentially a counting argument : the number of strings of length @xmath50 where the fraction of @xmath0s is at most @xmath38 , is bounded by @xmath51 and we show that the bound for the number of bad strings , @xmath52 is less than the total number of strings ( @xmath53 ) . still the compression metaphor makes the proof more intuitive , at least for some readers .
in our examples of probabilistic proofs we proved the existence of objects that have some property by showing that _ most _ objects have this property ( in other words , that the probability of this property to be true is close to @xmath0 under some natural distribibution ) .
not all probabilistic proofs go like that .
one of the exceptions is the famous lovsz local lemma ( see , e.g. , @xcite ) .
it can be used in the situations where the union bound does not work : we have too many bad events , and the sum of their probabilities exceeds @xmath0 even if probability of each one is very small .
still lovsz local lemma shows that these bad events do not cover the probability space entirely , assuming that the bad events are `` mainly independent '' .
the probability of avoiding these bad events is exponentially small , still lovsz local lemma provides a positive lower bound for it .
this means , in particular , that we can not hope to construct an object satisfying the requirements by random trials , so the bound provided by lovsz local lemma does not give us a randomized algorithm that constructs the object with required properties with probability close to @xmath0 . much later moser and tardos
@xcite suggested such an algorithm in fact a very simple one .
in other terms , they suggested a different distribution under which good objects form a majority .
we do not discuss the statement of lovsz local lemma and moser tardos algorithm in general .
instead , we provide two examples when they can be used , and the compression - language proofs that can be considered as ad hoc versions of moser tardos argument . these two examples are ( 1 ) satisfiability of formulas in conjunctive normal form ( cnf ) and ( 2 ) strings without forbidden factors .
a cnf ( _ conjunctive normal form _ ) is a propositional formula that is a conjuction of _ clauses_.
each clause is a disjunction of _ literals _ ; a literal is a propositional variable or its negation .
for example , cnf @xmath54 consists of two clauses .
first one prohibits the case when @xmath55 , @xmath56 , @xmath57 ; the second one prohibits the case when @xmath58 , @xmath59 , @xmath60 .
a cnf is _ satisfiable _ if it has a _ satisfying assigment _ ( that makes all clauses true , avoiding the prohibited combinations ) . in our example
there are many satisfying assigments .
for example , if @xmath61 and @xmath62 , all values of other variables are ok
. we will consider cnf where all clauses include @xmath4 literals with @xmath4 different variables ( from some pool of variables that may contain much more than @xmath4 variables ) .
for a random assignment ( each variable is obtained by an independent tossing of a fair coin ) the probability to violate a clause of this type is @xmath63 ( one of @xmath53 combinations of values for @xmath4 variables is forbidden ) .
therefore , _ if the number of clauses of this type is less than @xmath53 , then the formula is satisfiable_. this is a tight bound : using @xmath53 clauses with the same variables , we can forbid all the combinations and get an unsatisfiable cnf .
the following result says that we can guarantee the satisfiability for formuli with much more clauses .
in fact , the total number of clauses may be arbitrary ( but still we consider finite formulas , of course ) .
the only thing we need is the `` limited dependence '' of clauses .
let us say that two clauses are _ neighbors _ if they have a common variable ( or several common variables ) .
the clauses that are not neighbors correspond to independent events ( for a random assignment ) .
the following statement says that if the number of neighbors of each clause is bounded , then cnf is guaranteed to be satisfisable .
assume that each clause in some cnf contains @xmath4 literals with different variables and has at most @xmath64 neighbor clauses .
then the cnf is satisfiable .
note that @xmath64 is a rather tight bound : to forbid all the combinations for some @xmath4 variables , we need only @xmath53 clauses .
it is convenient to present a proof using the compression language , as suggested by lance fortnow .
consider the following procedure @xmath65 whose argument is a clause ( from our cnf ) . \
{ @xmath66 is false } + @xmath65 : + @xmath67 + * for * all @xmath68 that are neighbors of @xmath66 : + * if * @xmath68 is false * then * @xmath69 + \ { @xmath66 is true ; other clauses that were true remain true } here @xmath67 is the procedure that assigns fresh random values to all variables in @xmath66 .
the pre - condition ( the first line ) says that the procedure is called only in the situation where @xmath66 is false .
the post - condition ( the last line ) says that _ if the procedure terminates _ , then @xmath66 is true after termination , and , moreover , all other clauses of our cnf that were true before the call remain true .
( the ones that were false may be true or false . )
note that up to now we do not say anything about the termination : note that the procedure is randomized and it may happen that it does not terminate ( for example , if all resample calls are unlucky to choose the same old bad values ) .
* simple observation * : if we have such a procedure , we may apply it to all clauses one by one and after all calls ( assuming they terminate and the procedure works according to the specification ) we get a satisfying assignment . *
another simple observation * : it is easy to prove the `` conditional correctness '' of the procedure @xmath65 . in other words , it achieves its goal assuming that ( 1 ) it terminates ; ( 2 ) all the recursive calls @xmath69 achieve their goals . it
is almost obvious : the @xmath67 call may destroy ( = make false ) only clauses that are neighbors to @xmath66 , and all these clauses are fix - ed after that .
note that @xmath66 is its own neighbor , so the * for*-loop includes also a recursive call @xmath65 , so after all these calls ( that terminate and satisfy the post - condition by assumption ) the clause @xmath66 and all its neighbors are true and no other clause is damaged .
note that the last argument remains valid even if we delete the only line that really changes something , i.e. , the line @xmath67 . in this case
the procedure never changes anything but still is conditionally correct ; it just does not terminate if one of the clauses is false .
it remains to prove that the call @xmath65 terminates with high probability .
in fact , it terminates with probability @xmath0 if there are no time limits and with probability exponentially close to @xmath0 in polynomial time . to prove this
, one may use a compression argument : we show that _ if the procedure works for a long time without terminating , then the sequence of random bits used for resampling is compressible_. we assume that each call of @xmath70 uses @xmath4 fresh bits from the sequence . finally , we note that this compressibility may happen only with exponentially small probability .
imagine that @xmath65 is called and during its recursive execution performs many calls @xmath71 ( in this order ) but does not terminate ( yet ) .
we stop it at some moment and examine the values of all the variables . knowing the values of the variables after these calls and the sequence @xmath72
, we can reconstruct all the @xmath73 random bits used for resampling .
let us go backwards . by assumption we know
the values of all variables after the calls .
the procedure @xmath74 is called only when @xmath75 is false , and there is only one @xmath4-tuple of values that makes @xmath75 false . therefore we know the values of all variables before the last call , and also know the random bits used for the last resampling ( since we know the values of variables after resampling ) .
the same argument shows that we can reconstruct the values of variables before the preceding call @xmath76 , and random bits used for the resampling in this call , etc .
now we need to show that the sequence of clauses @xmath72 used for resampling can be described by less bits than @xmath77 ( the number of random bits used ) . here
we use the assumption saying each clause has at most @xmath64 neighbors and that the clauses @xmath68 for which @xmath69 is called from @xmath65 , are neighbors of @xmath66 .
one could try to say that since @xmath78 is a neighbor of @xmath79 , we need only @xmath80 bits to specify it ( there are at most @xmath64 neighbors by assumption ) , so we save @xmath81 bits per clause ( compared to @xmath4 random bits used by resampling ) .
but this argument is wrong : @xmath78 is not always the neighbor of @xmath79 , since we may return from a recursive call that causes resampling of @xmath79 and then make a new recursive call that resamples @xmath78 . to get a correct argument , we should look more closely at the tree of recursive calls generated by one call @xmath65 ( fig .
[ pic : treecall ] ) . in this tree
the sons of each vertex correspond to neighbor clauses of the father - clause .
( up to some moment ) ] the sequence of calls is determined by a walk in this tree , but we go up and down , not only up ( as we assumed in the wrong argument ) .
how many bits we need to encode this walk ( and therefore the sequence of calls ) ?
we use one bit to distinguish between steps up and down .
if we are going down , no other information is needed .
if we are going up ( and resample a new clause ) , we need one bit to say that we are going up , and @xmath80 bits for the number of neighbor we are going to . for accounting purposes we combine these bits with a bit needed to encode the step back ( this may happen later or not happen at all ) , and we see that in total we need at most @xmath82 bits per each resampling .
this is still less than @xmath4 , so we save one bit for each resampling . if @xmath83 is much bigger than the number of variables , we indeed compress the sequence of random bits used for resampling , and this happens with exponentially small probability .
this argument finishes the proof .
the next example is taken from word combinatorics .
assume that a list of binary strings @xmath84 is given .
these @xmath85 are considered as `` forbidden factors '' : this means that we want to construct a ( long ) string @xmath86 that does not have any of @xmath85 as a factor ( i.e. , none of @xmath85 is a substring of @xmath86 ) .
this may be possible or not depending on the list .
for example , if we consider two strings @xmath87 as forbidden factors , every string of length @xmath13 or more has a forbidden factor ( we can not use zeros at all , and two ones are forbidden ) .
the more forbidden factors we have , the more chances that they block the growth in the sense that every sufficiently long string has a forbidden factor . of course , not only the number of factors matters : e.g. , if we consider @xmath88 as forbidden factors , then we have long strings of ones without forbidden factors
. however , now we are interested in quantitative results of the following type : _ if the number of forbidden factors of length @xmath89 is @xmath90 , and the numbers @xmath90 are `` not too big '' , then there exists an arbitrarily long string without forbidden factors_. this question can be analyzed with many different tools , including lovsz local lemma ( see @xcite ) and kolmogorov complexity . using a complexity argument , levin proved that if @xmath91 for some constant @xmath92 , then there exists a constant @xmath93 and an infinite sequence that does not contain forbidden factors of length smaller than @xmath93 .
( see ( * ? ? ?
* section 8.5 ) for levin s argument and other related results . ) a nice sufficient condition was suggested by miller @xcite : we formulate the statement for the arbitrary alphabet size .
consider an alphabet with @xmath94 letters .
assume that for each @xmath95 we have @xmath90 `` forbidden '' strings of length @xmath89 .
assume that there exist some constant @xmath96 such that @xmath97 then there exist arbitrarily long strings that do not contain forbidden substrings .
* remarks*. 1 .
we do not consider @xmath98 , since this means that some letters are deleted from the alphabet .
\2 . by compactness
the statement implies that there exists an infinite sequence with no forbidden factors .
the constant @xmath31 should be at least @xmath99 , otherwise the right hand side is negative .
this means that @xmath100 should be small , and this corresponds to our intution ( @xmath90 should be significantly less than @xmath101 , the total number of strings of length @xmath89 ) .
the original proof from @xcite uses some ingenious potential function defined on strings : miller shows that if its value is less than @xmath0 , then one can add some letter preserving this property .
it turned out ( rather misteriously ) that exactly the same condition can be obtained by a completely different argument ( following @xcite ) so probably the inequality is more fundamental than it may seem !
this argument is based on compression .
here is the idea .
we start with an empty string and add randomly chosen letters to its right end .
if some forbidden string appears as a suffix , it is immediately deleted .
so forbidden strings may appear only as suffixes , and only for a short time .
after this `` backtracking '' we continue adding new letters .
( this resembles the famous `` tetris game '' when blocks fall down and then disappear under some conditions . )
we want to show that if this process is unsuccessful in the sense that after many steps we still have a short string , then the sequence of added random letters is compressible , so this can not happen always , and therefore a long string without forbidden factors exists .
let us consider a `` record '' ( log file ) for this process that is a sequence of symbols `` @xmath102 '' and `` @xmath103 '' ( for each forbidden string we have a symbol , plus one more symbol without a string ) . if a letter was added and no forbidden string appears , we just add ` @xmath102 ' to the record .
if we have to delete some forbidden string @xmath104 after a letter was added , we write this string in brackets after the @xmath102 sign .
note that we do _ not _ record the added letters , only the deleted substrings .
( it may happen that several forbidden suffixes appear ; in this case we may choose any of them . ) at every stage of the process the current string and the record uniquely determine the sequence of random letters used . having this information
, we can reconstruct the configuration going backwards .
this reversed process has steps where a forbidden string is added ( and we know which one , since it is written in brackets in the record ) , and also steps when a letter is deleted ( and we know which letter is deleted , i.e. , which random letter was added when moving forwards ) .
if after many ( say , @xmath105 ) steps we still have a short current string , then the sequence of random letters can be described by the record ( due to the lemma ; we ignore the current string part since it is short ) .
as we will see , the record can be encoded with less bits than it should have been ( i.e. , less than @xmath106 bits ) .
let us describe this encoding and show that it is efficient ( assuming the inequality @xmath107 ) .
we use arithmetic encoding for the lengths .
arithmetic encoding for @xmath93 symbols starts by choosing positive reals @xmath108 such that @xmath109 .
then we split the interval @xmath110 $ ] into parts of length @xmath108 that correspond to these @xmath93 symbols . adding a new symbol corresponds to splitting the current interval in the same proportion and choosing the right subinterval .
for example , the sequence @xmath111 corresponds to @xmath112th subinterval of @xmath113th interval ; this interval has length @xmath114 .
the sequence @xmath115 corresponds to interval of length @xmath116 and can be reconstructed given any point of this interval ( assuming @xmath108 are fixed ) ; to specify some binary fraction in this interval we need at most @xmath117 bits , i.e. , @xmath118 bits .
now let us apply this technique to our situation . for @xmath102 without brackets we use @xmath119 bits , and for @xmath120 where @xmath104 is of length @xmath89
, we use @xmath121 bits . here
@xmath122 are some positive reals to be chosen later ; we need @xmath123 .
indeed , we may split @xmath122 into @xmath90 equal parts ( of size @xmath124 ) and use these parts as @xmath125 in the description of arithmetical coding above ; splitting adds @xmath126 to the code length for strings of length @xmath89 . to bound the total number of bits used for encoding the record ,
we perform amortised accounting and show that the average number of bits per letter is less than @xmath127 .
note that the number of letters is equal to the number of @xmath102 signs in the record .
each @xmath102 without brackets increases the length of the string by one letter , and we want to use less that @xmath128 bits for its encoding , where @xmath129 is some constant saying how much is saved as a reserve for amortized analysis .
and @xmath120 for a string @xmath104 of length @xmath89 decreases the length by @xmath130 , so we want to use less than @xmath131 bits ( using the reserve ) .
so we need : @xmath132 together with @xmath133 technically is it easier to use non - strict inequalities in the first two cases and a strict one in the last case ( and then increase @xmath134 a bit ) : @xmath135 then for a given @xmath136 we take minimal possible @xmath134 : @xmath137 and it remains to show that the sum is less than @xmath0 for a suitable choice of @xmath136 .
let @xmath138 , then the inequality can be rewritten as @xmath139 or @xmath140 and this is our assumption . now we see the role of this mystical @xmath31 in the condition : it is just a parameter that determines the constant used for the amortised analysis . * acknowledgement*. author thanks his lirmm colleagues , in particular pascal ochem and daniel gonalves , as well as the participants of kolmogorov seminar in moscow .
9 n. alon , j.h .
spencer , _ the probabilistic method _ , wiley , 2004 .
d. gonalves , m. montassier , a. pinlou , _ entropy compression method applied to graph colorings _ , https://arxiv.org/pdf/1406.4380.pdf .
j. miller , two notes on subshifts , _ proceedings of the ams _ , * 140 * , 1617 - 1622 ( 2012 ) .
r. moser , _ a constructive proof of the lovsz local lemma _ , https://arxiv.org/abs/0810.4812 .
r. moser , g. tardos , a constructive proof of the general lovsz local lemma , _ journal of the acm _ , * 57*(2 ) , 11.111.15 ( 2010 ) .
p. ochem , a. pinlou , application of entropy compression in pattern avoidance , _ the electronic journal of combinatorics _ , * 21*:2 , paper p2.7 ( 2014 ) .
a. rumyantsev , a. shen , probabilistic constructions of computable objects and a computable version of lovsz local lemma , _ fundamenta informaticae _ , * 132 * , 114 ( 2013 ) , see also https://arxiv.org/abs/1305.1535 a. rumyantsev , m. ushakov , forbidden substrings , kolmogorov complexity and almost periodic sequences , _
stacs 2006 proceedings _ , lecture notes in computer science , * 3884 * , 396407 , see also https://arxiv.org/abs/1009.4455 .
a. shen , v.a .
uspensky , n. vereshchagin , _ kolmogorov complexity and algorithmic randomness _ , to be published by the ams , www.lirmm.fr/~ashen/kolmbook-eng.pdf .
( russian version published by mccme ( moscow ) , 2013 . )
there is one more sufficient condition for the existence of arbitrarily long sequences that avoid forbidden substrings .
here is it.[multiblock footnote omitted ] if the power series for @xmath141 ( where @xmath142 is the number of forbidden strings of length @xmath94 ) has all positive coefficients , then there exist arbitrarily long strings withour forbidden substrings .
moreover , in this case the number of @xmath4-letter strings without forbidden substrings is at least @xmath143 , where @xmath143 is the @xmath4th coefficient of this inverse series . to prove this result ,
consider the number @xmath144 of allowed strings of length @xmath3 .
it is easy to see that @xmath145 indeed , we can add each of @xmath94 letters to each of @xmath144 strings of length @xmath3 , and then we should exclude the cases where there is a forbidden string at the end .
this forbidden string may have length @xmath13 , then there are at most @xmath146 possibilities , or length @xmath81 , there are at most @xmath147 possibilities , etc .
( note that @xmath148 and @xmath149 ; note also that we can get a string with two forbidden suffixes , but this is ok , since we have an inequality . )
these inequalities can be rephrased as follows : the product @xmath150 has only non - negative coefficients .
denote the second term by @xmath151 ; if @xmath152 has only positive coefficients @xmath153 , ( as our assumption says ) , then the first term is a product of two series with non - negative coefficients .
the first factor ( @xmath154 ) starts with @xmath0 , so the @xmath4th coefficient of a product , i.e. , @xmath155 , is not less than @xmath4th coefficient of the second factor , i.e. , @xmath143 .
assume that the series in the denominator does not have a root , but the inverse series has all positive coefficients .
in fact , non - negative coefficients are enough to get a contradiction . for a series with all non - negative coefficients , or with finitely many negative coefficients ,
the radius of convergence is determined by behavior of the sum on the real line : when the argument approaches the convergence radius , the sum of the series goes to infinity .
now we have the product of two series @xmath156 that is equal to @xmath0 .
one of these series should have finite convergence radius , otherwise both are everywhere defined and the product is everywhere @xmath0 , but both are large for large @xmath31 .
look at the minimal convergence radius ( of two ) ; one of the series goes to infinity near the corresponding point on the real line , so the other one converges to zero , so it has bigger convergence radius and reaches zero at the real line .
finally , note that only the first factor ( the denominator ) may have a zero , since the other one has all non - negative coefficients .
now assume that the denominator has a zero ; we have to prove that the inverse series has only positive coefficients . in general , the following result is true ( d. piontkovsky ) : _ if the series @xmath157 has @xmath158 , and @xmath159 , and for some positive @xmath31 this series converges to @xmath18 , then the inverse series has all positive coefficients . _ to prove this statement , let @xmath160 be the root , so @xmath161 . recall the long division process that computes the inverse series .
it produce the sequence of remainders : the first @xmath162 is @xmath0 ; then we subtract from the @xmath3th remainder @xmath163 the product @xmath164 to cancel the first term , and get the next remainder @xmath165 . by induction
we prove that for each remainder @xmath166 : the second claim : by induction assumption the first coefficient in @xmath171 was positive , so we subtract the series @xmath172 with positive first and non - negative third , fourth , etc .
the first term cancels the first term in @xmath171 , the second term does not matter now , but all the subsequent coefficients are negative or zeros , since we subtract non - negative coefficients from non - positive ones . note that we have shown that if all coefficients of the series @xmath154 are non - negative , then they are positive .
also note that we get a bit stronger result compared to the entropy argument where we required the series to reach a negative value ( now the zero value is enough ) . | we consider several examples of probabilistic existence proofs using compressibility arguments , including some results that involve lovsz local lemma . |
the schrdinger equation for a combined system of electrons and nuclei enables us to predict most of the chemistry and molecular physics that surrounds us , including biophysical processes of great complexity .
unfortunately , this task is not possible in general , and approximations need to be made ; one of the most important and successful being the classical approximation for a number of the particles .
mixed quantum - classical dynamical ( mqcd ) models are therefore necessary and widely used .
we could say that , typically , the technique used to build mqcd models is a partial ` deconstruction ' of the quantum mechanics ( qm ) of the total system ( electrons and nuclei ) followed by a ` reconstruction ' that tries to recover the essential properties of the total schrdinger equation lost in the deconstruction process .
it is unrealistic to expect the reconstructed theory has the same predictive power as the schrdinger equation , so the reconstructed theory will apply with enough accuracy only to a subset of systems and questions ; a subset whose boundaries are difficult to predict a priori . in the literature , there are at least two common levels of deconstruction , one further away from the total schrdinger equation for electrons and nuclei , called born - oppenheimer molecular dynamics ( bomd ) , where electrons are assumed to remain in the ground state for all times , and another one closer to it , called ehrenfest dynamics ( ed ) , where nuclei are still classical ( as in bomd ) but the electrons are allowed to populate excited states ( some misleading notation used in the literature on ed is clarified in section 2 of ref . ) . in j. c. tully s surface hopping methods,@xcite for example , the deconstruction goes to bomd and the reconstruction proceeds by allowing the system to perform certain specially designed stochastic jumps between adiabatic states . in the decay of mixing formalism by d. g. truhlar and
coworkers,@xcite the deconstruction stops at the ed and the reconstruction is developed by adding decoherence to it .
this has been shown to be more accurate than surface hopping methods for non - born - oppenheimer collisions . when considered for a single system ,
ed is a fully coherent semiclassical method , and hence purity preserving . as decoherence must be a property of any realistic model,@xcite many mqcd models
have been reconstructed in order to produce electronic decoherence .
see , for example , refs . , that range from one of the most classic in this matter @xcite to one of the most recent.@xcite . in ref .
, we can find a recent study of decoherence in the context of surface hopping and an important conclusion : averaging over a swarm of initial conditions , decoherence can be measured ; but the method can not capture all the observed effects ( for example , the averaging is not enough to capture the physics of wave packet bifurcations on multiple surfaces ) . in the case of decay of mixing formalisms , where the trajectories in the swarm are considered as independent , decoherence phenomena are incorporated algorithmically ( see eq .
18 of ref . ) .
in this paper we study the problem from a different perspective .
first of all , we consider a complete statistical description of ed , which was introduced in ref . in full detail .
based on that construction , we develop a description of quantum statistical mechanics which can be adapted to mixed quantum - classical systems in a straightforward but rigorous manner .
this allows us to consider , in a simple way , the evolution of the purity of the quantum subsystem of our ehrenfest model .
we prove thus that , while a single ehrenfest system evolves preserving the purity of the quantum state , the behavior changes dramatically when a statistical distribution is considered and , in general , it introduces a change in the purity of the quantum system along the trajectory .
therefore , we can claim that our general statistical ed formalism is purity non - preserving ; a property which always accompanies decoherence phenomena ( see for example sec . 3.5 in ref . ) .
this work opens thus a new line of research in the direction of taking into account decoherence phenomena in ed .
of course , those ingredients still missing for a proper description of decoherence in our ehrenfest statistical formalism could be later added in a reconstruction process similar to the ones mentioned before , but this time starting from a , presumably better , purity non - preserving dynamics a full study of the decoherence process is a very complex task which includes deep quantum theoretic concepts as the measurement problem and the interpretations of qm ( see , for example , refs . for a general discussion , and ref . for the analysis of the decoherence phenomenon on molecular systems ) . in this paper
we voluntarily restrict ourselves to a simple property reflecting the decoherence phenomenon .
namely , the change in the ` degree of mixture ' of the quantum state in a mqcd model , as quantified by the purity @xmath2 .
as mentioned , we shall see in sections [ sec : purity - non - pres ] and [ sec : ejemplo ] that ed provides a framework where this change takes place .
the actual relation with the electronic decoherence in molecular systems requires a much more involved analysis which will be developed in the future . besides the incomplete description of decoherence , usual approaches to ed
have also been often criticized on the basis that it does not yield the boltzmann equilibrium distribution for the electrons exactly.@xcite the lack of this property , which we agree is desirable , is however not enough to rule out ed for all applications , as we recently argued .
@xcite the structure of the paper is as follows : sections from [ sec : notion - decoherence ] to [ sec : purity - non - pres ] introduce the mathematical formalism and the relevant definitions , which are then put into practice in the numerical example in sec .
[ sec : ejemplo ] .
[ sec : notion - decoherence ] reviews the notion of purity in qm and proves the well known fact that ed preserves the purity of the quantum subsystem when we consider the evolution of a single trajectory from perfectly determined initial conditions .
[ sec : geom - quant - stat ] presents a very brief summary of the formulation of geometric qm ( see for a more careful presentation ) and it provides an analogous formulation of a quantum statistical system within the same framework .
in particular , a suitable formulation of the purity of a quantum system is introduced .
[ sec : purity - non - pres ] presents the main contribution of the paper : first , we review the geometrical formulation of ed and its associated statistical equilibrium introduced in . then , we adapt the tools introduced in the previous section in order to be able to study the evolution of the purity of the quantum subsystem in a suitable way , and to show that ed is purity non - preserving . the use of the geometrical formalism , as we will see in what follows , allows to perform a very direct analysis of the problem . in sec .
[ sec : ejemplo ] , we numerically illustrate the change in purity produced by ed using a very simple but extremely useful example : a statistical system defined by a pure quantum state and an ensemble of initial conditions of the classical subsystem .
such a system has been used in the literature as a natural framework for molecular dynamics ( see for example refs . ) .
we use it as the simplest nontrivial ehrenfest statistical system where we can show how the purity of the quantum part of the system evolves in time depending on the coupling between the classical and quantum systems , the initial momentum of the classical particles , the dimension of the quantum state space and the number of trajectories considered in the initial conditions . in sec .
[ sec : concl - future - work ] we present our conclusions and our plans for future works .
given a hilbert space @xmath3 , we shall call * density states * to the elements @xmath4 obtained as convex combinations of rank - one projectors @xmath5 , each element satisfying @xmath6 with a probability vector , @xmath7 with @xmath8 and @xmath9 , @xmath10 .
the expression of a general density state is then @xmath11 the evaluation of some observable @xmath12 on this state is given by @xmath13 the state of the quantum system is said to be * pure * if the density matrix which represents it is a rank - one projector , i.e. , if the convex combination above contains only one term .
if this property does not hold , the system is said to be in a * mixed * state , since , from the physical point of view , there is a statistical mixture of the different pure states represented by the density matrices @xmath14 above .
being a hermitian operator , the matrix @xmath4 can be diagonalized .
its eigenvalues @xmath15 satisfy @xmath16 if the state is pure , there is one eigenvalue equal to one , the rest being zero . obviously , the rank of @xmath4 as a projector on the hilbert space @xmath3 coincides with the number of nonvanishing eigenvalues .
it is an immediate property that a state @xmath4 is pure if and only if @xmath17 the proof requires only eq . .
the description of a quantum system in terms of a density matrix uses von neumann s equation to introduce the dynamics .
then we know that , given the hamiltonian operator @xmath18 , the evolution of the state @xmath4 is given by @xmath19,\ ] ] where @xmath20 $ ] is the usual commutator of operators . using a simple proof which is formally identical to the one that we shall present in the next section , one can easily show that this dynamics is purity - preserving , i.e. , @xmath21 .
the ehrenfest equations @xcite for a system composed of a set of @xmath22 classical particles ( typically nuclei ; described by the phase space variables @xmath23 , @xmath24 ) and a set of @xmath25 quantum particles ( typically electrons ; described by a wavefunction @xmath26 , defined on the space parameterized by @xmath27 ) are : @xmath28 where @xmath29 and the electronic hamiltonian operator @xmath30 is related to the molecular one @xmath18 and it is defined as follows : @xmath31 where all sums must be understood as running over the whole natural set for each index , @xmath32 is the mass of the @xmath33-th nucleus in units of the electron mass , and @xmath34 is the charge of the @xmath33-th nucleus in units of ( minus ) the electron charge . at first sight ,
given the similarity between eq .
and the schrdinger equation for an isolated full - quantum system , one might erroneously think that the ehrenfest evolution for the quantum part of the system is unitary .
@xcite if this was correct , then it would be trivial to prove that ed is purity preserving , but this is not the case .
as is well known , for a one to one transformation , we can define unitarity as the property of preserving the scalar product , i.e. , given two arbitrary vectors @xmath35 and @xmath36 , we say that @xmath37 is unitary if @xmath38 .
one can easily see that any reversible transformation @xmath37 that enjoys this property is necessarily linear : @xmath39 as @xmath37 is reversible , @xmath40 is an arbitrary vector , and therefore , we must have , @xmath41 but , although the quantum part of the the equations of motion in resembles a typical schrdinger equation , the coupling with the classical part makes the evolution of the quantum system nonlinear . consequently , it can not be a unitary transformation as defined above . despite this non - unitarity
, it is very simple to prove that , if we consider the evolution of a single trajectory @xmath42 of an ehrenfest system , the quantum part is always in a pure state : let @xmath42 be the initial state of an ehrenfest system subject to the dynamics given by eqs .
. then , the quantum part of the system is always in a pure state we consider the density matrix @xmath43 corresponding to the quantum part of the ehrenfest system .
the evolution of @xmath44 is given by eq .
, which induces a von neumann - like evolution for the density matrix at every time @xmath45 @xmath46,\ ] ] being @xmath30 the electronic hamiltonian in
. then , @xmath47\rho ) \\ & = 2\left ( \mathrm{tr}(h_e\rho\rho ) - \mathrm{tr}(\rho h_e \rho ) \right ) = 0,\end{aligned}\ ] ] for all times @xmath45 . hence ,
if @xmath48 at @xmath49 , it will remain so .
the main goal of the rest of the paper is to prove that , when we consider the case of a statistical ensemble of ehrenfest trajectories , this is no longer the case : the evolution of an ensemble whose quantum part is a pure state at @xmath49 will become an ensemble in which the quantum part is mixed as long as the initial conditions for the nuclei are not perfectly determined .
thus , in such a case , the ed produces a purity change , a necessary condition for decoherence .
the aim of this section is to provide a very brief summary of the mathematical formalism more thoroughly introduced in and references therein .
classical mechanics can be formulated in several mathematical frameworks each corresponding to a different level of abstraction : newton s equations , the hamiltonian formalism , the poisson brackets , etc .
perhaps its more abstract and general formulation is geometrical , in terms of poisson manifolds .
similarly , qm can also be formulated in different ways , some of which resemble its classical counterpart ( see ref . for a classical reference in the context of molecular systems , ref . for a more recent one , and refs . for more mathematical approaches ) .
for example , the observables ( self - adjoint linear operators ) are endowed with a poisson algebra structure ( based on the commutators ) almost equal to the one that characterizes the dynamical variables in classical mechanics . moreover
, schrdinger equation can be recast into hamilton s equations form by transforming the complex hilbert space into a real one of double dimension .
the observables are also transformed into dynamical functions in this new phase space , in analogy to the classical one .
finally , a poisson bracket formulation has also been established for qm , which permits to classify both the classical and the quantum dynamics under the same heading .
this variety of formulations does not emerge from academic caprice ; the succesive abstractions simplify further developments of the theory , such as the step from microscopic dynamics to statistical dynamics : the derivation of liouville s equation ( or von neumann s equation in the quantum case ) , at the heart of statistical dynamics , is based on the properties of the poisson algebra.@xcite consider a basis @xmath50 for the hilbert space @xmath51 .
each state @xmath52 can be written in that basis with complex components ( or coordinates , in more differential geometric terms ) @xmath53 : @xmath54 now , we can just take the original vector space inherent to the hilbert space , and turn it into a real vector space ( denoted as @xmath55 ) , by splitting each coordinate into its real and imaginary parts : @xmath56 we will use real coordinates @xmath57 , @xmath58 , to represent the points of @xmath3 when thought of as real vector space elements . from this point of view , the similarities between the quantum dynamics and the classical one will be more evident .
it is important to notice , though , that despite the formal similarities these coordinates @xmath57 do not represent physical positions and momenta of any actual system .
they simply correspond to the real and imaginary parts of the complex coordinates used for the hilbert space vectors in a given basis .
the scalar product of the hilbert space is encoded in three tensors defined on the real vector space @xmath55 .
the interested reader is addressed to ref . for the details .
we just highlight here that two of these tensors correspond to a metric tensor @xmath59 and a symplectic one @xmath60 which allow us to write the expression of the schrdinger equation as a hamilton equation , in a form which is completely analogous to the hamiltonian formulation of classical mechanics .
it is precisely this similarity the key ingredient to successfully combine classical and quantum mechanics in a well - defined framework to describe the ehrenfest equations as a hamiltonian system , as we will summarize later and it can also be seen in ref . .
in this formalism , instead of considering the observables as linear operators ( plus the usual requirements , self - adjointness , boundedness , etc . ) on the hilbert space @xmath3 , we shall represent them as functions defined on the real space @xmath55 .
the reason for that is to resemble , as much as possible , the classical mechanical approach .
but we can not forget the linearity of the operators , and thus the functions must be chosen in a very particular way .
the usual choice is inspired in ehrenfest s description of quantum mechanical systems and defines , associated to any operator @xmath12 on @xmath3 , a function of the form : @xmath61 the operations which are defined on the set of operators can also be translated into this new language .
thus , the associative product of operators ( the matrix product when considered in a finite dimensional hilbert space ) , the commutator ( which encodes the dynamics ) and the anticommutator can be written in terms of the functions of the type defined in eq . . as an example , we can write the case of the commutator , which will be used later : given two operators @xmath12 and @xmath62 , with the corresponding functions @xmath63 and @xmath64 , the function associated to the commutator @xmath65=i(ab - ba)$ ] ( the imaginary unit is used to preserve hermiticity ) is written as @xmath66}=\ { f_a , f_b\}=\frac 12 \sum_k \left ( \frac { \partial f_a}{\partial q^k}\frac{\partial f_b}{\partial p_k}- \frac { \partial f_a}{\partial p_k}\frac{\partial f_b}{\partial q^k } \right ) .\ ] ] thus , from the formal point of view , the operation is completely analogous to the poisson bracket used in classical mechanics .
another important property in the set of operators of qm is the corresponding spectral theory . in any quantum system
, it is of the utmost importance to be able to find eigenvalues and eigenvectors .
we can summarize these properties in the following result : if @xmath67 is the function associated to the observable @xmath12 , then , as a consequence of ritz s theorem , @xcite * the eigenvectors of the operator @xmath12 coincide with the critical points of the function @xmath67 , i.e. , @xmath68 * the eigenvalue of @xmath12 at the eigenvector @xmath26 is the value that the function @xmath67 takes at the critical point @xmath26 . as usual
, the dynamics can be implemented in essentially two different forms ( but always in a way which is compatible with the geometric structures introduced so far ) : the so - called schrdinger and heisenberg pictures @xcite . in the heisenberg picture , which is the one we will use in what follows ,
the dynamics is introduced by translating the well - known heisenberg equation into the language of functions : @xmath69 being @xmath70 the function associated to the hamiltonian operator and @xmath12 any observable . a classical result in qm states that , given a quantum system , the average value of any observable @xmath12 can always be computed as the trace of the observable and some density state @xmath4 , as defined in section [ sec : noti - purity - quant ] : @xmath71 this result is known as gleason theorem ( see ref . for details ) . instead of using the density matrix @xmath4
, we can use an alternative approach which is formally closer to the description of classical statistical systems and is used , for instance , in ref . .
consider a probability distribution @xmath72 on @xmath55 and a volume element @xmath73 , satisfying the properties : * @xmath74 * expected values can be computed as @xmath75 for all @xmath63 of the form ; @xmath12 being a hermitian operator .
notice that we have chosen to integrate over all the states in @xmath55 and divide by the norm of the state , as it is done in the final section of ref . .
this is equivalent to integrate over the states of norm one as it was also done in the first sections of ref . . the canonical symplectic form of @xmath55 described in ref
provides a natural candidate for the volume form since it is also preserved by the quantum evolution ( see ref . for the technical details ) .
some simple examples for the distribution @xmath72 can also be provided : for the case of the pure state @xmath76 , we can use @xmath77 to satisfy the above two equalities .
analogously , a mixed state @xmath78 ( where @xmath79 and @xmath80 ) can be represented by @xmath81 in particular , it is straighforward to prove that , in this case , @xmath82 for @xmath83 and @xmath84 . the definition of the function @xmath72 contains a number of ambiguities which are explained in detail in ref . .
essentially , we can add to any @xmath72 a term which integrates to zero and has vanishing second - order momenta .
due to the structure of the observable functions @xmath63 , this modification will not change any average value computed as in eq .
( [ eq : exp_q_general ] ) , nor will it change the normalization condition for @xmath72
. this defines an equivalence class of distributions that produce the same average values , and ( through the relationship between distributions and density matrices ) gleason theorem implies that there is always a distribution in the class in which the fact that we are dealing with a convex combination of rank-1 projectors is visible .
this is what we used in the example above . in order to advance in the formulation and illustrate these facts
more precisely , we can consider , for every @xmath85 , the following function : @xmath86 now , it is easy to see that the following ( averaged , now @xmath26-independent ) function @xmath87 is the phase - space function associated to a density operator @xmath4 defined by @xmath88 i.e. , @xmath89 with these definitions , we can now prove the following result : [ sec : density ] let us consider a quantum system in a state described by a probability distribution @xmath72 . then , the equivalence class of distributions that produce the same expected values as @xmath72 contains as an element : @xmath90 where @xmath91 is the set of critical points of @xmath92 ( as defined in eqs .
( [ eq:11 ] ) and ( [ eq:11bis ] ) ) and @xmath93 it is immediate if we realize that @xmath94 indeed , the operator @xmath4 appearing in this expression and defined in eq .
( [ eq:14 ] ) can be shown to be a density matrix ( i.e. , @xmath95 , @xmath96 , and @xmath97 ) , and gleason s theorem guarantees it is unique .
we also know that , if we use the spectral decomposition of @xmath4 , i.e. , @xmath98 , with @xmath83 , being @xmath99 and @xmath100 its eigenvectors and eigenvalues , respectively , we also have that @xmath101 as we set out to prove . hence , from gleason theorem , we know that , among all the equivalent distributions , there is always one @xmath72 equal to a convex combination of dirac - delta functions .
notice that the function @xmath72 provides us with all the information encoded in the density matrix @xmath4 .
as a probability density , it allows us to define the average values of the observables , and in the form @xmath92 , it allows us to read the spectrum of @xmath4 from the set of critical points .
this result allows us to realize in terms of @xmath72 any quantum system : as the average values coincide with those obtained from the spectral decomposition of the density matrix , we can use it to implement any desired model .
we will see a practical example in the following sections .
finally , we would like to analyze purity in this geometrical context .
we saw in section [ sec : notion - decoherence ] that purity preservation is encoded in the behavior of @xmath4 as a projector .
if the evolution of the system preserves the purity of the density matrix we say that the evolution is * purity - preserving * , while in the other case , we call it * purity non - preserving*. our first task is to express in this geometrical language the concept of purity
. consider the following expression for any @xmath102 : @xmath103 then , @xmath104 if the state is pure and @xmath105 if the state is mixed .
this is so because , trivially , @xmath106 let us now mention how this change in the purity of the state can be detected in the measurement of average values of observables .
recall that a change in the purity as above produces a transformation at the level of the states such that a pure state corresponding to a distribution of the form @xmath107 becomes a distribution of the form @xmath108 , where there is more than one value of @xmath109 different from zero .
then , it is immediate to prove that the average value of a generic observable @xmath12 will be different between one case and the other .
in this section , we will now extend the previous construction to the ehrenfest case , by combining it with the approach introduced in ref . .
first , let the physical states of our ehrenfest system correspond to the points in the cartesian product @xmath110 where @xmath111 is the phase space of the classical system .
the physical observables will be now functions defined on that manifold . to define statistical averages of observables depending on classical and quantum degrees of freedom ( i.e. , functions as @xmath112 ) we consider @xmath113 where @xmath114 represents the classical degrees of freedom , @xmath115 the quantum ones , and
@xmath116 is the volume on the state space manifold @xmath22 .
we can now ask what properties we must require from @xmath117 in order for eq . to correctly define the statistical mechanics for the ehrenfest dynamics ( ed ) .
analogously to what happened in the quantum case , the conditions are as follows : * the expected value of any constant observable should be equal to that constant , which implies that the integral on the whole set of states is equal to one : @xmath118 * the average , for any purely quantum observable @xmath63 of the form in , associated to a positive definite hermitian operator @xmath12 , should be positive .
this implies the usual requirement of positive probability density in standard classical statistical mechanics . in ref .
, it was proved that the ed defined on the manifold @xmath22 is hamiltonian with respect to the poisson bracket @xmath119 where @xmath120 represents the usual poisson bracket of the classical degrees of freedom , and @xmath121 represents the poisson bracket defined by eq . .
being hamiltonian , we know that we can define an invariant measure on the space of states @xmath22 .
we shall denote such a measure by @xmath122 .
thus , the dynamics defined on the microstates is straightforwardly translated into the probability density @xmath117 as a liouville equation : @xmath123 where @xmath70 is the hamiltonian function of the ehrenfest system : @xmath124 analogously , the evolution of any function @xmath112 is given by @xmath125 again , this property provides us with a natural candidate for the volume element @xmath122 ( and , for analogous reasons , also for @xmath126 ) arising from the symplectic form which gives its hamiltonian structure to the liouville equation in this context .
as it happens in the pure quantum case , this volume form is preserved by the dynamics ( see ref . ) . with this in mind
, we can consider the analogue of the objects introduced in the previous section .
hence , given an ehrenfest system in a state described by a probability density @xmath127 , where @xmath128 and @xmath129 , we can consider the definition of the operator @xmath130 which still depends on the classical variables @xmath131 and , therefore , it can be interpreted as having turned the phase - space representation of the quantum part into the more familiar one based on density matrices .
also , since we have not integrated over @xmath131 , this object still represents in a certain way a probability density in the classical part of the space . as for any @xmath131-dependent operator ( see ref . ) , we can define the associated phase - space function : @xmath132 it is also possible to integrate again the object in eq .
( [ eq:25 ] ) , and define : @xmath133 which is a purely quantum object encoding the averaged information of the complete system .
notice that , as it is usually done and in order to lighten the notation , we will often use the same symbol for different objects ( as in @xmath134 and @xmath4 ) , understanding that it is the explicit indication of the variables on which they depend what distinguishes them notationally . also , for simplicity , we use the same symbols for operators in the full - quantum case in sec .
[ sec : gqsm ] , and for the ones in the quantum - classical scheme in this section .
we can now formulate the dynamics in terms of this operator .
after a brief computation , we obtain : @xmath135,\ ] ] where @xmath136 is the very electronic hamiltonian defined at the beginning of the paper . in ed , considered statistically , this equation represents the analogue for the mixed case of von neumann s equation . if we consider a single ( ` pure ' ) state of the classical system @xmath137 and a pure state of the quantum system @xmath138 , i.e. , @xmath139 we obtain @xmath140 analogously , the associated operator in eq
. reads @xmath141 and the one in eq .
is @xmath142 as expected .
we can also consider the two marginal distributions : * the distribution in @xmath55 obtained by integrating out the classical degrees of freedom : @xmath143 * and the corresponding classical version in @xmath111 : @xmath144 obviously both functions are distribution functions on the corresponding manifolds with analogous properties to @xmath117 , and they could be used to compute expected values of functions depending only on @xmath26 or @xmath131 , respectively .
it is immediate to check that the definitions make sense for distributions of the form @xmath145 i.e. , for ` pure ' classical part and a quantum - mixed one canonically expressed with deltas . in this case , the marginal distributions are of the form @xmath146 also notice that , in terms of the quantum marginal distribution @xmath72 , we can write the density matrix in eq . as @xmath147
once we have recovered the needed ingredients , we can discuss the quantum purity of a system governed by ed : we say that a quantum - classical system is * quantum - pure * if and only if @xmath148 being @xmath4 the one defined in eqs . and .
in case that the state of a system does not satisfy the condition above , we say that it is * quantum - mixed*. for the sake of completeness , and in order to better connect with the purely quantum case we discussed in sec .
[ sec : gqsm ] , we can consider now the function obtained from @xmath92 in eq .
averaging directly over @xmath111 , i.e. , @xmath149 this function plays the role of the function in in the pure case .
indeed , we have that @xmath150 and hence it corresponds to the quantum expected value of the operator @xmath4 in eqs . and
, whose full - quantum analogue is the one in in sec .
[ sec : gqsm ] . consider the following initial distribution evolving under ed : @xmath151 this system is completely deterministic and , therefore , the liouville equation will produce exactly , as a solution , the integral curves of ed @xmath152 .
thus we can write : @xmath153 consider now a slightly more complex system , consituted by a distribution of @xmath1 equally probable classical states , and a pure quantum state at @xmath154 : @xmath155 we can also write the marginal distributions as we did in section [ s:4.a ] : @xmath156 the evolution of such a system becomes @xmath157 where @xmath158 represents the ehrenfest trajectory having @xmath159 as initial condition .
the evolved marginal distribution in the quantum manifold is now : @xmath160 and then , using eq .
, we have : @xmath161 where , of course , @xmath162 the associated density matrix at time @xmath45 reads : @xmath163 and at time @xmath49 : @xmath164 now , the purity at time @xmath49 is @xmath165 but at a general time @xmath45 , @xmath166 hence , the purity of the system seems to evolve in time , in general in an involved way .
we can compute , though , its time derivatives , in order to get an idea about its initial evolution .
after some calculations which we detail in appendix [ app : derivatives ] , one can see that , for the initial state considered in this section , we have @xmath167 and @xmath168 which is negative definite unless the expectation value of @xmath169 at @xmath138 vanishes .
this means that the purity evolves from its initial value , and it does so by decreasing , which is entirely expected if we realized that we started from a state in which the purity is maximal .
thus , we can see that the evolution has made a quantum - pure system become a quantum - mixed one , and we can claim that : the statistical ehrenfest evolution is purity non - preserving for hamiltonians with quantum - classical couplings .
this is the main result of our paper : even though the ed of a single - trajectory state does preserve purity , when we consider a statistical state the behavior changes . in the case above ( and as we will see numerically in the next section ) , we show how it is possible to transfer uncertainty from the classical domain into the quantum one .
analogously , it is straightforward to see that an analogous process happens when we consider a single classical state coupled to an ensemble of quantum states .
this uncertainty transfer takes place whenever the dynamics of the quantum and the classical subsystems are coupled to each other , since such coupling produces a splitting of the trajectories from the given initial conditions and thus the mixing of the final states .
to illustrate the concepts introduced in the previous section and to see the purity change in a complex numerical case , let us consider a situation with an equally probable initial distribution as in eq . of classical particles with @xmath170 and a quantum part constituted by a 10-level system .
such a system has been used in the literature as a natural framework for molecular dynamics ( see for example refs . ) .
we consider a hamiltonian function for the quantum - classical system of the following form : @xmath171 where @xmath172
are canonically conjugated classical variables and @xmath12 and @xmath62 are hermitian matrices acting on the quantum vector space @xmath173 .
the classical part is written in action - angle coordinates to simplify the analysis .
the dynamics of the system is obtained from the solutions of the @xmath1 different trajectories with initial conditions defined by each one of elements of the initial classical distribution @xmath174 .
the resulting distribution takes thus the form given by eq . .
as it can be found in the supplementary material , for all the trajectories presented below the initial conditions for the classical subsystem are chosen as @xmath175 the points were fixed as five random points in the interval @xmath176 $ ] . choosing different initial conditions leads to equivalent results .
the quantum initial condition is chosen to be , for all five trajectories , @xmath177 the evolution defines thus a equiprobable distribution of @xmath1 quantum - classical single trajectories @xmath178 .
for each trajectory , the equations of motion are of ehrenfest type and they are given by eqs .
, with @xmath179 .
these dynamical equations exhibit two different regimes : * for @xmath180 or @xmath181 , the classical and the quantum subsystems evolve uncoupled .
the purity of the quantum subsystem is always equal to one . * for non - zero coupling constants and classical momenta ,
the behavior of the system depends sharply on the initial conditions : the evolution from different initial conditions for the classical subsystems is very different .
the purity tends then to evolve to the purity corresponding to a set of five random projectors on 1-dimensional subspaces of the hilbert space .
the evolution of the purity of the resulting system is obtained from eq . after integrating the dynamics numerically .
some interesting observations can be extracted from the results : * in fig .
[ fig:3 ] we represent the evolution of the purity for a fixed value of the coupling constant @xmath182 and increasing value for the initial condition of the classical momentum @xmath183 .
we see how this makes the system change its originally integrable behavior and become more and more chaotic .
+ and @xmath184 ( dashed green line ) , @xmath185 ( red line ) , @xmath186 ( blue line ) and @xmath187 ( brown line).we also depict the reference ( black line ) of the level of purity of a distribution of @xmath170 random projectors on @xmath173.,width=302 ] * instead , we can consider a fixed value of the initial classical momentum and increase the value of the coupling .
it can be remarked that the system reaches the level of purity of the set of random projectors much faster than in the previous case , ( see fig .
[ fig:4 ] ) .
+ and @xmath180 ( dashed green line ) , @xmath188 ( blue line ) , and @xmath189 ( red line ) .
again , the reference ( black line ) represents the level of purity of a distribution of @xmath170 random projectors on @xmath173.,width=302 ] the interesting behaviour of the purity shown for some values of the parameters in figs .
[ fig:3 ] ( brown line ) and [ fig:4 ] ( red line ) , in which its value rapidly decreases to a given low one and fluctuates around it after that , can be explained as a consequence of the dynamics mentioned before . if the system exhibits sensitive dependence on the initial conditions , after a small lapse of time , the different trajectories become decorrelated from one another in the hilbert space .
therefore , the density matrix becomes the normalized sum of @xmath1 rank one projectors chosen at random .
in this situation , it can be shown ( see appendix [ app : random ] ) that the purity is distributed around an expected value @xmath190 = \frac{n+d-1}{nd},\ ] ] which is represented by the black straight lines in the figures , with fluctuations of size @xmath191 - \operatorname{e}[\mathrm{tr}(\rho^2)]^2\right)^{1/2 } \nonumber \\ = & \frac{\sqrt{2}}{nd}\left(\frac{(n-1)(d-1)}{n(d+1)}\right)^{1/2},\end{aligned}\ ] ] where @xmath1 is the total number of trajectories used and @xmath0 is the dimension of the quantum hilbert space .
notice that if @xmath0 is very large , the expected value for the purity tends to its minimal value @xmath192 and the fluctuations tend to zero .
remember that the degree of mixture of a quantum system ranges from the pure state case ( i.e. , purity equals to one ) , for which the density operator is a projector on a one dimensional subspace of the hilbert space and the maximal mixture case ( thus minimum purity ) which corresponds to a density operator which is proportional to the identity matrix . as it must have trace equal to one , the proportionality factor is equal to the inverse of the dimension of the hilbert space . figs .
[ fig:3 ] and [ fig:4 ] show also the effect of the coupling between the classical and the quantum subsystems , as well as the effect of the momentum ( or equivalently , the energy ) of the classical particles on the evolution of purity
. the greater the strength of the coupling and the energy of the classical particle , the faster the system reaches its asymptotic behavior .
although clearly the coupling has a stronger effect .
thus , analogously to what happens in the case of molecular dynamics , where the velocity of the classical nuclei induces the coupling between all the eigenstates of the electronic hamiltonian , we see here that it also has the effect of mixing the quantum part if the ehrenfest system is treated statistically . from our analysis and example , we can then conclude that statistical ehrenfest dynamics provides a framework in which the evolution affects the quantum dynamics in at least one way decoherence does .
changing the degree of mixture of the quantum state is certainly one of the most relevant effects of electronic decoherence on molecular systems , although further work is required to analyze whether or not other decoherence effects can be explained by our construction .
an appropriate description of the electronic decoherence in molecular systems is on the wishlist of every quantum - classical dynamics scheme .
in which amount each theoretical model includes the sought effects is a complicated question whose answer will depend both on the model and on the intended application . loosely speaking
, we could expect the different models to range from ` no electronic decoherence at all ' ( e.g. , bomd ) to ` a perfect description of quantum electronic decoherence ' ( say , full quantum dynamics of electrons and nuclei ) , with most of them lying somewhere in between the two extremes .
several methods @xcite have been developed to deal with the problem but , to our knowledge , without a definitive solution .
pure ehrenfest dynamics ( ed for single states ) clearly lies in the ` no decoherence at all ' side of the spectrum , since it does not even allow the change of purity .
however , if we are willing to accept the possibility that the electrons evolve into a mixed state , i.e. , a statistically uncertain electronic state , then , to perform a coherent analysis , we should also allow the initial conditions of the nuclei to be described statistically .
this point of view was used in some of the papers mentioned above when considered within surface - hopping or decay - of - mixing formalisms . in this work
we implement the idea using the statistical description of ehrenfest dynamics introduced in ref .
combined with a specially convenient geometric formalism we introduce . using these methods , one can observe that a system starting from uncertain nuclear initial conditions and a pure electronic state evolves into a situation in which the electronic state is no longer pure but mixed , i.e. , ed is capable of transferring uncertainty between the nuclei and the electrons .
besides , our method provides us also with tools to compute in a simple but rigorous way the purity change , and it indicates an interesting dependence of the effect on the number of trajectories @xmath1 and the dimension of the hilbert space @xmath0 , as well as on the coupling and the velocity ( or energy ) of the classical system . ` how much ' decoherence the statistical ed contains , as related to interesting practical applications or to other mixed quantum - classical schemes,@xcite is a complex and important question that we shall explore in future works .
we thank a. castro for many fruitful discussions on these topics and many interesting suggestions on previous versions of the manuscript . this work has been supported by the research grants
e24/1 , 24/2 and e24/3 ( dga , spain ) , fis2009 - 13364-c02 - 01 , fpa2009 - 09638 and fis2009 - 12648 ( micinn , spain ) , and araid and ibercaja grant for young researchers ( spain ) .
in this appendix we derive the expected value of the purity and its fluctuations when the density matrix is obtained from the sum of decorrelated , random , rank - one projectors .
the purity for a density matrix of the form @xmath193 is @xmath194 where we have defined @xmath195 next , we will determine the probability density for @xmath196 when @xmath197 and @xmath198 are two decorrelated random vectors .
given the global @xmath199 symmetry of the problem we can take the first vector to be @xmath200 and the second chosen at random .
if we denote @xmath201 we get @xmath202 where @xmath203 .
we also need the adequate probability measure in @xmath55 that distributes the random vector @xmath198 .
it can be defined by @xmath204 where @xmath205 is a positive function chosen so that @xmath206 as we shall see , the actual form of @xmath205 is not relevant for the distribution of the purity
. it will be convenient to write the probability measure in @xmath207 in the following way : @xmath208 where @xmath209 represents polar coordinates in the plane @xmath210 , @xmath211 is the radial coordinate in @xmath212 ( i.e. , @xmath213 ) , while the volume element @xmath214 stands for the angular coordinates in @xmath212 . in these coordinates @xmath215 and @xmath216 .
the next step is to perform the change of variables from @xmath217 to @xmath218 . taking into account that the jacobian is @xmath219 one obtains @xmath220 and marginalizing out all variables except @xmath221 we find the needed measure : @xmath222 once we have determined the probability distribution for @xmath221 we can compute the average value for the purity . using the expression at the beginning of this appendix , @xmath223 = \frac1{n^2}\left ( n+2\sum_{j < k } \operatorname{e}[\chi_{ij}]\right ) = \frac1{n}\big ( 1+(n-1 ) \operatorname{e}[\chi]\big),\ ] ] where we have used that all random variables @xmath224 are identically distributed .
finally , given that @xmath225=\int_0 ^ 1\chi d\pi_\chi=\frac1d,\ ] ] we obtain the sought result : @xmath223=\frac{n+d-1}{nd}.\ ] ] as for the fluctuation , one has @xmath226 - \operatorname{e}[\chi_{jk}]^2 ) \nonumber \\ & = \frac{2(n-1)}{n^3 } ( \operatorname{e}[\chi^2 ] - \operatorname{e}[\chi]^2 ) = \frac{2(n-1)}{n^3}\frac{(d-1)}{d^2(d+1)},\end{aligned}\ ] ] where we have used that @xmath227=\frac2{d(d+1)}.\ ] ]
in sec . [ sec : transferring ] , we explained how an initial uncertainty in the classical part of the state of an ehrenfest system produces a change in the purity at @xmath49 , even if the initial state is quantum - pure . in this appendix , we present in more detail the calculations that led us to that conclusion .
first of all , we need the first time - derivative of the purity . we know that the evolution of the @xmath228 is given by eq . .
then , the evolution of the purity can be written as : @xmath229 \cdot p_{\psi ' } \big ) \nonumber \\ = & i\hbar^{-1 } \int d\mu d\mu ' f f ' \mathrm{tr}\big ( [ p_\psi , p_{\psi ' } ] \cdot h_e \big),\end{aligned}\ ] ] where the integral is taken over @xmath230 , and we have denoted [ eq : abb ] @xmath231 in the last step , we also used that @xmath232 \cdot p_{\psi ' } \big ) & = \mathrm{tr } \big ( p_\psi\cdot h_e\cdot p_{\psi ' } - h_e\cdot p_\psi\cdot p_{\psi ' } \big ) \\ & = \mathrm{tr } \big ( h_e\cdot p_{\psi'}\cdot p_\psi - h_e\cdot p_\psi\cdot p_{\psi ' } \big)\\ & = \mathrm{tr}\big ( h_e\cdot [ p_\psi , p_{\psi ' } ] \big ) \\ & = \mathrm{tr}\big ( [ p_\psi , p_{\psi ' } ] \cdot h_e \big ) .
\end{aligned}\ ] ] also , as it is common in statistical dynamics , we can assign the time - evolution of the state to the probability distribution @xmath117 and see the objects @xmath131 , @xmath233 and @xmath44 , @xmath234 just as the initial conditions , or we can alternatively think that @xmath117 is the static distribution of initial conditions and consider that the time - evolving objects are @xmath131 , @xmath233 and @xmath44 , @xmath234 . either dynamical image is valid , and the two of them produce , of course , the same result , but we have performed the calculation thinking in the second way , which looked to us slightly more direct .
using again eq . - , we can also compute the second derivative of the density matrix , @xmath238 \big ) \right .
\nonumber \\ & - \hbar^{-2 } \left .\big ( \big [ h_e , \big [ h_e , p_\psi ] \big ] \big ) \right),\end{aligned}\ ] ] where we have used the same notation in eqs . , and the integral is this time extended to @xmath239 . with this expression , we can compute the second derivative of the purity : @xmath240 now , using eqs . , and
we can calculate a more explicit form for this second derivatives in terms of the objects associated to the geometric formalism @xmath241 \cdot p_{\psi ' } \big ) \right .\nonumber \\ & - & \hbar ^{-2 } \mathrm{tr } \big ( \big [ h_e , [ h_e , p_\psi ] \big ] \cdot p_{\psi ' } \big ) \nonumber \\ & - & \left .\hbar ^{-2 } \mathrm{tr } \big ( \big[h_e , p_\psi\big ] \cdot \big[h'_e , p_{\psi'}\big ] \big ) \right ] \nonumber \\
= \int & d\mu & d\mu ' f f ' \left [ i\hbar^{-1 } \mathrm{tr } \big ( \big [ \ { h_e , f_h \ } , p_\psi \big ] \cdot p_{\psi ' } \big ) \right .\nonumber \\ & + & \hbar ^{-2 } \mathrm{tr } \big ( \big[h_e , p_\psi\big ] \cdot \big[h_e , p_{\psi'}\big ] \big ) \\ & - & \left .
\hbar ^{-2 } \mathrm{tr } \big ( \big[h_e , p_\psi\big ] \cdot \big[h'_e , p_{\psi'}\big ] \big ) \right ] \nonumber
\\ = \int & d\mu & d\mu ' f f ' \left [ i\hbar^{-1 } \mathrm{tr } \big ( \big [ \ { h_e , f_h \ } , p_\psi \big ] \cdot p_{\psi ' } \big ) \right .
\nonumber \\ & + & \left .\hbar ^{-2 } \mathrm{tr } \big ( \big[h_e , p_\psi\big ] \cdot \big[(h_e - h'_e ) , p_{\psi'}\big ] \big ) \right],\nonumber\end{aligned}\ ] ] where we used that @xmath242 \big ] \cdot p_{\psi ' } \big ) = \mathrm{tr } \big ( \big[h_e , p_\psi\big ] \cdot \big[h_e , p_{\psi'}\big ] \big).\ ] ] using this expression , it is finally straightforward to compute the second derivative at @xmath49 for the distribution @xmath117 given by eq .
: @xmath243 ^ 2 \right ) \nonumber \\ = & - \hbar^{-2 }
\langle \psi_0 |\big ( h_e(\xi_0 ^ 1 ) - h_e(\xi_0 ^ 2 ) \big)^2 |\psi_0\rangle \nonumber \\ & + \hbar^{-2 } \langle \psi_0 |\big ( h_e(\xi_0 ^ 1 ) - h_e(\xi_0 ^ 2 ) \big ) |\psi_0\rangle^2.\end{aligned}\ ] ] 38ifxundefined [ 1 ] ifx#1 ifnum [ 1 ] # 1firstoftwo secondoftwo ifx [ 1 ] # 1firstoftwo secondoftwo `` `` # 1''''@noop [ 0]secondoftwosanitize@url [ 0 ]
+ 12$12 & 12#1212_12%12@startlink[1]@endlink[0]@bib@innerbibempty @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) `` , '' in @noop _ _ , ( , ) pp .
* * , ( ) @noop * * , ( ) @noop * * , ( ) `` , '' in @noop _ _ , ( , ) p. `` , '' in @noop _ _ , ( , ) pp . @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop _ _ ( , ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) in @noop _ _ , ( , ) `` , '' ( , ) chap . , pp .
@noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop _ _ ( , ) @noop * * , ( ) @noop _ _ ( , ) @noop * * ( ) @noop * * , ( ) | we discuss the evolution of purity in mixed quantum / classical approaches to electronic nonadiabatic dynamics in the context of the ehrenfest model . as it is impossible to exactly determine initial conditions for a realistic system
, we choose to work in the statistical ehrenfest formalism that we introduced in ref . .
from it , we develop a new framework to determine exactly the change in the purity of the quantum subsystem along the evolution of a statistical ehrenfest system . in a simple case ,
we verify how and to which extent ehrenfest statistical dynamics makes a system with more than one classical trajectory and an initial quantum pure state become a quantum mixed one .
we prove this numerically showing how the evolution of purity depends on time , on the dimension of the quantum state space @xmath0 , and on the number of classical trajectories @xmath1 of the initial distribution .
the results in this work open new perspectives for studying decoherence with ehrenfest dynamics . |
over the last decades , our knowledge of the chemical properties of the universe and , in particular , of galaxies , has improved dramatically .
observations of the local universe ( e.g. garnett & shields 1987 ; skillman , kennicutt & hodge 1989 ; brodie & huchra 1991 ; zaritsky , kennicutt & huchra 1994 ; mushotzky et al . 1996 ; ettori et al . 2002 ; tremonti et al . 2004 ; lamareille et al .
2004 ) as well as at intermediate and high redshifts ( e.g. prochaska & wolfe 2002 ; adelberger et al . 2003 ; kobulnicky et al . 2003
; lilly , carollo & stockton 2003 ; shapley et al .
2004 ) have resulted in a quite detailed picture of the chemical history of the stellar populations and of the interstellar and intergalactic media .
it is an important challenge for the theory of galaxy formation to explain these observational results in the context of the current cosmological paradigm .
numerical simulations are an important tool to make specific predictions for galaxy formation theories and to confront them with observations , because they can accurately track the growth of structure in the dark matter and baryonic components .
it is hence natural to ask what such simulations predict for the chemical properties of the universe .
currently , hydrodynamic cosmological simulations including radiative cooling and star formation , at best , coarsely reproduce the observed properties of galaxies .
they fail when scrutinised in detail . among the most prominent problems are `` catastrophic angular momentum loss '' ( e.g. navarro & benz 1991 ; navarro & white 1994 ) and the difficulty in finding a consistent modelling of supernova ( sn ) feedback ( e.g. navarro & white 1993 ; metzler & evrard 1994 ; yepes et al . 1997 ; marri & white 2003 ; springel & hernquist 2003 , hereafter sh03 ) .
the first issue mainly affects galaxy morphology , size and kinematics , while the latter primarily influences the efficiency of star formation .
both are the subject of intense research . reproducing the chemical properties of observed galaxies in detail
may be viewed as an intertwined third problem . regarding the angular momentum problem ,
recent simulations have shown some promising advances .
domnguez - tenreiro , tissera & siz ( 1998 ) found that the formation of compact stellar bulges can stabilize the discs and thereby prevent the angular momentum loss during violent minor merger events , such that disc systems that resemble observed galaxies are obtained .
abadi et al . ( 2003 ) have also demonstrated the importance of a dense , slowly rotating spheroidal component , and pointed out its relevance for comparing simulated and observed galaxies consistently .
more recently , robertson et al .
( 2004 ) were able to produce large , stable disc systems in their cosmological simulations which are comparable in size to spiral galaxies . in their approach
, they introduced a subresolution model for star formation and feedback which pressurizes the star - forming gas and stabilizes discs against fragmentation .
we note that all these studies agree on the need for a self - consistent treatment of sn feedback as a crucial mechanism to regulate star formation and to reproduce discs similar to observational counterparts .
sn explosions are thought to play a fundamental role in the evolution of galaxies since they are considered the most efficient and ubiquitous mechanism for the ejection of metals and energy into the interstellar medium ( ism ) .
both chemical and energy feedback can affect the condensation of gas and consequently the evolution of galactic systems .
on one hand , the presence of metals in the ism affects gas cooling times because the radiative cooling rate depends sensitively on metallicity ( sutherland & dopita 1993 , hereafter sd93 ) . on the other hand ,
the release of energy is crucial for regulating star formation through the heating and disruption of cold gas clouds , and for producing outflows which can transport enriched material into the intergalactic medium ( e.g. lehnert & heckman 1996 ; dahlem , weaver & heckman 1998 ; frye , broadhurst & bentez 2002 ; rupke , veilleux & sanders 2002 ; martin 2004 ) . as suggested by observations and theory , the largest outflows ( e.g. larson 1974 ; white & rees 1978 ; dekel & silk 1986 ; white & frenk 1991 ) should be able to develop in small systems because of their shallower potential wells . because in hierarchical galaxy formation scenarios large systems are formed by the aggregation of smaller systems , energy feedback from sne
is then expected to have important effects for nearly all systems in the different stages of galaxy formation .
modelling of chemical feedback has been addressed already in numerous studies , mostly with the aim to reproduce the chemical properties of certain types of galaxies , or of clusters of galaxies ( e.g. larson 1976 ; tinsley & larson 1979 ; burkert & hensler 1988 ; white & frenk 1991 ; burkert , truran & hensler 1992 ; ferrini et al . 1992 ; theis , burkert & hensler 1992 ; steinmetz & mller 1994 ; kauffmann 1996 ; chiappini , matteucci & gratton 1997 ; kauffmann & charlot 1998 ; boisser & prantzos 2000 ; valdarnini 2003 ) . as first showed by mosconi et al .
( 2001 ) , in order to properly model the process of metal enrichment in galaxy formation it is necessary to consider the full cosmological growth of structure in which mergers and interactions have important effects on the star formation process and the dynamical evolution of galaxies ( see also kawata & gibson 2003 ; tornatore et al . 2004 ; okamoto et al .
2005 ) .
perhaps the most fundamental motivation for the relevance of metal enrichment for galaxy formation is based on the fact that the cooling rate of baryons depends on metallicity ( sd93 ) .
radiative cooling in turn allows the condensation of gas into dense and cold clouds , which form the reservoir of material available for the formation of stars . as chemical enrichment
is a result of star formation , a chemical feedback cycle emerges , in which metals can in principle significantly accelerate the transformation of baryons into stars . in hierarchical scenarios these processes depend significantly on the chemical prescription adopted ( kaellander & hultman 1998 ; kay et al .
these affect galaxy properties such as the luminosity function ( white & frenk 1991 ) . for these reasons , a detailed analysis of the effects of metal - dependent cooling is clearly important for galaxy formation studies .
substantial numerical challenges in simulations of galaxy formation come from the large dynamic range required , and from the multiphase character of the ism .
gas in very different physical states coexists in the ism , but the standard formulation of smoothed particle hydrodynamics ( sph , e.g. gingold & monaghan 1977 ; lucy 1977 ) is not well suited to deal with this situation .
a number of attempts have been made to solve this problem , ranging from an ad - hoc decoupling of phases ( e.g. pearce et al .
1999 , 2001 ; marri & white 2003 ) to simple analytic sub - resolution models ( sh03 ) that work with an effective equation of state .
our new model for sn feedback in sph cosmological simulations has been divided into two stages : chemical and energy feedback . in this work
we address the description of the chemical feedback , which is based on the previous model of mosconi et al .
( 2001 ) . in a forthcoming paper
, we will present the second part of this work : the implementation of energy feedback in the framework of a multiphase scheme for the ism , which extends the approach of marri & white ( 2003 ) . while a treatment of chemical enrichment without energy feedback is an incomplete picture
, we think the complexity of the problem merits a step - wise discussion of our model such that the primary physical effects can be better understood .
this paper is organized as follows . in section [ code ] ,
we summarize the numerical implementation of the chemical model , and in section [ isolated - spheres ] we analyse the dependence of the results on numerical resolution and input parameters , using simulations of isolated galaxies .
we then study in section [ cosmological ] the properties of galaxies formed in cosmological simulations and examine how they depend on the chemical model .
finally , in section [ conclusions ] we give our conclusions .
our new model for the production and ejection of chemical elements by sne is based in part on the approach in mosconi et al .
we have implemented our scheme in the treepm / sph code gadget-2 , an improved version of the public code gadget ( springel , yoshida & white 2001 ) which manifestly conserves energy and entropy where appropriate ( springel & hernquist 2002 ) .
note that we do not use the multiphase treatment and the feedback model developed by sh03 for this code , but we do include their treatment of ultraviolet ( uv ) background . also , we use a different parameterisation for star formation , as described in detail below .
chemical elements are synthesized in stellar interiors and ejected into the ism by sn explosions . in order to model the production and distribution of metals in the context of the simulations , there are three ingredients we have to consider : the sn rate ( i.e. number of exploding stars per time unit ) , the chemical yields ( i.e. chemical material ejected in explosions ) and the typical lifetimes of stars , which determine the characteristic time of the metal release . in our model , we include a separate treatment of type ii and type ia sne ( snii and snia ) .
these two types of sne originate from different stellar populations , and have different rates , yields and typical time - scales .
we use different yields for snii and snia which , in the case of snii , are metal - dependent . hence , we will treat snii and snia separately , owing to the need to model their different characteristics . for example , snii produce most of the chemical elements , except for iron , which is mainly produced by snia .
another difference between these two types of sn is that snii are the endpoint of the evolution of massive stars with short lifetimes , in contrast to snia , which result from the evolution of binary systems with lifetimes of @xmath1 1 gyr .
we assume that initially gas particles have primordial abundances : @xmath2 and @xmath3 , and we consider the enrichment by the following elements , assuming production only by snii and snia : h , @xmath4he , @xmath5c , @xmath6o , @xmath7 mg , @xmath8si , @xmath9fe , @xmath10n , @xmath11ne , @xmath12s , @xmath13ca and @xmath14zn .
note that we do not consider nucleosynthesis by intermediate - mass stars .
consequently , we will concentrate our analysis on elements such as o or fe where restricting to sn production may be a good approximation . in the rest of this section , we describe the most important aspects of the chemical model , namely the star formation and cooling prescriptions , and our scheme for the production and ejection of metals .
we assume that gas particles are eligible for star formation if they are denser than a critical value ( @xmath15 , where @xmath16 denotes gas density ) and lie in a convergent flow ( @xmath17 ) . for these particles , we assume a star formation rate ( sfr ) per unit volume equal to @xmath18 where @xmath19 is a star formation efficiency ( we adopt @xmath20 ) and @xmath21 is the dynamical time of the particle .
we base the creation of new stellar particles on the stochastic approach of sh03 ( see also lia , portinari & carraro 2002 ) . to this end , each gas particle eligible for star formation in a given time - step is assigned a probability @xmath22 of forming a star particle given by @xmath23,\ ] ] where @xmath24 is the integration time - step of the code , @xmath25 is the mass of the gas particle , and @xmath26 is defined as @xmath27 with @xmath28 being the original mass of gas particles at the beginning of the simulation and @xmath29 a parameter which determines the number of ` generations ' of stars formed from a given gas particle ( we assume @xmath30 ) . if a random number drawn from a uniform distribution in the unit interval is smaller than @xmath22 , we form a new stellar particle of mass @xmath26 and reduce the mass of the gas particle accordingly .
once the gas particle mass has become smaller than @xmath31 , we instead turn it into a star particle .
newly born stars are assumed to have the same element abundances as the gas mass from which they form .
the main advantage of this stochastic approach is that all particles are either purely stellar or purely gas , which avoids the use of ` hybrid ' particles with both gaseous and stellar components .
the latter would artificially force the stellar component to evolve dissipately like the gas .
note that , as a consequence of the star formation scheme , the number of baryonic particles in the simulation does not remain constant .
unlike sh03 , we also exchange the masses of heavy elements explicitly between gas and star particles . as a result ,
the mass spectrum of stellar and gas particles is slightly washed out around otherwise sharp discrete values .
our numerical implementation of the enrichment model has three main components : snii element production , snia element production and metal distribution .
_ snii element production _ in order to estimate the snii rate we use a salpeter initial mass function ( imf ) with lower and upper mass cut - offs of 0.1 and 40 m@xmath32 , respectively , and assume that stars more massive than @xmath33m@xmath32 end their lives as snii . for the chemical production , we adopt the yields of woosley & weaver ( 1995 ) .
these are metal - dependent yields and vary differently for different elements .
we use half of the iron yield of ww95 , as it is often adopted ( e.g. timmes , woosley & weaver 1995 ) . for the sake of simplicity , in most of our simulations
we have assumed that snii explode ( i.e. produce and eject the chemical elements ) within an integration time - step of the code ( which is typically @xmath34 yr ) , because snii originate in massive stars which have very short characteristic lifetimes of the order of @xmath35 yr .
however , we have also analysed the effect of relaxing this instantaneous recycling approximation ( ira ) for snii ( see section [ snparameters ] ) .
_ snia element production _ as a progenitor model , we adopt the w7 model of thielemann , nomoto & hashimoto ( 1993 ) , which assumes that snia explosions originate from co white dwarf systems in which mass is transferred from the secondary to the primary star until the chandrasekhar mass is exceeded and an explosion is triggered .
it is generally assumed that the lifetime of such a binary system is in the range @xmath36[@xmath37 gyr , depending on the age of the secondary star ( greggio 1996 ) . in our model , the material produced by a snia event
is ejected when a time @xmath38 after the formation of the exploding star has elapsed .
we choose this time randomly within a given range ( in most of our experiments we assume @xmath39 $ ] gyr ) . in order to estimate the number of snia
, we adopt an observationally motivated range for the relative ratio of snii and snia rates ( e.g. van den bergh 1991 ) .
we use the chemical yields of thieleman et al .
( 1993 ) for the metal production itself . _ metal distribution _
we distribute chemical elements ejected in sn explosions within the gaseous neighbours of exploding star particles , using the usual sph kernel interpolation technique ( mosconi et al .
each neighbouring gas particle receives a fraction of the ejected metals according to its kernel weight . for this purpose , gas particle neighbours of stellar particles
are identified whenever metal distribution has to take place , and smoothing lengths for these stars are calculated by setting a desired number of gas neighbours that should be enclosed in the smoothing length ( we have assumed the same criterium used for the smoothing lengths of gas particles ) .
this is particularly important for the enrichment by snia because of the time delay between the formation of the stars and the ejection of metals .
note that after a star particle is created , it is only subject to gravitational forces , unlike the gas particles , which continue to be subject to hydrodynamical forces as well .
consequently , as the system evolves , star particles can release metals associated with snia explosions in a different environment from where they were born .
this , together with the different lifetimes and yields of snii explosions , can then produce non - trivial variations in abundance ratios .
it is well known that the cooling rate of thin astrophysical plasmas depends sensitively on metallicity , in such a way that at a given density , the cooling rate is higher for a higher metallicity . in fig .
[ figcooling ] we show the cooling function computed by sd93 from primordial ( i.e. no heavy elements ) to suprasolar abundance ( [ fe / h]=@xmath40 ) . as one can see from this figure
, the cooling function can show very large variations with the metal abundance of the gas , depending on the temperature range considered .
for example , the usual estimates of the cooling time ( @xmath41 ) for a gas particle at the critical density ( @xmath42 ) and at a temperature of @xmath43 k yield that @xmath41 is 50 times larger for primordial gas than for a mildly suprasolar medium ( [ fe / h]@xmath44 ) , 18 times larger than that of a solar abundance one , and still twice as large as the corresponding time for gas with [ fe / h]@xmath45 .
the temperature range of [ @xmath46 k shows the largest differences , right in the temperature range relevant to the abundant population of dwarf galaxies . including effects owing to metal line cooling
is therefore crucial for realistic modelling of galaxy formation because it affects directly the relation between cooling time and dynamical time of forming galaxies .
we therefore include a treatment of radiative cooling consistent with the metal content of the gas component produced by the enrichment model .
we use the tables computed by sd93 for the metal - dependent cooling functions , and interpolate from them for other metallicities as needed . for particles with iron abundance
larger than @xmath47 , we adopt the suprasolar cooling rate ( i.e. that for [ fe / h]@xmath48 ) . in cosmological simulations
, we also include a redshift - dependent photo - heating uv background , with an intensity evolution that follows the model of haardt & madau ( 1996 ) .
abundance , from suprasolar ( [ fe / h]=0.5 ) to primordial .
the plot is based on data adapted from sd93.,width=317 ]
as first tests of our model , we consider simulations of isolated haloes set up to form disc galaxies at their centre . this allows us to examine the dependence of the results on resolution and on different parameters of the chemical enrichment model .
the initial conditions consist of a static dark matter potential corresponding to a navarro , frenk & white ( 1996 , 1997 ) profile of concentration @xmath49 , and a baryonic gas phase initially in hydrostatic equilibrium in this potential .
our typical system has virial mass @xmath50 ( @xmath51 ) , @xmath52 per cent of which is in the form of baryons ( @xmath53 ) .
the initial radius of the system is @xmath54 and the gas has an initial angular momentum characterized by a spin parameter @xmath55 .
we have followed the evolution of this system for @xmath56 , where @xmath57 gyr is the dynamical time at @xmath58 .
these idealized initial conditions yield a simple model for disc formation , which is an ideal test bench for the performance and validity of the code , and the dependence of the results on the free parameters .
we stress however that these models are not meant to provide a realistic scenario for the whole galaxy formation process .
results for full cosmological simulations will be discussed in section 4 , where we analyse the properties of galactic objects formed in @xmath59 cold dark matter ( cdm ) models with virial masses of @xmath60 , comparing them also to the results of the idealized tests discussed here .
both the star formation process and the distribution of metals can be affected by numerical resolution . particularly important for our model is the fact that the smoothing length of the particles ( i.e. the radius which encloses @xmath61 gaseous neighbours ) determines the region where the newly released chemical elements are injected .
mosconi et al .
( 2001 ) found that chemical abundances could be oversmoothed in low resolution simulations because metals are distributed over a comparatively large volume and , consequently , are more efficiently mixed with surrounding gas . on the other hand , high numerical resolution could also produce spurious results if the metal mixing mechanism becomes too inefficient . provided other physical mixing mechanisms are not operating
, the simulation results might show an artificially large dispersion in gas metallicities . in order to test the effects of numerical resolution
, we carried out simulations of the isolated disc system described above using three different initial numbers of gas particles : 2500 ( r1 ) , 10000 ( r2 ) and 40000 ( r3 ) , with mass resolutions of @xmath62 , @xmath35 and @xmath63 for r1 , r2 and r3 , respectively .
we have set the gravitational softening to @xmath64 kpc . in these three runs
we use metal - dependent cooling functions for the gas particles according to the tables by sd93 . in table
[ simulations ] we summarize the main characteristics of these simulations . in fig .
[ sfr - resolution ] , we show the evolution of the sfrs obtained for r1 ( dotted line ) , r2 ( solid line ) and r3 ( dashed line ) . most of the stellar mass ( @xmath65 , @xmath66 and @xmath67 per cent in r1 , r2 and r3 , respectively ) is formed at times @xmath68 gyr ( @xmath69 per cent of the dynamical time ) in the three runs . only at later times , we find that the lowest resolution test simulation ( r1 ) yields a significantly lower result for the sfr , indicating that the number of gas particles has dropped so much by the conversion into stars that discrete effects become appreciable
. however , the final stellar mass fractions ( i.e. stellar mass formed divided by the total baryonic mass ) are very similar in r1 , r2 and r3 .
this can be also appreciated from fig .
[ perfiles - resolution ] , where we show the mass - weighted metallicity profiles for r1 ( dotted line ) , r2 ( solid line ) and r3 ( dashed line ) , in the inner region and for the stellar ( left column ) and gaseous ( right column ) components and for three different times , @xmath70 gyr ( upper panels ) , @xmath71 gyr ( middle panels ) and @xmath72 gyr ( lower panels ) .
we can see that the stellar abundance profiles are insensitive to resolution .
the mean iron abundance differs by less than @xmath73 dex between the different simulations .
recall that the stellar metallicity of newly forming stars is given by the local gas metallicity .
this is directly visible at the beginning of the simulation where the gaseous and stellar components show a similar level of enrichment in star - forming regions ( upper panels in fig .
[ perfiles - resolution ] ) , while at later stages of the evolution ( middle and lower panels ) this constraint becomes hidden as a result of the superposition of stars of different ages .
however , as can be seen from the right panels of fig
. [ perfiles - resolution ] , the mass - weighted metallicity profiles for the gas components show stronger differences with resolution than their stellar counterparts . at the beginning of the evolution ,
the profiles of the three test simulations are still very similar , although the enriched gas is more spread in the highest - resolution test . however , the gas profiles differ more at later times when factors such as variations in the consumption time of the gas and differences in the spatial distribution of stars begin to affect the radial distribution of metals . at the end of the simulations ( lower right panel ) ,
the mean gaseous iron abundance for r1 is significantly lower than those for the other two simulations .
this decrease in the metallicity is produced by the infall of less enriched gas .
this gas is not becoming dense enough to trigger star formation activity at the level of r2 and r3 , and the infall tends to dilute the metal content in the ism .
hence , the fact that the gas component in r1 shows a lower iron abundance is driven by its poorer gas resolution .
note that as a result of the consumption of most of the gas , the final number of gas particles in r1 is very low ( @xmath74 in the inner @xmath75 kpc ) and , as a consequence , the distribution of metals in the gas component is strongly affected by numerical noise . on the other hand , the results for the tests with 10000 and 40000 particles converge quite well both in their stellar and gaseous chemical properties .
note that the rms scatter of particles within each radial bin is similar for the different tests indicating that , on average , the metal mixing is not significantly affected by resolution if we consider at least an initial number of particles of @xmath76 or more . [
cols="<,>,^,^,^,<",options="header " , ] [ glos ] in fig .
[ sfr - glos ] , we compare the star formation histories for the selected galactic objects in c1 ( dotted line ) , c2 ( solid line ) and c3 ( dashed line ) , as a function of look - back time , i.e. @xmath77^{-3/2}$ ] .
note that the global features of the sfrs are preserved in the three runs , although the detailed levels of activity are different .
in general , the sfrs of the galactic objects formed in c3 are systematically higher than the corresponding ones in c2 , as expected .
note that because galactic objects formed in c3 have higher sfrs than those in c2 , their final stellar masses are much larger than the corresponding ones in c2 ( see table [ glos ] ) .
if we compare the sfrs obtained for c1 and c2 , although the differences are small , the sfrs in c1 are systematically lower than those in c2 .
it is noteworthy that the impact of metallicity on the star formation history seems to depend on the particular evolutionary histories of the systems , as can be appreciated from fig .
[ sfr - glos ] . in fig .
[ mean - perfiles - mstar - obj ] , we show the mean stellar mass fraction @xmath78 as a function of radius , estimated from the six selected galactic objects in c1 ( dotted line ) , c2 ( solid line ) and c3 ( dashed line ) . as in section [ cooling ] , the distribution function for each galactic object was calculated by summing up the stellar mass in bins of radius normalized to the total baryonic mass within the corresponding bin . as we found for the tests of the isolated galaxies ,
there are clear differences between the results for the three runs .
although in the inner @xmath79 kpc the simulations yield similar stellar fractions , within @xmath80 kpc , galactic objects in c2 have transformed a larger fraction of the gas into stars compared with those in c3 , while galactic objects in c1 show the opposite behaviour . at larger radii ,
the mean distribution function of stellar mass for the galactic objects in c3 is significantly larger than its counterpart for c2 , indicating that as a result of the higher cooling efficiency , the gas in c3 has been transformed into stars further away from the centre of mass of the objects .
conversely , the primordial run shows fewer stars in the outer region .
note that efficient energy feedback is needed to prevent excessive star formation in the outer regions , as it can be observed from fig .
[ mean - perfiles - mstar - obj ] . in order to assess how these differences affect the chemical properties of the formed galactic objects , we analysed the metal indicator @xmath81 for the gaseous and stellar components in c1 , c2 and c3 .
consistent with our results for the idealized disc tests , we found that the metallicity of the stellar components is not sensitive to the cooling function adopted , in contrast to the gaseous components . in fig .
[ perfiles - glos ] we compare the gaseous oxygen profiles for the galaxy samples in c1 ( dotted lines ) , c2 ( solid lines ) and c3 ( dashed lines ) .
although there is a trend for systems in c3 to have a higher level of enrichment and flatter mass - weighted abundance profiles compared to their counterparts in c2 and c1 , how much metals affect these relations seems to depend on the particular history of evolution of each galaxy , as can be inferred from figs .
[ sfr - glos ] and [ perfiles - glos ] .
finally , for a comparison with observational results , we performed linear regression fits to the gaseous oxygen profiles of our simulated galactic objects within the optical radius . by definition , the optical radius encloses @xmath82 per cent of the baryonic stellar mass within the virial radius . in fig .
[ slope - zero ] , we show the zero - point of these fits as a function of the corresponding slopes for the galactic objects in c1 ( triangles ) , c2 ( diamonds ) and c3 ( asterisks ) .
we also include observational results from zaritsky et al .
( 1994 , filled circles ) .
in general , the properties of the simulated galactic objects agree reasonably well with the range obtained in observational studies . however , the simulated galactic objects identified in c2 tend to have steeper profiles and higher zero - points compared with observations .
conversely , galactic objects in c3 exhibit flatter profiles . a similar behaviour was also found in our tests for the isolated disc galaxies run with primordial ( pc , squared triangle ) , metal - dependent ( r2 , squared diamond ) and suprasolar ( sc , squared asterisk ) cooling functions .
the fact that these simulations do not reproduce the whole observed range of gradients could be related to the lack of a treatment of energy feedback by sne .
the latter , if realistically modelled , is expected to trigger mass outflows from the star - forming regions into the halo .
since these regions are highly enriched , metals may be transported outwards , affecting the metallicity distributions and flattening the metallicity profiles .
our simulation run with suprasolar cooling function was performed in order to assess the effects of the presence of metals in these outer regions on the gas cooling , mimicking an extreme case of metal enrichment by energy feedback .
we have implemented a new model for chemical evolution within the cosmological sph code gadget-2 .
we account for nucleosynthesis from snia and snii separately .
radiative cooling of the gas according to self - consistent metallicity values is realized by means of precomputed look - up tables .
our implementation was designed to allow a reasonably accurate treatment of the main features of stellar evolution while still not becoming computationally too expensive . using a number of test simulations both of isolated galaxies and of cosmological structure formation
, we have validated our implementation and established some first basic results for the effects of chemical enrichment .
an important consequence of the presence of metals is the increase in the cooling efficiency that it can cause , leading to accelerated star formation . in our isolated galaxy simulations ,
the sfr for test runs with metal - dependent cooling function can be up to a factor of @xmath83 higher than the corresponding one for a test with a primordial cooling function , and up to a factor of @xmath84 lower than the one for suprasolar cooling .
we found that these differences do not originate in the gas in the central regions , because it is already cold and dense anyway .
instead , the differences are caused by more diffuse gas at outer radii which has yet to cool . here
the increase of the cooling efficiency allows condensation of gas and star formation to occur further away from the centre of the systems .
an important implication of this result is that strong effects owing to metal - dependent cooling can only be expected when heavy elements are efficiently transported and mixed into diffuse gas that has yet to cool . in our cosmological simulations , which lacked a physical model for energy feedback
, such a large - scale spreading of metals only occurs at a low level .
as perhaps was to be expected , we found that simulations with metal - dependent cooling produced similar results as simulations with a primordial cooling function .
we expect however that this could be very different if a model for efficient metal distribution is added to our simulations .
an explicit demonstration of this has been provided by our test simulation where we assumed suprasolar cooling rate functions , resulting in substantially elevated star formation activity and different abundance profiles .
note that in this paper we explored galaxy scale systems which cool very efficiently .
larger effects from chemical mixing are expected in more massive haloes such as clusters of galaxies ( white & frenk 1991 ; de lucia , kauffmann & white 2003 ) .
we have also tested the impact of a number of assumptions made in our model , and the relevance of numerical parameters .
relaxing the ira for snii has only a negligible effect and does not change the chemical properties of the simulated systems .
however , as also shown by previous work ( e.g. raiteri , villata & navarro 1996 ; chiappini et al .
1997 ; mosconi et al .
2001 ) , the chemical properties of both the stellar populations and the ism are sensitive to the time - delay associated with snia explosions . in agreement with previous results
, we found that the latter must be included in order to be able to reproduce the observed chemical properties of galaxies .
we have shown that chemical enrichment can significantly modify the chemical and dynamical properties of forming galactic systems . however , the strength of the influence of metals on galaxy formation might also depend on the physics of energy feedback .
if the latter is neglected , as we have done here , metals are not mixed widely in gas that has yet to cool .
effects due to metal - dependent cooling remain then moderate . in a forthcoming paper
, we will complement our model for chemical enrichment with a new description of a multiphase ism and energy feedback , which is capable of reproducing strong outflows and regulating the star formation activity .
this combined model should provide a powerful tool to exploit observational data on the chemical properties of galaxies to constrain galaxy formation .
we thank the anonymous referee for comments that helped to improve the paper .
this work was partially supported by the european union s alfa - ii programme , through lenac , the latin american european network for astrophysics and cosmology .
simulations have been perfomed on ingeld pc - cluster funded by fundacin antorchas .
we acknowledge support from consejo nacional de investigaciones cientficas y tcnicas , agencia de promocin de ciencia y tecnologa , fundacin antorchas , secretara de ciencia y tcnica de la universidad nacional de crdoba and the daad exchange programme .
c.s . thanks the alexander von humboldt foundation , the federal ministry of education and research and the programme for investment in the future ( zip ) of the german government for partial support . greggio l. , 1996 , in kunth d. , guiderdoni b. , heydari - malayeri m. , thuan t. , eds , the interplay between massive star formatin , the ism and galaxy evolution .
editions fronti@xmath85res , gif - sur - ivette , p.89 | we discuss a model for treating chemical enrichment by type ii and type ia supernova ( snii and snia ) explosions in simulations of cosmological structure formation .
our model includes metal - dependent radiative cooling and star formation in dense collapsed gas clumps .
metals are returned into the diffuse interstellar medium by star particles using a local smoothed particle hydrodynamics ( sph ) smoothing kernel .
a variety of chemical abundance patterns in enriched gas arise in our treatment owing to the different yields and lifetimes of snii and snia progenitor stars .
in the case of snii chemical production , we adopt metal - dependent yields . because of the sensitive dependence of cooling rates on metallicity , enrichment of galactic haloes with metals can in principle significantly alter subsequent gas infall and the build up of the stellar components .
indeed , in simulations of isolated galaxies we find that a consistent treatment of metal - dependent cooling produces @xmath0 per cent more stars outside the central region than simulations with a primordial cooling function . in the highly - enriched central regions , the evolution of baryons is however not affected by metal cooling , because here the gas is always dense enough to cool .
a similar situation is found in cosmological simulations because we include no strong feedback processes which could spread metals over large distances and mix them into unenriched diffuse gas .
we demonstrate this explicitly with test simulations which adopt suprasolar cooling functions leading to large changes both in the stellar mass and in the metal distributions .
we also find that the impact of metallicity on the star formation histories of galaxies may depend on their particular evolutionary history .
our results hence emphasise the importance of feedback processes for interpreting the cosmic metal enrichment .
[ firstpage ] methods : numerical - galaxies : abundances - galaxies : formation - galaxies : evolution - cosmology : theory . |
one of the surprises which has emerged from the heavy ion experiments at rhic is that the medium or plasma which is produced in a heavy ion collision appears to display collective behavior analogous to a fluid with a very small viscosity ( the so - called `` quark - gluon liquid '' ) @xcite . in particular
, hydrodynamic treatments taking the medium to be an ideal fluid give the correct flow description @xcite . to clarify ,
we remind the reader that `` ideal fluid '' is the opposite of `` ideal gas '' ; it means that scatterings or interactions so efficiently maintain local equilibrium ( or at least isotropy ) , that the stress tensor remains everywhere isotropic when measured in the local rest frame .
gases behave like ideal fluids on distance and time scales large compared to transport mean free paths and mean free times .
the most popular , and perhaps most likely , explanation for the small viscosity is that the quark - gluon plasma at the energy densities achieved at rhic is very far from weak coupling .
the strong coupling @xmath2 is large , interactions are very strong , and the collective behavior is natural . to secure this interpretation , we would need to see that the observed behavior really differs from the expected weakly coupled behavior .
most treatments to date of weakly coupled hot qcd show that equilibration is slow @xcite ( see however @xcite ) .
however , we have recently shown @xcite that even the most complete of these , the `` bottom - up '' scenario of baier , mueller , schiff and son @xcite , is incorrect , because it ignores plasma instabilities @xcite , which in fact dominate the physics of weakly - coupled anisotropic plasmas , in qcd or in ordinary qed .
plasma instabilities arise as a result of two pieces of physics .
first , lorentz contraction of the nuclei means that the initial region of plasma is a flat pancake shape .
it is reasonable to expect at weak coupling that quarks and gluons will for a time fly in nearly straight lines at near the speed of light ( at least , once the system has expanded enough that the densities of quarks and gluons are perturbative ) .
consider particles scattered in all directions in the initial moments of the collision .
the starting geometry dictates that the momentum distribution of particles will subsequently become highly anisotropic with time , as illustrated in fig .
second , in the presence of such anisotropic momentum distributions , certain soft gauge field configurations grow exponentially , at least if they are small enough for a perturbative treatment to be reliable .
= 0.9 not only have plasma instabilities not been fully taken into account in studying the evolution of the quark - gluon plasma ; instabilities in yang - mills theory in general are not yet well understood or well characterized .
therefore , at the current juncture we can not say whether the experimental results at rhic are in contradiction with weakly coupled predictions , because we simply do not know what the predictions of isotropization in a weakly coupled quark - gluon plasma actually are .
this rather embarrassing gap in our knowledge needs to be filled .
in particular , the behavior of plasma instabilities in a nonabelian theory such as qcd are expected to be intrinsically different from the case of an ordinary qed plasma , because nonabelian gauge fields can directly interact with each other , and because charge carriers can have their colors rotated by intervening qcd fields .
recently there has been substantial progress on this problem , though it is far from settled .
the growth of qcd plasma instabilities when the soft colored gauge fields are small has been well characterized @xcite .
arnold and lenaghan @xcite conjectured that exponential growth would continue after qcd fields became large .
1 + 1 dimensional simulations @xcite confirmed this behavior , but 3 + 1 dimensional studies @xcite showed something different : when soft colored gauge fields become large due to instability growth , exponential growth shuts off and the energy in soft gauge fields instead grows _ linearly _ with time .
recent partial attempts to explore the consequences of instabilities for thermalization may be found in refs .
@xcite .
the goal of this paper is to further investigate the behavior of an anisotropic quark - gluon plasma , after soft gauge fields have become large .
we will show that the soft gauge fields with wave - numbers which should make them unstable instead enter a dynamical quasi - steady state , gaining energy from the instability but losing energy , via their nonabelian interactions , to more ultraviolet field excitations .
the energy released into infrared gauge fields by the instability then cascades towards higher wave - number gauge field modes , with occupation number @xmath3 up to a time - dependent cutoff @xmath4 . through simulations
, we measure the spectral index to be @xmath1 .
we will carry out a numerical study of nonabelian , classical yang - mills theory coupled to an anisotropic bath of high momentum particles .
the numerical tools were already presented in detail in @xcite , and this paper is a continuation of that work . to make our presentation more
self - contained , we will briefly review the setup and formalism here .
our treatment is founded on two assumptions about the system under study .
the first is that the coupling @xmath5 .
the second is that there is a separation of scales between the momentum of the typical excitation , @xmath6 , and the _ screening scale _ @xmath7 .
the screening scale is set by the momentum @xmath6 and number density @xmath8 of the typical excitations as @xmath9 .
the number density is given by @xmath10 , where @xmath11 is the occupancy of momentum state @xmath12 and @xmath13 is its angular average .
therefore there is a scale separation @xmath7 provided that the angular - averaged occupancy @xmath13 of typical excitations is @xmath14 .
this is essentially a diluteness condition on the typical excitations in the system . in the context of heavy ion collisions ,
the condition @xmath5 is formally valid for the collision of extremely large and high - energy heavy ions @xcite .
the initial angular - averaged occupancy is at most @xmath15 , which is the saturation limit @xcite , then the magnetic screening length they would provide would be larger than the imputed typical momentum .
however , the concept of momentum for an excitation of a gauge dependent field is only well defined and robust under gauge fixing prescriptions to an accuracy set by the magnetic screening length .
these ideas are central to the idea of the saturation scale .
however this argument does not depend on the saturation scenario as such . ]
( corresponding to number and energy densities @xmath16 and @xmath17 ) .
the density @xmath8 of the original particles , and so @xmath18 , will fall with time as the system expands .
so the scale separation condition @xmath19 will be satisfied at least for @xmath20 ( up until the much later time when the particles of momentum @xmath6 lose their energy and the plasma thermalizes see , for example , the discussion in the original bottom - up scenario of ref .
@xcite ) .
if we have a clean separation between the screening scale @xmath21 and typical momenta @xmath6 , then we can describe the physics on length scales @xmath22 in terms of ( i ) classical fields with wave vectors @xmath23 and ( ii ) a distribution of classical particles ( possessing well defined momenta and positions ) representing excitations with momenta @xmath24 .
this is a vlasov equation treatment . with a little more work @xcite
, one may work in terms of a classical field variable @xmath25 , a variable @xmath26 , which represents the net ( adjoint ) color of all particles moving in direction @xmath27 at point @xmath28 , and a background colorless particle density with angular distribution @xmath29 and polarizability characterized by a screening mass @xmath30 , @xmath31 with the sum over spin and particle type ( including anti - particles ) .
as @xmath32 , but here we will drop the subscript . ] here @xmath33 is a color group factor [ defined in terms of color generators @xmath34 by @xmath35 and equal to 3 for gluons ] . for an isotropic medium , @xmath36 , where @xmath37 is the debye mass and
@xmath38 is the plasma frequency .
we argued in @xcite that the intrinsic length and time scales for the instability to develop are short compared to the length and time scales on which the heavy ion system as a whole evolves .
therefore we will restrict our attention in this ( still somewhat exploratory ) study to a non - expanding system with spatially uniform @xmath39 .
the evolution equations for @xmath40 and @xmath41 are @xmath42 \omega(\v ) \ , .
\label{eq : w}\end{aligned}\ ] ] the first equation is the yang - mills field equation , with @xmath43 giving rise to a current . the second equation , derived in ref .
@xcite , shows how electric and magnetic fields can polarize the colorless distribution of particles to create a net color moving in each direction .
the dynamics of the soft fields is equivalent to that of hard - loop effective theory @xcite . for a fully nonperturbative study of this system
, we implement these equations on a lattice .
the treatment of the classical , minkowski space gauge fields on the lattice is the standard one @xcite .
the representation of the @xmath41 fields is developed in @xcite and consists of replacing first derivatives with forward - backwards ( covariantized ) differences in , and expanding the space of velocities @xmath27 in spherical harmonics , truncated at a finite but large @xmath44 . in this work
we make one additional modification which improves approach to the @xmath45 limit and so reduces the memory and time requirements for simulations : we apply a weak damping term to the large @xmath46 spherical harmonic components of the @xmath41 field .
the reason this is an improvement is that very large @xmath46 excitations of the @xmath41 field tend to cascade to still larger @xmath46 , due to the @xmath47 term in .
term mixes neighboring @xmath46 s because @xmath48 has non - zero cases when @xmath49 .
] with a finite @xmath44 cutoff , some of this excitation energy `` bounces off ''
the @xmath44 cutoff and re - enters the infrared . this effect becomes smaller as @xmath44 is raised , but can be largely eliminated with the damping term we add .
this point is discussed and justified in greater detail in appendix [ sec : damp ] .
the goal of this paper is to understand the behavior of a would - be unstable system after the unstable modes have already grown to have nonperturbative occupancy . in the context of a heavy ion collision
, this is probably the only relevant question ; since the exponential growth rate is larger than the inverse system age for all times after the formation time of the plasma , there has always been time for unstable gauge field modes to grow to nonperturbative size .
therefore we will begin our system with initial conditions for which the infrared ( ir ) fields are already nonperturbatively large .
again , since this is a qualitative exploratory work , we will work in su(2 ) rather than su(3 ) gauge theory and we will only consider a single anisotropic particle distribution , the same as the one considered in ref .
@xcite .
in addition to our focus on large rather than small initial conditions , there are slight technical differences in our choice of initial conditions compared to those of ref .
we start the system with vanishing @xmath50 and @xmath41 fields and with the @xmath51 field selected from the thermal ensemble at temperature @xmath52 , but then field - smeared to a scale @xmath53 .
this is a gauge - invariant procedure which corresponds perturbatively to multiplying a thermal spectrum for @xmath54 by @xmath55 .
we will explain the field smearing procedure below because it also constitutes one of our best measurables for understanding , in a gauge - invariant fashion , what corresponds to infrared phenomena and what to ultraviolet .
our first conclusion , already presented in our previous paper @xcite , is that the energy in soft electromagnetic fields grows linearly with time .
this is displayed in fig .
[ fig : lin ] , which shows that the linear behavior is common to electric and magnetic fields and is robust to changes in the lattice spacing . however , this result leaves it unclear whether or not this represents continued growth of the soft unstable fields , whether these fields `` abelianize '' @xcite , and whether they retain long time scale coherence .
we now present evidence that the answer to all three questions is , `` no . ''
= 0.7 first , consider the behavior of chern - simons number , @xmath56 chern - simons number is useful because it characterizes nonperturbative physics . in an abelian theory , or a nonabelian theory where the fields are weak
, @xmath57 may fluctuate about zero but can not drift away from zero permanently .
that is because permanent change requires topology change ( the minkowski version of instantons ) .
therefore , ignoring small fluctuations , the time evolution of @xmath57 is purely indicative of the dynamics of nonperturbatively large fields .
fully topological algorithms for tracking @xmath57 in a real - time gauge field evolution already exist .
we use the one from ref .
@xcite , which is a modification of a technique due to ambjrn and krasnitz @xcite .
a sample chern - simons number trajectory , from the @xmath58 evolution shown in fig .
[ fig : lin ] , is shown on the left in fig.[fig : ncs ] .
chern - simons number is changing by large amounts ( indicating nonperturbatively large fields ) , and in a chaotic fashion ( indicating that the fields have no long time - scale stability ) .
this means that the soft fields remain large , but evolve dynamically , changing on a time scale set by the scale @xmath59 . to see this better ,
consider the auto - correlator @xmath60 , shown on the right in that figure .
this correlator indicates over what time scale coherent changes to the gauge fields occur .
the figure presented represents an average over time and over 10 independent initial conditions drawn from the same ensemble , using a @xmath61 lattice with @xmath62 and @xmath63 .
the errors were determined by the jackknife method .
= 0.46 = 0.46 although we have not shown it in the figure , we have also checked that the evolution of @xmath57 does not speed up or slow down during the course of the linear rise in magnetic energy .
for instance , comparing the first and second halves of the evolutions used to make fig.[fig : ncs ] , the chern - simons number diffusion rate ( sphaleron rate ) is consistent between the two halves to within @xmath64 error bars , and the full auto - correlator is also consistent within errors .
what this tells us is that the ir fields truly are nonperturbative , that they evolve with a characteristic time scale of order @xmath65 , and that the time scale and the size of the nonperturbative fields is constant throughout the linearly growing regime .
if ir fields are not growing , then the linear growth of magnetic energy must reflect growth of higher - momentum modes of the soft gauge field .
to clarify the situation , it would be useful to know the power spectrum of the soft gauge fields as a function of wavenumber @xmath21 and time @xmath66 .
we will start with a direct but gauge - dependent measurement of the power spectrum in coulomb gauge .
later we will discuss how the spectrum can be accessed indirectly through gauge - invariant measurement involving smearing ( also known as cooling ) of the field configurations .
our picture ( to be supported by data below ) is that , during the linear growth phase of the total energy in soft fields , the soft fields consist of ( i ) a non - perturbative ir component plus ( ii ) a perturbative component in the form of higher - momentum plasmons .
we would like to know the distribution function @xmath67 of these plasmons as a function of @xmath21 .
we fix coulomb gauge using the standard algorithm @xcite adopted to the real - time case @xcite .
then we ( i ) extract the fourier spectrum of @xmath68 and @xmath69 , ( ii ) evaluate the two point function , averaging over @xmath70 vectors in narrow blocks of @xmath71 , technically , we use the lattice @xmath72 , @xmath73 . and ( iii ) define the distribution function as determined by @xmath68 and as determined by @xmath69 through ) plasmons .
this could be written in terms of @xmath74 as @xmath75 or in terms of the fields as @xmath76 .
since @xmath77 for such plasmons , and @xmath78 in coulomb gauge , we can also write the energy as @xmath79 or @xmath80 . comparison of these expressions leads to the identification ( [ eq : fdef ] ) .
] @xmath81 here , @xmath82 is the total spatial volume and @xmath83 accounts for the two transverse polarization states and the @xmath84 adjoint color states in su(2 ) gauge theory . on scales where the fields are nonperturbatively large , these distribution functions are difficult to interpret , depending somewhat on the gauge fixing procedure .
we do not expect @xmath85 to equal @xmath86 ; indeed , if the dominant fields are slowly evolving or unstable magnetic fields , we expect @xmath87 , perhaps by a large margin . on scales , presumably at larger @xmath21 , where the fields are perturbative , we should see @xmath88 if the relevant degrees of freedom are behaving as plasmons with @xmath89 .
therefore , the ratio @xmath90 serves as a diagnostic of whether the physics is nonperturbative and whether the degrees of freedom are primarily independently propagating plasmons or something else , such as the magnetic fields associated with hard particle currents . to test this , we evolved the system for a time of @xmath91 in a @xmath92 box with @xmath63 and lattice spacing @xmath93 , corresponding to a ( large ) physical volume of @xmath94 , tracking the distribution functions after the initial transient had died and the magnetic field energy was undergoing linear growth .
the occupancy after initial transients is displayed on the left in fig .
[ fig : coulomb1 ] , showing that , as expected , the ir has nonperturbatively large fields , while fields are perturbative at larger wave number . on the right in the figure
, we show the time development .
each curve is time - averaged over an interval of @xmath95 , and the central times of consecutive curves are spaced apart by @xmath96 .
the ir occupancies remain nonperturbative but with stable amplitude , while the uv occupancy increases . at any @xmath21
, the occupancy rises and eventually saturates ; the saturation point progresses to larger @xmath21 .
this looks like a momentum - space ( kinetic ) cascade .
= 0.48 = 0.48 cascades of energy from the infrared to the ultraviolet are familiar from turbulence in hydrodynamics and from many other physical systems .
there are also weakly coupled examples such as weak plasma turbulence in traditional plasma physics @xcite , and theoretical studies of post - inflationary thermalization in the early universe @xcite .
such cascades typically lead to a steady - state , power law distribution for the power spectrum @xmath67 , usually referred to as a kolmogorov spectrum ( in honor of the application to hydrodynamic turbulence ) .
different microphysics leads to different powers of @xmath21 .
a thermal spectrum is @xmath97 .
cascades in scalar field theories during `` preheating '' after inflation , for example , typically display a power spectrum with various power laws at different stages , such as @xmath98 and @xmath99 @xcite .
obviously we do not expect power behavior at values of @xmath21 where the field is nonperturbative ; indeed , it is not even clear whether to use @xmath85 or @xmath86 in this region . however , we do expect power behavior for @xmath21 large enough that @xmath88 , but small enough and at late enough times that @xmath67 has become nearly time independent . the first figure in fig .
[ fig : coulomb2 ] repeats the righthand figure from fig.[fig : coulomb1 ] and shows that the cascade region is well fitted by a power law with @xmath100 .
unfortunately , the most ultraviolet wave numbers involved in the cascade are already at large enough @xmath21 that lattice spacing effects may be a concern . at @xmath101 . for @xmath93 ,
this is the lowest @xmath102 value where the lattice group velocity can vanish , corresponding to a van hove singularity . ] as a check on the robustness of the result , we performed a second evolution , also in a @xmath92 box , with the same choice of @xmath39 , but with @xmath103 rather than @xmath104 , and going out to a time of @xmath105 .
the smaller spacing means that the physical volume was somewhat smaller .
nevertheless , it was large enough : we have checked agreement within errors of the @xmath57 diffusion coefficients , and close agreement in the energy growth rates . in any case
the physics of the cascade is presumably more ultraviolet than the physics of the instability , and should not show severe volume sensitivity .
the power spectrum from this evolution is shown on the right in fig .
[ fig : coulomb2 ] , and is in very good agreement with the larger volume figure .
the line superposed is drawn to guide the eye and is not actually a fit ; it is precisely the same line in each figure .
= 0.49 = 0.49 we have found power - law fall - off @xmath106 with spectral index @xmath1 . to get a crude idea of the error in our determination of @xmath107 , we show a more detailed view in fig . [
fig : errors ] of the late - time distribution on the finer lattice .
we show least - square fits of various power laws ( @xmath107 = 1.6 , 1.8 , 2.0 , 2.2 , 2.4 ) to the data points data .
there is only a slight difference between @xmath108 and @xmath109 , which is at the ir end of the range chosen .
the reason for slightly preferring @xmath108 over @xmath109 will become obvious in sec .
[ sec : cool ] below . ] in the range @xmath110 .
the @xmath111 fit has the longest span of agreement , and we take our final result to be @xmath112 .
= 0.60 we argued that the coulomb gauge spectrum of fig .
[ fig : coulomb2 ] should be trustworthy away from the ir because the higher - momentum components of the field are perturbative ( as can be seen from the figures by the drop of occupation number with increasing momentum ) .
however , in order to be sure that results are not artifacts of gauge fixing , it is usually preferable to investigate gauge theories with gauge - invariant observables .
it is possible to probe aspects of power spectra in a gauge - invariant way by calculating the energy of smeared ( also known as cooled ) gauge fields .
smearing is a gauge - covariant process which is a function of a parameter @xmath113 known as the smearing depth .
define @xmath114 to be the actual gauge - field configuration @xmath115 at a given physical time @xmath66 .
then evolve in @xmath113 according to @xmath116 here @xmath117 is the covariant derivative and @xmath118 is the field strength , also using the smeared field @xmath119 .
this is a gauge - invariant procedure which has a straightforward lattice implementation .
such smearings have a long history in lattice gauge theory studies ; for instance they are also used extensively in our technique for measuring @xmath57 .
other fields can also be smeared .
for instance , we define a smeared @xmath41 field through , @xmath120 where again @xmath121 is the covariant derivative using the smeared gauge field at the same smearing depth @xmath113 .
note that we do not introduce smearing into the dynamical evolution of the fields in time ( @xmath66 ) ; we only use smearing for making measurements . to answer the question `` what is the smeared @xmath122 at time @xmath66 ? ''
, we make a copy of the fields at time @xmath66 , apply smearing , and measure @xmath122 on it .
smeared fields are good at telling us whether the energy going into soft electromagnetic fields is appearing in very long wavelength , nonperturbative fields , or in plasmons with larger wave number . to study this , we consider the magnetic field energy density @xmath123 as a function of time and of smearing depth , shown in fig.[fig : smear1 ] . very roughly , smearing to a depth @xmath113 eliminates fields with @xmath124 and leaves fields with smaller @xmath21 . but near @xmath125 , the fields are only partially removed , so the real story is slightly more complicated ; smearing is similar to a laplace transform of the power spectrum with @xmath102 playing the role of time .
the figure shows that the infrared energy is stable through the evolution .
( it bounces around , which shows that the fields are evolving and that it is a small number of degrees of freedom contributing to the infrared energy .
this bouncing would average out in larger volumes because of incoherent averaging . )
the total energy ( @xmath126 ) rises linearly .
= 0.46 based on the cascade picture of fig .
[ fig : coulomb2 ] , what behavior should we have expected for the intermediate case of moderate @xmath113 ? for any fixed smearing @xmath113 , the energy should eventually stop growing once the modes with @xmath127 have grown to reach their steady - state distribution in the cascade .
the smaller @xmath113 , the longer it should take to reach the steady state .
the @xmath128 curve in fig .
[ fig : smear1 ] is a good example of an initial rise in energy that then tapers off and is plausibly approaching a steady - state value .
it is unclear from our data whether the smaller ( non - zero ) @xmath113 curves will eventually reach steady - state values .
the problem is that cooling does not select out a single @xmath21 but gives a superposition in the form of a laplace transform .
note , for instance , that the @xmath129 behavior of the coulomb spectrum of fig .
[ fig : coulomb1 ] is limited to the relatively narrow range of @xmath130 . with infinite computing resources
, one could run long enough , on fine enough lattices , to extend this region over a huge range of @xmath21 , and then one would expect to see the predicted behavior in fig . [
fig : smear1 ] for a wide range of @xmath113 .
but there is a way to check that our coulomb gauge results are trustworthy .
we can laplace transform the coulomb gauge spectrum and see if it agrees with the ( gauge - invariant ) smeared measurements of fig .
[ fig : smear1 ] .
perturbatively , the smeared magnetic energy density should be related to the distributions @xmath67 by @xmath131 note that the growth of @xmath132 with @xmath21 typically more than compensates for the fall of @xmath67 with @xmath21 in fig .
[ fig : coulomb2 ] , so that these integrals are dominated by @xmath133 at late times @xmath66 .
we replot fig .
[ fig : smear1 ] in fig .
[ fig : compare ] , superposed with dashed lines that show the results of ( [ eq : compare ] ) with @xmath134 .
the results are very close until one cools deep into the infrared .
( we also show one sample curve based on @xmath109 , which gives slightly less accurate results than @xmath108 .
the @xmath109 have a more significant ir contribution than @xmath108 , as can be seen in fig . [
fig : coulomb1 ] . )
we conclude that coulomb - gauge distributions provide a reasonably accurate description of the physics of the cascade far from the ir , supporting our conclusions based on figs .
[ fig : coulomb1 ] and [ fig : coulomb2 ] .
= 0.46 the instability preferentially excites gauge fields with @xmath70 vector along the @xmath135 axis ( the axis about which the particle momentum distribution is oblate ) @xcite .
such modes have primarily transverse magnetic fields , so @xmath136 for the fields excited by the instability
. therefore we might expect this behavior of the infrared gauge fields .
if the higher momentum fields represent a nearly thermalized bath , they will be close to isotropic and the ratio @xmath137 will be 1/2 .
but if they scatter predominantly off the ir fields , they may also carry a momentum space anisotropy . to study this , fig .
[ fig : smear2 ] shows the ratio @xmath137 as a function of smearing , for the same lattice parameters as in fig .
[ fig : smear1 ] .
indeed , the soft fields are anisotropic as expected .
the unsmeared fields are ( on average ) less so , but still have a definite anisotropy along the @xmath135 axis .
the picture that has emerged is that there are soft , anisotropic , nonperturbatively large gauge fields with higher wave - number plasmons superposed .
the size of the soft gauge fields fluctuates about a steady mean , and the plasmons become more numerous , populating higher and higher wave numbers .
one way to check whether the interpretation of the high wave number excitations as plasmons is correct , is to look at the @xmath41 fields .
we will concentrate on the @xmath138 component of @xmath41 , which is the same as the particle current up to a factor : @xmath139 @xcite . a plasmon with @xmath140 carries almost all its energy in @xmath141 and @xmath122 , roughly equipartitioned , and only a subdominant amount in currents . therefore ,
if the energy growth really represents a growing number of plasmons with @xmath142 , then @xmath143 should grow slowly if at all , and should be ir dominated .
[ fig : jsmear ] shows our measurements , and the lack of growth with time is clear . in that reference , as we discuss in the appendix . ]
[ following ref .
@xcite , we have normalized the curves by plotting @xmath144 , which is the contribution of the @xmath145 components of @xmath41 to what would be a conserved energy @xmath146 for isotropic systems . ]
@xmath147 falls more slowly with cooling depth than the magnetic energy of fig .
[ fig : smear1 ] , indicating less power in the ultra - violet .
we also find , in fig .
[ fig : jzsmear ] , that @xmath148 , nearly independent of time and smearing depth . a small ratio of @xmath149 to @xmath150 and @xmath151 is expected if the currents are primarily associated with long wavelength , transverse modes in the directions which are perturbatively unstable .
= 0.6 = 0.6
putting our numerical results together , the physical picture which emerges is the following .
plasma instabilities drive ir gauge fields with @xmath152 to grow .
nonperturbatively strong interactions between these soft field modes remove energy from these unstable modes , moving it instead into less - ir gauge field modes .
the size of the soft nonperturbative fields reaches a quasi - steady state ; if it grows larger , the nonperturbative physics removing energy gets more efficient , and if it gets smaller , the instability drives it back up . the energy absorbed via the instability from the hard particles
thereby powers a cascade of soft gauge field excitation energy towards the ultraviolet .
the cascade has power spectrum @xmath3 with @xmath1 .
this is not a thermal spectrum , as also evidenced by the failure of @xmath153 to approach @xmath154 .
we give a theoretical explanation of the value @xmath155 in another work @xcite .
these same instability - powered energy cascades should appear in the early stages of arbitrarily high - energy heavy ion collisions .
these cascades will only be a temporary feature along the path to thermalization and will disappear by the time the plasma is finally fully thermalized .
we leave to future work the complete integration of the physics of instabilities into quark - gluon plasma thermalization at arbitrarily high energies .
we would like to thank larry yaffe for useful conversations , especially in the intial development of this project .
this work was supported , in part , by the u.s .
department of energy under grant nos .
de - fg02 - 97er41027 , by the national sciences and engineering research council of canada , and by le fonds nature et technologies du qubec .
in this appendix we argue that the large @xmath44 limit is achieved more quickly by applying weak damping on large @xmath46 modes than by not doing so , and we present numerical evidence that this is the case . at the perturbative level ,
the behavior of the isotropic version of the @xmath51 and @xmath41 field system has been investigated by bdeker , moore , and rummukainen @xcite .
the correct analytic structure of the gauge field propagator in the presence of hard thermal loops ( the infinite @xmath44 , isotropic theory ) is that there should be a propagating `` plasmon '' pole at a frequency @xmath157 , and a cut in the spectral weight for all frequencies @xmath158 .
physically , the cut reflects landau damping .
it means that most of the energy of a long wavelength magnetic field should be absorbed by the particle degrees of freedom represented by the @xmath41 fields , never to return .
however , the finite @xmath44 system is a non - dissipative hamiltonian system . therefore it is not possible for it to contain cuts in the ( leading order ) propagator .
instead , the region which should contain the cut contains a series of poles and zeros in the propagator , illustrated in fig.[fig : propagator ] , which is taken from ref .
@xcite .
what this means is that the excitation energy present in the magnetic field , which is supposed to be landau damped away , instead appears only in periodic oscillatory modes .
most of the energy is stored in the @xmath41 fields , as is supposed to happen under landau damping . however
, a fraction of it periodically reappears as magnetic field energy , an effect which is unphysical .
the size of this effect is determined by the number of @xmath41 field modes , @xmath159 , which sets the heat capacity of these modes to absorb and store excitation energy .
the larger @xmath44 is , the more effectively the @xmath41 fields can retain the energy rather than feeding it back to the magnetic fields .
the problem in the current context is that , in the anisotropic system , quite large amounts of @xmath41 field excitation are generated .
the derivative term @xmath160 in should cause this excitation energy to migrate to higher and higher @xmath46 values .
this is shown in the left - hand plot of fig .
[ fig : lvalue ] , which shows how @xmath161 varies with time and with @xmath46 , if we start with large magnetic fields but with no initial excitation in the @xmath41 fields
. however , with a finite @xmath44 cutoff , the excitation can not migrate to arbitrarily large @xmath46 , as it should ; some of it instead moves back into lower @xmath46 and re - enters the gauge fields and small @xmath46 @xmath41 fields .
this causes a fake increase in the energy of these fields , which becomes more severe as @xmath44 is decreased . our approach to solve this problem is to assume that any excitation energy which reaches very large @xmath46 values should continue to arbitrarily large @xmath46 and be lost to the low @xmath46 system .
we can make this happen artificially by applying a damping term to the highest @xmath46 modes , modifying the @xmath41 equation of motion via @xmath162 that is , we add a term which causes exponential shrinkage with rate @xmath163 in all @xmath46 at or above a cutoff @xmath164 .
the damping should only affect the high @xmath46 modes , so we choose @xmath165 .
note that the damping we apply is gauge invariant : so long as @xmath166 , the current remains exactly conserved , because the current is determined by the @xmath167 modes and the charge density by @xmath168 .
we also have to choose a value for @xmath163 .
if the value is too large , then the @xmath41 fields with @xmath169 are effectively forced to be zero , which is equivalent to lowering @xmath44 to @xmath164 .
therefore we should choose @xmath170 , since @xmath171 is the intrinsic scale of the dynamics .
we should also make sure we pick @xmath172 ; otherwise the damping is too slow to do anything , as excitation energy can get from @xmath164 to @xmath44 , reflect , and go back below @xmath164 before being damped away . with this in mind , we have chosen @xmath173 . of course , it remains to test whether this procedure of damping large @xmath46 modes really gives the same behavior as choosing an extremely large value for @xmath44 ( which is numerically impractical , since the number of degrees of freedom rises as @xmath174 ) .
to check , we have performed evolutions with the same initial random seed and other parameters but with different values of @xmath44 , and either with or without the damping term added .
the results are shown in fig.[fig : lmax ] for @xmath62 , @xmath175 , and the same @xmath39 used in the main body of the paper @xcite .
on the left , we see magnetic energy growth vs. time . with and without damping , the growth rates converge ( from opposite sides ) towards the same large @xmath44 limit .
but the damped results converge much faster : with damping , the magnetic energy growth does not change between @xmath63 and @xmath176 , and appears to coincide with the very large @xmath44 , undamped behavior .
now return to the flat result of fig .
[ fig : jsmear ] for the time dependence of @xmath147 . on the right of fig .
[ fig : lmax ] , we see that this flat behavior is obtained only very gradually in the large @xmath44 limit unless damping is implemented , in which case it occurs immediately . for investigating current growth
, damping is an essential numerical tool for practical simulations .
naturally , if @xmath44 is too small , we will see errant behavior whether or not we implement damping . for the choice of @xmath39 used here ( _ i.e. _ for the degree of hard particle anisotropy in our simulations ) , fig .
[ fig : lmax ] shows large deviation of the magnetic energy growth for @xmath44 below about 10 .
we have therefore conservatively used @xmath63 for the bulk of the studies presented in the main text .
e. shuryak , `` why does the quark gluon plasma at rhic behave as a nearly ideal fluid ? , '' prog .
phys . * 53 * , 273 ( 2004 ) [ hep - ph/0312227 ] ; m. gyulassy and l. mclerran , `` new forms of qcd matter discovered at rhic , '' nucl .
a * 750 * , 30 ( 2005 ) [ nucl - th/0405013 ] .
p. f. kolb , p. huovinen , u. w. heinz and h. heiselberg , `` elliptic flow at sps and rhic : from kinetic transport to hydrodynamics , '' phys .
b * 500 * , 232 ( 2001 ) [ hep - ph/0012137 ] ; d. teaney , j. lauret and e. v. shuryak , `` flow at the sps and rhic as a quark gluon plasma signature , '' phys .
lett . * 86 * , 4783 ( 2001 ) [ nucl - th/0011058 ] ; u. w. heinz , `` thermalization at rhic , '' aip conference proceedings * 739 * , _ hadon physics and relativistic aspects of nuclear physics _
( 2004 ) 163 [ nucl - th/0407067 ] .
r. baier , a. h. mueller , d. schiff and d. t. son , phys .
b * 539 * , 46 ( 2002 ) ; d. molnar and m. gyulassy , nucl .
a * 697 * , 495 ( 2002 ) [ erratum - ibid .
a * 703 * , 893 ( 2002 ) ] .
z. xu and c. greiner , `` thermalization of gluons in ultrarelativistic heavy ion collisions by including three - body interactions in a parton cascade , '' phys .
c * 71 * , 064901 ( 2005 ) [ hep - ph/0406278 ] .
p. arnold , j. lenaghan and g. d. moore , `` qcd plasma instabilities and bottom - up thermalization , '' jhep 08 ( 2003 ) 002 [ hep - ph/0307325 ] .
r. baier , a. h. mueller , d. schiff and d. t. son , `` ` bottom - up ' thermalization in heavy ion collisions , '' phys .
b * 502 * , 51 ( 2001 ) [ hep - ph/0009237 ] .
s. , `` stream instabilities of the quark - gluon plasma , '' phys .
b * 214 * , 587 ( 1988 ) ; y. e. pokrovsky and a. v. selikhov , `` filamentation in a quark - gluon plasma , '' jetp lett . *
47 * , 12 ( 1988 ) [ pisma zh .
* 47 * , 11 ( 1988 ) ] ; `` filamentation in quark plasma at finite temperatures , '' sov .
j. nucl .
phys . *
52 * , 146 ( 1990 ) [ yad .
* 52 * , 229 ( 1990 ) ] ; `` filamentation in the quark - gluon plasma at finite temperatures , '' sov .
j. nucl .
phys . * 52 * , 385 ( 1990 ) [ yad .
fiz . * 52 * , 605 ( 1990 ) ] ; o. p. pavlenko , `` filamentation instability of hot quark - gluon plasma with hard jet , '' sov .
j. nucl .
phys . * 55 * , 1243 ( 1992 ) [ yad . fiz . *
55 * , 2239 ( 1992 ) ] ; s. , `` plasma instability at the initial stage of ultrarelativistic heavy ion collisions , '' phys .
b * 314 * , 118 ( 1993 ) ; `` color collective effects at the early stage of ultrarelativistic heavy ion collisions , '' phys .
c * 49 * , 2191 ( 1994 ) ; `` color filamentation in ultrarelativistic heavy - ion collisions , '' phys .
b * 393 * , 26 ( 1997 ) [ hep - ph/9606442 ] .
p. romatschke and m. strickland , `` collective modes of an anisotropic quark gluon plasma , '' phys .
d * 68 * , 036004 ( 2003 ) [ hep - ph/0304092 ] .
p. arnold and j. lenaghan , `` the abelianization of qcd plasma instabilities , '' phys .
d * 70 * , 114007 ( 2004 ) [ hep - ph/0408052 ] .
a. rebhan , p. romatschke and m. strickland , `` hard - loop dynamics of non - abelian plasma instabilities , '' phys .
lett . * 94 * , 102303 ( 2005 ) [ hep - ph/0412016 ] .
a. dumitru and y. nara , `` qcd plasma instabilities and isotropization , '' phys .
b * 621 * , 89 ( 2005 ) [ hep - ph/0503121 ] .
p. arnold
, g. d. moore and l. g. yaffe , `` the fate of non - abelian plasma instabilities in 3 + 1 dimensions , '' phys . rev .
d * 72 * , 054003 ( 2005 ) [ hep - ph/0505212 ] .
a. rebhan , p. romatschke and m. strickland , `` dynamics of quark - gluon plasma instabilities in discretized hard - loop approximation , '' hep - ph/0505261 .
a. h. mueller , a. i. shoshi and s. m. h. wong , `` a possible modified ` bottom - up ' thermalization in heavy ion collisions , '' hep - ph/0505164 ; d. bdeker , `` the impact of qcd plasma instabilities on bottom - up thermalization , '' hep - ph/0508223 . l. v. gribov , e. m. levin and m. g. ryskin , `` semihard processes in qcd , '' phys .
* 100 * , 1 ( 1983 ) ; j. p. blaizot and a. h. mueller , `` the early stage of ultrarelativistic heavy ion collisions , '' nucl .
b * 289 * , 847 ( 1987 ) ; l. d. mclerran and r. venugopalan , `` computing quark and gluon distribution functions for very large nuclei , '' phys .
d * 49 * , 2233 ( 1994 ) [ hep - ph/9309289 ] ; `` green s functions in the color field of a large nucleus , '' * 50 * , 2225 ( 1994 ) [ hep - ph/9402335 ] ; j. jalilian - marian , a. kovner , l. d. mclerran and h. weigert , `` the intrinsic glue distribution at very small x , '' phys . rev .
d * 55 * , 5414 ( 1997 ) [ hep - ph/9606337 ] ; e. iancu , `` effective theory for real - time dynamics in hot gauge theories , '' phys .
b * 435 * , 152 ( 1998 ) .
d. bdeker , g. d. moore and k. rummukainen , `` chern - simons number diffusion and hard thermal loops on the lattice , '' phys .
d * 61 * , 056003 ( 2000 ) [ hep - ph/9907545 ]
. s. mrowczynski , a. rebhan and m. strickland , `` hard - loop effective action for anisotropic plasmas , '' phys .
d * 70 * , 025004 ( 2004 ) [ hep - ph/0403256 ] .
j. ambjorn , t. askgaard , h. porter and m. e. shaposhnikov , `` sphaleron transitions and baryon asymmetry : a numerical real time analysis , '' nucl .
b * 353 * , 346 ( 1991 ) .
g. d. moore , `` measuring the broken phase sphaleron rate nonperturbatively , '' phys
. rev .
d * 59 * , 014503 ( 1999 ) [ hep - ph/9805264 ] .
j. ambjorn and a. krasnitz , `` improved determination of the classical sphaleron transition rate , '' nucl .
b * 506 * , 387 ( 1997 ) [ hep - ph/9705380 ] .
j. e. mandula and m. ogilvie , `` the gluon is massive : a lattice calculation of the gluon propagator in the landau gauge , '' phys .
b * 185 * , 127 ( 1987 ) .
g. d. moore and n. turok , `` lattice chern - simons number without ultraviolet problems , '' phys .
d * 56 * , 6533 ( 1997 ) [ hep - ph/9703266 ] .
v. n. tsytovich , _ lectures on non - linear plasma kinetics _ ( springer - verlag , 1995 ) .
d. t. son , `` reheating and thermalization in a simple scalar model , '' phys . rev .
d * 54 * , 3745 ( 1996 ) [ hep - ph/9604340 ] ; r. micha and i. i. tkachev , `` turbulent thermalization , '' phys . rev .
d * 70 * , 043538 ( 2004 ) [ hep - ph/0403101 ] .
p. arnold and g. d. moore , in preparation . | magnetic plasma instabilities appear to play an important role in the early stages of quark - gluon plasma equilibration in the high energy ( weak coupling ) limit .
numerical studies of the growth of such instabilities from small seed fluctuations have found initial exponential growth in their energy , followed by linear growth once the associated color magnetic fields become so large that their non - abelian interaction are non - perturbative . in this paper , we use simulations to determine the nature of this linear energy growth .
we find that the long - wavelength modes associated with the instability have ceased to grow , but that they cascade energy towards the ultraviolet in the form of plasmon excitations of ever increasing energy .
we find a quasi - steady - state power - law distribution @xmath0 for this cascade , with spectral index @xmath1 .
# 1eq .
( [ # 1 ] ) l v c |
in search for new fundamental structures in theoretical physics and mechanics the physicists are increasingly turning their attention to algebraic structures which are based on ternary multiplication law ( more generally on @xmath0-ary multiplication law ) . in 1973
y. nambu proposed a generalization of hamiltonian mechanics where he replaced a canonically conjugate pair of variables by a triple of canonical variables and the usual poisson bracket by a ternary operation ( nambu bracket ) @xcite . a geometric formalism for this generalization of hamiltonian mechanics based on a notion of nambu bracket of order @xmath0 , the fundamental identity and a concept of nambu - poisson manifold was developed by l. takhtajan in @xcite .
an important part of this geometric formalism is a notion of an @xmath0-lie algebra which was also studied by v. t. filippov in @xcite .
j. arnlind , a. makhlouf and s. silvestrov constructed an @xmath0-lie algebra by means of a binary lie algebra endowed with an analog of a trace @xcite and this @xmath0-lie algebra was called an @xmath0-lie algebra induced by lie algebra .
they also studied the cohomologies of @xmath0-lie algebra induced by a lie algebra and found the relation of these cohomologies to the cohomology of initial binary lie algebra . in this paper
we propose a notion of super @xmath0-lie algebra and follow an approach of j. arnlind , a. makhlouf and s. silvestrov to construct a super @xmath0-lie algebra with the help of a binary super lie algebra which is equipped with an analog of a supertrace , and this super @xmath0-lie algebra is called the super @xmath0-lie algebra induced by a super lie algebra .
we apply this approach to the super lie algebra of clifford algebra ( with even number of generators ) with the matrix representation given by a supermodule of spinors and for any even integer @xmath1 we construct a super 3-lie algebra .
in this section we remind the definition of @xmath0-lie algebra .
let @xmath3 be a field of real or complex numbers , @xmath4 be a vector space over @xmath5 , and @xmath6 .
a vector space @xmath7 is said to be an @xmath0-lie algebra if @xmath7 is endowed with a multilinear skew - symmetric mapping @xmath8:{\mathfrak g}^n\to { \mathfrak g}$ ] which satisfies the identity @xmath9 = \nonumber\\ & & \qquad\quad\sum_{k=1}^{n}[x_1,x_2,\ldots , x_{k-1},[y_1,y_2,\ldots , y_{n-1},x_{k}],x_{k+1},\ldots , x_n ] , \label{identity for n - lie algebra}\end{aligned}\ ] ] where @xmath10 .
particularly if @xmath11 then the above definition yields the definition of a ( binary ) lie algebra and the identity ( [ identity for n - lie algebra ] ) takes on the form of jacobi identity .
let @xmath12 be a ( binary ) lie algebra with a lie bracket @xmath13:{\frak h}\times{\frak h}\to { \frak h}$ ] , and @xmath14 be an @xmath5-multilinear skew - symmetric @xmath15-form which satisfies the condition @xmath16,x_{m+1},\ldots , x_k)=0,\ ] ] where @xmath17 are elements of @xmath12 . particularly if @xmath18 then a form @xmath19 satisfies @xmath20)=0 $ ] for any @xmath21 .
let @xmath22 be an integer , @xmath23 be a permutation of the integers @xmath24 such that @xmath25 , @xmath26 and @xmath27 be the parity of @xmath28 .
define @xmath29=\sum_{\sigma } ( -1)^{|\sigma|}\phi(x_{i_1},x_{i_2},\ldots , x_{i_k})[x_{i_{k+1}},x_{i_{k+2 } } ] .
\label{n - bracket for lie algebra}\ ] ] then @xmath12 endowed with ( [ n - bracket for lie algebra ] ) is the @xmath0-lie algebra , where @xmath30 .
it is worth noting that in the particular case of @xmath18 this theorem yields the 3-lie algebra with ternary lie bracket @xmath31=\phi(x)[y , z]+\phi(y)[z , x]+\phi(z)[x , y ] , \label{ternary lie bracket}\ ] ] and a form @xmath19 which for any @xmath21 satisfies @xmath20)=0 $ ] can be viewed as an analog of a trace .
the 3-lie algebras of this kind induced by a binary lie algebra are introduced and studied @xcite .
in this section we give a definition of a super @xmath0-lie algebra and prove that one can construct a super @xmath0-lie algebra by means of a super trace .
let @xmath32 be a finite - dimensional super vector space , where @xmath33 is the subspace of even elements and @xmath34 is the subspace of odd elements . if @xmath35 is a homogeneous element then its degree will be denoted by @xmath36 , where @xmath37 and @xmath38 .
let @xmath39 .
for any @xmath40 , where @xmath41 are homogeneous elements , and for every integer @xmath42 we define @xmath43 let @xmath44 be the super vector space of endomorphisms of a super vector space @xmath45 .
the composition of two endomorphisms @xmath46 determines the structure of superalgebra in @xmath44 , and the graded binary commutator @xmath47=a\circ b-(-1)^{|a||b|}$ ] induces the structure of super lie algebra in @xmath44 .
the supertrace of an endomorphism @xmath48 can be defined by @xmath49 for any endomorphisms @xmath50 it holds @xmath51)=0 $ ] .
given @xmath40 and a permutation @xmath52 of the integers @xmath53 we assign to them the element @xmath54 defined by @xmath55 let @xmath56 be an element of permutation @xmath28 .
assume that @xmath57 are the elements of permutation @xmath28 which precede an element @xmath56 in @xmath28 and greater than @xmath56 .
obviously @xmath58 is the number of inversions of an element @xmath56 in @xmath28 .
define the integer @xmath59 by @xmath60 a super vector space @xmath61 is said to be a super @xmath0-lie algebra if it is endowed with an @xmath0-ary bracket @xmath62 : { \frak g}\times { \frak g}\times\ldots\times{\frak g}($]n@xmath63 which satisfies the following conditions : 1 .
@xmath0-ary bracket is a multilinear mapping and the degree of @xmath0-ary bracket of @xmath0 homogeneous elements is equal to the sum of degrees of these elements , i.e. for any homogeneous element @xmath64 it holds @xmath65|=|{\bf x}|_n,\ ] ] 2 .
@xmath0-ary bracket is a graded skew - symmetric multilinear mapping , i.e. for any homogeneous elements @xmath66 and for any @xmath67 it holds @xmath68=-(-1)^{|x_k||x_{k+1}|}[x_1,x_2,\ldots , x_{k+1},x_{k},\ldots , x_n],\ ] ] 3 .
@xmath0-ary bracket satisfies the identity @xmath9 = \nonumber\\ & & \;\;\sum_{k=1}^{n}(-1)^{|{\bf x}|_{k-1}|{\bf y}|_{n-1}}[x_1,x_2,\ldots , x_{k-1},[y_1,y_2,\ldots , y_{n-1},x_{k}],x_{k+1},\ldots , x_n],\nonumber\end{aligned}\ ] ] where @xmath69 are homogeneous elements .
let @xmath44 be the super vector space of endomorphisms of a super vector space @xmath45 and @xmath70 be a sequence of @xmath0 endomorphisms of @xmath45 .
define @xmath71=\sum_{\sigma}(-1)^{|\sigma|+|\bf{a}_\sigma|}a_{i_1}\circ a_{i_2}\circ\ldots\circ a_{i_n } , \label{n - bracket for endomorphisms}\ ] ] where @xmath52 is a permutation of integers @xmath53 and @xmath27 is the parity of this permutation .
then the super vector space @xmath44 endowed with the @xmath0-ary bracket ( [ n - bracket for endomorphisms ] ) is the super @xmath0-lie algebra .
a representation of a super @xmath0-lie algebra @xmath4 is a linear mapping @xmath72 , where @xmath45 is a super vector space ( a representation space of @xmath4 ) , which satisfies : 1 . for
any homogeneous element @xmath73 its image @xmath74 in @xmath44 is the homogeneous element and @xmath75 , 2 . for any elements @xmath66 it holds @xmath76)=[\rho(x_1),\rho(x_2),\ldots,\rho(x_n)].\ ] ]
let @xmath77 be a super @xmath0-lie algebra with @xmath0-ary bracket @xmath8 $ ] , and @xmath72 be a representation of @xmath4 .
for any homogeneous element @xmath78 of @xmath4 we define the @xmath79-ary bracket by @xmath80_\rho & = & \sum_{k=1}^{n+1}(-1)^{k-1}(-1)^{|x_k||{\bf x}|_{k-1}}\mbox{str}(\rho(x_k))\nonumber\\ & & \qquad\qquad\qquad\quad\times [ x_1,x_2,\ldots,\hat x_k,\ldots , x_{n+1 } ] , \label{n - bracket with supertrace}\end{aligned}\ ] ] where the hat over an element @xmath81 means that this element is omitted .
a super @xmath0-lie algebra @xmath4 equipped with the @xmath79-ary bracket ( [ n - bracket with supertrace ] ) is the super @xmath79-ary lie algebra .
[ theorem for n+1 lie algebra with super trace ]
a clifford algebra is a unital associative algebra , which can be equipped with an @xmath82-graded structure , and it provides a well known example of a superalgebra . taking the graded commutator of two elements of this superalgebra one can consider it as the super lie algebra . in this section
we consider a supermodule over clifford algebra with even number of generators , and this supermodule provides us with a representation of mentioned above super lie algebra .
making use of a supertrace of this representation we construct a super 3-lie algebra with the help of the formula ( [ n - bracket with supertrace ] ) .
we remind that clifford algebra @xmath83 is the unital associative algebra over @xmath84 generated by @xmath85 which obey the relations @xmath86 where @xmath87 is the unit element of clifford algebra .
let @xmath88 be the set of integers from 1 to @xmath0 .
if @xmath89 is a subset of @xmath90 , i.e. @xmath91 where @xmath92 , then one can associate to this subset @xmath89 the monomial @xmath93 . if @xmath94 one defines @xmath95 .
the number of elements of a subset @xmath89 will be denoted by @xmath96 .
it is obvious that the vector space of clifford algebra @xmath83 is spanned by the monomials @xmath97 , where @xmath98 .
hence the dimension of this vector space is @xmath99 and any element @xmath100 can be expressed in terms of these monomials as @xmath101 where @xmath102 is a complex number .
it is easy to see that one can endow a clifford algebra @xmath83 with the @xmath103-graded structure by assigning the degree @xmath104 to monomial @xmath97 .
then a clifford algebra @xmath83 can be considered as the superalgebra since for any two monomials it holds @xmath105 .
let @xmath89 be a subset of @xmath88 , and @xmath97 be a symbol associated to @xmath89 .
let @xmath83 be the vector space spanned by the symbols @xmath97 .
define the degree of @xmath97 by @xmath106 , where @xmath96 is the number of elements of @xmath89 , and the product of @xmath107 by @xmath108 where @xmath109 , @xmath110 is the number of elements of @xmath89 which are greater than @xmath111 , and @xmath112 is the symmetric difference of two subsets .
then @xmath83 is the unital associative superalgebra , where the unit element @xmath87 is @xmath113 .
this theorem can be proved by means of the properties of symmetric difference of two subsets .
we remind a reader that the symmetric difference is commutative @xmath114 , associative @xmath115 and @xmath116 .
the latter shows that @xmath117 is the unit element of this superalgebra .
the symmetric difference also satisfies @xmath118 .
hence @xmath83 is the superalgebra .
the superalgebra @xmath83 can be considered as the super lie algebra if for any two homogeneous elements @xmath119 of this superalgebra one introduces the graded commutator @xmath120=xy-(-1)^{|x||y|}yx$ ] and extends it by linearity to a whole superalgebra @xmath83 .
we will denote this super lie algebra by @xmath121 .
then @xmath122 are the generators of this super lie algebra @xmath121 , and its structure is entirely determined by the graded commutators of @xmath97 .
then for any two generators @xmath107 we have @xmath123=f(i , j)\;\gamma_{i\delta j } , \label{binary commutators}\ ] ] where @xmath124 is the integer - valued function of two subsets of @xmath90 defined by @xmath125 it is easy to verify that the degree of graded commutator is consistent with the degrees of generators , i.e. @xmath126=|\gamma_i|+|\gamma_j|.$ ] indeed the function @xmath127 satisfies @xmath128 and @xmath129 hence @xmath126=-(-1)^{|i||j|}[\gamma_j,\gamma_i]$ ] which shows that the relation ( [ binary commutators ] ) is consistent with the symmetries of graded commutator .
it is obvious that if the intersection of subsets @xmath130 contains an even number of elements then @xmath131 , and the graded commutator of @xmath107 is trivial . particularly if at least one of two subsets @xmath130 is the empty set then @xmath131 .
thus any graded commutator ( [ binary commutators ] ) containing @xmath87 is trivial . as an example , consider the super lie algebra @xmath132 .
its underlying vector space is 4-dimensional and @xmath132 is generated by two even degree generators @xmath133 and two odd degree generators @xmath134 the non - trivial relations of this super lie algebra are given by @xmath135=[\gamma_2,\gamma_2]=2\,e,\ ; [ \gamma_1,\gamma_{12}]=2\,\gamma_2,\ ; [ \gamma_2,\gamma_{12}]=-2\,\gamma_1.\end{aligned}\ ] ] now we assume that @xmath136 is an even integer . the super lie algebra @xmath121 has a matrix representation which can be described as follows .
fix @xmath11 and identify the generators @xmath137 with the pauli matrices @xmath138 , i.e. @xmath139 then @xmath140 where @xmath141 let @xmath142 be the 2-dimensional complex super vector space @xmath143 with the odd degree operators ( [ pauli matrices 1,2 ] ) , where the @xmath82-graded structure of @xmath142 is determined by @xmath144 .
then @xmath145 , and @xmath142 can be considered as a supermodule over the superalgebra @xmath146 .
let @xmath147 .
then @xmath148 can be viewed as a supermodule over the @xmath149-fold tensor product of @xmath146 , which can be identified with @xmath83 by identifying @xmath137 in the @xmath150th factor with @xmath151 in @xmath83 .
this @xmath83-supermodule @xmath148 is called the supermodule of spinors @xcite .
hence we have the matrix representation for the clifford algebra @xmath83 , and this matrix representation or supermodule of spinors allows one to consider the supertrace , and it can be proved @xcite that @xmath152 now we have the super lie algebra @xmath153 with the graded commutator defined in ( [ binary commutators ] ) and its matrix representation based on the supermodule of spinors . hence we can construct a super 3-lie algebra by making use of graded ternary commutator ( [ n - bracket with supertrace ] ) . applying the formula ( [ n - bracket with supertrace ] )
we define the graded ternary commutator for any triple @xmath154 of elements of basis for @xmath153 by @xmath155 = \mbox{str}(\gamma_i)\,[\gamma_j,\gamma_k]-(-1)^{|i||j|}\mbox{str}(\gamma_j)\ , [ \gamma_i,\gamma_k]\nonumber\\ & & \qquad\qquad\qquad\qquad\qquad\qquad\quad + ( -1)^{|k|(|i|+|j|)}\mbox{str}(\gamma_k)\,[\gamma_i,\gamma_j ] , \label{ternary graded commutator with super trace}\end{aligned}\ ] ] where the binary graded commutator at the right - hand side of this formula is defined by ( [ binary commutators ] ) .
according to theorem [ theorem for n+1 lie algebra with super trace ] the vector space spanned by @xmath156 and equipped with the ternary graded commutator ( [ ternary graded commutator with super trace ] ) is the super 3-lie algebra which will be denoted by @xmath157 . making use of ( [ binary commutators ] ) we can write the expression at the right - hand side of the above formula in the form @xmath155 = f(j , k ) \mbox{str}(\gamma_i)\,\gamma_{j\delta k}-(-1)^{|i||j|}f(i , k)\mbox{str}(\gamma_j)\ , \gamma_{i\delta k}\nonumber\\ & & \quad\qquad\qquad\qquad\qquad\qquad\qquad\quad + ( -1)^{|k|(|i|+|j|)}f(i , j)\mbox{str}(\gamma_k)\,\gamma_{i\delta j}.\nonumber\end{aligned}\ ] ] from the formula for supertrace ( [ super trace ] ) it follows immediately that the above graded ternary commutator is trivial if none of subsets @xmath158 is equal to @xmath159 .
similarly this graded ternary commutator is also trivial if all three subsets @xmath160 are equal to @xmath90 , i.e. @xmath161 , or two of them are equal to @xmath159 .
the graded ternary commutators of the generators @xmath162 of the super 3-lie algebra @xmath157 are given by @xmath163=\left\ { \begin{array}{ll } ( 2i)^mf(i , j)\gamma_{i\delta j } & \text{if } i\neq { \mathcal n } , j\neq { \mathcal n } , k={\mathcal n},\\ 0 & \text{in all other cases } .
\end{array } \right . \label{proposition}\ ] ] the author gratefully acknowledges that this work was financially supported by the estonian science foundation under the research grant etf9328 and this work was also financially supported by institutional research funding iut20 - 57 of the estonian ministry of education and research .
1 j. arnlind , a. kitouni , a. makhlouf , s. silvestrov , _ structure and cohomology of 3-lie algebras induced by lie algebras _ , arxiv : 1312.7599v . j. arnlind , a. makhlouf , s. silvestrov , _ construction of @xmath0-lie algebras and @xmath0-ary hom - nambu - lie algebras _ , j. math .
* 52 * ( 2011 ) , 123502 , 13 pp . v. t. filippov , _
@xmath0-lie algebras _ , siberian math .
j. * 26 * ( 1985 ) , 879 891 .
y. nambu , _ generalized hamiltonian dynamics _ , phys .
d ( 3 ) 7 ( 1973 ) , 2405 2412 .
v. mathai , d. quillen , _ superconnections , thom classes , and equivariant differential forms _ ,
topology * 25 * , 1 ( 1986 ) , 85110 . l. takhtajan , _ on foundation of the generalized nambu mechanics _ , commun.math.phys .
160 ( 1994 ) , 295 - 316 | we propose a notion of a super @xmath0-lie algebra and construct a super @xmath0-lie algebra with the help of a given binary super lie algebra which is equipped with an analog of a supertrace .
we apply this approach to the super lie algebra of a clifford algebra with even number of generators and making use of a matrix representation of this super lie algebra given by a supermodule of spinors we construct a series of super 3-lie algebras labeled by positive even integers . |
nonlinear optical materials are the active constituent for many applications such as light modulators , contrast agents for medical imaging and therapy , optical solitons , phase conjugation mirrors and optical self - modulation . in each of these ,
the performance of the system is improved by using a material with a stronger nonlinear response , quantified by various nonlinear susceptibilities defined by expanding the induced polarization @xmath8 in a power series in the applied electric field , @xmath9 here , @xmath10 is the linear susceptibility familiar from dielectric materials ; @xmath11 and @xmath12 are the nonlinear susceptibilities and are referred to as the first and second hyperpolarizability respectively .
these quantities are in general frequency - dependent tensors that depend on the electronic structure of the constituent molecules , their symmetry , ordering and the material in which they are embedded . in the present work we focus on the off - resonant molecular contribution .
much effort has been expended over the years in synthesizing new molecules with higher @xmath11 or @xmath12 .
comparisons between materials must be made carefully however , because these quantities increase trivially with the size of the molecule .
important progress on developing suitable figures - of - merit for comparison was made by kuzyk , who showed that fundamental quantum mechanics requires that @xmath11 and @xmath12 for the off - resonant case are bounded by the inequalities , @xmath13{4}\left(\frac{e\hbar}{\sqrt{m}}\right)^{3}\frac{n^{3/2}}{e_{10}^{7/2}}\equiv\beta_{0}^{max},\\ \left(\frac{e\hbar}{\sqrt{m}}\right)^{4}\frac{n^{2}}{e_{10}^{5}}\le & \gamma & \le4\left(\frac{e\hbar}{\sqrt{m}}\right)^{4}\frac{n^{2}}{e_{10}^{5}}\equiv\gamma_{0}^{max}\label{eq : kuyzklimit}\end{aligned}\ ] ] where @xmath14 is the difference between the ground and first excited state and @xmath0 is the number of electrons participating . from the maximum values
@xmath15 and @xmath16 , one defines intrinsic hyperpolarizabilities , @xmath17 the intrinsic quantities have the property that they remain invariant under a simultaneous rescaling of energy and length , @xmath18 and hence are useful quantities for comparing materials because they remove the irrelevant scaling with size .
analysis of extant materials following the derivation of the bounds in eq .
( [ eq : kuyzklimit ] ) revealed that all of them fell short of the fundamental limits by more than an order of magnitude , an observation that has catalyzed a great deal of research over the past decade on how to create materials that approach these fundamental limits .
the derivation of the bounds provides some guidancefor optimal @xmath11 and @xmath12 only three states are assumed to significantly contribute to the hyperpolarizabilities and the optimum can be achieved by tuning the dipole transition matrix elements and energy level spacingsbut does not construct an explicit molecule or potential that has these properties .
subsequent work , thoroughly reviewed in @xcite has attempted to determine whether these predictions are universal and to explicitly construct potentials that approach them .
one approach has been to conduct monte carlo searches of hamiltonians with arbitrary spectra and dipole transition elements to identify those with large @xmath6 and @xmath7 .
the results support the three - state hypothesis , though the optima found in such calculations need not correspond to a local potential . to address this ,
several authors have numerically optimized the intrinsic hyperpolarizabilities with respect to the potential function for 1 electron , using various representations of the potential , including power laws@xcite , elementary functions with a superimposed fourier series@xcite , piecewise linear potentials@xcite and quantum graphs@xcite .
the best potentials from these different studies possess hyperpolarizabilities within the bounds of eq .
( [ eq : kuyzklimit ] ) with the best known values found of @xmath19 and @xmath20 achieved in several studies with qualitatively different potentials .
it has therefore been speculated that the fundamental limits may require exotic potentials and not be achievable with local potential functions .
the effect of including multiple electrons on the optimized potentials has , however , received relatively little attention .
watkins and coworkers @xcite found for @xmath21 electrons that the best intrinsic hyperpolarizabilities are somewhat lower than for the @xmath22 electron case , but even with electron - electron interactions included , the universal features identified in other studies remained the same . in this paper
, we apply the potential optimization technique to potentials with @xmath23 electrons that interact only through pauli exclusion .
this is a key step towards simulating realistic molecules .
we find that the best values of @xmath6 and @xmath7 fall off with increasing @xmath0 from the @xmath22 electron case , but rapidly converge to a universal value .
the small number of parameters in our potentials allows a detailed exploration of the `` landscape '' of @xmath6 and @xmath7 around the maximum . as in previous work ,
the hyperpolarizabilities are more sensitive to one parameter than the other .
dimensional and approximate analytical arguments allow us to provide physical interpretations of these two parameters in terms of the wavefunctions of the highest occupied molecular orbital .
the paper is organized as follows : in section [ sec : model ] the choice of potential , calculation and optimization techniques are briefly reviewed ; in section [ sec : results ] we present results for @xmath6 and @xmath7 separately together with some discussion of the implications of our results for identifying the features of potentials most important to the hyperpolarizabilities ; brief conclusions are presented in section [ sec : conclusion ] .
schematics of the potential classes in which @xmath24 and @xmath25 are optimized . ( a ) asymmetric triangle will with a delta function at the center .
the slope on the right side is fixed at @xmath26 while the slope of the side @xmath27 and the strength of the delta function @xmath10 can be varied .
@xmath27 can be fixed at 1 to study a symmetric 1-parameter potential .
( b ) a 2 parameter symmetric linear piecewise potential .
the first slope to the right of the origin is fixed at 1 .
the position of the boundary between the first and second elements @xmath28 and the second slope @xmath29 can be varied .
the left side of the potential is constrained to be the reflection of the right side . ] following a similar approach to that established in our earlier papers@xcite , we optimize @xmath6 and @xmath7 with respect to the shape of a piecewise linear potential dressed with dirac delta functions for @xmath0 electrons interacting only through pauli exclusion .
we perform this optimization for two carefully chosen potentials depicted in fig . [
fig : schematic ] as well as a more general potential .
the first type is an asymmetric triangular well with a delta function at the center [ fig .
[ fig : schematic](a ) ] , @xmath30 parametrized by the left hand slope @xmath31 and the strength of the @xmath32-function @xmath10 .
the effect of the @xmath32 function is to introduce a sudden change in the phase of the wavefunction .
_ @xcite recently showed that the addition of a @xmath32 function to a 1d potential has an equivalent effect on the wavefunction to adding a side chain on a quantum graph .
this correspondence suggests that 1d potentials dressed with @xmath32-functions could be engineered in molecules by the addition of appropriate side groups . the second type of potential we consider , depicted in fig .
[ fig : schematic](b ) , contains only linear elements defined for @xmath33 , @xmath34 and for @xmath35 defined by enforcing @xmath36 symmetry , i.e. @xmath37 ; this potential is specified by two parameters @xmath28 the position of the boundary between the two elements and @xmath29 the slope of the outermost element .
these potentials were motivated by our results in @xcite that only 2 parameters at most were important to the optimization of both @xmath6 and @xmath7 ; they have been designed to achieve the known limits for @xmath22 electrons with residual flexibility .
for example , we showed in @xcite that a triangular well with a @xmath32-function of variable strength at the center , i.e. taking the potential ( [ eq : potentialdelta ] ) and fixing @xmath38 , was able to reach within @xmath39 of the upper bound for @xmath7 with @xmath40 as the only free parameter .
the parametrization has also been chosen to eliminate variables irrelevant to @xmath6 and @xmath7 associated with translations of the potential and rescalings of the form ( [ eq : rescaling ] ) .
we also minimized @xmath6 for a piecewise linear potential with @xmath41 elements , @xmath42 with positions @xmath43 and slopes @xmath44 as the parameter set describing the potential .
we used this potential in our earlier paper on maximizing @xmath6 for one electron@xcite .
there are some necessary constraints on the parameters : the @xmath43 are strictly ascending ; @xmath45 and @xmath46 with no loss of generality and @xmath47 with the remaining constants @xmath48 given by , @xmath49 the energy scale associated with the potential is also fixed , as in the previous paper , by setting @xmath50 .
having imposed these constraints , there remain @xmath51 free parameters . for each of these potentials ,
@xmath6 and @xmath7 were calculated for @xmath0 electrons as follows : first , the schrdinger equation is written for each segment as , @xmath52\psi_{n}=e\psi_{n}\ ] ] where @xmath44 and @xmath48 are the slope and offset in the @xmath53th segment and @xmath54 is the applied electric field .
this can be solved analytically using the well - known airy functions , @xmath55{2}\left(b_{n}-e+x\left(a_{n}+\epsilon\right)\right)}{\left(a_{n}+\epsilon\right)^{2/3}}\right]\nonumber \\ & & + d_{n}\text{bi}\left[\frac{\sqrt[3]{2}\left(b_{n}-e+x\left(a_{n}+\epsilon\right)\right)}{\left(a_{n}+\epsilon\right)^{2/3}}\right].\end{aligned}\ ] ] to solve for the coefficients @xmath56 and @xmath57 in each element , a set of boundary conditions are assembled at the edge of each element from the usual conditions , i.e. , @xmath58 where @xmath59 is the strength of the delta function centered at @xmath60 .
the requirement that @xmath61 as @xmath62 eliminates two coefficients .
the boundary conditions can be written as a set of linear equations , @xmath63 where @xmath64 is a vector comprised of the @xmath56 and @xmath57 coefficients and @xmath65 is a matrix that depends on @xmath66 , @xmath54 and the parameters @xmath44 and @xmath60 .
the single electron energy levels @xmath67 for the potential are determined by numerically finding the roots of , @xmath68 setting @xmath69 .
having determined these , we construct the non - interacting @xmath0 electron ground state from the single electron states by successively filling the energy levels using the _ aufbau _ principle ; we similarly determine the first excited state by promoting an electron from the highest occupied orbital to the lowest unoccupied orbital . in this work
we focus only on even values of @xmath0 , as this simplifies determining which electron to promote .
the @xmath0-electron ground state energy @xmath70 , and that of the first excited state @xmath71 , are then determined by summing over the energies of the individual single electron energies , @xmath72 where @xmath73 is the occupation number for the @xmath74-th single electron level in the @xmath53-th multi - electron state .
the hyperpolarizabilities @xmath11 and @xmath12 are obtained by differentiating the ground state energy as a function of @xmath54 , @xmath75 the necessary derivatives can be related to the single - electron energy levels by differentiating ( [ eq : energyoccupance ] ) with respect to @xmath54 .
an important advantage of the piecewise linear potential is that the necessary derivatives of the single electron energy levels @xmath67 are conveniently obtained by repeated differentiation of the determinant eq .
( [ eq : determinant ] ) using the jacobi formula , @xmath76 where @xmath77 is the adjugate matrix of @xmath65 , together with the chain rule , @xmath78 to avoid repetition , formulae are available for these derivatives in references @xcite and @xcite . having evaluated these derivatives , the intrinsic hyperpolarizabilities are easily calculated numerically and a program to do so was implemented in _
mathematica 10_. in subsequent sections , we present the optimized results as a function of @xmath0 , together with the corresponding best potentials .
( a ) maximum @xmath24 achieved in 2-parameter asymmetric @xmath32-function potential .
results for the asymmetric delta function potential are shown as grey circles ; results for an arbitrary linear potential are shown as red crosses .
the dashed line represents the largest values of @xmath24 for one electron found to date
. optimized potential and wavefunctions for the arbitrary linear potential with ( b ) @xmath21 electrons and ( c ) @xmath79 electrons . for the asymmetric @xmath32-function potential and @xmath21 , ( d ) the objective functions @xmath24 plotted versus the @xmath32-function potential shape parameters @xmath27 and @xmath10 and ( e ) the optimized potential and wavefunctions .
corresponding plots for @xmath79 electrons are shown in ( f ) and ( g ) .
for all subfigures , wavefunctions plotted in black are occupied in the ground state ; those in grey correspond to the two lowest unoccupied states . ]
> m0.5in > m1in > m0.5in > m0.5in > m0.75in > m0.75in > m1in > p0.5in > p0.5 in value of @xmath6 & potential and wavefunctions & @xmath27 & @xmath10 & hessian eigenvalues & hessian principal eigenvector & @xmath80 & @xmath81 & @xmath66 + 0.404 ( 100% ) & & 1.334 & -2.229 & -6.84 , -0.099 & @xmath82 & & 0.573 & 0.500 + -0.372 ( 96% ) & & 1.227 & 1.928 & 7.70 , 0.130 & @xmath83 & & 0.570 & 0.500 + 0.370 ( 92% ) & & 2.568 & 2.307 & -1.24 , -0.014 & @xmath84 & & 0.569 & 0.523 + we optimized @xmath6 with respect to the parameters of the asymmetric @xmath32-function potential of eq .
( [ eq : potentialdelta ] ) as well as the more general piecewise linear potential with of eq .
( [ eq : generalpotential ] ) with four linear elements and @xmath85 parameters .
the number of linear elements was chosen because earlier work showed no improvement in @xmath6 after 6 parameters@xcite .
the highest values found for increasing @xmath0 are displayed in fig .
[ fig : betaresults](a ) for both of these .
it is apparent that the best achievable result diminishes for @xmath23 but rapidly reaches a plateau of @xmath86 and that both potentials give consistent results .
for @xmath87 electrons , the two choices of potential give very consistent results , suggesting that we have indeed found a likely global optimum .
for @xmath88 electrons , however , the optimization procedure failed to find a maximum of @xmath6 for the general linear potential that approached that of the asymmetric @xmath32-function potential .
we speculate that this is because the number of local maxima increases with @xmath0 ; we shall show that this is true later for the more carefully chosen potentials at least .
because of this , we did not consider the general piecewise linear potential further but display the optimized potential and wavefunctions for @xmath21 and @xmath79 respectively in fig .
[ fig : betaresults](b ) and ( c ) . that these potentials achieve similar values of @xmath6 to the asymmetric @xmath32-function potentials , despite visually appearing very different , supports our conclusion in previous work@xcite that @xmath6 is poorly determined in potential space with many irrelevant directions . due to the small parameter space of the asymmetric @xmath32-function potential , it is possible to directly visualize the objective function ;
this is displayed in fig .
[ fig : betaresults](d ) for @xmath22 or 2 electrons ( the magnitude is scaled by @xmath89 for the @xmath21 electron case ) , with corresponding optimized potential and wavefunctions shown in fig .
[ fig : betaresults](e ) . the objective function for @xmath79 electrons and the optimized potential and wavefunctions
are shown in fig .
[ fig : betaresults](f ) and ( g ) respectively .
it is immediately apparent that as @xmath0 increases , the objective function acquires additional local extrema . for the @xmath22 case ,
the global optimum found is @xmath90 at @xmath91 and @xmath92 , which is only marginally short of the best known value of @xmath93 found from optimizing many different classes of potential @xcite .
the global maximum lies at the top of the long , narrow ridge viewed in fig .
[ fig : betaresults](d ) , a feature of the objective function that was also seen in earlier studies of more complicated potentials @xcite .
its presence implies that @xmath6 is much less sensitive to one of the parameters than the other , which can be quantified by computing the hessian matrix of @xmath6 with respect to the parameters , @xmath94 and finding the eigenvalues and eigenvectors .
these quantities respectively measure the curvature and principal directions of the objective function around the maximum .
we previously used this technique in @xcite to show that that while the best known value of @xmath93 was obtained by optimizing a piecewise linear potential with 6 free parameters , in fact @xmath6 was effectively only sensitive to 2 - 3 parameters at the optimum . here , the eigenvalues of the hessian evaluated at the global maximum of @xmath6 are @xmath95 and @xmath96 and the associated eigenvectors are @xmath97 and @xmath98 .
hence , @xmath6 is very sensitive to the value of @xmath10 , the second parameter and much less sensitive to the value of @xmath27 ; since the eigenvectors are nearly parallel to basis vectors @xmath99 and @xmath100 in parameter spaceas is visible from the orientation of the ridge in fig .
[ fig : betaresults](d)it is clear that @xmath6 is sensitive to these features of the potential specifically , and not some combination of them .
in contrast , the eigenvectors in @xcite were not well aligned with the parameter space and so it was not possible to ascribe high @xmath6 to particular features of the potential .
the design advice from this study is much clearer : to optimize @xmath6 , create an asymmetric potential well with a steep wall on one side , i.e. set @xmath101 ; then add an attractive group in the center and tune the strength of attraction , i.e. carefully adjust @xmath10 as this largely determines @xmath6 .
a similar analysis was applied to the multi - electron case . in table
[ tab : beta ] , we display the three extrema with largest @xmath6 for @xmath79 , together with a plot of the potential and wavefunctions ; parameter values of @xmath27 and @xmath32 at the optimum and the results of the eigenanalysis . unlike the @xmath22 case , the global optimum has a repulsive @xmath32-function ; the next two solutions have attractive @xmath32-functions .
the existence of both attractive and repulsive extrema supports the paradigm proposed by lytel _
_ @xcite in their work on optimization of quantum graphs .
they suggest that large @xmath6 is achieved by introducing a disturbance at some point in the @xmath102-electron chain of molecule , e.g the addition of a side group .
the disturbance then induces a phase shift in the wavefunction , producing a change in dipole moments sufficient to achieve large @xmath6 . in our work
, the @xmath32-function serves to provide the disturbance ; the insight of lytel _ et al .
_ is that it is the overall phase shift at the disturbance that is the relevant parameter , not its detailed nature .
hence both attractive and repulsive features can provide an appropriate phase shift .
just as for the @xmath22 case above , eigenanalysis of the hessian for the multi - electron case shows that the eigenvectors remain well aligned with the parameter basis vectors for the multi - electron case .
surprisingly , while @xmath10 was found to be the most important parameter for @xmath22 , it is @xmath27 that appears to be most significant for the @xmath79 case .
the ratio of eigenvalues is also less extreme , around @xmath103 rather than @xmath104 as before .
these results suggest that when applying the design approach herein proposed , i.e. an asymmetric well with a phase shift - inducing feature , to real systems , tuning both parameters may be important to achieve high @xmath6 .
we also performed eigenanalysis for the more general linear potential ; as a prototypical example for @xmath79 electrons the eigenvalues were @xmath105 indicating that only one parameter is important .
interestingly , with increasing @xmath0 the lowest @xmath106 eigenvalues remained roughly constant while the largest eigenvalue strongly increased : for @xmath21 ; the principal eigenvalue was found to be @xmath107 , for @xmath108 it was @xmath109 and for @xmath110 it was @xmath111 .
this progression is interesting because it suggests that for large @xmath0 , the problem in some sense becomes simpler as the important parameter dominates the others to an ever increasing extent .
unfortunately , as with previous work@xcite , the eigenvectors are not clearly aligned with the parameter space , so the general linear potential provides less useful information than the highly constrained asymmetric @xmath32-function potential .
we also display for each potential a visualization of the first few position matrix elements @xmath112 .
these are important because , as discussed more fully in the appendix below , the hyperpolarizabilities can be expressed as a sum over states involving @xmath80 as well as the energy - level differences @xmath113 .
it is therefore natural to examine this matrix to determine which transitions contribute most to the hyperpolarizabilities .
the interpretation of this matrix is , however , complicated by the fact that many combinations of these parameters are individually irrelevant to the hyperpolarizabilities . as is well - known
, for example , the three state model@xcite achieves the bounds quoted in equation ( [ eq : kuyzklimit ] ) , and only requires two parameters @xmath114 and @xmath115 with @xmath116 . the @xmath80 matrices are dominated by the tridiagonal terms , and the diagonal elements can be eliminated from the expressions for the hyperpolarizabilities , e.g. by using the dipole - free sum over states ( dfsos ) formula . to aid inspection ,
we have therefore omitted the diagonal terms and plotted the first off - diagonal terms , i.e. those with @xmath117 , in greyscale .
the remaining terms are plotted in a scheme where intensity corresponds to magnitude and red or blue refers to the positive or negative sign of the term respectively . reflecting the potential in space and changing the signs of odd - indexed wavefuctions changes such plots only by changing the signs , and for @xmath118 the colors of @xmath80 for @xmath119 even . to avoid confusion ,
we have chosen the potential or its mirror image in such a way that @xmath80 is positive for @xmath53 the homo and @xmath41 the lumo+1 .
the off - tridiagonal terms are important to @xmath6 , because a tridiagonal matrix would yield @xmath120 as for the harmonic oscillator .
clearly , however , for all the local optima displayed in table [ tab : beta ] , the @xmath121 terms are much larger than the remaining @xmath122 terms . for each local maximum in table [
tab : beta ] , we display calculated values of the @xmath66 and @xmath81 parameters .
for the three state model , these parameters yield optimal @xmath123 for values of @xmath124 and @xmath125 .
however , past studies@xcite of optimized potential functions for @xmath22 electrons find values of @xmath126 and @xmath127 regardless of the starting potential .
optimization of quantum graphs@xcite produces mildly different values of @xmath128 and @xmath129 . for the optima presented here
, we also find @xmath130 , but the results seem to favor a value of @xmath131 . this result is consistent with the results of eigenanalysis of the hessian , which suggests only one of the parameters @xmath66 and @xmath81 can be important .
the dipole matrix plots , together with the value of @xmath81 or @xmath132 allows us to appreciate the compromises made in this optimization .
all contributions to @xmath11 in the dipole - free sos formula involve three states , the product of the transition moments between them and a function of the energies . from the x - matrix plots it is clear that matrix elements become smaller very quickly moving away from the diagonal .
thus the sos for @xmath11 is expected to be dominated by contributions from terms that involve only one off - tridiagonal element .
there are exactly two sets of three states involving only the one off - tridiagonal matrix element : the ground state , the excited state in which one electron has been excited from the homo to the lumo and one of two doubly excited states : the state in which one electron has been excited from the homo to the lumo+1 or that in which one electron has been excited from the homo-1 to the lumo . as the homo to lumo+2 and homo-1 to lumo transition matrix elements have opposite signs , these contributions have opposite signs .
generically the transition moments are larger between higher energy states to the homo to lumo+1 and so are expected not to dominate .
the smaller result for many electrons seems , at least in part , to be explained by the negative contribution of the homo-1 to lumo contribution .
this contribution is lacking for 1 or two electrons when there is no homo-1 .
other than this , the pattern of matrix elements looks substantially similar for all the maximaincluding ones we have not displayed .
thus it seems that the smaller results for @xmath6 for more than two electrons may be explained by the need to include four states rather than two , given the near degeneracy of the two doubly excited states .
( a ) @xmath25 versus @xmath32 function strength @xmath10 for varying electron number @xmath0 in the symmetric @xmath32 function potential .
each time @xmath0 is increased by 2 , the sign of @xmath10 for which the maximum or minimum occurs changes sign .
( b ) the energy level structure of the symmetric @xmath32 function potential as a function of @xmath32 function strength @xmath10 .
, width=288 ] the symmetric triangular well with a @xmath32-function , i.e. eq . ( [ eq : potentialdelta ] ) with @xmath38 , is known to achieve near - optimal results for @xmath7 for @xmath22 despite only containing one free parameter . due to the simplicity of this potential , it is possible to directly visualize @xmath7 as a function of this parameter , the strength of the delta function @xmath10 , for different values of @xmath0 .
the results are displayed in fig .
[ fig : gammadelta](a ) . immediately apparent
is an alternation of the sign of the @xmath7 curves as @xmath0 is increased by 2 : for @xmath133 the maximum value of @xmath7 occurs for positive @xmath10corresponding to a @xmath32-function wall in the middle of the well , while for @xmath134 the maximum value of @xmath7 occurs for negative @xmath10 .
the magnitude of @xmath10 for which the maxima occur increases only slightly with @xmath0 , however .
the behavior of the minimum value of @xmath7 is similar , but shifts even less .
some insight into these results is obtained by examining the effect of the @xmath32-function on the single electron energy level spectrum for the potential , shown in fig .
[ fig : gammadelta](b ) .
clearly , only the even wavefunctions are affected by the @xmath32-function due to the @xmath36-symmetry of the potential . while the placement of @xmath135 can be freely adjusted by changing @xmath10 ,
the higher even energy levels are bounded by their invariant odd neighbors , e.g. @xmath136 .
the freedom of selecting @xmath135 relative to @xmath137 allows the @xmath21 case to achieve @xmath12 a high fraction of the kuzyk maximum , while the restricted higher levels only permit a lower value of @xmath7 to be achieved .
the reason for the alternation is also apparent . for @xmath21 , the highest occupied molecular orbital ( homo ) in the ground state
is @xmath135 while the lowest unoccupied molecular orbital ( lumo ) is @xmath137 ; increasing @xmath10 serves to _ widen _ the homo - lumo gap . on the other hand , for @xmath108
the homo is @xmath137 and the lumo is @xmath138 ; increasing @xmath10 for this case serves to narrow the homo - lumo gap .
this trend continues with @xmath10 widening the homo - lumo gap for @xmath133 and narrowing the homo - lumo gap for @xmath139 ; the alternating effect of @xmath10 explains the different signs of @xmath7 with @xmath0 .
maximum ( a ) and minimum ( b ) @xmath25 achieved for varying increasing numbers of electrons ; results for the symmetric linear piecewise potential are shown in red and for the symmetric @xmath32 function potential in black .
( c ) for @xmath79 , the objective function @xmath25 plotted versus the linear piecewise potential shape parameters @xmath29 and @xmath28 for positive @xmath140 and ( d ) the best potential and wavefunctions obtained .
( e ) results for negative @xmath140 and ( f ) the corresponding potential and wavefunctions .
( g ) results for the @xmath32 function potential shape parameters @xmath10 and @xmath29 and ( h ) optimized wavefunctions and potential . ] to determine whether these results are universal , we also optimized @xmath7 for the asymmetric @xmath32-function potential eq .
( [ eq : potentialdelta ] ) as well as the linear piecewise potential given by eq .
( [ eq : potentiallinear ] ) . shown in fig .
[ fig : gammaall](a ) and ( b ) are the best maximum and minimum @xmath7 obtained as a function of @xmath0 .
although there are small differences between results obtained with different potentials , the same trend is clear : that the best @xmath7 falls off with increasing @xmath0 but rapidly reaches a constant value , yielding an apparent feasible range of @xmath141 .
these apparent bounds are shared by all three potentials .
we plot the objective function results for @xmath79 for several different scenarios : for the linear piecewise potential , the sign of @xmath140 must be chosen to be positive or negative prior to optimization .
the objective function is shown for @xmath142 in fig .
[ fig : gammaall](c ) revealing several local maxima ; the corresponding optimized potential , which maximizes @xmath7 , and wavefunctions are shown in fig . [ fig : gammaall](d ) . if alternatively , @xmath143 , a minimum of @xmath7 is obtained ; the objective function is rather simpler as apparent in fig .
[ fig : gammaall](e ) and the corresponding potential and wavefunctions are shown in fig .
[ fig : gammaall](f ) .
note that despite the arbitrariness of the linear potential , the optimized potentials strongly resemble the symmetric @xmath32-function potential , validating its use as _ ansatz _ earlier .
> m0.5in > m1in > m0.5in > m0.5in > m0.75in > m0.75in > m1in > m0.5in > m0.75 in value of @xmath7 & potential and wavefunctions & @xmath27 & @xmath10 & hessian eigenvalues & hessian eigenvectors & @xmath80 & @xmath81 & @xmath66 + 0.161 ( 100% ) & & 1.000 & -2.963 & -0.597 , -0.048 & @xmath144 & & 0.615 & 0.765 + 0.134 ( 83% ) & & 1.375 & 2.534 & -1.926 , -0.039 & @xmath145 & & 0.516 & 0.638 + 0.127 ( 79% ) & & 2.043 & -3.765 & -1.597 , -0.039 & @xmath146 & & 0.511 & 0.638 + -0.0613 ( 100% ) & & 1.630 & -2.107 & 6.970 , 0.0187 & @xmath147 & & 0.655 & 0.500 + -0.0609 ( 99% ) & & 1.000 & 1.524 & 13.78 , 0.0311 & @xmath148 & & 0.654 & 0.500 + -0.0502 ( 82% ) & & 3.260 & 2.467 & 1.162 , 0.0123 , & @xmath149 & & 0.639 & 0.500 + the asymmetric @xmath32-function potential has a much more complicated landscape for @xmath7 as is evident in fig .
[ fig : gammaall](g ) .
many local minima and maxima exist .
the best three for both positive and negative @xmath12 are displayed in table [ tab : asymmetricgamma ] , together with the parameter values and results of the hessian eigenanalysis .
many of the results above for @xmath11 are also seen for @xmath12 : both attractive and repulsive @xmath32-functions lead to large @xmath12 , in agreement with the phase interruption paradigm .
eigenvalues of the hessian differ by a similar ratio of @xmath150 and again the eigenvectors are well aligned with the basis vectors of the parameter space . just as for the first hyperpolarizability with @xmath79 electrons
, it appears that @xmath27 is the most important parameter .
intriguingly , the best negative @xmath7 obtained does not possess @xmath36 symmetry in sharp contrast to the @xmath22 case , although a @xmath36-symmetric solution within 1% of this value also exists .
it is possible this is due to the freedom of choosing the relative position of the lowest energy level @xmath135 relative to @xmath137 ; less freedom exists for the higher single particle energy levels .
position matrix elements @xmath80 are shown in [ tab : asymmetricgamma ] for each potential .
note that these matrices for @xmath36 symmetric potentials have a characteristic checkerboard structure .
elements immediately off the diagonal @xmath117 are plotted on a greyscale while non - tridiagonal elements are plotted in red and blue and it is evident , as for @xmath6 , that the matrices are diagonally dominant . for positive @xmath7 ,
the @xmath36-symmetric global optimum has large terms on the @xmath151 off diagonal , which is consistent with a few - state hypothesis ; here the homo@xmath152lumo+2 and homo-1@xmath152lumo+1 terms make the strongest contribution .
the transition matrices for the non @xmath36-symmetric solutions , however , seem to have little in common with the global optimum , which is perhaps to be expected since the values of @xmath7 for these secondary maxima fall some @xmath153 below it . for negative @xmath7 ,
where the two best solutions have nearly identical values , the transition matrix for the @xmath36-symmetric optimum resembles the solution for positive @xmath7 , i.e. homo@xmath152lumo+2 and homo-1@xmath152lumo+1 are the dominant off - tridiagonal terms .
the non @xmath36-symmetric solution has an interesting structure : the homo@xmath152lumo element is large , while the homo@xmath152lumo+1 and homo-1@xmath152lumo elements are significantly weaker than the other @xmath121 elements .
a similar pattern , where off - tridiagonal elements are suppressed around the fermi surface , is also seen in the third highest optimum that is also non @xmath36-symmetric .
it is for negative @xmath7 , therefore , that we find the clearest correspondence with the transition matrix elements of the three - state ansatz .
we now connect these results to the sum - over states formula for @xmath12 , which includes two types of terms : there are both terms involving the ground state and three other states , @xmath41 , @xmath53 , and @xmath154 , energy factors and the matrix element product @xmath155 or @xmath156 which we shall hereafter call `` four state '' terms .
there are also terms involving the ground state and two excited states , @xmath53 , @xmath41 , referred to hence as `` three state '' terms .
a three state term proportional to @xmath157 makes a positive contribution to @xmath12 , one proportional to @xmath158 will make a negative contribution .
this is more easily seen from eq .
( 73 ) in @xcite , if @xmath41 is considered the unconstrained summation variable , @xmath53 and @xmath154 are ( as before ) both considered to be different from @xmath41 , and the energy factors are made symmetric in @xmath53 and @xmath154 .
moreover , only three state terms involving at least one off - tri - diagonal matrix element can contribute .
if there are no off - tri - diagonal matrix elements , the matrix elements and energies are constrained to be those for the harmonic oscillator , for which @xmath159 .
the largest terms in @xmath12 from four - state terms involves three excited states , three - tridiagonal matrix elements and one matrix element for which @xmath151 .
the largest contributions to @xmath12 from three state terms involves the square of a tri - diagonal matrix element and the square of an @xmath160 matrix element .
these last matrix elements , and so also these terms , are forbidden by @xmath36 symmetry .
now looking at the @xmath161 matrices , we see that the the @xmath36-symmetric potentials have three large contributions , with the signs of these three contributions oscillating with period 2 in the empty state in the highest or lowest singly occupied state .
the resultant cancellations may again partly explain the smaller @xmath7 achieved for these potentials .
the difference between positive and negative @xmath12 is not the kinds of terms involved ( as suggested in @xcite ) but rather in the signs of the off - tridiagonal matrix elements .
the three - state terms that are allowed for non @xmath36-symmetric potentials are more similar to those considered in @xcite , except that these can have either sign . to facilitate comparisons with other work
, we again computed values of the kuzyk three - state model parameters @xmath66 and @xmath81 . for the lower bound
, all three extrema favor @xmath162 and @xmath163 .
an earlier study by watkins _ _ et al.__@xcite , who optimized @xmath7 for different potentials and @xmath22 , found for negative @xmath7 @xmath164 and two clusters of solutions with @xmath165 or @xmath166 .
our results are therefore in excellent agreement . for the upper bound , our results display more variation with @xmath167 and @xmath168 for the best potential found .
et al . _ similarly found more variation in positive @xmath7 , obtaining results around @xmath169 with @xmath170 . as for @xmath11 ,
the discrepancy is explicable because eigenanalysis reveals that effectively only one parameter is truly important , seemingly the slope of the left hand boundary .
objective functions ( a ) @xmath6 and ( b ) @xmath7 for the asymmetric @xmath32-function potential shown in the natural parameter space @xmath27 and @xmath10 ; wavefunction parameters at the classical turning points ( c ) @xmath171 and ( d ) @xmath172 plotted with respect to @xmath27 and @xmath10 .
a selected region of the objective function , enclosed in the grey volume in ( a ) and ( b ) , is show replotted with respect to @xmath173 and @xmath174 in ( e ) and ( f ) respectively . ]
the above subsections have presented optimized potentials for @xmath6 and @xmath7 with increasing @xmath0 , expressed in a parameter space that _ a posteriori _ is found to coincide with what is important to the hyperpolarizabilities .
as discussed , this is a significant development over previous work , where the parameter space was larger the optimized potentials gave a less clear sense of which features are important . nonetheless , the parameters @xmath27 and @xmath10 are an artifact of the parameterization chosen , and it is therefore desirable to identify quantities that are invariant under reparametrization and rescaling of the form in eq .
( [ eq : rescaling ] ) .
this is particularly important because , if we generalize to the space of arbitrary potentials , there are many other solutions with identical values of the hyperpolarizability . for the @xmath22 case , we know that the hyperpolarizabilities can be completely reconstructed from the ground state wavefunction@xcite . extending this to the @xmath1 case , it is easy to see that the hyperpolarizablities can be constructed for this problem as appropriate integrals of the occupied single particle wavefunctions alone .
as the largest term is expected to the homo , the important parameters ought to be expressible in terms of parameters crucial to the homo .
moreover , there is a need to connect the dipole - free sum - over - states view , which prescribes certain values of the dipole transition matrix elements , and the potential view that we pursued above .
we therefore performed an approximate analysis of the problem , inserting wkb ansatz wavefunctions into expressions for @xmath175 and @xmath80 to determine what details of these wavefunctions most significantly affect these parameters .
the full calculations presented in the appendix below support this argument and suggest that the three parameters introduced in the next paragraph are the crucial ones .. on the basis of approximate analytic arguments made in the appendix , and a numerical experiment that follows , we believe that there are three length scales that are important to the hyperpolarizabilities around the maximum .
one is @xmath176 , the separation between the turning points of the homo .
the other two are length scales @xmath177 for @xmath178 and an element of @xmath179 that characterizes the rate at which the wavefunction varies near the right and left turning points of the homo .
a variety of nearly equivalent forms can be given for these lengths and dimensionless ratios containing them , including @xmath180 , and more global formulae related the derivatives of the wavefunction at the turning points or to the normalization of the homo .
from these three lengths , it is possible to construct two dimensionless parameters that are arguably important to the hyperpolarizability , for example , @xmath181 are invariant under reparameterization and rescaling . here
@xmath176 is the distance between the turning points , @xmath182 and @xmath183 are the slopes of the wavefunctions where the subscripts @xmath154 and @xmath184 refer to the left and right turning point respectively .
as some of the definitions of @xmath185 are zero when @xmath186 is infinite at the turning points , and it is still possible for the hyperpolarizabilities to approach the maximum in this case , appropriate dimensionless combinations should not be infinite in this limit .
as @xmath6 requires asymmetry in the potential , it is expected that a parameter odd under @xmath36 symmetry , such as @xmath173 , mostly controls @xmath11 while one even under @xmath36 symmetry , such as @xmath174 , mostly controls @xmath7 . in fig .
[ fig : reparam ] we show the objective functions ( a ) @xmath6 and ( b ) @xmath7 plotted in the parameter space of the asymmetric @xmath32-function potential , i.e. @xmath27 and @xmath10 . in the same parameter space ,
we show in [ fig : reparam](c ) and ( d ) the values of the wavefunction parameters @xmath173 and @xmath174 from eq .
( [ eq : newparams ] ) . from these plots , we see that much of the structure in the objective functions @xmath6 and @xmath7 is attributable to these new parameters : notice that the position of the ridges in [ fig : reparam](c ) corresponds to the ridges in [ fig : reparam](a ) , while those in [ fig : reparam](d ) corresponds , more roughly , those in [ fig : reparam](b ) . in figs . [
fig : reparam](e ) and ( f ) , the portion of the objective function in figs .
[ fig : reparam](a ) and ( b ) that is enclosed within the grey cuboid is reprojected into the new ( @xmath173 , @xmath174 ) parameter space .
plots of the objective functionsnot shown herein @xmath187 space for the full range of @xmath188 from fig .
[ fig : reparam](a ) and ( b ) closely resemble the structure observed in this reduced reduced region . from these results ,
we conclude that @xmath6 is largely determined by @xmath173 while @xmath7 is largely determined by @xmath174 , and the remaining parameter in each case must be tuned less precisely to achieve the optimum .
it is also now clear that the asymmetric @xmath32-function performs so well due to a fortuitous correspondence : the @xmath27 parameter directly controls the ratio of the slope of the potentials at the turning points , which is readily related to the slope of the wavefunctions at the classical turning points , i.e @xmath27 and @xmath173 are simply related .
.[tab : dimensionlessparams]values of dimensionless parameters @xmath173 and @xmath174 constructed from the homo wavefunction for each of the optimized @xmath79 solutions displayed in tables [ tab : beta ] and [ tab : asymmetricgamma ] [ cols="^,^,^,^",options="header " , ] we display in table [ tab : dimensionlessparams ] the values of @xmath173 and @xmath174 calculated for each of the optima of @xmath6 and @xmath7 shown in table [ tab : beta ] and [ tab : asymmetricgamma ] . for each of the optimization problems ,
i.e. @xmath189 , @xmath190 and @xmath191 the values of at least one of these parameters are internally quite consistent with each other .
this is particularly so for the negative @xmath7 results where the secondary optima are close to the global optimum .
values of @xmath173 and @xmath174 for the global optimum as a function of @xmath0 are shown in table [ tab : dimensionlessfunctionn ] .
these reveal several trends : first @xmath192 where @xmath193 is the number of nodes in the homo .
otherwise , it is expected that @xmath36 symmetry just changes the sign of @xmath6 and @xmath173 .
also , a sign alternation occurs in @xmath173 for @xmath6 ; @xmath174 seems to increase with @xmath0 for @xmath7 , while @xmath173 is found alternately @xmath194 or some small value .
the _ a posteriori _ consistency of these parameters supports the argument above that these are the `` real '' parameters of the optimization problem .
we have optimized the intrinsic hyperpolarizabilities @xmath6 and @xmath7 for non - interacting multi - electron systems with respect to the shape of several classes of potential : a piecewise linear potential , and an asymmetric triangular well with a @xmath32-function .
the best values obtained for @xmath6 and @xmath7 drop from the @xmath22 case and approach an apparent feasible range of @xmath195 and @xmath141 for @xmath0 larger than around eight electrons .
the asymmetric @xmath32-function potential achieves these bounds and , due to the small number of parameters , and _ a posteriori _ verification that the parameters are indeed relevant , provides a design prototype for synthesis of new chromophores . for @xmath6 , a molecule should have asymmetric walls and possess an attractive or repulsive feature in the middlea main chain functional group or side - chainthat promotes a rapid change in the phase of the wavefunction .
the asymmetry of the boundary and the strength of the attraction or repulsion should then be tuned to achieve high @xmath6 . for @xmath7 , the molecule should be essentially @xmath36-symmetric with a central attractive or repulsive feature that should similarly be tuned . by approximate analysis
, we also determined that the _ ad hoc _ parameters of our potential can be re - expressed in terms of the dimensionless , scale invariant parameters derived from the shape of the homo wavefunction at the classical turning points .
these new parameters are important both because they explain the success of our original parametrization and because they provide a new wavefunction - centered approach to screen potential chromophores for large hyperpolarizabilities .
the results also provide important information on how well the many theoretical studies of single electron systems might describe real molecules that posses multiple electrons
. while the apparent bounds quoted above are more restrictive than the @xmath22 case , it seems , encouragingly , that the overall design paradigms described above apply equally well to both cases .
information from other studies on target values of the @xmath66 and @xmath81 parameters of the three state model for @xmath22 also seems to remain valid with increasing @xmath0 .
the insights of @xcite that only a very small number of parameters are required remain valid , and happily , this appears to be increasingly so with large @xmath0 . because of this and
because the objective function for multiple electrons acquires many more local extrema , it should in principle be _ easier _ to tune multi - electron systems .
in this appendix we use approximate techniques to argue that the dimensionless parameters identified in the main text constructed from the separation between the turning points of the homo and the slopes of the potential at these turning points ought to mostly explain the hyperpolarizability .
the simple , zeroth order argument is given in the next two paragraphs . a more detailed but still very approximate calculation follows . for all potentials with large hyperpolarizabilities ,
the transition matrix elements between the frontier wavefunctionsthose that are close to the fermi levelmust be large .
when these matrix elements are large , there are very strong constraints on the energy differences between the various states , and hence we can regard these parameters are fixed by the matrix elements and we need only calculate the transition matrix elements to predict the hyperpolarizabilities . in order to calculate these transition matrix elements approximately , we note that wavefunctions fall rapidly in the classically forbidden region and oscillate in the classically allowed region .
thus most of the transition matrix integral between two states necessarily comes from the region between or very close to the turning point of the lower energy state .
moreover , far from the turning points , each wavefunction becomes relatively small , in effect because the particles are ( classically ) moving relatively quickly .
thus most of the contribution to these integrals comes from the region close to the turning point of the lower energy wavefunction . from this
it is clear that the distance between the turning points of the homo is an important parameter .
the wavefunction near a turning point is constrained by the slope of the potential at the turning point . moreover , the ratio of the amplitudes of the wavefunction at the two turning points is limited by general principles , described below . while the wavefunction of the higher energy state near the turning points of the lower energy states is more free to vary , provided the potential is not too far from linear , they are still largely constrained by the slope of the potential .
thus the slopes of the potential at the turning points of the homo also seems to be a very important parameter . from these parameters ,
it is possible to construct two dimensionless parameters , and it is known from numerical experiments that only two parameters seem to be important to maximizing hyperpolarizabilities for model potentials . we now provide a more detailed , but nonetheless approximate , analysis .
we begin with the sum - over - states formulae for the hyperpolarizabilities .
for instance , the off - resonant expression for the second hyperpolarizability is , which contains three kinds of quantity : the energy level differences @xmath113 , matrix elements @xmath112 and barred quantities @xmath197 that contain dipole terms , @xmath198 as is well known , the dipole terms can be eliminated from these expressions using the sum rules@xcite , leaving only the transition elements @xmath80 and energy level differences @xmath175 .
numerous previous studies , as well as the results above , have shown that only a few statesin fact 2 - 3near the fermi energy contribute significantly to the hyperpolarizabilities .
hence , the question of what is important to the hyperpolarizability maybe be addressed by understanding how the potential and wavefunctions affect these quantities . in the remainder of the appendix , we will answer this question by constructing wkb ansatz wavefunctions , inserting them into expressions for these quantities and examining the form of the results . in order to proceed
, we make some simplifying assumptions .
first , we assume that the true optimum potential is sufficiently smooth that the wkb approximation yields good approximations , at least near the turning points , to the relevant wavefunctions , i.e. those near the fermi surface .
this is justified by previous studies@xcite that have shown the addition of small rapidly varying perturbations to the potential does nt affect the hyperpolarizabilities .
the ansatz potentials studied in this work , with delta functions and changes in slope at isolated points , all satisfy this criterion .
second , we shall assume that for wavefunctions near the fermi energy there are only two classical turning points , except possibly for isolated delta functions in the potential that may violate this rule .
we justify this because the presence of multiple turning points would result , at least approximately , in roughly independent particles in the separate classically allowed regions .
this would lead to hyperpolarizabilities that grow @xmath199 rather than @xmath200 ( for @xmath11 ) or @xmath201 ( for @xmath12 ) as the kuzyk bounds imply .
the turning points shall be denoted @xmath202 and @xmath203 , corresponding to the left and right turning point respectively , and are found by solving @xmath204 . in this expression , and hereafter , @xmath154 indexes the left or right turning point . within the above assumptions it is possible to write an ansatz wavefunction for the @xmath53th state that is valid except for the region near the classical turning points , @xmath205 , & e > v\\ a_{n}(x)\exp\left[-\gamma_{n}^{l}(x)\right ] , &
e < v \end{cases}\label{eq : psi}\ ] ] where @xmath206 and @xmath207 refer to the classically allowed and forbidden regions respectively .
the functions @xmath208 are defined by , @xmath209 and the function @xmath210 is given by , @xmath211 the remaining functions @xmath212 and @xmath213 are smooth and slowly varying , except where the potential has delta functions or sharp changes ; nonetheless they are necessarily smoother than the potential . for the smooth potentials considered here
, @xmath214 is of order @xmath102 or smaller .
@xmath215 is relatively close to the semiclassical result @xmath216 where @xmath217 is the classical velocity and @xmath218 changes only by amounts that are asymptotically small where the wkb approximation is valid , except very close to the turning points .
the dependence of @xmath218 can easily be made more precise by using the uniform asymptotic wkb approximation to the wavefunction@xcite though this does little to illuminate the discussion . moreover , even if wkb is invalid somewhere between the turning points , the magnitude , @xmath219 is , roughly speaking , the rate at which the electron turns at the turning point and so is expected to be the same at both turning points . in wkb
there is no reflection in classically allowed regions and hence it is obvious that @xmath219 can not change .
more generally , however , any structure between the turning points must have equal magnitude for the transmission and reflection from each side .
this implies that @xmath219 must have the same values at both turning points .
moreover , @xmath218 can be chosen to be real near both turning points , and the ratio of @xmath220 near the two turning points alternates sign as we increase the energy of the wavefunctions .
@xmath227-\delta x_{t , m}^{2}\left[e_{m}-v(\delta x_{m}y+x_{m}^{l}\right]}{\delta x_{n}\left[e_{n}-v(\delta x_{n}y+x_{n}^{l})\right]+\delta x_{t , m}\left[e_{m}-v(\delta x_{m}y+x_{m}^{l})\right]}.\label{eq : wkbnmfinal}\ ] ] the form of the integrand in ( [ eq : wkbnmfinal ] ) is instructive : near the turning points , i.e. as @xmath228 and @xmath229 , the numerator goes to zero linearly in @xmath230 while the denominator goes to zero as a square root ; the integrand therefore vanishes like @xmath231 near the turning points .
hence , the majority contribution to this integral comes from the spatial region far from the classical turning points , particularly if the two energies are similar and the potential near the turning points is slowly varying .
moreover , this integral tends to smooth out small , high frequency variations in the potential ; it follows that if @xmath232 is small , then @xmath233 is largely determined by the form of the potential far from the turning points .
we now turn to the dipole matrix elements @xmath80 , which can be computed using the position formula , @xmath234 we shall restrict our analysis to small values of @xmath232 , because only these states contribute significantly to the hyperpolarizability . moreover , approximate analysis of these integrals is complicated for large @xmath232 because , while small , they are strongly dependent on the detailed analytic behavior of the potential and wavefunctions . .
this integral is then the integral of a function that has @xmath236 nodes , roughly evenly spaced through the interval , and has a somewhat ( algebraically ) larger magnitude near the turning points . from this formula , it is apparent that the transition matrix elements depend primarily on the nature of the wavefunctions near the classical turning points , i.e. separation between the two turning points and as well as the shape of the wavefunctions in their vicinity .
as the shape of the wavefunctions are most dependent on the slopes of the potential at the turning points and there must be a connection between the energy differences and the transition matrix elements , this strongly argues that the separation between the turning points of the homo and the slopes of the potential at these turning points should be among the most important heuristic parameters that determine the hyperpolarizabilities .
18ifxundefined [ 1 ] ifx#1 ifnum [ 1 ] # 1firstoftwo secondoftwo ifx [ 1 ] # 1firstoftwo secondoftwo `` `` # 1''''@noop [ 0]secondoftwosanitize@url [ 0 ] + 12$12
& 12#1212_12%12@startlink[1]@endlink[0]@bib@innerbibempty @noop * * , ( ) @noop `` , '' ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) link:\doibase 10.1103/physreva.87.043824 [ * * , ( ) ] @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) link:\doibase 10.1364/ol.32.000942 [ * * , ( ) ] @noop * * , ( ) @noop * * , ( ) @noop _ _ , advanced mathematical methods for scientists and engineers ( , ) | we optimize the first and second intrinsic hyperpolarizabilities for a 1d piecewise linear potential dressed with dirac delta functions for @xmath0 non - interacting electrons .
the optimized values fall rapidly for @xmath1 , but approach constant values of @xmath2 , @xmath3 and @xmath4 above @xmath5 .
these apparent bounds are achieved with only 2 parameters with more general potentials achieving no better value .
in contrast to previous studies , analysis of the hessian matrices of @xmath6 and @xmath7 taken with respect to these parameters shows that the eigenvectors are well aligned with the basis vectors of the parameter space , indicating that the parametrization was well - chosen .
the physical significance of the important parameters is also discussed . |
modern physics , specifically , quantum mechanics ( qm ) and special relativity ( sr ) , have brought three notions into common currency , namely : `` non - locality , '' `` asymmetric aging '' and `` advanced interaction , '' that defy accommodation with classical physics , common sense and , we hold , basic logic . how this came about
historically is well documented in the literature and so will not be belabored here .
although asymmetric aging , a.k.a . the `` twin paradox , '' has been a foil for ` nondomesticated ' newcomers and numerous outsiders for 90 years , all of these notions nowadays enjoy solid ensconcement in the corpus of `` verified physics . ''
one reason for this appears to be their appeal as harbingers from the preternatural ; or as a foot - in - the - door leading out from the stolid , unromantic material world to a great mystical beyond .
another reason is that too little effort has been devoted to seeking alternatives out of respect for the profound successes of both qm and sr .
it is this last deficit , we aim to attack .
the case _ for _ instantaneous interaction is based on two arguments .
one was originated by bell with his now famous derivation of inequalities.@xcite the other exploits arguments pertaining to the algebraic properties of spin operators .
the later argument has a rather more complicated pedigree , but the key ideas are most often credited to kochen and specker.@xcite@xcite bell s main argument , and the drift of all of his analysis , is in the form of a _ reductio ad absurdum .
_ that is , he argues that by assuming that a fully local , realistic extention of qm with hidden variables exists , then certain inequalities derived on the assumption of locality ( as he encodes it ) constrain the coincident probabilities for the einstein , podolsky and rosen ( epr ) gedanken experiment in a way that can be tested .
in fact , these inequalities are violated in experiments , so that one arrives at a empirical contradiction with bell s assumptions.@xcite the case _ against _ such instantaneous interaction , on the other hand , stands on three legs .
the first consists of detailed analysis of the logical structure of the basic hypotheses supporting the existence of non - locality .
such arguments turn mainly on minutia pertaining to the definitions of statistical statements and their application to epr gedanken - type experiments .
the second leg goes straight to constructing counterexamples .
if a valid , classical model for epr experiments can be found in direct conflict with the conclusions of bell s analysis then obviously the hypotheses going into this analysis must be wrong even when the exact error remains unidentified .
such models can consist of either calculations of plausible setups , or `` monte carlo '' simulations of such setups .
the final leg consists of experiments in non traditional regimes .
optical experiments based on epr considerations have most often been done in the visible part of the spectrum where `` photon '' phenomena are most evident .
but , because the reasoning behind the bohm variation of epr setups is not dependant on the wave length , these same setups can be executed in other parts of the em spectrum in which the peculiarities of `` photodetectors '' and therefore `` photons '' are irrelevant . to the best of this writer s information ,
the fundamental error in bell s reasoning was first identified by jaynes.@xcite ( there are , in addition , those who claim that the point made by jaynes was used already before bell s seminal paper . in other words , had bell focused on such work in connection with his analysis of epr correlations , he never would have authored his now famous paper . )
subsequent to jaynes , the kernel of his observation was rediscovered independently in various styles at least three times.@xcite-@xcite the core of jaynes point is that bell misapplied bayes formula .
the marginal probability distribution of a coincident probability dependant on three variables takes the form @xmath0 by basic probability theory , the integrand of this equation is to be decomposed in terms of individual detections in each arm according to bayes formula @xmath1 where @xmath2 is a conditional probability . in turn , the integrand of eq .
( [ coincidence ] ) can be converted for use in integrand of bell s _ ansatz _ in which he considers the correlation of epr variables : @xmath3 iff @xmath4 ( in bell s notation , @xmath5(a[b],\ : \lambda ) = a[b]p(a[b],\ : \lambda ) $ ] where all @xmath6 s are probabilities corresponding to the moduli of wave functions ; in accord with current practice , even better notation would be : @xmath7 . )
bell argued , in jaynes terms , that strict locality implies that the dependence of @xmath8 on @xmath9 , implies a _ causative _ relationship between the measuring stations .
this is clearly not the case for bayes formula as the correlation can just as well arise from a _ common cause _ at a point in the intersection of the past light cones of both detectors.@xcite thus , we conclude that bell s factorization , is just not tenable . bell s analysis would have been standard statistical analysis were is possible to introduce a @xmath10-meter .
however , if @xmath10 is a ` hidden variable ' intrinsically unavailable for observation but whose existence is to be inferred by observing the patterns in the values revealed by @xmath9- and @xmath11-meters , then whatever @xmath10 encodes must be evident in those values , but bell s encoding of locality precludes this by inadvertent assumption .
although this argument against bell s factorization is clean and indisputable in the optical case ( actually for any em phenomena ) , it is unfair to the total challenge faced by bell or any physicist with an eye to the consistency of all of physics .
their problem here is this : particle beams are seen to `` navigate '' as if they had a wave character but register at detectors as if they are composed of collections of ` particles . '
that is , the dualistic behavior of particle beams implies that in some concrete sense that the beams are ontologically ambiguous while they are underway and resolved only at measurement , in other words , their wave packet is collapsed .
the particle character of waves , on the other hand , can be attributed to the conversion of the continuous energy stream of the beam into a digitized photocurrent comprised of electrons .
thus , this ambiguity is not necessary for radiation beams but seems to have been imposed anyway for the sake of overall uniformity .
the final point of these considerations is , that without a means of rationalizing beam behavior , for the particle beam equivalent of coincident probabilities considered by bell for optical , i.e. , radiation beams , the implied causality relationship must be respected .
this is so , because when one arm of such a wave function is ` collapsed ' by measurement , then its partner also collapses at the same instant . in other words , for particle beams
, a wave function should not be just an epistemological _ aid dmemoire _ but somehow also substance .
therefore , jaynes argument is incomplete without an accompanying classical model for particle beam wave - like behavior .
in fact , such exists@xcite ; so his argument stands .
the earliest local realistic model for epr correlations known to this writer is barut s model for the original spin variant of bohm s rendition of the epr setup.@xcite some other attempts were mislead by the presumed correctness of bell s arguments and tried simultaneously to satisfy both classical physics and bell s inequalities.@xcite@xcite these latter models can be considered rigorous technical counterexamples to bell s reasoning but they still fail to be convincing because they also incorporate features that are at odds with some other seemingly necessary characteristic .
a systematic study providing models for various epr and higher order ( i.e. , ghz ) experiments is only recently published.@xcite herein only the most basic variant is considered as an illustration . to model the prototypical epr experiment with ` entangled ' polarization states , the source is assumed to emit a double signal for which individual signal components are anticorrelated and , because of the fixed orientation of the excitation , confined to orthogonal polarization modes ; i.e. @xmath12 where @xmath13 takes on the values @xmath14 and @xmath15 with an even , random distribution .
the transition matrix for a polarizer is given by , @xmath16 , \ ] ] so the fields entering the photodetectors are given by : @xmath17 coincidence detections among @xmath18 photodetectors ( here @xmath19 ) are proportional to the single time , multiple location second order cross correlation , i.e. : @xmath20 the final result of the above is : @xmath21 this is immediately recognized as the so - called ` quantum ' result .
( of course , it is also malus law , thereby being in total accord with one premise of this report . ) likewise , epr correlations can be simulated fully on the basis of classical physics.@xcite the `` quantum '' character of epr experiments resides in the peculiarities of `` photons . '' with respect to these experiments , however , the view point can be taken that photons are just a means of accounting for the fact that continuous radiation is converted in a photodetector to a digitized photocurrent .
thus , where epr experiments are done in a part of the spectrum in which it is possible to track the time evolution of an electric field , then the photocurrent can be raised to such a high intensity that it can be regarded as a continuous entity like radiation .
the epr correlations then become simply those among current intensities .
such experiments have been done and the results are in full conformity with the so - called `` quantum '' results .
perhaps the first experiment of this nature was done by evdokimov et al . with radar gear.@xcite recently , a four - fold `` ghz '' experiment using hetrodyning techniques has also been done , again giving results in full conformity with those from qm and in direct contradiction with the conclusion of bell s `` theorem.''@xcite in short , these experiments provide purely classical examples of the origin of epr correlations .
in classical mechanics , the 3-d vector position is the dependant variable while time is an independent parameter . likewise , in special relativistic mechanics , the 4-d vector of space - time ` location ' is the dependant variable and proper time is the independent parameter , _ so long as : _ a single particle in a field is under consideration .
when two mutually interacting particles are taken into consideration , this structure seems to break down because it is held that the proper time intervals on separate world lines between two crossings are unequal.@xcite the most renown illustration of this situation is known as : `` the twin paradox.''@xcite it is , however , the contention herein that this situation is the result of error .
when this error is corrected , asymmetric aging is seen not to occur.@xcite the cause of the error is found in a nonintuitive property of the lorentz transformation : it induces nonuniform scale changes . although this latter fact is well known , its effect on what can be called `` space - time '' perspective , is still oft ill understood and misapplied
. customarily analysis of the twin paradox has not carefully taken into account the determination of the distance to the turn - around point ( which for brevity , we denote the ` pylon ' ) of the traveling twin .
this distance is not really a vector on a minkowski diagram but rather the space - like separation of two entire world lines , namely those of the terminus and pylon of the trip .
the pylon , that is , its place in the world , is not an event but a location , in other words , a worldline .
the turn - around itself is , of course , an event in the usual meaning of that word for special relativity . for the traveling twin , however , the turn - around event _ per se _ is a secondary matter as far as his navigational needs are concerned .
his primary concern is to travel to the designated point in space , regardless of the time taken , before reversing course .
how can he do this ?
first , he and his stay - at - home sibling would chart a course before the beginning of the trip ; that is , they would select an object in the world , a star say , and designate it as the pylon . from standard references
they find that this star is located in a particular direction at a distance @xmath22 this distance is not the length of a unique lorentz vector but the proper length of the displacement from the home location of the twins ; i.e. , the length of all space - like lorentz vectors connecting these two world lines .
for parallel world lines , this value is invariant starting from any arbitrary point on either world line . with this in hand ,
the traveling twin then determines the speed capabilites of his craft and calculates the anticipated arrival time at the pylon .
the distance to the pylon star is not an apparent distance , the length of a moving rod as seen from a second frame , for example , but the proper length to the whole world line of the selected star .
such a length is a scaler and a lorentz invariant .
the location of the world line of the pylon on a minkowski diagram depends on the axis to which it refers .
that is , this world line with respect to the stationary twin passes through the space coordinate at ` @xmath23 ' on the his abscissa .
likewise , this world line must pass through the traveler s abscissa also ; but , because of the difference in the scale of the traveler s axis , this same world line , although still parallel to the stay - at - home s world line , will not be congruent to the line referred to the stay - at - home s axis but is displaced by the scale factor .
( it is this displacement that has been overlooked in previous analysis and which distinguishes this approach . )
the consequence of this displacement is that , the intersection of the traveler s world line with the world line of the turn -around point is also further out on the traveler s world line ; i.e. , the proper time taken to reach the pylon is seen to be greater than heretofore estimated .
in fact , it is equal to the proper time of the stay - at - home as he himself computes it for the time taken by the traveler to reach the pylon . thus , when the whole trip is completed , both twins agree that they have experienced equal portions of proper time since the start of the trip ; i.e. , their internal clocks , ages , are equal
. their reports to each other via light signals on the passage of time , in the usual way do not agree , however .
the final consequence of these considerations is that , contrary to oft expressed opinion , proper time can serve in a self consistent way as the independent variable for relativistic mechanics .
these points can be depicted as follows on a minkowski chart .
( see figure 1 ) [ ] [ ] [ 0.8]@xmath24 [ ] [ ] [ 0.8]@xmath25 [ ] [ ] [ 0.8]@xmath26 [ ] [ ] [ 0.8]@xmath27 [ ] [ ] [ 0.8]@xmath28 [ ] [ ] [ 0.80]@xmath29 [ ] [ ] [ 0.80]@xmath30 the same conclusion can be won also as follows : let @xmath31 be the minkowski configuration four - vector with components @xmath32 of the @xmath33-th particle .
let @xmath34 be a differential displacement along the @xmath33-th particle s orbit ; i.e. , a differential of arc length .
two such differentials tangent to arbitrary points @xmath35 and @xmath36 on orbits @xmath33 and @xmath37 are related to each other by the lorentz transformation @xmath38 , between the instantaneous rest frames of @xmath33 and @xmath37 ; i.e. , given @xmath39 , @xmath40 is essentially defined by @xmath41 thus , the differential of arc length,@xmath42 is invariant because at each point it satisfies @xmath43 all such differentials may , therefore be set equal to the common differential @xmath44 , where @xmath45 is the speed of light and @xmath46 is the independent parameter which assumes the units of time ; i.e. , @xmath47 dividing ( 2.3 ) by @xmath45 and rewriting yields @xmath48 where @xmath49 in the customary notation .
alternately , this conclusion also follows directly from standard formulas .
it is known that all four - velocity vectors satisfy @xmath50 , so that differentiating by @xmath46 leads to the conclusion that four - acceleration is always orthogonal to the four - velocity.@xcite this means that acceleration does not change the modulus of the velocity , in other words , the ` four - lengths ' of all velocity vectors for all particles equal each other so that by dividing each by @xmath51 and taking the square root , one obtains eq .
( [ 2.3 ] ) again .
all standard works on special relativity cite experiments attesting to the `` reality '' of time dilation and the consequent aging - discrepancy .
how are they to be understood in view of the above considerations ?
first , note that to date no experiment meets the conditions leading to the twin - paradox .
certain experiments , those involving muon decays , for example , are described by linear transformations but are not round trips .
`` clocks - around - the - world '' experiments did involve round trips , but not linear ( acceleration free ) motion .
further , note that time dilation is ` real ' in the sense that it actually occurs with respect to signals , and is really no more puzzling than the fact that one s hand ( @xmath52@xmath53@xmath54 ) can shade the sun ( @xmath55@xmath54 ) .
it is an effect attendant to ` perspective ' in space - time .
thus , all physical effects resulting from the ` appearance ' ( i.e. , the way in which light signals transmit information or momentum - energy ) will be modified by the the perspective .
so any test of time dilation which involves a report from , or the interaction between , disparate frames will exhibit phenomena resulting from relative positions and times of emitter and receiver ; i.e. , space - time perspective .
some experiments seem exempt from the effects of perspective .
the two customary examples are the muon decay curve in the atmosphere , and the transport of atomic ` clocks - around - the - world . ' here the situation is less clear .
each of these experiments , however , is afflicted with features that allow contest.@xcite muon decay , for example , largely seems to ignore possible cross - section dependence on the velocity of the projectile and secondary production .
the clocks - around - the - world experiment has been strongly criticized for its data reduction techniques . in particular
, the existence of time delay effects for transported clocks has been questioned.@xcite without access to the details of these experiments and their subsequent data analysis , one is not in position to do deep critical analysis ; nevertheless , there is sufficient information in the literature to reasonably justify considering conclusions drawn on their basis as disputable .
moreover , experience with contemporary communication technology seems to present numerous practical reasons to question the conventional understanding of time delay effects for transported clocks.@xcite on the other hand , there are also experimental results completely in accord with this result .
an attempt by phipps to observe the so called ehrenfest effect
fitzgerald contraction of the circumference of a disk as a consequence of high tangential velocity due to rotation gave unambiguous null results , for example.@xcite the lack of radial dependence of element abundance and star species in observed galaxies can be taken as cosmic scale confirmation of phipps result .
crucial to a test of this formulation is that the aging of ` twins ' be compared directly rather than via reports conveyed between frames . because customary experiments rely on signals sent from the moving to the fixed frame one way or another , it is not possible to exclude ` space - time ' perspective effects .
perhaps this can be overcome .
consider a variation of a pound - rebka experiment employing a material with an element whose nucleus is naturally unstable .
let a sample of this material be divided and then hold half at a high temperature and half at a low temperature long enough such that the calculated time dilation of the more rapidly moving atoms of the heated half is great enough to yield a detectable difference in decay products .
the ratios of decay products then should be compared finally in the same frame , i.e. , at the same temperature .
an experiment of this structure would not be dependant on the transmission of signals from frame to frame but simply internally tally the total passage of eigen time in terms of decay half - lives in each frame for subsequent comparison .
( note : this scheme can be considered only conceptual inspiration .
in fact the shape of decay curves conceals rather than enhances differences in the accumulation of proper ages . )
electrodynamics as field theory ( i.e. , maxwell s equations ) does not result in a closed formulation .
that is , the interaction between two charged particles is describe by considering one charge as a current and solving for the fields at the position of the second which is then ` moved ' according to the lorentz force law .
then , the second particle is considered a source current which generates perturbing fields back on the first charge .
thereafter the first charge s motion is corrected and used to recalculate its fields at the position of the second_ad infinitum , _ or to the desired accuracy.@xcite fokker developed a closed formulation for the electromagnetic force by incorporating light - cone into action - at - a - distance mechanics .
essentially he found a lagrangian which is not merely the sum of of individual lagrangians whose variation yields coupled equations of motion.@xcite this lagrangian , however , produced yet another complexity : it led to simultaneous advanced and retarded interaction for each particle .
this feature is problematic on two levels .
first , it raises questions of causality because it would mean that the present is always partially conditioned by all of the future , contrary to observation .
secondly , it introduces the calculational complication of precluding the known methods of integrating the equations of motion ( this point will be discussed below ) . no resolution for the causality difficulties of the _ pure _
two - particle problem appear to have been proposed ; in fact , apparently the only attempt at resolution immerses the problem in a many body universe by invoking radiation absorbers at infinity.@xcite moreover , although integration of the pure two - particle equations has been attempted , thus far the proposed schemes are clearly approximation techniques or useful in severely restricted circumstances .
taking advantage , however , of the integrity of proper - time , @xcite we can formulate direct interaction mechanics as follows : let four - velocities be defined as @xmath56 and momenta as @xmath57 , where @xmath58 is the @xmath33-th particle s rest mass . with these definitions , the four - vector version of hamilton s principle @xmath59 where ( for @xmath18 ( number of particles ) = 2 ) @xmath60 yields equations of motion coupled by only two interactions ( because of the upper bound on the integral , advanced interactions are excluded . ) : @xmath61 where @xmath62 the features peculiar to this formulation can best be delineated by comparison with fokker s .
the most outstanding difference is that fokker s formulation does not exploit eq .
( [ 2.3 ] ) and therefore employs a separate independent parameter for each particle , which leads to a number of problems , including synchronization of these parameters.@xcite fokker s lagrangian is not simply the sum of individual patched together in an _ ad hoc _ manner ; he argued that a truly fundamental formulation should proceed from the variation of a _ single _ system lagrangian to a set of coupled equations of motion .
the lagrangian @xmath63 , @xmath64 satisfies these criteria and leads , by means of the variation @xmath65 to the equations of motion @xmath66 these equations , however , can not be integrated by a local procedure as is obvious is one imagines attempting a machine integration of the @xmath33-th equation at a given value of @xmath67 .
such an integration ; i.e. , a calculation of the an incremental extention of the world line for an incremental increase in @xmath67 , requires knowledge of the @xmath37-th world line on the forward light cone of the @xmath33-th particle , which , in order to be computed , requires knowledge of the @xmath68-th world line on the forward light cone of the @xmath33-th particle , but this portion of this orbit is yet to be computed , etc .
, _ ad infinitum . _ in effect , the solution is needed as initial data in order to compute the solution in this way . of course , advanced interaction could be precluded by changing the upper limit of integration in eq .
( [ 3.1 ] ) to @xmath69 , where @xmath69 is that value of @xmath67 which includes only the retarded potential from the @xmath33-th particle ; however , as @xmath69 would then also appear in eq .
( [ 3.1 ] ) , it could be written as the sum of individual lagrangians and therefore would not qualify as a system lagrangian . schemes can be imagined which circumvent this problem by some sort of global approach ; i.e. , by seeking the whole solution at once . for example ,
perhaps the solution could be found as the limit of a technique each successive step of which gave a closer approximation to the entire world line . at present , however , such techniques appear to have not been developed
( [ 3.3 ] ) are in general numerically and analytically unsolvable .
( [ 2.9 ] ) , on the other hand , can always be integrated by machine because the information needed to compute each incremental increase of any world line has already been computed . also by imagining a machine calculation
, it is clear that if each particle s world line between the past and the future with respect to the same but otherwise arbitrary light cone is given as initial data , then the system of world lines can be extended by calculations indefinitely into the future or the past .
( note , however , that retrodicting is not simply equivalent to reversing @xmath46 because the active and passive ends of the interaction do not thereby also exchange roles . in other words ,
this formulation with differential - delay equations of motion but advanced interaction has an intrinsic ` time arrow . ' ) although this type of initial data is greater that the customary cauchy data @xmath70 , it is a general characteristic of differential - delay equations that cauchy data are insufficient to determine a particular solution as enough initial data must be given to span the delay.@xcite
the main results of this work are twofold : 1 . ) strong doubt is cast on the validity of the notion that at a fundamental level , nature is nonlocal ; and 2 . )
it is shown that invariant proper time has the logical integrity required in order to be the independent parameter for special relativistic mechanics . while the first conclusion is without empirical contest , the second must still be reconciled with several experiments whose current interpretation seems to be in conflict .
the human psyche being what it is , it is in exactly those areas where certain knowledge is the least likely , that compensation perversely induces the strongest convictions . while matters of `` sex , politics and religion '' deliver the least contestable examples of this effect , fundamental physics , because of its `` deep '' reputation and cultural affiliation with virtually transcendental wisdom , runs a close second .
further , it seems to this writer that ` non - locality ' and ` asymmetric aging ' tempt many who would like to find a confirmation of mystical , religious or just exotic - futuristic beliefs , to see exactly that in logic - defying phenomena .
real , honest science , however , demands uncompromised logical consistency . likewise ,
real , honest philosophy does not revel in antilogy as a portal on the preternatural , rather the opposite ( see various contributions in , e.g. , @xcite , wherein time contortions are found wanting ) .
the arguments supporting our conclusions are strictly points of logic , rather than physical analogies or intuitive _
anstze_. internal logical consistency is even more demanding a standard than empirical verification whenever interpretation intervenes .
internal consistency is vital for developing meaningful theory ; it is well known that if an axiom set contains an inconsistent element , it is possible to prove as correct any statement whatsoever .
in fact the test for consistency consists essentially of finding a statement that can not be proven true . to many
it seems doubtful that modern physics could pass such a test ; thus , the continued development of physics theories best proceeds only on a basis purged of all antinomy among its basic definitions and hypotheses .
this is our aim
. 10 bell , j. s. ( 87 ) speakable and unspeakable in quantum mechanics , _ _ _ _ cambridge university press , cambridge . _ _ mermin , n. d. ( 1993 ) hidden variable and the two theorems of john bell , rev .
* 65 * ( 3 ) , pp .
803 - 815 .
kochen , s. and specker , e. p. ( 1967 ) the problem of hidden variables in quantum mechanics , j. math .
_ _ _ _ * * 17 , * * pp .
afriat a. and selleri , _ f. _ ( 1999 ) the einstein , podolski and rosen paradox , _ _ _ _ plenum , new york .
jaynes , e. t. ( 1989 ) clearing up mysteries - the original goal , in skilling , j. ( ed . ) maximum entropy and baysian methods _ _ , _ _ kluwer , dordrecht , pp . 1 - 29 .
perdijon , j. ( 1991 ) localit des comptages et dpendance des concidences dans les tests e.p.r .
avec polariseurs , ann .
l. de broglie , * 16 * ( 3 ) , * * * * pp .
281 - 286 .
kracklauer , a. f. ( 2000 ) `` la ` thore ' de bell , est - elle la plus grande mpris de lhistoire de la physique ? '' , ann .
l. de broglie , * 25 * ( 2 ) , pp .
193 - 207 .
hess k. and philipp , w. ( 2001 ) a possible loophole in the theorem of bell , proc .
* 98 * ( 25 ) , pp .
14224 - 14226 .
feller , w. ( 1950 ) an introduction to probability theory and its applications , wiley , new york , 1950 , cp .
v. kracklauer , a. f. ( 1999 ) pilot wave steerage : a mechanism and test , found .
_ _ _ _ * * 12 * * ( 5 ) , pp .
441 - 453 .
barut , a. o. ( 1992 ) how to avoid `` quantum paradoxes '' , found .
_ _ _ _ * * 22 * * ( 1 ) , pp .
137 - 142 .
santos , e. ( 1996 ) unreliability of performed tests of bell s inequality using parametric down - conversion photons , phys .
a , * 212 * pp . 10 - 14 .
aerts , s. , kwiat , p. , larson , j - a . and zukowski , m. ( 1999 ) two - photon franson type experiments and local realism , phys .
lett , * 83 * ( 15 ) , pp .
2872 - 2875 .
kracklauer , a. f. ( 2002 ) is ` entanglement ' always entangled ?
, _ _ _ _ j
b , quant .
semiclassical , _ _ _ _ * * 4 , * * pp .
hofer , w. ( 2001 ) visibility , efficiency and bell violations in real einstein , podolski and rosen experiments , http://arxiv.org/quant-ph/0106120 .
evdokimov , n. v. , klyshko , d. n. , komolov , v. p. and and yarochkin , v. a. ( 1996 ) bell s inequalities and epr - bohm correlations : working classical radiofrequency model , phys .- uspk .
* 39 * ( 1 ) , pp .
lee , k. f. and thomas , j. e. ( 2002 ) experimental simulation of two - particle quantum entanglement using classical fields , phys .
lett . * 88 * , 097902 , pp . 1 - 4 .
trump , m. and scheive , w. c. , ( 1997 ) the synchronization problem in covariant relativistic dynamics , fond .
phys__. _ _ * 27 * ( 1 ) , pp . 1 - 17 .
einstein , a. ( 1905 ) zur elektrodynamik bewegte krper , ann .
_ , _ * * 17 * * _ _ , _ _ pp . 891 - 921 .
kracklauer , a. f. ( 2000 ) on the twin non - paradox , http://arxiv.org/physics/0012041 .
_ _ rindler , w. ( 2001 ) relativity _ _ , _ _ oxford university press , oxford , p. 100 .
galeczki , g. and marquart , _ p. _ ( 1997 ) requiem fr die spezielle relativitt _ _ , _ _ ( haar + herchen , frankfurt a.m. ; present a concise survey ( in german ) of critical analysis of experiments widely understood to support the conventional understanding of special relativity .
kelly , _ _
( 2000 ) special relativity , right or wrong ? , electronics world , _ _ _ _ * * 9 , * * pp .
_ 722 - 723_. van flandern , t. ( 1998 ) what the global positioning system tells us about relativity , in : open questions in relativistic physics , _
selleri ( ed . ) aperion , montreal , 1998 , pp .
phipps jr .
, t. e. ( 1974 ) kinematics of a `` rigid '' rotor , lettere al nuovo cimento _ _ , _ _ * 9 , * pp .
467-xxx . rohrlich , f. ( 1965 ) classical charged particles , _ _ _ _ addison - wesley , reading .
fokker , a. d. ( 1929 ) ein invarianter variationssatz fr die bewegung meherer electronischer massenteilchen , z. phys .
* 58 , * pp . 386 - 393 .
wheeler , j. a. and feynman , r. p. ( 1945 ) interaction with the absorber as the mechanism of radiation , rev .
* 17 , * pp . 157 - 181 ; classical electrodynamics in terms of direct interparticle action , rev .
* 21 , * pp .
425 - 433 .
kracklauer , a. f. ( 1978 ) a theory of the electromagnetic two - body interaction , j. math .
_ _ _ _ * * 19**(4 ) , pp .
838 - 842 .
bellman , r. e. and cooke , k. l. ( 1963 ) differential - difference equations , academic press , new york .
wegener , m. ( ed . ) ( 1999 ) time , creation and world - order , aarhus university press , aarhus . | as a basis for epistemological study of `` time , '' we analyze three suspect phenomena introduced by modern physics : non - locality , asymmetric aging and advanced interaction .
it is shown that all three arise in connection with what has to be taken as arbitrary ideosyncrasies in formulation .
it is shown that minor changes result in internally consistent variations of both quantum mechanics and special relativity devoid of these phenomena .
the reinterpretation of some experiments though to confirm the existence of non - locality and asymmetric aging is briefly considered and a possible test is proposed .
key words : non - locality , asymmetric aging , advanced interaction , quantum mechanics , relativity _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ draft for : _ proceedings , the nature of time : geometry , physics & perception ; + may 21 - 24 , 2002 , tatranska lomnica , slovak republic _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ |
the exact renormalization group ( rg ) equation @xcite ( erge ) also called non - perturbative or functional rg equation can not be concretely used without recourse to approximation ( for modern reviews or introductory lectures see , e.g. , @xcite ) . the best known approximation framework for
the erge is the derivative expansion @xcite .
the leading order of that expansion , @xmath3-order , also named the local potential approximation ( lpa ) @xcite , completely discards any momentum dependence from the study . in principle
the lpa amounts to projecting the rg flow of the complete action @xmath4 $ ] ( a functional of the field @xmath5 ) onto the space of simple functions @xmath6 of a uniform field @xmath7 by assuming that:@xmath8 = \omega _ { d}\,u\left ( \phi \right ) \label{1}\]]where @xmath9 is the volume of the @xmath10-dimensional space . due to its simplicity and
because it is thought that it qualitatively involves most of the properties of the complete erge in the large distance regime ( e.g. , stability properties and number of fixed points ) , the lpa is currently utilized in many studies .
numerically , the lpa is considered as a reasonable approximation because the estimations of the critical properties would only be vitiated by the obligatory zero value of the critical exponent @xmath0 ( characterizing the large distance behavior of the two - point correlation function at the critical point ) which , in many circumstances , is actually a small parameter . in the early studies ,
the condition @xmath2 in the lpa has been justified as a consequence of the neglect of the detailed momentum dependence in the rg @xcite ( the same kind of justification of @xmath2 may be found in @xcite ) . though this argumentation by default is sometimes reused @xcite
, it is not very strong .
it has been argued that , in the lpa , _ it is not possible to consistently determine _
@xmath0 , or @xmath0 _ is set to zero as there is no mechanism to determine
_ its value @xcite ( see also @xcite ) .
the alert reader could express some surprise and argue that the vanishing of @xmath0 in the lpa has been clearly demonstrated a long time ago by hasenfratz and hasenfratz @xcite as often put forward in current studies ( see , e.g , @xcite ) .
unfortunately , the arguments are not unassailable because they rely , at least allusively ( see section [ invalidity ] ) , on the following truncation of @xmath11 $ ] : @xmath8 = \omega _ { d}\,u\left ( \phi \right ) + \frac{\bar{z}}{2}% \int d^{d}x\left ( \partial _ { x}\phi \left ( x\right ) \right ) ^{2 } \label{2}\]]in which the coefficient @xmath12 of the kinetic term would be maintained unaltered ( equal to unity ) along a rg flow of @xmath13 . a condition which would imply @xmath2 @xcite .
pending to show that the argument is actually artificial ( see section [ invalidity ] ) , we may already notice that truncation ( [ 2 ] ) differs in nature from the pure lpa ( [ 1 ] ) since it refers partly to the @xmath14-order of the derivative expansion .
consequently , assuming it was correct , this currently accepted argument , basis of what is referred to in the following as the conventional lpa , spoils the logic of the expansion based on a systematic projection of the complete erge onto the space of actions successively truncated according to the number of the derivatives of the field @xmath5 .
normally the lpa should correspond to ( [ 1 ] ) and not to ( [ 2 ] ) so that the supposedly proof of @xcite , even true , would be inappropriate .
hence , only remains the poor default argument .
it is then legitimate to wonder whether the condition @xmath2 is actually obligatory in the pure lpa .
it is a matter of fact that the conventional value @xmath2 in the lpa not accidentally but associated with a systematic absence of field renormalization raises some questions : 1 .
the absence of any field renormalization precludes in the lpa the eventual setting up of a non - classical power law behavior of correlation functions at criticality @xcite other than that purely induced by a diverging correlation length .
according to the theory of critical phenomena , two critical exponents are necessary to determine all the other critical indices of a second order critical point . with the erge considered around a wilson - fisher - like ( wf ) of the correlation length . ]
fixed point @xcite , these two exponents are @xmath0 and @xmath15 .
the two exponents arise differently in the erge . the index @xmath15 ( which characterizes the divergence of the correlation length @xmath16 when the temperature @xmath17 approaches its critical value @xmath18 ) occurs as a positive eigenvalue of the rg equation linearized around a fixed point ( the number of such positive values determines the order of the transition ) .
the role of the index @xmath0 is more subtle .
it is associated with the field renormalization allowing for a non - classical power law behavior of the correlation function at criticality .
usually one introduces @xmath0 in order to reproduce the critical behavior of the correlation function at large distances ( momenta going to zero ) and @xmath19 ; this manner of doing tightly links the field renormalization to the momentum dependence and suggests that no field renormalization is required when the momentum dependence is neglected @xcite implying @xmath20 .
however , the fluctuation - dissipation theorem relates the correlation function to the susceptibility .
this allows another introduction of @xmath0 via the critical exponent @xmath21 characterizing the critical behavior of the two - point correlation function at zero momenta and @xmath22 .
no reference to an explicit momentum is required in that case but the field should be renormalized nonetheless ( to take this eventual non - trivial power law with @xmath23 into account ) . in the lpa ( conventional or not )
@xmath15 takes on a non - classical value 6137 when the fixed point is a non - trivial one .
then , there is a priori no reason why @xmath24 would take on a classical value at this fixed point ( a priori no reason for keeping the field unrenormalized ) .
2 . [ invariance]although broken by the derivative expansion
, the reparametrization invariance of the complete erge is expected to be progressively restored as the order of the expansion grows .
when it is satisfied , this invariance specifies , in particular , that a change of normalization of the field by a pure constant [ like the parameter @xmath12 in ( [ 2 ] ) ] generates a line of equivalent fixed points characterized by a unique set of critical exponents with the joint existence of a zero eigenvalue mode in the solutions of the erge linearized around those fixed points .
the breaking of that invariance by the derivative expansion has been concretely observed at next - to - leading order [ @xmath14-order ] @xcite ; it is such that , for a given smooth cutoff function , a line of fixed points is well generated by the change of normalization of @xmath7 but those fixed points are not equivalent .
nonetheless , in agreement with a remark of bell and wilson @xcite in such a situation , one observes the existence at @xmath25-order of a vestige of the invariance via an extremum by a constant factor @xmath26 .
] of @xmath0 @xcite accompanied by the presence of a zero mode .
this gives a preferred estimate for @xmath0 ( one sometimes also refers to a principle of minimal sensitivity ) .
] , see , e.g , 2267,5469,5805 ) . because the conventional lpa offers no opportunities to look at the state of the invariance being arbitrarily fixed to zero there is no line of fixed points to be observed . ] , then no signs of progressive restoration of the invariance may be observed by going from the @xmath27-order to the @xmath14-order of the derivative expansion .
thus , considered as the leading order of that expansion , the conventional treatment creates confusion about the convergence property of the derivative expansion .
this is a pity because the issue is of some importance .
3 . [ etadet]having defined the rg - time @xmath28 ( with @xmath29 the running cutoff scale and @xmath30 an arbitrary fixed momentum scale ) the critical exponent @xmath31
is actually defined in the rg as the limit of a function @xmath32 on approaching a given fixed point ( when @xmath33 ) . according to wilson s prescriptions @xcite ,
the function @xmath32 is determined by keeping fixed the coefficient of one particular term in the action @xmath34 $ ] with the initial condition @xmath35 .
it is customary to keep constant the coefficient @xmath12 of the kinetic term because , due to a symmetry of the action linked to the reparametrization invariance , the flow of such a term is non - essential ( redundant ) so that it may be constrained without altering the model integrity .
when the kinetic term is not part of the approximation , as in the pure lpa , the redundancy still exists at least formally and it seems logical to wonder what the state of the reparametrization process is in the pure lpa . to this end one should introduce a function @xmath36 that would maintain fixed a particular monomial of @xmath37 . the line of fixed points mentioned in point [ invariance ]
would presumably be generated and this would give back the status of genuine leading order [ @xmath3-order ] of the derivative expansion to the lpa ( see section [ revisedlpa ] ) . in section [ revisedlpa ]
we present and discuss a version of the lpa with @xmath38 .
we show that it satisfies all the required conditions for being a genuine @xmath3-order of the derivative expansion .
in particular we study , in section [ fixedpoints ] , the structure of the fixed points for any value of the dimension @xmath10 and show explicitly , for the first time in the lpa , how the reparametrization invariance is broken .
we also introduce , in section [ lt ] , a legendre transformation of the potential adapted to the case studied ( @xmath38 ) .
this allows us to utilize easy quasi - analytic methods ( section [ tsm ] ) of integration of an ordinary differential equation ( ode ) well adapted to the obtention of the eigenvalues of the rg flow linearized around a fixed point .
it is then shown that the principle of minimal sensitivity ( pms ) does not necessarily indicate preferred values of the critical exponents ( section pms ) . in section [ versuspseudolpa ] ,
we first show that the conventional argument which is generally put forward to justify @xmath2 in the lpa , is actually artificial ( section [ invalidity ] ) .
we then discuss briefly , according to the rg rules , how the reparametrization invariance could be studied with the partial truncation ( [ 2 ] ) ( section pseudolpa ) . in appendix [ movable ] , we illustrate some reason why the quasi - analytic methods of integration of ode of section [ tsm ] do not work in the case of the potential of the action whereas they work after a legendre transformation is performed . we conclude in section [ conclusion ] .
let us consider the rg flow equation of the polchinski erge extended to include the parameter @xmath38 in the lpa @xcite , it reads@xmath39where @xmath40 stands for a simple function of @xmath7 and @xmath41 , @xmath42 , @xmath43 , @xmath44 , @xmath10 is the spatial dimension , and @xmath36 the field renormalization parameter which , at a fixed point , takes on the value @xmath45 . in principle and with the complete erge , @xmath45 should coincide with the critical index @xmath0 .
the flow equation ( [ lpa0 ] ) has already been studied by kubyshin et al @xcite for the derivative @xmath46 .
but they have considered @xmath47 in the lpa for technical reasons exclusively @xcite .
hence , in accordance with the conventional lpa , they have left @xmath36 undefined and focused their interest on @xmath45 considered as an arbitrarily adjustable parameter while emphasizing that physically @xmath45 should be zero at this order of the derivative expansion . with a view to study the fixed point equation ( @xmath48 ) for any @xmath10 at one time , we perform the following change of normalization of @xmath7:@xmath49then eq .
( [ lpa0 ] ) transforms into @xcite:@xmath50 \label{lpa } \\
\mu \left ( t\right ) & = & \frac{d-2+\eta \left ( t\right ) } { 2d } \label{mu}\end{aligned}\]]for a given @xmath10 , @xmath51 plays the role of @xmath32 and for any @xmath10 , the fixed point equation involves only one parameter instead of two in the preceding case of ( [ lpa0 ] ) . considering exclusively the issue of finding a non - singular solution @xmath52 to the fixed point equation corresponding to ( [ lpa ] ) [ or ( [ lpa0 ] ) ] , gives no possibility for determining a value of @xmath53 . then @xmath54 may rightly be considered as an extra parameter .
indeed @xmath52 is a solution of the following two - point boundary value problem of a second order non - linear ode:@xmath55now the whole of the two integration constants are fixed by the property of parity ( [ cond1 ] ) and by the adjustment of @xmath56 so as to get a non - singular @xmath57 in the whole range @xmath58 -\infty , + \infty \right [ $ ] as prescribed by condition ( [ cond2 ] ) ; then there is no room for determining @xmath59 ( @xmath45 at fixed @xmath10 ) without a supplementary condition . in the conventional lpa ,
the supplementary condition is merely @xmath60 which would be obtained ( see section [ invalidity ] however ) by an explicit reference to a larger space of truncation functions [ see eq .
( [ 2 ] ) ] .
however , even correct , this procedure would not be justified because the rg theory gives precise rules to determine both the function @xmath36 and its fixed point value @xmath45 . as recalled in point [ etadet ] of the introduction
, the function @xmath61 is determined by keeping one particular term of the action fixed along the rg flows @xcite ; then @xmath45 is the value reached by @xmath36 in approaching a given fixed point .
this procedure is a direct consequence of the reparametrization invariance of the complete action which induces the redundancy of the flow of one term of the action . in absence of any kinetic term , as in the pure lpa
, it is logical , and coherent with the rg theory , to define @xmath36 by keeping constant the coefficient of the quadratic term @xmath62 let us examine , on a general ground , the approach of the wf fixed point @xmath63 with the flow equation ( [ lpa ] ) starting at @xmath64 with the following simple potential:@xmath65where the coefficient of the quadratic term has been intentionally noted @xmath66 . because @xmath63 has only one relevant direction , to make the flow approaching @xmath63 only one coefficient of ( [ vdephi0 ] ) , say @xmath67 , must be fine - tuned ( in terms of the other three coefficients ) @xcite .
this adjustment is necessary to place the initial potential on the critical surface ( within the domain of attraction ) of @xmath63 @xcite . in order to follow a rg flow
, the function @xmath36 must be defined .
we do it such that @xmath68 all along the rg flow , with the initial condition @xmath35 and @xmath69 a constant independent of @xmath70 . at the fixed point ( reached at infinite rg - time
provided the initial potential lies in the domain of attraction of the fixed point ) , @xmath71 takes on the value of @xmath72 and this defines a line of fixed points ( parametrized by @xmath26 ) . if it was satisfied
, the reparametrization invariance would imply that @xmath73 be independent of @xmath26 and equal to @xmath0 .
of course , in the pure lpa one rather expects to observe the breaking of that marvelous property and the true question is : to which extent is that invariance broken in the lpa ? to look at this question , suffices to express the variation of @xmath74 in terms of @xmath75 .
this is precisely what kubyshin et al have done in @xcite : they studied eq .
( lpa0 ) for @xmath76 and @xmath77 ( for @xmath78 in @xcite ) .
the purpose of kubyshin et al was not the status of the reparametrization invariance in the lpa however .
in fact , having considered the flow equation for the derivative @xmath79 , the connection parameter of their fixed point equation was not @xmath80 but instead @xmath75 that they have noted @xmath81 .
then they have naturally drawn the variation of @xmath82 on changing the value of @xmath45 ( or the reverse ) without relating this variation to the reparametrization process .
it is however clear that , with our prescription of keeping @xmath83 fixed along a rg flow , the fixed point is reached with @xmath84 where obviously @xmath26 may be considered as the normalization of the field .
consequently , the evolutions of @xmath85 drawn by kubyshin et al are nothing but illustrations of the breaking of the reparametrization invariance in the lpa .
let us redo the study of kubyshin et al using our own conventions . using a standard numerical shooting method , we have looked for regular solutions ( the values of @xmath86 ) of the two - point boundary value problem ( [ fp]-[cond2 ] ) .
clearly , those solutions are parametrized by @xmath87 . since the coefficient of the quadratic term @xmath88
is linked to @xmath87 via the differential equation as @xmath89 , one easily gets functions @xmath90 corresponding to the functions ) the value of @xmath91 in the study of kubyshin et al is related to our @xmath26 as @xmath92 . ]
@xmath93 of kubyshin et al 5255 .
the four first solutions are displayed in fig ( [ figpf1 ] ) as continuous curved lines ; this figure involves simultaneously the two graphs of fig .
( 1 ) of kubyshin et al @xcite and displays the same features . let us discuss them . to @xmath94 obtained as regular solutions of eqs .
( [ fp]-[cond2 ] ) .
the full circles represent the thresholds of instability of the gaussian fixed point .
the horizontal lines indicate three values of @xmath10 where such instabilities occur when @xmath95 see text for more details.,width=453 ] each curved line drawn on fig .
( [ figpf1 ] ) corresponds to a line of fixed points of a particular nature .
it appears as bifurcating from the gaussian fixed point ( full circles ) on varying @xmath59 each time @xmath59 falls below the thresholds @xmath96 $ ] , @xmath97 [ horizontal lines on fig .
( [ figpf1 ] ) , corresponding to the usual dimensional thresholds @xmath98 ( for @xmath99 ) ] .
in particular fig .
( [ figpf1 ] ) shows the well - known fact that the gaussian fixed point is stable for @xmath100 and @xmath101 ( @xmath102 where there is no regular solution to eq .
( [ fp ] ) .
a new fixed point bifurcates each time the gaussian fixed point acquires a new direction of instability ; then the fixed points belonging to a line @xmath90 have @xmath103 directions of instability each . fig .
( [ figpf1 ] ) shows the four first lines of fixed points @xmath104 to @xmath105 ) having respectively one , two , three and four directions of instability .
all the lines accumulate at the horizontal line @xmath106 as @xmath107 along which @xmath26 reaches @xmath108 or stop at @xmath109 in agreement with the analytical solution found by kubyshin et al @xcite for @xmath110 .
let us focus our attention on the line @xmath111 which is a line of wf fixed points . because , for a given @xmath10 , @xmath112 varies along the line , it is obvious that the reparametrization invariance is broken . of course , this was expected in the lpa but surprisingly had never been explicitly emphasized before the present study . because the line is smooth and monotonous
, there is no vestige of the invariance , the breaking is perfect except at the limiting value @xmath113 where , for @xmath78 , @xmath45 takes on the values @xmath114 .
notice that nothing particular distinguishes the value @xmath101 from the other values lies at the intersection of @xmath111 and the horizontal line @xmath115 , @xmath101 with @xmath116 ( i.e. @xmath117 ) . ] except at the limiting cases @xmath106 , @xmath76 and
@xmath118 , @xmath119 . one observes also that @xmath120 on the whole line of fixed points @xmath121 ( except the gaussian fixed point ) ; this means that the basin of attraction of a non - trivial wf fixed point implies the condition @xmath122 on the initial potential @xmath123 ( otherwise the rg flow goes away from the critical surface towards the trivial high - temperature fixed point ) .
notice that the perfect breaking of the reparametrization invariance does not completely spoil the universal character of the critical behavior since the infinite number of initial potentials with a given @xmath122 lying on the critical surface are characterized by the same critical behavior governed by a unique value of @xmath45 and , subsidiarily , of the other critical exponents . from the line @xmath111 , it is also interesting to notice that non - trivial fixed points may be formally considered as existing for @xmath119 provided that @xmath45 be strictly negative .
usually such fixed points are rejected but in the present study there is no reason to reject them a priori since it is a consequence of the breaking of the reparametrization invariance to generate also negative values of @xmath124 .
it would be puzzling however if a vestige of that invariance led us to choose such a negative value of @xmath45 .
fortunately we observe no sign of such a preferred value of @xmath45 along the line @xmath125 except the limit case @xmath106 . at this level
we conclude that the lpa does not allow one to determine any estimate of @xmath45 but only ranges of possible values .
for example , if one excludes negative values of @xmath126 then for @xmath78 this range would be @xmath127 for the only line @xmath128 , the other lines being excluded .
from this example taken alone , one could be inclined to conclude that the conventional lpa , by imposing @xmath101 , would be merely a reasonable choice since one knows that @xmath45 is most often small
. however , fig .
( [ figpf1 ] ) shows that from @xmath129 down to @xmath76 , emerges a rich structure of various fixed points with possible different positive and growing values of @xmath45 for which the conventional lpa would impose , increasingly poorly as @xmath10 decreases , the same zero value ( one knows that at @xmath76 , @xmath130 what is not small ) .
notice that , for @xmath131 we have excluded the limit case @xmath132 on the line @xmath133 . ] from the range of possible values of @xmath45 though it corresponds to the only point on @xmath111 where a zero eigenvalue exists ( see section [ pms ] ) . indeed , since @xmath134 at this point , this zero mode is not a vestige of the reparametrization invariance ; instead it indicates that the nature of the gaussian fixed point is going to change by losing one direction of instability .
hence , if a direct derivative of the fixed point equation with respect to @xmath26 shows that an extremum of @xmath135 implies the appearance of a zero eigenvalue , the reverse is not true . in order to better illustrate the role of the zero mode in the process of restoration of the reparametrization invariance ,
let us look at the critical exponents in the lpa and at their variations on changing the normalization of the field . to this end
, we perform a legendre transformation of the potential which will allow us to make use of user - friendly quasi - analytic methods of integration of ode .
the interest of using some quasi - analytic methods to solve the rg flow equation in the lpa is the extremely easy access that they offer to estimate : 1 . the fixed point value @xmath136 of the connection parameter 2 . a set of critical ( and subcritical ) exponents at one time . on the contrary ,
the purely numerical shooting method necessitates a skillful adjustment of an initial guess of the final value of @xmath136 or , independently , of each critical exponent sought .
the use of quasi - analytic methods based on taylor series , in solving a two - point boundary value problem like ( [ fp]-[cond2 ] ) , has been recently reviewed and illustrated in @xcite . among such methods
is an extremely simple procedure @xcite ( named the simplistic method in @xcite ) that merely consists of imposing the vanishing of the last term @xmath137 of the maclaurin series of the truncated solution : @xmath138the coefficients @xmath139 being determined such that the edo considered be satisfied order by order in powers of @xmath140 . the auxiliary condition @xmath141 gives a condition from which one tries to extract an estimate of the connection parameter @xmath142 corresponding to the only regular solution of ( [ fp]-[cond2 ] ) . of course , because it is too simple , this simplistic method is not always ( most often never ) efficient .
firstly the finite character of the radius of convergence of the series limits the accuracy of the method @xcite .
secondly the method may simply not work at all ( in the sense that even a rough estimate of @xmath136 may not be approachable ) .
indeed , in trying to solve ( [ fp]-[cond2 ] ) , the issue we are faced with amounts to pushing a movable singularity to infinity .
the efficiency of the simplistic method then depends on whether or not that singularity lies within the circle of convergence of the maclaurin series or not ( see appendix [ movable ] ) .
a variant of the simplistic method , referred to below as the taylor method , is frequently used which is based on a taylor expansion around the minimum of the potential , as proposed in @xcite ( see also @xcite ) .
the solution of the ode is thus expressed as : @xmath143 where @xmath144 is the expansion point chosen to coincide with the minimum of the potential arbitrarily and the two unknowns would have been @xmath67 and @xmath145 .
this would offer the possibility of improving the apparent convergence of the taylor method by varying @xmath146 , see @xcite . ] since @xmath147 , and @xmath148 , whereas the original connection parameter is , by definition , given by:@xmath149there are two unknowns ( @xmath67 and @xmath146 ) to be determined .
the taylor method consists of imposing the vanishing of the two last terms of the taylor series to get two auxiliary conditions on @xmath67 and @xmath146 .
this method may improve considerably the simplistic method ( it has provided excellent estimates of the critical exponents in the lpa @xcite ) . the reason is due to the fact that , other things being equal compared to the simplistic method , one starts closer to the movable singularity .
but the taylor method requires that the expansion point ( the minimum of the potential ) lies within the circle of convergence of the maclaurin series ( otherwise one could not get a reliable estimate of @xmath136 by summing the series back to the origin ) . also , the accuracy of the method is naturally limited by the finite range of convergence of the taylor expansion .
it is a matter of fact that , in the conventional lpa with @xmath60 , the two quasi - analytic methods presented just above do not work when they are applied to the polchinski rg flow equation of @xmath13 but they work if one first performs a legendre transformation ( @xmath150 ) as that defined in @xcite . with a view to make use of these user - friendly quasi - analytic methods
that allow anyone to easily verify the content of the present paper , let us introduce a legendre transformation appropriate to the case @xmath38 . to begin with , we consider the legendre transformation originally introduced for @xmath152 in @xcite and we apply it to the flow equation of @xmath153 extended to include @xmath38 , namely:@xmath154 according to @xcite , the legendre transformation reads @xmath155from which we have:@xmath156 thus applied to ( [ lpamorris0 ] ) we get the following flow equation for @xmath13 : @xmath157which differs from the usual polchinski equation ( [ lpa0 ] ) when @xmath158 .
the appearance of this coefficient may be seen as the consequence of a non - linear introduction of @xmath61 in the erge @xcite instead of the linear introduction of @xcite that corresponds to ( [ lpa0 ] ) .
though , near a fixed point , the coefficient @xmath159 may be removed from ( lpa0bis ) through the change @xmath160 to get the same equation as ( [ lpa0 ] ) , we have numerically studied ) corresponding to applying the legendre transformation ( [ tl1]-[tl3 ] ) on ( [ lpa0 ] ) , but this would have made the quasi - analytic methods heavier and thus less attractive . ]
( [ lpa0bis ] ) explicitly for @xmath78 ( using a shooting method ) .
we have , this way , verified explicitly ( in the case of wf fixed points ) both that we get the same kind of line of fixed points as previously ( monotonous function @xmath161 ) and that the simplistic and taylor methods applied to ( lpamorris0 ) work well also for @xmath38 [ at least for the values of @xmath162 shown in fig ( [ fignudez ] ) ] .
let us focus our interest on the eigenvalue problem corresponding to eq .
( [ lpamorris0 ] ) linearized around a fixed point @xmath163 [ solutions of ( [ lpamorris0 ] ) such as @xmath164 .
we get the following second order linear ode ( once @xmath165 is known):@xmath166where @xmath167 is the eigenfunction and @xmath168 the eigenvalue parameter . for
a given set of initial conditions , such as @xmath169 , @xmath170 in the even case , one expects to obtain an infinite set of discrete couples @xmath171 ordered according to the magnitude of @xmath172 .
the number of positive values depends on the fixed point considered . except the trivial eigenvalue @xmath173 , the wf fixed point
is characterized by the existence of only one positive value @xmath174 corresponding to the critical exponent @xmath175 , the next eigenvalue @xmath176 is negative and corresponds to the leading subcritical exponent @xmath177 characterizing the leading correction - to - scaling with @xmath178 .
these two exponents have been estimated , for @xmath78 , with a very high accuracy in the conventional lpa ( with @xmath20 ) to get @xcite:@xmath179of course , due to the legendre transformation , the same set of critical exponents is obtained in both cases of the flow equations of @xmath153 and @xmath13 @xcite .
it is clear that this is also the case in the present study with @xmath38 , provided the methods used converge .
we have determined the evolution in terms of @xmath180 of the two first exponents @xmath15 and @xmath181 using both the simplistic and taylor methods and obtained the curves shown in figures ( [ fignudez ] , [ figomegadez ] ) . , of the critical exponent @xmath182 as function of the field - normalization @xmath26 .
the full line corresponds to calculations done using the taylor method , open circles correspond to results obtained with the simplistic method .
the point located at @xmath183 corresponds to the gaussian fixed point with @xmath184.,width=377 ] , of the subcritical exponent @xmath185 as function of the field - normalization @xmath26 ( full line ) obtained using the taylor method .
the point located at @xmath186 @xmath187 is the only possibility of having a zero mode ; it is located far from the value corresponding to @xmath188 for which @xmath189 undergoes a minimum [ see fig .
( [ fignudez])].,width=377 ] fig .
( [ fignudez ] ) shows that @xmath190 undergoes a minimum at @xmath191 corresponding to @xmath192 whereas , at this point we get @xmath193 [ see fig .
( [ figomegadez ] ) ] . according to a principle of minimal sensitivity ( pms ) @xcite sometimes used in calculations at higher orders of the derivative expansion of the erge ( see , e.g
, @xcite) one could be inclined to propose those values as being the preferred estimates of the critical exponents in the lpa for @xmath78 .
however , one may observe that those values are not designated as the consequence of a vestige of the reparametrization invariance which is the only reason that fundamentally led us to vary @xmath26 . indeed
no zero eigenvalue is obtained at this point as shown in table [ table1 ] .
( [ figomegadez ] ) clearly shows that the only point where a zero mode occurs corresponds to the gaussian fixed point which is losing one direction of instability ( @xmath194 , @xmath195 , @xmath132 , for @xmath78 ; or @xmath194 , @xmath196 , @xmath101 , for @xmath119 ) but at this point @xmath197 does not vanish [ see fig .
( [ figpf1 ] ) ] .
notice that this is only a confirmation of the absence of any extrema in the function @xmath73 .
indeed , if one performs a derivation with respect to @xmath26 of the fixed point equation corresponding to ( [ lpamorris0 ] ) , assuming @xmath198 , then one gets ( [ vplitim ] ) with @xmath199 .
the reverse is not true however : the presence of a zero mode may reveal instead the change of the stability properties of the fixed point .
.comparison of the six first eigenvalues of eq .
( [ vplitim ] ) obtained for @xmath200 , @xmath78 and with the two quasi - analytic methods considered in the study .
the first line of numbers corresponds to @xmath201 ( 1/@xmath202 ) , the second to @xmath203 ( @xmath204 ) , etc .
no zero eigenvalue is present . [
cols="<,<",options="header " , ] we may thus conclude that , because it has no link with the reparametrization invariance , the observed minimum of @xmath15 occurs accidentally and that the pms can not be utilized in the circumstances as a tool to determine a preferred set of values of the critical indices .
in this section , we first show that the argument of hasenfratz - hasenfratz @xcite , by which the rg flow projected on ( [ 2 ] ) would imply @xmath205 if @xmath12 is kept unaltered by the flow of @xmath13 , is artificial and reduces to a triviality that poorly justifies the default argument .
then we briefly illustrate that , correctly treated , the projection of the erge on ( [ 2 ] ) gets an intermediate order [ between the @xmath3 and @xmath14 ] of the derivative expansion that we name pseudo - lpa .
this partial @xmath25-order differs from the approximation introduced in @xcite in that we try to account for the reparametrization invariance . to get the rg flow equations of the wilson - polchinski erge correctly projected on ( [ 2 ] ) , suffices to consider the complete @xmath25-order equations available in the literature as , e.g. , in 3491,3836 , and to impose within them that the kinetic term is a pure number that remains constant along a rg flow of the potential .
for example , let us consider eqs .
( 12 ) of @xcite for the derivative @xmath207 and a function @xmath208 reduced to @xmath209 , it comes : @xmath210 \,f^{\prime } \left ( 0,t\right ) \label{plpaz}\end{aligned}\]]where @xmath211 is a constant parameter depending on the choice of cutoff function . up to inessential changes , eq .
( [ plpav ] ) is the same flow equation as ( [ lpa ] ) of the pure potential @xmath13 discussed previously in section revisedlpa .
( [ plpaz ] ) shows that the flow of @xmath13 induces a flow of @xmath12 so that keeping it constant , i.e. imposing @xmath212 , yields:@xmath213 \,f^{\prime } \left ( 0,t\right ) = 0 \label{p1}\]]this condition considered at a fixed point of ( [ plpav ] ) may be rewritten as:@xmath214 \,\gamma = 0 \label{p0}\]]where , as seen in section [ revisedlpa ] ( footnote [ gamma ] ) , @xmath215 is a function of @xmath45 as that given implicitly by the lines of fixed points @xmath216 drawn in fig .
( [ figpf1 ] ) where @xmath217 was playing the role of @xmath12 . for @xmath218 and @xmath219 ( the conventional values ) , eq .
( [ p0 ] ) implies @xmath220 .
( [ figpf1 ] ) shows that this is not possible along the line of wf fixed points @xmath128 except trivially at @xmath119 where the fixed point is gaussian@xmath221 the conventional argument of @xcite actually relies upon the arbitrary requirement that no contribution of @xmath13 must alter the flow of @xmath12 so that the right - hand - side of ( [ plpaz ] ) would be reduced to the first term exclusively , thus implying @xmath152 for a constant @xmath222 . clearly , this is only an illustration of an obvious fact : the non - necessity of renormalizing the field ( here forced by the obligatory absence of contribution coming from @xmath13 ) , induces @xmath223 ( what is true by definition ) .
we add that , not only truncation ( [ 2 ] ) is incompatible with a pure @xmath3-order , it is also in contradiction with the default argument by which @xmath61 is absent because there is _ no momenta _ in the lpa .
actually , the conventional argument is merely artificial . as shown in section [ revisedlpa ] , it is the basis of a conventional lpa which is misleading concerning the concept of reparametrization invariance and in contradiction with the logic of the derivative expansion . considered as an actual truncation of the action @xmath11 $ ] , eq .
( [ 2 ] ) gives access to an intermediate approximate order of the erge [ between @xmath3 and @xmath14 ] . by allowing the coefficient of the kinetic term to flow
( whereas it remains independent of @xmath7 ) one obtains the partial truncation first used by tetradis and wetterich @xcite in order to easily have @xmath224 in an improved(conventional ) lpa .
but the way of determining @xmath45 , in the original proposal , is limited to the erge for the effective average action @xmath225 $ ] ( for a review see @xcite ) .
this is because one utilizes the available momentum dependence of the exact propagator @xmath226 to determine the function @xmath227 yielding a value for @xmath45 .
that way of doing is not convenient to the polchinski erge , with which an easy access to @xmath226 is not possible .
moreover , the tetradis and wetterich approach does not give a clear account of the reparametrization invariance ( being in the spirit of the conventional lpa criticized in the present paper ) .
let us look at truncation ( [ 2 ] ) for the polchinski erge by strictly applying the basic rules of the rg theory .
the rg rules prescribe a field renormalization in order to maintain constant one term of the action . with truncation ( [ 2 ] ) , we choose it to be the kinetic term - order we could have pursued the process of maintaining constant the quadratic term of the action ( instead of the kinetic term ) . in the present pseudo - lpa
this procedure would give nothing new compared to the pure lpa .
this underlines the particular character of that truncation.[pseudofoot ] ] ( the only momentum - dependent monomial of the approximation ) .
hence we get eq .
( [ p1 ] ) and , at a fixed point , eq . ( [ p0 ] ) . considering @xmath12 as a free parameter at hand ,
@xmath45 and @xmath82 appear to be functions of @xmath222 .
we thus get lines of fixed points parametrized by @xmath12 .
the relation between @xmath45 and @xmath82 being unchanged compared to the lpa [ eq .
( [ plpav ] ) is the derivative with respect to @xmath7 of ( lpa0 ) ] , we have:@xmath228\]]where @xmath229 $ ] has been determined at leading order in section [ fixedpoints ] [ implicitly through the lines of fixed points displayed by fig .
( [ figpf1 ] ) ] .
it is then easy to seek for a vestige of the reparametrization invariance eventually displayed by the new lines of fixed points ( parametrized by @xmath230 ) .
suffices to look at possible values of @xmath45 where @xmath231 . differentiating ( [ p0 ] ) with respect to @xmath12 , we get:@xmath232 \,\gamma \,_{\mathrm{lpa}}^{\prime } \right\ } + \eta ^{\ast } + 4\,\gamma _ { \mathrm{lpa}}=0\]]where @xmath233 .
finally , imposing the required condition gives a preferred value of @xmath45 defined by:@xmath234 notice that this condition is independent of the choice of the cutoff function , contrary to what is observed with the complete @xmath25-order @xcite ( for another difference with the complete order , see footnote [ pseudofoot ] ) . in terms of the quantity @xmath235 defined by ( [ mu ] ) and taking into account the change of field variable ( [ changephi ] ) , this condition writes , for @xmath78 and @xmath236 ( see footnote [ gamma])@xmath237 from the calculations done in section [ fixedpoints ] , we obtain:@xmath238which are not excellent values .
this result may however be considered as an improvement compared to the lpa for which no vestige of the reparametrization invariance was observed .
we have justified the presence of a non - vanishing value of @xmath0 in the lpa as a strict consequence of the general principles of the rg theory . without field renormalization , as usually prescribed in the conventional lpa ( with @xmath2 ) , the approximation could not be considered as the genuine @xmath27-order of the derivative expansion . if no particular estimate of @xmath0 can actually be proposed at this order , @xmath31 is not an arbitrary parameter , it varies monotonously ( within some limits ) on changing the normalization of the field as a consequence of a perfect breaking of the reparametrization invariance .
the situation is coherent with the idea that , if the derivative expansion converges then , at least , a progressive and smooth restoration of the invariance must be observed from the few first terms of the expansion ( this was not possible with the conventional view ) . we have done explicit calculations of the lines of fixed points generated by the change of the normalization of the field by a constant @xmath26 using both purely numerical and quasi - analytic methods in order to offer the possibility to anyone to easily redo the calculations .
we also emphasize that , despites the minimum observed ( for @xmath78 ) in the evolution of the critical exponent @xmath15 on varying @xmath26 , the principle of minimal sensitivity can not be applied being not compatible with a possible vestige of the reparametrization invariance in the lpa .
we have shown ( in section [ invalidity ] ) that it is purely artificial the conventional argument stating that @xmath0 should vanish if one keeps the kinetic term unchanged along a rg flow of the potential .
we have illustrated ( in appendix movable ) the respective roles of the movable and fixed singularities of the fixed point solutions in the convergence property of the quasi - analytic methods of integration utilized in the study .
in this appendix we illustrate the role of the movable singularity of the solution of the fixed point equation of the lpa in the convergence and efficiency of the simplistic method ( see section [ tsm ] ) .
let us consider the fixed point equations of respectively the rg flow ( lpa0 ) of @xmath13 and the rg flow ( [ lpamorris0 ] ) of the legendre transformed potential @xmath153 [ see ( [ tl1]-[tl3 ] ) ] . for @xmath78 and @xmath239 the common value of the connection parameter @xmath240 , corresponding to the respective regular fixed point solutions ,
is known with a huge number of digits @xcite to be:@xmath241 that value corresponds precisely to the only solution of the two - point boundary problem ( [ fp]-[cond2 ] ) with @xmath242 .
if one forgets about the condition at infinity , then it exists a solution involving a singularity for each value , is named movable singularity .
] of @xmath243 different from @xmath244 . hence getting the value ( [ kstar ] ) may be viewed as the consequence of pushing that movable singularity to infinity .
in addition to a movable singularity , a solution of ( [ fp]-[cond1 ] ) displays also fixed singularities . ] .
potentially , those singularities control the convergence properties of the maclaurin series of the ultimate @xmath245 .
we have performed pad approximants on the maclaurin series of solutions for various @xmath246 of the fixed point equations in the two cases of @xmath13 and @xmath153 .
the complex zeros of the denominators of the corresponding rational fractions give an approximate image of the location of the singularities in the complex plane of the variables @xmath247 and @xmath248 of respectively @xmath13 and @xmath153 .
, of the singularities of the solutions @xmath249 of the fixed point equation , for @xmath250 and @xmath78 , of the rg flow eq .
( [ lpamorris0 ] ) in the process of determining @xmath251 @xmath252 the value of which is given by eq .
( [ kstar ] ) .
two steps are shown : @xmath253 ( top ) and @xmath254 ( bottom ) .
as one approaches @xmath136 , the movable singularity is pushed on the right ( ideally up to infinity ) .
the efficiency of the simplistic method ceases when the movable singularity is about to leave the disc of convergence of the maclaurin series ( determined by the location of the fixed singularity the closest to the origin ) .
the method yields a rather high accuracy on the estimate of @xmath255 with 9 accurate figures .
the radius of convergence of the maclaurin series for @xmath256 is @xmath257.,width=377 ] figure ( [ figsinlitim ] ) shows the singularity structure of @xmath153 for two values of @xmath246 on approaching @xmath136 .
one clearly sees that the movable singularity still lies within the circle of convergence of the maclaurin series though @xmath246 is already close to @xmath136 . on approaching closer to @xmath258 the movable singularity
is pushed to the right and the simplistic method ceases to converge when the singularity comes out of the disc of convergence of the series .
concretely the method provides an estimate of @xmath258 with 9 accurate figures .
, of the singularities of the solution of eq .
( [ fp ] ) , for @xmath250 and @xmath78 , in the process of determining @xmath251 @xmath259 given by eq .
( [ kstar ] ) .
same presentation as in fig .
( [ figsinlitim ] ) .
the step corresponds to @xmath260 which is relatively far from @xmath136 whereas the movable singularity is already outside the disc of convergence of the maclaurin series .
the simplistic method does not work in that case .
the radius of convergence of the maclaurin series for @xmath261 is @xmath262,width=377 ] on the contrary , fig .
( [ figsingpol ] ) shows that the movable singularity is already well outside the circle of convergence of the maclaurin series of @xmath13 when @xmath246 is still far from @xmath136 . in that case
the simplistic method can not even give a poor estimate of @xmath136 .
k. i. aoki , int .
b * * 14 * * ( 2000 ) 1249 ; j. polonyi , cent .
* 1 * ( 2003 ) 1 .
[ arxiv : hep - th/0110026 ] ; j. m. pawlowski , ann .
( n.y . ) * 322 * ( 2007 ) 2831 .
[ arxiv : hep - th/0512261 ] ; b. delamotte , in _ order , disorder and criticality .
advanced problems of phase transition theory , vol 2 _ , p. 1 , _ ed . by _
yu . holovatch ( world scientific , publ .
co. , singapore , 2007 ) ; also in lecture notes in physics vol .
* 852 * ( 2012 ) 49 .
[ arxiv : cond - mat/0702365 ] ; o. j. rosten , phys . rep .
* 511 * ( 2012 ) 177 .
[ arxiv:1003.1366 ] ; a. wipf , lecture notes in physics * 864 * ( 2013 ) 257 .
c. wetterich , z. phys .
c * 57 * ( 1993 ) 451 ; t. r. morris , nucl .
b ( proc . suppl . )
* 42 * ( 1995 ) 811 .
[ arxiv : hep - lat/9411053 ] ; int . j. mod
b * 12 * ( 1998 ) 1343 .
[ arxiv : hep - th/9610012 ] ; phys . lett .
b * * 329 * * ( 1994 ) 241 .
[ arxiv : hep - ph/9403340 ] t. r. morris , nucl .
b * 458 * ( 1996 ) 477 .
[ arxiv : hep - th/9508017 ] ; j. comellas and a. travesset , nucl .
b * 498 * ( 1997 ) 539 .
[ arxiv : hep - th/9701028 ] ; t. r. morris and m. d. turner , nucl . phys .
b * * 509 * * ( 1998 ) 637 .
[ arxiv : hep - th/9704202 ] y. kubyshin , r. neves and r. potting , in _ the exact renormalization group _ , p. 159 , _ ed . by _
a. krasnitz , y. a. kubyshin , r. potting and p. s ( world scientific , publ .
co. , singapore , 1999 ) .
[ arxiv : hep - th/9811151 ] | the conventional absence of field renormalization in the local potential approximation ( lpa ) implying a zero value of the critical exponent @xmath0 is shown to be incompatible with the logic of the derivative expansion of the exact renormalization group ( rg ) equation .
we present a lpa with @xmath1 that strictly does not make reference to any momentum dependence .
emphasis is made on the perfect breaking of the reparametrization invariance in that pure lpa ( absence of any vestige of invariance ) which is compatible with the observation of a progressive smooth restoration of that invariance on implementing the two first orders of the derivative expansion whereas the conventional requirement ( @xmath2 in the lpa ) precluded that observation .
local potential approximation , derivative expansion , exact renormalization group equation , reparametrization invariance , anomalous dimension 05.10.cc , 11.10.gh , 11.10.hi , 64.60.ae , 02.60.lj |
a random graph is a set of nodes that are connected by random links @xcite .
when the number of links exceeds a certain threshold , a giant component with a finite fraction of all nodes emerges @xcite .
therefore , random graphs are equivalent to a mean - field percolation process @xcite .
random graphs underlie many natural phenomena from polymerization @xcite to the spread of infectious diseases @xcite , and they are also used to model social networks @xcite as well as complexity of algorithms @xcite .
the classical random graph has no restrictions on the degree of a node . in many situations ,
the number of connections is limited , and specifically , the degree of each node must be bounded .
examples include the percolating network of contacts in a bead pack @xcite , computer networks @xcite , and communication networks @xcite .
we modify the canonical model for an evolving random graph @xcite by imposing a strict bound on the degree of a node @xcite . in an evolving random graph ,
two nodes are picked at random , and regardless of their degrees , the nodes are connected by a link . at the end of this process , a complete graph forms . in this study , we focus on a constrained process where the two nodes are linked only if both degrees are smaller than @xmath0 . starting with a set of @xmath2 isolated nodes , the final state is a regular random graph @xcite , where all nodes have degree @xmath0 ( fig . [ fig - process ] ) .
the evolving graph includes two types of nodes , active nodes with degree smaller than the maximum , and inactive nodes with the maximal degree .
we obtain the density of active nodes from the full degree distribution , which is shown to be a truncated poisson distribution . in general ,
the graph is a set of disjoint connected components @xcite . in each component ,
any two nodes are connected by a path . as in classical random graphs
, our graphs could be in one of two phases a non - percolating phase where all components are finite , and a percolating phase where a single _ giant _ component with a macroscopic number of nodes coexists with finite components .
the transition between these two phases occurs when the number of links exceeds a threshold .
for example , when @xmath3 , the critical link density , @xmath4 , is @xmath5 when @xmath6 the final state consists of dimers , while when @xmath7 the final state includes multiple rings @xcite . in both cases ,
the system does not condense into a single component .
we study the more interesting situation where the system undergoes a percolation transition and the final state is a fully connected graph . as a result of the linking process
, connected components undergo an aggregation process . in the traditional binary aggregation framework @xcite
, connected components are characterized by the total number of nodes @xcite . in our process , however , this minimal description does not lead to closed evolution equations .
instead , we describe clusters by the total number of nodes of degree @xmath8 , @xmath9 , @xmath10 , @xmath0 . this complete description naturally lends itself to closed evolution equations .
we use the elegant hamilton - jacobi method to obtain a formal solution for the density of connected components , and find the percolation thresholds from this solution .
we also use the degree distribution to illuminate the behavior during the late stages of the evolution .
the regular random graph forms via a two - step process .
first , the giant component takes over the entire system .
this step occurs after @xmath11 linking attempts , with @xmath12 . at this point ,
only a small minority of the nodes in the giant component remain active .
the total number of active nodes , @xmath13 , scales logarithmically with system size , @xmath14 eventually all nodes become inactive , and the regular random graph is complete .
the number of linking attempts required to reach this final state is proportional to the system size .
hence , the evolution of the regular random graph is much slower than the evolution of the classical random graph , where @xmath11 is logarithmic in @xmath2 @xcite .
the rest of this paper is organized as follows .
the degree distribution is derived in section [ dd ] .
we then show that the linking process is equivalent to a multivariate aggregation process ( sec .
a formal solution is obtained in sec .
[ hj ] using the hamilton - jacobi method .
the emergence of the giant component is discussed in section [ gc ] . in particular , the mass of the giant component is obtained as a function of time .
the late - time kinetics including the emergence of a fully connected graph and a perfect regular random graph are described in section [ lt ] .
we discuss the results in section [ disc ] . in appendix
a , we rederive relevant properties of classical random graphs using the hamilton - jacobi method .
we study a random process that generates a regular random graph .
the process starts at time @xmath15 with @xmath2 disconnected nodes . at each elementary step ,
two nodes are chosen at random .
if the degrees of the two nodes are both smaller than @xmath0 , a new link connects them ; otherwise , no link is added .
time is augmented by @xmath16 after each linking attempt , , so that every node participates in one linking attempt per unit time , on average . linking is repeatedly attempted until a regular random graph forms ( fig . [ fig - process ] ) .
there are two types of nodes active nodes with degree smaller than @xmath0 and inactive nodes with degree equal to @xmath0 . a new link between two active nodes of degrees @xmath17 and @xmath18 augments the degrees : @xmath19 this random process occurs with unit rate .
the linking process transforms the system from a set of active nodes into a set of inactive nodes , as illustrated on fig .
[ fig - process ] .
let @xmath20 be the degree distribution , that is , the fraction of nodes with degree @xmath21 at time @xmath22 .
the degree distribution is properly normalized , @xmath23 , and its partial sum gives the total density of active nodes , @xmath24 , @xmath25 the density of active nodes controls the process : the degree distribution obeys the rate equations @xmath26 the initial condition is @xmath27 and the `` boundary '' condition @xmath28 ensures that .
equations are nonlinear because is a two - body process . yet
, since the density @xmath24 merely sets the overall linking rate , we can linearize eqs . by introducing the time variable @xmath29 or equivalently , @xmath30 . in terms of the time variable @xmath31 ,
the rate equations become @xmath32 we use the initial condition and solve recursively to find @xmath33 , @xmath34 , , etc . therefore , the degree distribution is a truncated poisson distribution @xcite @xmath35 for @xmath18 .
the density of inactive nodes , , follows from the normalization condition .
from the definitions and , the time variable @xmath31 obeys the nonlinear ordinary differential equation @xmath36 with the initial condition @xmath37 .
equations and specify the degree distribution . versus time @xmath22 for @xmath3.,scaledwidth=45.0% ] the time evolution of the degree distribution , obtained by numerical integration of for the case @xmath3 , is shown in fig .
[ fig - nt ] . as expected , the density of isolated nodes , @xmath38 , declines monotonically , while the density of inactive nodes , @xmath39 , increases monotonically .
isolated nodes initially dominate , then nodes with degree one are most numerous , and eventually , inactive nodes take the lead for good .
the average degree , @xmath40 , characterizes the overall connectivity .
since every link connects two nodes , the density of links , @xmath41 , is proportional to the average degree , @xmath42 . by multiplying by @xmath21 and summing over @xmath21 ,
we see that the average degree satisfies @xmath43 using the initial and final values , @xmath44 and @xmath45 , we have the integral identity @xmath46 .
every node belongs to a connected component , or `` cluster '' .
initially , there are @xmath2 clusters , and eventually , a single cluster remains .
the number of clusters dwindles as a result of a cluster - cluster aggregation @xcite process : any link between two active nodes in separate clusters merges the two clusters .
since the aggregation rate depends on the total number of _ active _ nodes , characterization of clusters by their total size is insufficient . adding the number of active nodes as a second variable does not lead to a closed description either , because the number of active nodes may change once a link is added .
therefore , we use a complete description and characterize clusters by @xmath47dimensional vectors @xmath48 where @xmath49 is the number of nodes of degree @xmath21 .
the cluster size , @xmath50 , is simply .
finite clusters are typically trees @xcite .
since there are links in a tree of size @xmath50 , we have the constraint @xmath51 in particular , minimal clusters with trivially satisfy this condition . as a result of the linking process
, clusters undergo the multivariate aggregation process @xmath52 with @xmath17 and @xmath18 . here
, the vector @xmath53 describes the degree augmentation @xmath54 , @xmath55 the aggregation rate @xmath56 in equals the product of the number of nodes of degree @xmath57 in the first cluster and the number of nodes of degree @xmath21 in the second cluster .
the density @xmath58 of clusters with `` size '' @xmath59 at time @xmath22 obeys the rate equation @xmath60 where @xmath24 is the density of active nodes . henceforth , the time dependence is implicit .
the gain term directly reflects the aggregation process , and the factor @xmath61 prevents over - counting .
the loss rate is proportional to the total number of active nodes in the cluster , as well as the density of active nodes in the entire system .
a crucial feature of the above rate equation is that the gain term and the loss term are different in nature .
the gain term , which describes creation of clusters with finite size , is proportional to the product of cluster densities .
the loss term , which describes loss of finite clusters due to linking , is proportional to the product of the cluster density and the total density of active nodes in the entire system , which does take into account the giant component , if one exists .
for this reason , the rate equation is valid in the non - percolating phase as well as in the percolating phase . in terms of the time variable @xmath31 ,
the loss term becomes linear @xmath62 the initial condition is @xmath63 .
the rate equations for the cluster size distribution are nonlinear , in contrast with the linear equations for the degree distribution
. moreover , the rate equations are substantially more challenging than those previously analyzed in studies of multivariate aggregation processes @xcite .
the generating function approach is generally useful in aggregation problems @xcite , and in our case , we must use the multivariate generating function @xmath64 here @xmath65 and @xmath66 is a shorthand for the product of monomials . multiplying by @xmath67 and performing the summation over @xmath59 , we find that the generating function evolves according to @xmath68 the initial condition is @xmath69
the generating function satisfies a non - linear partial differential equation ( pde ) with non - constant coefficients .
crucially , equation is a _ first - order _ pde , and hence , it can be reduced to a set of coupled ordinary differential equations ( odes ) by applying the method of characteristics @xcite .
this is a significant simplification as a pde is essentially a collection of infinitely many odes .
moreover , in the present case , we conveniently treat the governing equation as a _ hamilton - jacobi _ equation , thereby bypassing the method of characteristics @xcite .
then , the canonical hamilton equations coincide with the equations for the characteristics .
first , we convert into a hamilton - jacobi equation @xmath70 and view @xmath71 and @xmath72 as `` action '' and `` hamiltonian '' , respectively .
second , we treat @xmath73 as `` coordinate '' , and @xmath74 as `` momentum '' , ; in components , @xmath75 for all @xmath21 .
the hamiltonian is @xmath76 where for convenience , we introduced a set of auxiliary functions , @xmath77 for all @xmath21 .
we stress that the hamiltonian is not a conserved quantity because of the explicit dependence on time @xmath31 , which enters via @xmath78 .
the hamilton - jacobi equation is equivalent to the canonical hamilton equations @xmath79 hence , the non - linear partial differential equation reduces to @xmath80 coupled ordinary differential equations for the coordinates @xmath81 and the momenta @xmath82 . from and , the evolution equations for the momenta are @xmath83 the initial condition @xmath84 follows from @xmath85 and , and the boundary condition is @xmath86 . the structure of the _ linear _
equations for @xmath82 is identical to that of equations for @xmath87 , and the initial conditions are the same , too . by integrating equations recursively
, we find @xmath88 for @xmath18 , with function @xmath89 defined by @xmath90 or alternatively , @xmath91 . to determine @xmath92 , we rewrite the last equation in as @xmath93 , and therefore , @xmath94 . formally , the momenta @xmath82 resemble the degree distribution @xmath87 . using and , the evolution equations for the coordinates are @xmath95 the initial condition is @xmath96 .
one of the coordinates does not change , @xmath97 . we may integrate and solve for the final coordinate @xmath98 as a function of the initial coordinate @xmath99 . yet , as shown below , we need to solve the inverse problem find @xmath99 as a function of @xmath98 .
fortunately , the auxiliary functions defined in enable us to accomplish this task .
we first note that at time @xmath100 , these sums coincide with the initial coordinates , @xmath101 . using the evolution equations and
, we calculate the evolution equations for @xmath102 , @xmath103 when @xmath104 . here , the overdot is shorthand for time derivative , @xmath105 .
a similar calculation shows that @xmath106 is conserved , @xmath107 .
hence , the functions @xmath102 obey @xmath108 the initial condition is @xmath109 . since the coordinate @xmath110 is a constant , integration of these evolution equations is immediate , and using the initial condition , we find @xcite @xmath111 ( the lower expression holds up to @xmath112 ; indeed , using @xmath113 we recover @xmath114 . ) equations , together with the known momenta , specify the initial coordinates in terms of the final coordinates .
remarkably , we formally solved the inverse problem ! to complete the solution of the hamilton - jacobi equations , we must specify the function @xmath89 in . using the definition
, we calculate the second derivative of @xmath89 , @xmath115 here , we used @xmath116 , as follows from . by multiplying the above equation by @xmath24 , and by substituting and
, we find that @xmath89 satisfies a _ closed _ second - order ordinary differential equation , @xmath117 the initial conditions are @xmath118 and @xmath119 .
this equation is linear if and only if @xmath120 .
however , we still do not know @xmath121 because the initial coordinates are coupled to the function @xmath89 , according to . to find @xmath121
, we have to integrate equation , starting with trial initial conditions @xmath118 and , and find the root of the integral equation @xmath122 the second order ode and the integral equation specify the variable @xmath89 and hence , the momenta @xmath82 and the initial coordinates @xmath123 via the explicit solutions and .
we stress that the function @xmath124 depends on the coordinate @xmath98 . finally , we obtain the generating function by evaluating its complete derivative with respect to @xmath31 , @xmath125 by integrating subject to the initial condition @xmath126 , and recalling the conservation law
, we get @xmath127 this expression , supplemented by equations and for @xmath89 , as well as the explicit expressions for the momenta , constitutes a formal solution for the generating function . therefore , representation of the linking process as a multivariate aggregation process enables analytical treatment of the cluster - size distribution .
by definition , the size distribution @xmath128 describes finite connected components . however , the system may be in one of two phases a _ non - percolating _ phase in which finite components contain all the mass , and a _ percolating _ phase in which a giant component , containing a fraction @xmath129 of the entire mass , coexists with finite components .
we expect that the giant component emerges suddenly , at a finite time @xmath130 ( see appendix a ) .
the mass of the giant component @xmath129 follows directly from the momenta when @xmath131 . indeed , using
, we have @xmath132 or alternatively , @xmath133 , as follows from the conservation law @xmath134 .
hence , we substitute @xmath135 into equation , and obtain the evolution equation for @xmath89 , @xmath136 the initial conditions are @xmath137 and @xmath119 . also , when @xmath131 , the integral equation becomes @xmath138 with @xmath82 given by .
first , there is the trivial solution @xmath139 and @xmath140 for all @xmath21 , for which the momenta equals the degree distribution , @xmath141 .
for this solution , @xmath142 , and the system is in the finite component phase . when @xmath120 , the trivial solution holds at all times .
in the case @xmath6 the final state is merely a collection of @xmath143 dimers ; in the case @xmath7 , the final state includes only rings @xcite .
in both of these cases the system does not condense into a single component . when @xmath1 , there is a second nontrivial solution when @xmath144 or equivalently , @xmath145 . to find the nontrivial solution , we solved equation numerically , using adams method @xcite , starting with a trial initial condition @xmath118 and @xmath146 , and then found the root of the integral equation using the bisection method @xcite .
( conveniently , the integral equations may be converted into first - order differential equations @xcite . )
we then obtained the mass of the giant component using . to evaluate time , @xmath22 , and the average degree , @xmath147 , in terms of the modified time variable @xmath31 , we simply integrated @xmath148 and equation . for the case @xmath3
, we find the critical time that corresponds to @xmath149 when the giant component emerges , the average degree of a node is @xmath150 .
moreover , the critical density of links quoted in is obtained from @xmath151 .
thus , the fraction @xmath152 of successful linking attempts is @xmath153 generally , the percolation parameters characterizing the phase transition from the finite component phase to the giant component phase are nontrivial .
table i lists the percolation parameters for .
remarkably , the restriction on the degree of a node has a small effect on the percolation parameters already when @xmath154 .
moreover , as the maximal degree @xmath0 increases , the percolation parameters quickly converge to the classical random graph values .
figure [ fig - gt ] shows the mass of the giant component for and @xmath155 . at moderate times
, the quantity @xmath129 converges to the classical random graph behavior . yet , as will be shown in the next section , the long - time evolution of the regular random graph is much slower compared with the classical random graph . versus time @xmath22 .
results of the hamilton - jacobi theory are shown for @xmath3 , @xmath156 , @xmath157 .
also shown as reference , is the solution to eq . for the classical random graph , @xmath155,scaledwidth=45.0% ] .percolation parameters for @xmath158 and . listed
are the time , @xmath130 , the modified time , @xmath159 , the average degree , @xmath160 , the fraction of successful linking attempts , @xmath152 , and the density of active nodes , @xmath161 , when the giant component emerges . [ cols="<,<,<,<,<,<",options="header " , ]
we now turn to the large - time behavior .
this behavior largely follows from the degree distribution . we start with the crude estimate , @xmath162 , obtained by replacing the right - hand - side of with the dominant exponential term , @xmath163 .
then , we establish the corrections to this logarithmic behavior by keeping the leading term in @xmath164 from this equation , we find the relation @xmath165 when @xmath166 .
substituting this expression into gives the leading asymptotic behavior of the degree distribution @xmath167 for @xmath18 .
the densities of active nodes with degree @xmath168 decay algebraically with time , albeit with a logarithmic correction .
furthermore , the degree distribution is an increasing function of @xmath21 . since @xmath169 , the quantity @xmath24 exhibits the universal power - law decay @xmath170 the smaller the degree @xmath21 , the faster the decay of @xmath87 . in particular
, the density of isolated nodes exhibits the fastest decay , @xmath171 each isolated node represents a minimal cluster with size @xmath172 .
therefore gives the density of minimal clusters .
the asymptotic behavior is much slower than the fast exponential decay found in ordinary random graphs @xcite .
hence , the restriction on the degree leads to much slower long - time kinetics . the asymptotic results and provide useful information about the final stages of the evolution .
amongst all clusters , minimal clusters survive the longest . hence ,
when all minimal clusters disappear , a single cluster remains , and the giant component takes over the entire system . the time to reach a single component , @xmath11 , grows with @xmath2 according to @xmath173 as follows from the `` depletion '' criterion @xmath174 .
this time is not self - averaging : it fluctuates from realization to realization . in particular , the average @xmath175 does not give the second moment @xmath176 ; namely , @xmath177 this breakdown of self averaging is in sharp contrast with the behavior of ordinary random graphs where @xmath11 is a self - averaging quantity .
the conclusion is based on the observation that a slight variation in the depletion criterion , say @xmath178 , gives a different estimate for @xmath11 .
when the giant component contains all @xmath2 nodes , the vast majority of the nodes are inactive . yet
, a small majority of nodes remain active .
let @xmath179 be the average number of nodes with degree @xmath104 when the graph becomes fully connected for the first time @xcite .
this quantity grows logarithmically with system size ( figure [ fig - mj ] ) @xmath180 with @xmath181 , as follows from @xmath182 , along with equations and .
the majority of active nodes have the largest possible degree , @xmath183 . as stated in eq .
, the total number of active nodes , when the giant component first takes over the entire system , @xmath13 , is logarithmic . to test these predictions , we performed monte - carlo simulations . in the simulations ,
we start with @xmath2 isolated nodes . at each step
, we pick two _ active _ nodes and connect them .
keeping track of the active nodes maximizes computational efficiency , and allows us to simulate large systems with @xmath184 .
we performed simulations for the case @xmath3 , and measured the average number of nodes with degree @xmath9 and degree @xmath185 when the graph condenses into a single component for the first time ( figure [ fig - mj ] ) .
also , we verified that the time @xmath11 fluctuates from realization to realization by measuring the ratio in . and
@xmath186 as a function of system size @xmath2 , for the case @xmath3 .
the results are from an average over @xmath187 independent simulation runs.,scaledwidth=45.0% ] the asymptotic behavior indicates that there are several intermediate steps until the regular random graph fully forms .
isolated nodes disappear first , and at this time , there are @xmath188 active nodes .
nodes with degree @xmath9 disappear next , and at this point , @xmath189 active nodes remain .
there are @xmath0 such steps , and in the last step , nodes with degree @xmath190 disappear .
now , the regular random graph is complete . the time it takes to form the regular random graph is linear in system size , @xmath191 as follows from the depletion criterion @xmath192 , and the algebraic decay . like the quantity @xmath11 , the formation time @xmath193 is a fluctuating quantity , even in the limit . for completeness
, we also mention the asymptotic behavior of the cluster - size distribution , @xmath194 in the long - time limit ( @xmath7 is a special case of a model considered in @xcite ) . for minimal clusters , @xmath195 , and
this behavior follows from the density of isolated nodes .
the behavior follows from the master equation @xmath196 for @xmath197 . except for the loss rate , which is proportional to the overall density of active nodes , @xmath24 ,
this rate equation is the same as .
clusters with @xmath197 contain only active nodes , and hence , the loss term is proportional to cluster size , @xmath50 . starting with @xmath198 , the leading asymptotic behavior can be established recursively from equation . in general , we must classify clusters by their topology , and construct specific rate equations for each topology .
the resulting rate equations have the same overall structure as and as a consequence , the asymptotic behavior appears to hold also when @xmath199 .
the proportionality constants depend on cluster topology , however .
finally , we note that the criterion @xmath200 shows that the asymptotic behavior is relevant only in a limited range of sizes , @xmath201 with the logarithmic cutoff @xmath202 .
the cluster size distribution is sharply suppressed beyond the cutoff @xmath203 .
in conclusion , we studied the evolution of random graphs with a bounded degree using a dynamic linking process
. first , a giant component , which contains a macroscopic number of nodes , emerges in a finite time .
we obtained the nontrivial thresholds for this percolation transition .
then , the giant component takes over the entire system .
the evolution still proceeds as some nodes have yet to reach the maximal degree .
while the number of linking attempts before the giant component emerges is a self - averaging quantity , the number of steps it takes to form a fully connected graph or to complete the regular random graph are both fluctuating ( non - self - averaging ) quantities .
the regular random graph is completed via multiple steps . in the first step ,
nodes with degree @xmath8 disappear and at the point , the graph is fully connected .
then , nodes with degree @xmath9 disappear , etc .
the total number of nodes with degree @xmath190 diminishes by a factor @xmath204 in each step , @xmath205 similarly , the number of nodes of degree @xmath206 shrinks according to @xmath207 .
furthermore , the asymptotic behavior indicates a multitude of time scales .
clusters of size @xmath50 disappear at time @xmath208 for all @xmath201 .
the time @xmath209 is simply the largest in a series of time scales , @xmath210 .
therefore , multiple finite - size scaling laws characterize kinetics of regular random graphs .
the evolution equation for the cluster - size density is not a closed equation .
to address this challenge , we introduced the multivariate size distribution , where the number of nodes with a given degree is specified .
for this quantity , the evolution equations are closed . using the hamilton - jacobi method , also useful for analyzing large fluctuations in population dynamics @xcite , we constructed a formal solution for the multivariate size distribution .
moreover , the cumbersome rate equations reduce to a single second order differential equation , and numeric integration of this equations gives the percolation thresholds with , essentially , arbitrary accuracy .
our analysis shows that the rate equation description extends to a broader class of evolving random graphs .
certainly , the multivariate aggregation analysis can be generalized to situations where the linking rate is degree dependent .
the multivariate aggregation framework can also be used to study the structure of clusters @xcite , and in particular , cycles .
we thank wolfgang losert for useful discussions .
this research was supported by doe grant de - ac52 - 06na25396 and nsf grant ccf-0829541 . 99 r. solomonoff and a. rapaport , bull .
* 13 * , 107 ( 1959 ) .
p. erds and a. rnyi , publ .
sci . * 5 * , 17 ( 1960 ) .
b. bollobs , _ random graphs _ ( academic press , london , 1985 ) .
s. janson , t. uczak and a. rucinski , _ random graphs _
( john wiley & sons , new york , 2000 ) . s. janson , d. e. knuth , t. uczak , and b. pittel , rand .
* 3 * , 233 ( 1993 ) .
b. pittel , j. spencer , and n. wormald , j. combin .
b * 67 * , 111 ( 1996 ) .
d. stauffer and a. aharony , _ introduction to percolation theory _
( taylor & francis , london , 1992 ) .
g. grimmett , _ percolation _ ( springer , berlin , 1999 ) .
p. j. flory , j. amer .
soc . * 63 * , 3083 ( 1941 ) .
w. h. stockmayer , j. chem . phys . * 11 * , 45 ( 1943 ) .
j. flory , _ principles of polymer chemistry _ ( cornell university press , ithaca , 1953 ) .
p. grassberger , math . biosci . *
63 * , 157 ( 1983 ) .
m. e. j. newman , phys .
e * 66 * , 016128 ( 2002 ) .
t. tome and r. m. ziff , phys .
e * 82 * , 051921 ( 2010 ) .
s. s. wasserman , j. math .
soc . * 5 * , 61 ( 1977 ) .
m. e. j. newman , d. j. watts , and s. h. strogatz , proc .
sci . * 99 * , 2566 ( 2002 ) .
m. girvan and m. e. j. newman , proc .
sci . * 99 * , 7821 ( 2002 ) .
b. bollobs , c. borgs , j. t. chayes , j. h. kim , and d. b. wilson , rand .
. alg . * 18 * , 201 ( 2001 ) .
m. herrera , s. mccarthy , s. slotterback , e. cephas , w. losert , and m. girvan , phys . rev .
e * 83 * , 061303 ( 2011 ) . c. dwork , d. peleg , n. peppenger , and e. upfal , siam j. comp . * 17 * , 975 ( 1988 ) .
y. wang and x. y. li , mob .
net . appl . * 11 * , 161 ( 2006 ) .
e. ben - naim and p. l. krapivsky , phys .
e * 71 * , 026129 ( 2005 ) .
a. a. lushnikov , physica d * 222 * , 37 ( 2006 ) .
l. krapivsky , s. redner , and e. ben - naim , _ a kinetic view of statistical physics _
( cambridge university press , cambridge , 2010 ) . k. t. balinska and l. v. quintas , j. math . chem .
* 8 * , 39 ( 1991 ) .
a. ruciski and n. c. wormald , comb .
prob . comp .
* 1 * , 169 ( 1992 ) .
n. c. wormald , j. alg . * 5 * , 247 ( 1984 ) .
r. w. robinson and n. c. wormald , rand .
struct . alg .
* 5 * , 363 ( 1994 ) .
g. chartrand , _ introductory graph theory _ ( dover , new york , 1984 ) .
e. ben - naim and p. l. krapivsky , phys .
e * 83 * , 061102 ( 2011 ) .
e. ben - naim and p. l. krapivsky , unpublished ( 2011 ) .
m. von smoluchowski , z. phys .
* 17 * , 557 ( 1916 ) .
s. chandrasekhar , rev .
* 15 * , 1 ( 1943 ) .
wormald , in _ lectures on approximation and randomized algorithms _ , m. karoski and h.j .
prmel ( eds ) , pp .
pwn , warsaw , 1999 .
f. leyvraz , phys . rep . * 383 * , 95 ( 2003 ) .
more complicated clusters do occur , but they are very rare , e.g. in classical random graphs their total number scales logarithmically with system size , see e.g. @xcite .
p. l. krapivsky and e. ben - naim , phys .
e * 53 * , 291 ( 1996 ) .
t. matsoukas and c. l. marshall , epl * 92 * , 46007 ( 2010 ) .
r. m. ziff , m. h. ernst , and e. m. hendriks , j. phys .
a * 16 * , 2293 ( 1983 ) .
e. ben - naim and p. l. krapivsky , j. phys .
a * 38 * , l417 ( 2005 ) .
v. i. arnold , _ geometrical methods in the theory of ordinary differential equations _
( new york , springer - verlag , 1988 ) .
j. d. logan , _ an introduction to nonlinear partial differential equations _ ( hoboken , n.j . , wiley - interscience , 2008 ) .
i. m. gelfand and s. v. fomin , _ calculus of variations _
( englewood cliffs , n.j . : prentice - hall , 1963 ) . for a
_ fixed _ set of final coordinates @xmath81 , the initial coordinates @xmath123 satisfy the backward evolution equations .
f. bashforth and j. c. adams , _ theories of capillary action _
( cambridge university press , london , 1883 ) .
d. zwillinger , _ handbook of differential equations _
( academic press , london , 1989 ) . w. h. press , s. a. teukolsky , w. t. vetterling , and b. p. flannery , _ numerical recipes _
, ( cambridge university press , cambridge , 1992 ) .
the density @xmath211 is twice differentiable at the percolation point ; the third derivative undergoes a jump @xcite .
h. e. stanley , _ introduction to phase transitions and critical phenomena _ ( oxford university press , new york , 1971 ) .
s. redner , _ a guide to first - passage processes _ ( cambridge university press , new york , 2001 ) .
m. mobilia , p.
l. krapivsky , and s. redner , j. phys .
a * 36 * , 4533 ( 2003 ) .
m. i. freidlin and a. d. wentzell , _ random perturbations of dynamical systems _ , ( springer - verlag , new york , 1998 ) . m. i. dykman , e. mori , j. ross , and p. m. hunt , j. chem . phys . * 100 * , 5735 ( 1994 ) .
v. elgart and a. kamenev , phys .
e * 70 * , 041106 ( 2004 ) .
a. kamenev and b. meerson , phys .
e * 77 * , 061107 ( 2008 ) .
e. ben - naim and p. l. krapivsky , j. phys .
a * 37 * , l189 ( 2004 ) .
r. m. ziff , e. m. hendriks , and m. h. ernst , phys .
lett . * 49 * , 593 ( 1982 ) .
this appendix briefly describes how to obtain relevant properties of the classical random graph , @xmath155 , using the hamilton - jacobi formalism .
the density of connected components with size @xmath50 obeys @xcite @xmath212 with the initial condition @xmath213 .
the generating function @xmath214 satisfies @xmath215 with the initial condition @xmath216 .
using the momentum @xmath217 , the hamiltonian is @xmath218 the hamiltonian does not depend explicitly on time .
hence , @xmath219 is a conserved quantity and consequently , @xmath220 is also constant .
the evolution equations are @xmath221 with the initial conditions @xmath222 and @xmath223 .
we can verify that @xmath224 , and therefore , @xmath225 . by integrating ,
we obtain the coordinate and the momentum , @xmath226 the evolution equation for generating function is @xmath227 hence , the generating function is a quadratic function of the initial coordinate , @xmath228 with @xmath229 .
the mass of the giant component , @xmath230 , follows from the momentum @xmath231 at @xmath232 .
the conservation law @xmath225 implies that @xmath233 if @xmath232 .
in analogy with , we have @xmath234 , and using , we find a closed equation , @xmath235 there is a trivial solution @xmath142 , valid at all time , and a second nontrivial solution , @xmath236 , valid only when @xmath237 .
the latter is physical when @xmath237 , and hence , the giant component emerges at time @xmath238 . to obtain the critical size distribution ,
that is , the size density at @xmath239 , we note that the critical generating function is given by with @xmath240 .
we then focus on the behavior for @xmath241 .
we substitute the expansion into the equation @xmath242 , and solve for the coefficients @xmath243 and @xmath244 .
this calculation gives @xmath245^{1/2}-\frac{2}{3}(1-x)+\cdots.\ ] ] this expansion , together with eq .
, gives the leading behavior of the critical generating function , @xmath246 in the limit @xmath241 .
thus , the critical size distribution has the power - law tail @xcite @xmath247 as follows from the leading singular component @xmath248 . | we investigate the dynamic formation of regular random graphs . in our model , we pick a pair of nodes at random and connect them with a link if both of their degrees are smaller than @xmath0 . starting with a set of isolated nodes ,
we repeat this linking step until a regular random graph , where all nodes have degree @xmath0 , forms .
we view this process as a multivariate aggregation process , and formally solve the evolution equations using the hamilton - jacobi formalism .
we calculate the nontrivial percolation thresholds for the emergence of the giant component when @xmath1 . also , we estimate the number of steps until the giant component spans the entire system and the total number of steps until the regular random graph forms .
these quantities are non self - averaging , namely , they fluctuate from realization to realization even in the thermodynamic limit . |
nonequilibrium thermal fluctuations for a small system in contact with heat bath in equilibrium become very large and exhibit interesting properties that are universal over different systems .
the underlying principle for the universal properties is the fluctuation theorem ( ft ) , which was first discovered for a deterministic system thermostatted so as to conserve kinetic energy @xcite .
the ft was later proven to hold for a wide class of stochastic systems @xcite .
it deals with the thermal fluctuations of thermodynamic quantities produced persistently in time , such as work and heat in a nonequilibrium process . a typical form of the ft is given by @xmath0 , where @xmath1 is the rate of work production piled over a period @xmath2 in nonequilibrium dynamics driven by a nonconservative force acting on the system initially prepared in equilibrium with heat bath .
@xmath3 is the inverse temperature @xmath4 for the boltzmann constant @xmath5 and the temperature @xmath6 of the heat bath .
the bracket denotes the average over all possible fluctuations . to investigate the ft and related issues , it is required to accurately deal with the probability distribution of the thermal fluctuations in a nonequilibrium process .
work and heat are path - dependent quantities . the thermal fluctuations of such quantities arise from many different paths ( trajectories ) along which the system evolves in time .
one essentially needs the ensemble average over all paths from the past to the present , unlike the usual ensemble average over all states at a certain time .
the path integral theory to deal with trajectory - dependent fluctuations was developed by onsager and mathlup based on the langevin equation @xcite , and was used successfully to prove the ft @xcite . in the experimental side ,
the fluctuations around the average value of a thermodynamic quantity for a small system become so predominant that they are measurable with high accuracy by using modern technologies .
interesting experiments were carried out to confirm the ft for various systems such as a colloidal particle in a moving optical trap @xcite , a molecule in the afm pulled by an external force @xcite , an electrical circuit driven by a small current @xcite , a harmonic oscillator under an external force @xcite , an rna molecule unfolded and refolded by optical tweezers @xcite , a rotating motor protein @xmath7-atpase @xcite , and a colloidal particle in breathing harmonic potential @xcite .
an alternative approach to the study on large fluctuations for a small system is molecular dynamics ( md ) .
md deals with both a small system and a heat bath consisting of an extremely high number of molecules by taking into account the interactions in detail among them , and it mimics an experiment as realistically as possible .
it plays the role of the bridge between experiment and phenomenological theory such as the langevin dynamics . the first md study in this field
was carried out accompanying an experiment for a colloidal particle in a moving optical trap @xcite .
recently , md studies were done to confirm the ft for non - equilibrium motions such as the effusion of an ideal gas through a hole between two compartments @xcite and the motion of a particle in a moving billiard @xcite . in our work
, we investigate via md simulation the motion of a colloidal particle in a liquid driven out of equilibrium by an external non - conservative force generating torque .
such external torque was found to produce non - zero circulating current even in steady state , characterizing nonequilibrium steady state @xcite .
energetically , it was found to produce work and heat incessantly in time @xcite .
such system was suggested as a heat engine @xcite and investigated as a model system to examine optimal efficiency for maximum power @xcite .
we examine and confirm various nonequilibrium properties including the ft .
we compare the results from the md simulation with those found from the langevin dynamics @xcite .
the particles employed in the md are composed of molecules in liquid and the colloidal particle immersed in them
. the liquid plays the role of a heat bath kept in equilibrium at a fixed temperature @xmath6 .
liquid molecules are designed to interact with each other via the lennard
jones ( lj ) potential : @xmath8~ , \label{lj_potential}\ ] ] where @xmath9 and @xmath10 define the effective diameter ( radius of cross section ) of a pair of molecules and the intensity of interactions , respectively . to maintain the liquid in equilibrium
, we use the langevin thermostat in which the same sort of dissipative and fluctuating forces are used as in the langevin equation @xcite .
then , the equation of motion of the @xmath11-th molecule with mass @xmath12 for position @xmath13 and velocity @xmath14 are given as @xmath15 where @xmath16 is the position of the colloidal particle .
for the lj potential between the colloidal particle and a liquid molecule , a different parameter @xmath17 is used with the same @xmath10 .
@xmath17 is the radius of cross section for a pair of the colloidal and a liquid molecule .
then , @xmath18 is the diameter of the colloidal particle .
the last two terms in the first line are given from the langevin thermostat .
the fluctuating force @xmath19 is white noise with mean zero and the variance given by @xmath20 for @xmath21 denoting the components in 3 dimensions .
the strength of the white noise , @xmath22 , relates the dissipation coefficient @xmath23 and the inverse temperature @xmath3 of the heat bath .
this relation is called the einstein relation or the fluctuation - dissipation relation , which guarantees that in the absence of the last term in eq .
( [ reservoir ] ) , the molecules reach an equilibrium with the boltzmann distribution associated with unperturbed energy @xmath24 , which is given by @xmath25 where the subscript @xmath26 denotes the pair of molecules @xmath11 and @xmath27 . even in the presence of the interaction with the colloidal particle , the molecules approximately maintain equilibrium since the interaction force gives @xmath28-contribution to the dynamics while @xmath29 .
this happens in experiments where the temperature of the liquid is kept well under the brownian motion of a colloidal particle in the liquid .
this can also be confirmed in md simulations if the period of measurement is not too long .
the equation of motion of the colloidal particle with the mass @xmath30 immersed in the liquid for position @xmath16 and velocity @xmath31 is given as @xmath32 where @xmath33 is an applied force acting exclusively on the colloidal particle , which is possible if an electromagnetic force is applied to a charged colloidal particle in an electrically neutral liquid .
@xmath34 is the sum of interaction forces exerted by molecules .
it plays the equivalent role of dissipating and fluctuating forces in the langevin dynamics .
heat is defined as work done by this force acting by molecules .
we mimic an optical trap by harmonic force @xmath35 with stiffness @xmath36 , which is applied to confine the colloidal particle .
choosing @xmath37-axis to be perpendicular to the surface of the liquid , one can consider a non - conservative and linear force @xmath38 in horizontal direction to @xmath39-@xmath40 plane .
then the total applied force in horizontal direction is given by @xmath41 @xmath33 becomes non - conservative for @xmath42 , which is the source for nonequilibrium .
it yields torque to produce nonzero circulation current even in the steady state .
divergence - less circulation current maintaining probability distribution is an important characteristics for nonequilibrium steady state @xcite .
energetically , nonzero current produces work and heat incessantly in time even in steady state @xcite . in reality
there is a confining force in @xmath37-direction between the interfaces at top and bottom of the liquid .
however , it usually depends on @xmath37 independent of @xmath33 , giving only simple equilibrium relaxation in @xmath37-space . for simplicity
, we use the same harmonic force @xmath43 . non - equilibrium motion due to this @xmath33 was studied via the overdamped langevin equation and many interesting properties beyond the ft were found @xcite , with which we will compare our md simulation results .
experimentally , this kind of non - conservative force can be induced by time - dependent magnetic field @xmath44 .
the resultant force is @xmath45 for charge @xmath46 and vector potential @xmath47 . for constant @xmath48 ,
the induced force is given by @xmath49 in the overdamped limit with large friction coefficient @xmath50 of the colloidal particle and short - time ( or small @xmath48 ) limit , the first term can be neglected compared to effective dissipating force @xmath51 in the regime @xmath52 . in this regime , @xmath53 .
in our work , we will carry out md simulations for the non - equilibrium motion due to @xmath54 in the part of eq .
( [ f_app ] ) and confirm the results from the overdamped langevin dynamics .
the central quantities characterizing non - equilibrium motion are work @xmath55 done on the colloidal particle and heat @xmath56 flowing into the heat bath , which are produced incessantly in time .
the production rates of the two quantities are given by @xmath57 the first law of thermodynamics is given by @xmath58 where @xmath59 .
we expect our study to serve as an alternative approach compared to the langevin dynamics and to provide a basis to extend to more general cases for underdamped motion and long - time regime with @xmath60 or larger .
to compare the results from md simulations with those from the overdamped langevin dynamics , we first estimate the friction coefficient @xmath50 from simulations , which is a relative quantity between the colloidal particle and the liquid . in the following
, we will present the simulation set - up to prepare a viscous liquid and a colloidal particle with large mass and size maintaining the overdamped limit , and how to estimate @xmath50 between the two systems . in our simulations , the liquid molecules and the colloidal particle
are initially placed in a simulation box of size @xmath61 .
periodic boundary conditions are imposed . since the lj
potential is long - ranged , it will cost an enormous running - time to sum all the interactions .
instead , we set a cut - off distance @xmath62 , above which the lj potential can be truncated with negligible correction compared to thermal energies . we adopt the parameters from the well - established md studies on the lj liquids @xcite .
the particle number density is set by @xmath63 and the temperature by @xmath64 , which was found to represent a compressed liquid above the melting temperature .
@xmath65 was found for these parameters .
we also use @xmath66 for the interactions between the colloidal particle and a liquid molecule .
we set the size of @xmath67 and about @xmath68 molecules are employed in simulations .
we introduce dimensionless variables and parameters used for the simulation .
we rescale @xmath69 , @xmath70 , @xmath71 , @xmath72 .
then @xmath12 , @xmath10 , and @xmath9 are set to unity , and @xmath73 to @xmath17 .
given the above choice of temperature , @xmath74 goes to @xmath75 . in this rescaling
, @xmath76 goes to @xmath77 and @xmath78 to @xmath79 .
we also change @xmath80 , @xmath81 .
@xmath82 and @xmath83 are set in the unit of @xmath84 as @xmath36 .
the white noises having the unit of @xmath85 are rescaled as @xmath86 .
all changed variables and parameters become dimensionless . in the following sections
, we will use the dimensionless variables and parameters , if not mentioned otherwise .
@xmath23 is not a real coefficient of the friction exerted in the liquid , but a mathematical parameter of the langevin thermostat .
then , we can choose @xmath87 as long as the temperature of the heat bath is fixed from the einstein relation , so the random force is written as @xmath88 with @xmath89 in eq .
( [ reservoir ] ) .
the experimental condition in most cases is consistent with the overdamped limit where the inertial effect is negligible . our lj - liquid itself
is expected to have a large friction coefficient . however
, a colloidal particle with lager size than that of liquid molecules will experience a much larger friction obeying the stokes law : @xmath90 for the viscosity @xmath91 and the radius of the colloidal particle @xmath92 . in the md simulation ,
liquid is not continuous but discrete , so we expect @xmath93 where @xmath17 is the radius of the cross section of a pair of the colloid and a liquid particle .
there is restriction on the increase of @xmath17 in simulations . under periodic boundary condition
, the box size must be large enough to avoid possible unrealistic hydrodynamic interactions between the colloidal particle and its periodic replicas .
a working criterion is @xmath94 , otherwise the chance of the colloidal particle interacting molecules near the boundary becomes too high . for @xmath67 used in our simulation , @xmath95 is about the maximal value we can obtain . from the point of view based on the langevin dynamics , the equation of motion of the colloidal particle corresponding to eq .
( [ colloidal ] ) is given as @xmath96 , where @xmath50 is the coefficient of the friction exerting on the colloidal particle in a viscous liquid and @xmath97 is the same gaussian noise as in eq .
( [ reservoir ] ) . before investigating nonequilibrium motion
, one can prepare the simulation set - up for the overdamped limit in an equilibrium situation .
we consider an equilibrium case for @xmath98 . in the overdamped limit , the fast - varying velocity is expected to relax more rapidly than the slowly - varying position .
one can find the relaxation times of the velocity and position are given by @xmath99 and @xmath100 , respectively .
then , the criterion for the overdamped limit is given by @xmath101 , which can be achieved either by small @xmath30 or by large @xmath50 .
in reality , @xmath30 is very large compared to the mass of liquid molecules , so @xmath50 should be chosen large enough for the overdamped limit .
we can estimate the friction coefficient @xmath50 of the colloidal particle from the correlation function for the position in time which can be derived from the overdamped langevin dynamics as @xmath102 which becomes @xmath103 for @xmath89 and @xmath104 .
it is a special case with @xmath105 for the general formula derived in eq .
( [ c11 ] ) .
we repeat the simulations for @xmath106 . figure .
[ fig1 ] confirms that @xmath107 from the simulations fits the above theoretical equation very accurately and the estimated @xmath50 s satisfy the stokes law .
@xmath50 ranges from @xmath108 to @xmath109 for @xmath106 and meets the criterion for the overdamped limit , @xmath110 , for @xmath111 used to mimic the large mass of the colloidal particle . in eq .
( [ auto_corr_eq ] ) for @xmath104 , @xmath105 for various values of @xmath17 show good agreements with @xmath112 for fitted values of @xmath50 .
( b ) the plot for @xmath50 versus @xmath17 confirms the stokes law well . ]
we run simulations in discrete time steps with an interval @xmath113 and take @xmath114@xmath115 samples produced by random noises in the langevin thermostat and random initial conditions .
the motion in @xmath37-direction is a simple equilibration process in an harmonic trap .
the nonequilibrium motion occurs in @xmath39-@xmath40 plane where the applied force @xmath116 is given from eq .
( [ f_app ] ) where @xmath117 the dynamics is shown to be stable so as to reach a new steady state if @xmath118 is positive - definite @xcite , i.e. , @xmath119 . if @xmath118 is symmetric with @xmath120 , the force is conservative .
then , the steady - state distribution is boltzmann associated with the potential energy @xmath121 . if @xmath118 is asymmetric with @xmath42 , the force is nonconservative and the colloidal particle reaches a nonequilibrium steady state .
we use @xmath104 , @xmath111 , and @xmath95 .
first , we carry out simulations with @xmath105 for a period much longer than the relaxation time @xmath122 in order for the colloidal particle to reach an equilibrium state .
then , we turn on a nonequilibrium protocol , @xmath42 , which drives the colloidal particle out of equilibrium . first , we examine the correlation function for the position in time and compare to the formula from the overdamped langevin dynamics , given as @xmath123~ , \label{auto_corr_neq}\ ] ] which is derived in detail in eq .
( [ c11 ] ) . since the colloidal particle is initially equilibrated for @xmath124 , @xmath125 .
figure [ fig2 ] shows a good agreement of the md simulation with the above equation .
we estimate @xmath126 for @xmath95 , which agrees with the estimation from the simulation in equilibrium shown in fig .
[ fig1](b ) . in nonequilibrium state for @xmath89 , @xmath104 , @xmath127 , and @xmath95 .
the plot from the md simulation agrees with that from eq .
( [ auto_corr_neq ] ) .
then , we estimate @xmath128 . ]
thermodynamic quantities characteristic to nonequilibrium are work and heat accumulated in time , which are expected to be produced persistently .
the work and heat productions are found as @xmath129 where @xmath130 ( @xmath131 ) is the work ( heat ) rates at time step @xmath132 , which are obtained from eqs .
( [ work ] ) and ( [ heat ] ) .
the average values are found from @xmath114-@xmath115 samples .
figure [ fig3 ] shows the average rate of heat production in time for equilibration process where there is no work production for @xmath105 .
as expected , it decays as the system approaches equilibrium and there is no persistent production of heat . and @xmath133 .
@xmath134 goes to zero as time goes larger than the relaxation time @xmath135 .
@xmath50 is given from the data in fig .
[ fig2 ] . ]
figure [ fig4 ] shows the average rates of work and heat production in time for @xmath127 .
time step @xmath132 is started after the simulation with @xmath105 is performed so that the colloidal particle is initially in equilibrium .
the nonequilibrium steady state is characterized by the incessant productions of work and heat . since @xmath136 does not change in the steady state , the work and heat production become the same asymptotically for large @xmath2 . in fig .
[ fig4 ] , we observe that @xmath137 and @xmath138 converge to the same value as the system approaches steady state . from a recent study via the overdamped langevin equation @xcite
, it was found that @xmath139 ^ 2 \beta^{-1}/(k\gamma_c)$ ] as @xmath140 . in our dimensionless units , it is equal to @xmath141 for @xmath142 , @xmath95 , and @xmath89 . from the data in fig .
[ fig4 ] , we estimate @xmath143 .
all of the estimations of @xmath50 from different methods presented in figs .
[ fig1 ] , [ fig2 ] , and [ fig4 ] agree very well . and @xmath138 , for @xmath111 , @xmath95 , @xmath89 , and @xmath127 .
the two rates approach the same value for large @xmath2 .
the converging value is approximately @xmath144 and the estimate value of @xmath50 is equal to 29 , which agrees very well with the values in fig .
[ fig1](b ) and fig .
[ fig2 ] . ] . [ fig4 ]
the repeated measurements of work and heat produced during a period @xmath2 present large fluctuations over samples around the average values .
the ft is a mathematical principle about the distribution functions for such fluctuations of thermodynamics quantities accumulated in time .
it is convenient to consider the rates of work and heat , @xmath145 and @xmath146 , since @xmath147 and @xmath148 increase linearly in large @xmath2 , as shown in fig .
[ fig4 ] .
we obtain the distribution functions for @xmath1 and @xmath46 , given by @xmath149 , and @xmath150 , respectively , from @xmath114@xmath115 samples of the md simulations .
figure [ fig5 ] shows the distribution functions for @xmath1 and @xmath46 .
the two distributions show clear difference even for large @xmath2 where @xmath151 becomes equal to @xmath152 .
the thermodynamic second law reads : @xmath153 where @xmath154 is the change in total entropy of the system and heat bath and @xmath155 the change in system entropy , which leads to @xmath156 in steady state . in thermodynamic limit ,
since thermal fluctuations are negligible , the statistical average is not necessary .
however , for small systems with large fluctuations , the second law can be shown to hold only in the average sense . from the ft for total entropy production : @xmath157 , one can get @xmath158 using the schwarz inequality .
this means that the second law can be violated in individual events . in fig .
[ fig5 ] , the probability of @xmath159 or @xmath160 does not vanish in long - time limit . .
the two distributions show clear difference even for large @xmath2 .
the non - vanishing distribution for @xmath159 or @xmath160 for large @xmath2 shows that the thermodynamic second law can be violated in individual events . ] in previous works @xcite , the distribution function @xmath161 for @xmath1 was extensively investigated by means of the generating function @xmath162 where the bracket denotes the average over @xmath1 . the so - called gallavotti
cohen ( gc ) symmetry was confirmed as @xmath163 , which is known to be another representation of the ft .
@xmath164 was found to be finite only for @xmath165 and @xmath166 otherwise where @xmath167 and @xmath168 . from the gc symmetry , @xmath169 .
it was found that the divergence of @xmath164 at @xmath170 determines an exponential decay in the tails of the work distribution .
it was assumed that @xmath171 then , near @xmath172 , the dominant contribution to @xmath164 comes from the integral in positive branch of @xmath1 , given as @xmath173 the same divergence of @xmath164 near @xmath174 can be obtained from the divergent integral of @xmath161 in negative @xmath1 .
the three types of tail behaviors were found as @xmath175 depending on the form of @xmath118 .
our study belongs to the first type , while the other two occur for unequal diagonal elements of @xmath118 , corresponding to an anisotropic optical trap .
we only restrict ourselves to isotropic optical trap and expect eq .
( [ tail ] ) .
figure [ fig6 ] shows a clear view of the exponential decay of the distribution function for work , which is an indicator of the first type .
the estimated values of @xmath176 are shown to confirm the expected gc symmetry . for various @xmath2
show exponential decays in both tails . in the inset
, the table for the values of @xmath176 from the slopes of the tails for various @xmath2 shows the gallavotti cohen symmetry , @xmath177 . ]
all the theoretical derivations of the ft are based on the langevin equations or the master equations .
we expect the ft to hold in the md simulations as good as in the real experiments @xcite .
we examine the ft for work , which can be expressed in either integral form as @xmath178 or detailed form as @xmath179 in this work , we examine the detailed ft . in fig .
[ fig7 ] , we plot @xmath180 $ ] , which is expected be @xmath1 .
the figure shows the ft for work holds for all times .
$ ] for @xmath181 .
the examination of the ft for ( a ) the work production and ( b ) heat production from the md simulations . in the panel ( a ) ,
the data points are in good agreement with @xmath182 , which confirms the ft for work . in the panel ( b )
, the guide lines , drawn using the fitting function @xmath183 , show that the ft for heat does not hold . ]
we also examine the ft for heat .
it was found that the ft for a certain time - accumulated quantity holds only if a proper initial distribution is given @xcite .
for the case of work , the initial distribution should be boltzmann in equilibrium , as we prepare in md simulations . for heat to present the ft
, the initial distribution should be uniform ( @xmath166-temperature distribution ) , which is hardly achievable in experiments .
in steady state as @xmath2 goes to @xmath166 , heat and work grow as @xmath2 while the energy difference remains finite .
moreover , the effect of the initial distribution decays exponentially in time .
therefore , one might expect that the ft for heat will approximately hold as @xmath2 goes large .
however , equivalence between heat and work is only true in average sense . in fig .
[ fig5 ] , the two distributions for large @xmath2 present clear difference in spite of the same average .
the two path - dependent quantities are related by the first law as @xmath184 . by rare chances ,
however , @xmath185 can be comparable to work and heat for unbound state with @xmath186 , as in our case .
if an initial ( final ) state is found from the tail with large energy far from the center of the distribution , @xmath185 can be negatively ( positively ) large so that it can significantly affect the positive ( negative ) tail of the heat distribution .
recent works revealed that the initial memory ever lasts in the positive tail of the heat distribution @xcite .
this so - called boundary effect on the heat distribution due to @xmath185 was investigated in recent works @xcite . indeed ,
figure [ fig7](b ) shows the violation of the ft for heat even if @xmath2 increases . for large @xmath2
, the slope of @xmath187 $ ] near the center of the distribution goes to unity while the tails remain nonlinear , as shown for the other nonequilibrium system @xcite .
we investigate the motion of a colloidal particle moving in a liquid driven by a nonconservative force producing a torque by using md simulations .
the liquid composed of many molecules are designed to play the role of a heat bath at a fixed temperature by using the langevin thermostat . the colloidal particle and the liquid molecules are designed to interact with each other via the lj potentials .
we mimic an experiment in the overdamped limit due to a large friction .
we assign larger size to the colloidal particle than that of the liquid molecules .
we estimate the values of the friction coefficient @xmath188 from three methods : the equilibrium and nonequilibrium correlation functions for position in time and the asymptotic production rate of work for infinite @xmath2 .
the three values are found to be identical with high accuracy and to hold the stokes law , which assures us of the reliability of our md simulations . after running the simulations for a sufficiently long time in the absence of nonconservative force for the colloidal particle to reach equilibrium
, we turn on a nonconservative force with @xmath42 and perform measurements for work and heat produced for period @xmath2 . from @xmath114@xmath115 samples due to random numbers used for the langevin thermostat and random initial conditions , we obtain the distribution functions for work and heat .
we observe that the second law of thermodynamics can be violated in individual events while satisfied in average .
the distribution functions for work for various @xmath2 are found to decay exponentially in the both tails of the distribution .
we confirm the gallavotti
cohen symmetry inherent in the tails .
we also confirm the detailed ft to hold for work , but not to hold for heat , as expected for nonequilibrium processes starting initially from equilibrium .
we discuss the boundary effect on the heat distribution to yield the violation of the ft for heat .
we are interested in the md simulations for long - time regime with @xmath189 or larger that yields a nonvanishing time - dependent lorentz force in addition to the torque - driving force in this work
. it will be interesting to investigate the nonequilibrium motion beyond the overdamped limit by increasing the mass of the colloidal particle or decreasing the friction coefficient and other examples driven by different nonequilibrium sources .
this work was supported by the basic science research program through the nrf grant no .
2013r1a1a2011079 ( ck ) and 2015r1d1a1a09057469 ( yj ) , and also by the korea institute of science and technology information with supercomputing resources including technical support ( ksc-2015-c1 - 001 ) .
the equation of motion of the colloidal particle in the overdamped limit is given by the langevin equation : @xmath190 where @xmath191 and @xmath192 for @xmath193 .
let @xmath194 be the correlation matrix with components @xmath195 .
then , one can find @xmath196 by multiplying eq .
( [ overdamped_langevin ] ) by @xmath197 , where @xmath198 is used .
the solution is given by @xmath199 .
@xmath118 has two eigenvalues @xmath200 and orthonormalized left ( right ) eigenvectors @xmath201 ( @xmath202 ) given as @xmath203 then , one can show @xmath204 with the property @xmath205 . using this property
, one can obtain @xmath206 , which gives @xmath207 where @xmath208 and @xmath209 . in our work
, we consider nonequilibrium protocols @xmath210 which are turned on at @xmath211 as the system is initially in equilibrium for @xmath105 with @xmath212 and @xmath213 .
in particular , we consider the case for @xmath214 and find @xmath215,\label{c11}\\ c_{12}(t)&=&c_{11}(0)\sqrt{\left|\frac{\kappa_1}{\kappa_2}\right|}e^{-\gamma_c^{-1}kt}\sin\left[\frac{\sqrt{-\kappa_1\kappa_2}}{\gamma_c}t\right],\\ c_{21}(t)&=&c_{11}(0)\sqrt{\left|\frac{\kappa_2}{\kappa_1}\right|}e^{-\gamma_c^{-1}kt}\sin\left[\frac{\sqrt{-\kappa_1\kappa_2}}{\gamma_c}t\right],\end{aligned}\ ] ] with @xmath216 .
we examine @xmath217 by md simulation for an equilibrium case with @xmath105 and a nonequilibrium case with @xmath218 , given in eq .
( [ auto_corr_eq ] ) and eq .
( [ auto_corr_neq ] ) , respectively .
99 g. gallavotti and e. g. d. cohen , phys . rev .
lett . * 74 * , 2694 ( 1995 ) ; j. stat . phys .
* 80 * , 931 ( 1995 ) . c. jarzynski , phys .
* 78 * , 2690 ( 1997 ) ; phys .
e * 56 * , 5018 ( 1997 ) .
l. onsager and s. machlup , phys . rev . *
91 * , 1505 ( 1953 ) ; s. machlup and l. onsager , _ ibid . _ * 91 * , 1512 ( 1953 ) .
g. m. wang , e. m. sevick , e. mittag , d. j. searles , and d. j. evans , phys .
* 89 * , 050601 ( 2002 ) .
e. h. trepagnier , c. jarzynski , f. ritort , g. e. crooks , c. j. bustamante , and j. liphardt , proc .
sci . * 101 * , 15038 ( 2004 ) .
g. hummer and a. szabo , proc .
sci . * 98 * , 3658 ( 2001 ) .
n. garnier and s. ciliberto , .
e * 71 * , 060101 ( 2005 ) .
f. douarche , s. joubaud , n. b. garnier , a. petrosyan , and s. ciliberto , phys .
. lett . * 97 * , 140603 ( 2006 ) .
s. joubaud , n. b. garnier and s. ciliberto , europhys . lett . * 82 * , 30007 ( 2008 ) .
j. liphardt , s. dumont , s. b. smith , i. tinico jr , and c. bustamante , science * 296 * , 1832 ( 2002 ) .
d. collin , f. ritort , c. jarzynski , s. b. smith , i. tinoco , jr and c. bustamante , nature * 437 * , 231 ( 2005 ) .
k. hayashi , h. ueno , r. iino , and h. noji , phys .
lett . * 104 * , 218103 ( 2010 ) .
d. y. lee , c. kwon , and h. k. pak , phys .
lett . * 114 * , 060603 ( 2015 ) . b. cleuren and c. van den broeck , phys .
e * 74 * , 021117 ( 2006 ) .
m. schmick , q. liu , q. ouyang , and m. markus , phys .
e * 76 * , 021115 ( 2007 ) . c. kwon , p. ao , and d. thouless , proc .
* 102 * , 13029 ( 2005 ) . c. kwon , j. d. noh , and h. park , phys .
e * 83 * , 061145 ( 2011 ) . j. d. noh , c. kwon , and h. park , phys .
* 111 * , 130601 ( 2013 ) .
r. fillinger and p. riemann , ohys .
lett . * 99 * , 230602 ( 2007 ) .
m . park , j. s. lee , and j. d. noh , phys .
rev e * 93 * , 032146 ( 2016 ) ; j .-
park , h - m .
chun , and j. d. noh , phys .
rev e * 94 * , 012127 ( 2016 ) .
g. s. grest and k. kremer , phys .
rev . a * 33 * , 3628(r ) ( 1986 ) . c. kwon , j. d. noh , and h. park , phys .
e * 88 * , 062102 ( 2013 ) .
p. a. thompson and m.o .
robbins , phys .
a * 41 * , 6830 ( 1990 ) ; p. a. thompson and s. m. troian , nature * 389 * , 360 ( 1997 ) . m. cieplak , j. koplik , and j. r. banavar , phys .
lett . * 86 * , 803 ( 2001 ) .
t. speck and u. seifert , j. phys .
a * 38 * , l581 ( 2005 ) ; u. seifert , phys .
* 95 * , 040602 ( 2005 ) .
m. esposito and c. van den broeck , phys .
. lett . * 104 * , 090601 ( 2010 ) .
j. farago , j. stat . phys . * 107 * , 781 ( 2002 ) ; physica a * 331 * , 69 ( 2004 ) .
r. van zon and e. g. d. cohen , phys .
e * 67 * , 046102 ( 2003 ) ; phys .
lett . * 91 * , 110601 ( 2003 ) .
a. puglisi , l. rondoni , and a. vulpiani , j. stat .
p08010 ( 2006 ) .
j. d. noh and j .- m .
park , phys .
* 108 * , 240603 ( 2012 ) .
j. s. lee , c. kwon , and h. park , phys .
e * 87 * , 020104(r ) ( 2013 ) ; j. stat .
p11002 ( 2013 ) .
k. kim , c. kwon , and h. park , phys .
e * 90 * , 032117 ( 2014 ) . | we investigate the motion of a colloidal particle driven out of equilibrium by an external torque .
we use molecular dynamics simulation as an alternative to the langevin dynamics .
we prepare a heat bath composed of thousands of particles interacting with each other through the lennard - jones potential and impose the langevin thermostat to maintain heat bath in equilibrium .
we consider a single colloidal particle to interact with the particles of the heat bath also by the lennard - jones potential , without applying any types of dissipative and fluctuating forces used in the langevin dynamics .
we set up simulation protocol fit for the overdamped limit as in real experiments by increasing the size and mass of the colloidal particle .
we study nonequilibrium fluctuations for work and heat produced incessantly in time and compare results with those obtained from the previous studies via the overdamped langevin dynamics .
we confirm the gallavotti - cohen symmetry and the fluctuation theorem for the work production . |
weakly interacting massive particles ( wimps ) are a promising dark matter candidate as they are generically produced in the early universe with roughly the right density . furthermore supersymmetry ( susy ) provides a well - motivated concrete wimp candidate in the form of the lightest neutralino ( for reviews see e.g. refs . ) .
wimps can be directly detected in the lab via their elastic scattering off target nuclei @xcite .
numerous direct detection experiments are underway and they are probing the regions of wimp mass - cross - section parameter space populated by susy models ( see e.g. ref . ) .
the realisation that uncertainties in the velocity distribution , @xmath0 , will affect the direct detection signals dates back to some of the earliest direct detection papers written in the 1980s ( e.g. ref . ) .
we first discuss the standard halo model and other approaches to modelling the milky way halo ( sec . [ modelling ] ) .
we then discuss what observations and simulations tell us about the dark matter distribution ( secs .
[ astro1 ] and [ astro2 ] respectively ) .
we then examine the direct detection signals ( sec .
[ signals ] ) and how the astrophysical uncertainties affect these signals ( sec . [
affect ] ) .
we conclude in sec .
[ strategies ] by discussing strategies for handling the uncertainties , including both ` astrophysics independent ' approaches and parameterizing the wimp speed distribution .
the steady - state phase space distribution function , @xmath1 , of a collection of collisionless particles is given by the solution of the collisionless boltzmann equation : @xmath2 in cartesian coordinates this becomes @xmath3 where @xmath4 is the potential .
the standard halo model ( shm ) is an isotropic , isothermal sphere with density profile @xmath5 . in this case
the solution of the collisionless boltzmann equation is a so - called maxwellian velocity distribution , given by @xmath6 where @xmath7 is a normalisation constant .
the isothermal sphere has a flat rotation curve at large radii and the velocity dispersion is related to the asymptotic value of the circular speed ( the speed with which objects on circular orbits orbit the galactic centre ) @xmath8 .
it is usually assumed that the rotation curve has already reached its asymptotic value at the solar radius , @xmath9 , so that @xmath10 where @xmath11 is the local circular speed . in the shm
the peak speed @xmath12 and the circular speed are identical , @xmath13 , and these parameters are often used interchangeably .
however this is not the case in general , for instance for a nfw @xcite density profile @xmath14 at @xmath15 ( where the scale radius , @xmath16 , is the radius at which the logarithmic slope of the density profile is equal to -2 ) @xcite . the density distribution of the shm is formally infinite and hence the velocity distribution extends to infinity too . in reality
the milky way halo is finite , and particles with speeds greater than the escape speed , @xmath17 , will not be gravitationally bound to the mw .
this is often addressed by simply truncating the velocity distribution at the measured local escape speed , @xmath18 , so that @xmath0 is given by eq .
( [ max ] ) for @xmath19 and @xmath20 for @xmath21 .
this sharp truncation is clearly unphysical .
an alternative , but still ad hoc , approach is to make the truncation smooth : @xmath22 \ , , & 0 \ , , & |{\bf v}| \geq v_{\rm esc } \ , .
\end{cases}\ ] ] another approach , used in the king model / lowered isothermal sphere , is to modify the shm distribution function so that it becomes zero for large energies @xcite .
the standard parameter values used for the shm are a local density @xmath23 , a local circular speed @xmath24 @xcite , and a local escape speed @xmath25 @xcite .
we will discuss the determination of these parameters and the associated uncertainties in sec .
[ astro1 ] . for spherical , isotropic systems
there is a unique relationship between the density profile and the distribution function known as the eddington equation ( see e.g. ref . ):
@xmath26\,,\ ] ] where @xmath27 and @xmath28 , where @xmath29 is the kinetic energy .
the eddington equation has been used to calculate the speed distribution for a range of spherically symmetric density profiles @xcite .
ref . found , using a bayesian analysis incorporating various dynamical constraints , speed distributions that are not too dissimilar to the standard maxwellian ( when the same value of the circular speed is used ) .
ref . used the eddington equation to motivate a phenomenological form for @xmath30 which matches the high speed tail of the speed distributions found in simulated halos ( see sec .
[ astro2 ] ) .
the osipkov - merritt models @xcite assume that the distribution function depends on the energy , @xmath31 , and angular momentum , @xmath32 , only via a single parameter , @xmath33 , @xmath34 where the constant @xmath35 is the anisotropy radius
. in this model the velocity anisotropy @xmath36 where @xmath37 are the velocity dispersions in the @xmath38 and @xmath39 directions , has a particular , radially dependent , form : @xmath40 and the distribution function can be calculated from a modified version of the eddington equation ( see e.g. refs . ) . in these models
the peak of the speed distribution becomes narrower and there is an excess of high speed particles ( e.g. ref . ) .
in general there is no unique relationship between the density profile and the velocity distribution for triaxial and/or anisotropic systems . in these cases
a common approach is to use the jeans equations to calculate the lower order moments of the distribution function .
multiplying the collisionless boltzmann equation , eq .
( [ cme ] ) , by one of the velocity components and integrating gives the jeans equations , which in cartesian coordinates take the form @xmath41 we now have three equations ( corresponding to @xmath42 ) for six unknowns ( @xmath43 ) . to make further progress it is therefore necessary to make assumptions about the alignment of the velocity ellipsoid , i.e. to choose coordinates such that @xmath44 for @xmath45 .
the velocity distribution is then approximated by a multivariate gaussian in these coordinates : @xmath46 where @xmath47 .
evans , carollo and de zeeuw @xcite studied the logarithmic ellipsoidal model , the simplest triaxial generalisation of the isothermal sphere .
they argue that aligning the velocity ellipsoid with conical coordinates is physically well motivated . in the planes of the halo , conical coordinates are locally equivalent to cylindrical polar coordinates .
hence , in this model , the velocity distribution can be approximated by a multi - variate gaussian in cylindrical polar coordinates , with velocity dispersions depending on the shape of the halo , the velocity anisotropy and also location within the halo .
both the width of the speed distribution and the peak speed can change ( e.g. refs . ) .
analytic models inevitably make assumptions ( regarding the shape and anisotropy of the halo , and their radial dependence ) which are almost certainly not completely valid .
the relationship between the observed properties of dark matter halos and the velocity distribution is non - trivial ; models with the same bulk properties ( e.g. shape and local velocity anisotropy ) can have velocity distributions with very different forms . furthermore all analytic approaches to halo modelling rely on the assumption that the phase space distribution function has reached a steady state . to some extent analytic halo models
have been superseded by results from numerical simulations .
it should be emphasized ( see sec .
[ micro ] and [ baryon ] ) that velocity distributions from numerical simulations also involve approximations and extrapolations .
the event rate is directly proportional to the local density , @xmath48 .
the standard value used is @xmath49 . as discussed in ref .
, the origin of the use of this particular value , rather than say @xmath50 , is unclear .
the local density is calculated via mass modelling of the milky way ( mw ) .
this involves constructing a model of the mw ( including its luminous components ) and then finding the range of values of the local density that are consistent with all observational data ( including , for instance , rotation curve measurements , velocity dispersions of halo stars , local surface mass density , total mass ) @xcite . as emphasised in ref .
the shape of the mw halo has a crucial effect on the local density extracted . for a fixed circular speed , in a flattened halo the local density , in the galactic plane ,
is higher .
this work has recently been updated to include new data sets , models for the mw halo motivated by numerical simulations , and , in some cases , bayesian statistical techniques .
ref . , assuming a spherical halo with a cuspy density profile ( @xmath51 as @xmath52 ) found @xmath53 , while ref . , assuming spherically symmetric nfw @xcite and einasto @xcite profiles for the mw halo found , @xmath54 . while these determinations have @xmath55 errors , they differ from each other by significantly more than this .
this suggests that the systematic errors are bigger than the statistical errors .
. finds , considering a range of density profiles including cored profiles ( @xmath56 as @xmath52 ) , values in the range @xmath57 .
other recent work has investigated ` model independent ' techniques , which do nt involve global mass - modelling of the galaxy , and hence have smaller hidden systematic errors .
ref . proposed using the equation of centrifugal equilibrium and subtracting the contribution of the stellar component of the mw .
the resulting determination of the local density is @xmath58 , where the uncertainties come from the uncertainty in the slope of the circular speed at the solar radius and the ratio of the solar radius to the length scale of the thin stellar disk .
ref . introduced a method which involves solving the jeans and poisson equations with minimal assumptions .
using hipparcos data they find @xmath59 , at @xmath60 confidence , assuming the stellar tracer populations are isothermal and @xmath61 if they have a non - isothermal profile .
an accurate determination of the vertical dispersion profile of the tracer population is therefore required to make an unbiased estimate of the local density using this method .
ref . used a high - resolution simulation of a milky way like galaxy , including baryons @xcite , to investigate the effect of halo shape on the local density .
specifically they examined how the local density varies from the density averaged in a spherical shell , as determined by observations .
they find that the density within the stellar disk , at a distance @xmath62 from the centre , is a factor of between 1.01 and 1.41 , with an average of 1.21 , larger than the value averaged in a spherical shell . the local circular speed , @xmath63 , is an important quantity to determine .
it appears in the galilean transformation of the velocity distribution into the lab frame ( see sec .
[ signals ] ) .
it is also related to the radial component of the velocity dispersion , @xmath64 , by one of the jeans equations ( e.g. ref . ):
@xmath65 where the anisotropy parameter @xmath66 is defined in eq .
( [ beta ] ) .
the shm has @xmath5 and the the velocity distribution is isotropic ( @xmath67 so that @xmath68 ) and independent of radius , so that @xmath69 .
the standard value of @xmath70 of @xmath71 dates back to a 1980s review of the galactic constants @xcite and was found by taking an average of the values found from a wide range of different analyses .
note that the ratio @xmath72 is better determined than either @xmath70 or @xmath73 individually ( e.g. refs . ) .
a recent analysis using measurements of the motions of galactic masers , found a significantly higher value , @xmath74 , @xcite .
reanalyzed the data using a more general model for the maser velocity distribution ( including allowing a non - zero velocity dispersion tensor ) and found that the maser data places only a relatively weak constraint on @xmath70 .
when combined with other measurements ( from the proper motion of sgr @xmath75 , and the orbit of the gd-1 stellar stream ) , they found @xmath76 , assuming a flat rotation curve . meanwhile ref . found the value of @xmath70 determined from the maser data depends strongly on the mw model
used . using a range of models for the rotation curve , including a power - law with free slope , they found values in the range @xmath77 to @xmath78 .
this illustrates that , as in the case of the local density , systematic , modelling errors are important . the escape speed is the speed required to escape the local gravitational field of the mw , @xmath79
the local escape speed , @xmath18 , is estimated from the speeds of high velocity stars . to do this
it is necessary to parameterise the shape of the high speed tail of the velocity distribution . assuming that the velocities are isotropic and the jeans theorem applies , the asymptotic form of the velocity distribution can be written as @xcite : @xmath80^{k } \ , , & |{\bf v}| < v_{\rm esc } \
, , \\ 0 \ , , & |{\bf v}| \geq v_{\rm esc } \ , .
\end{cases}\ ] ] traditionally a value @xmath81 , corresponding to the upper @xmath60 confidence limit from ref . , has been used .
ref . has updated these measurements , using additional data from the rave survey and using a prior on @xmath82 , @xmath83 $ ] ( motivated by analysis of the speed distributions of stellar particles in simulated halos ) .
they find that the escape speed lies in the range @xmath84 at @xmath60 confidence , with a median likelihood of @xmath85 .
a number of high resolution , dark matter only , simulations of the formation of milky way - like halos in a cosmological context have been carried out ( e.g. aquarius @xcite , ghalo @xcite and via lactea @xcite ) .
the velocity distributions of these halos deviate systematically from a multivariate gaussian @xcite .
there are more low speed particles and the peak in the distribution is lower and flatter in shape ( i.e. the distribution is platykurtic ) .
several fitting functions have been considered ; a tsallis distribution ( which arises from non - extensive statistical mechanics ) @xcite and a modified maxwellian @xcite @xmath86 } \
f(v_{\rm t } ) & = & \frac{v_{\rm t}}{n_{\rm t } } \exp { \left [ - \left ( \frac{v_{\rm t}^2}{\bar{v}_{\rm t}^2 } \right)^{\alpha_{\rm t } } \right ] } \,,\end{aligned}\ ] ] where @xmath87 and @xmath88 are the radial and tangential components of the velocity , @xmath89 are normalisation factors and @xmath90 are free - parameters which parameterise the deviations from a maxwellian distribution .
the most likely speed deviates from the circular speed : for vl2 and ghalo @xmath91 and @xmath92 respectively @xcite . the velocity distributions also have stochastic features at high speeds .
there are broad bumps which vary from halo to halo , but are independent of position within a given halo and are thought to reflect the formation history of the halo @xcite .
kuhlen et al .
@xcite also find narrow spikes in some locations , corresponding to tidal streams .
ref . presented an ansatz for the velocity distribution @xmath93^{k }
\theta ( v_{\rm esc } - component of the speed distributions found in simulations .
the shape of the high speed tail of the distribution is determined by the slope of the density profile at large radii . using the eddington equation
the parameter @xmath82 can be related to the outer slope of the density profile , @xmath94 , ( @xmath95 for large @xmath96 ) , by @xmath97 for @xmath98 @xcite . as @xmath99 ( the value corresponding to the nfw profile @xcite ) the calculation breaks down , and numerical fits to eq .
( [ k ] ) find @xmath100 .
more generally for outer slopes in the range @xmath101 , @xmath82 lies in the range @xmath102 $ ] .
( [ k ] ) with @xmath82 in this range provides a good fit to the high speed tails of the speed distributions found in simulation and has less high speed particles than the tail of the standard maxwellian distribution with a sharp cut - off .
note , however , that eq .
( [ k ] ) does not match the low and moderate speed behaviour of the simulation speed distributions , possibly due to the assumption of isotropy @xcite .
the simulations discussed above contain dark matter only , while baryons dominate in the inner regions of the milky way .
simulating baryonic physics is extremely difficult , and producing galaxies whose detailed properties match those of real galaxies is a major challenge .
some recent simulations have found that late merging sub - halos are preferentially dragged towards the disc by dynamical friction , where they are destroyed leading to the formation of a co - rotating dark disc ( dd ) @xcite . ref . modelled the dd velocity distribution as a gaussian with isotropic dispersion , @xmath103 and lag @xmath104 , matching ( roughly ) the kinematics of the milky way s stellar thick disc and considered dd densities in the range @xmath105 , where @xmath106 is the local halo density .
the properties ( and even existence ) of the dark disc are highly uncertain .
ref . argues that to be consistent with the observed morphological and kinematic properties of the milky way s thick disc , the milky way s merger history must be quiescent compared with typical @xmath107cdm merger histories , and hence the dd density must be relatively small , @xmath108 .
the total ( halo + disc ) local density can be probed by the kinematics of stars ( e.g. ref . ) .
there is no evidence for a dark matter disc ( i.e. the data are well fit without a dark disc ) and a thick , dense , @xmath109 , dark disc is excluded @xcite .
refs . and also argue that the dd velocity dispersion is likely to be substantially larger than that of the stellar thick disc .
baryonic physics will also affect the halo speed distribution .
for instance gas cooling makes halos more spherical ( e.g. ref . ) . the local speed distribution found in ref .
is well fit by a tsallis distribution , but appears to deviate less from the standard maxwellian than the distributions found in dark matter only simulations .
a further caution is that the scales resolved by simulations are many orders of magnitude larger than those relevant for direct detection experiments
. the earth moves at @xmath110 , therefore direct detection experiments probe the dark matter on sub mpc scales , while the numerical simulations discussed above have gravitational softening of order @xmath111 .
vogelsberger and white have developed a new technique to follow the fine - grained phase space distribution in simulations as it stretches and folds under the action of gravity @xcite .
they find that the median density of the resulting streams is of order @xmath112 the local halo density .
schneider et al .
@xcite have studied the evolution of the first , and smallest , roughly earth mass @xcite , microhalos to form in the universe .
they find that tidal disruption and encounters with stars produce tidal streams with average density @xmath113 the local halo density .
these results ( see also ref . ) suggest that the ultra - local dark matter density and velocity distribution should not be drastically different to those on the scales resolved by simulations .
the ultra - local velocity distribution may , however , contain some features or fine - grained structure @xcite .
the differential event rate for elastic scattering , assuming spin - independent coupling with identical couplings to the proton and neutron , is given by ( e.g. refs . ):
@xmath114 where @xmath48 is the local wimp density , @xmath115 the wimp velocity distribution in the lab frame , @xmath116 the wimp scattering cross section on the proton , @xmath117 the wimp - proton reduced mass , @xmath118 and @xmath119 the mass number and form factor of the target nuclei respectively and @xmath120 is the recoil energy . the lower limit of the integral , @xmath121 , is the minimum wimp speed that can cause a recoil of energy @xmath120 : @xmath122 where @xmath123 is the atomic mass of the detector nuclei and @xmath124 the wimp - nucleon reduced mass .
the event rate for inelastic scattering of wimps @xcite also depends on the wimp density and velocity distribution .
however , due to the altered kinematics the relationship between @xmath125 and @xmath120 is changed , and hence the effects of changes in the velocity distribution are different @xcite .
the wimp speed distribution must be transformed from the galactic rest frame to the lab frame .
this is done by carrying out , a time dependent , galilean transformation : @xmath126 .
the earth s motion relative to the galactic rest frame , @xmath127 , is made up of three components : the motion of the local standard of rest ( lsr ) , @xmath128 , the sun s peculiar motion with respect to the lsr , @xmath129 , and the earth s orbit about the sun , @xmath130 , .
the motion of the lsr is defined as @xmath131 , where @xmath70 is the local circular speed ( see sec .
[ vc ] for a discussion of recent determinations )
. the most recent determination of the sun s motion with respect to the lsr , taking into account the effects of the metallicity gradient in the disc , finds @xmath132 @xcite in galactic co - ordinates ( where @xmath133 points towards the galactic center , @xmath134 is the direction of galactic rotation and @xmath135 towards the north galactic pole ) . note that the value of @xmath136 is significantly ( @xmath137 ) larger than previously found @xcite .
the new larger value is also supported by the analysis of the motions of masers @xcite .
accurate expressions for the earth s orbit can be found in ref . .
simpler expressions , which are acceptable for most practical purposes , can be found in refs . .
the differential event rate , eq . ( [ drde ] ) , depends on the target nuclei mass , the ( a priori unknown ) wimp mass and the integral of the velocity distribution .
it is therefore useful to rewrite eq .
( [ drde ] ) as @xmath138 where @xmath139 and the prefactor @xmath140 contains the wimp and target dependent terms and is independent of the astrophysical wimp distribution .
we will now discuss the energy ( sec .
[ er ] ) , time ( sec .
[ am ] ) and direction ( sec . [ dd ] ) dependence of the differential event rate . in order to make concrete statements we will for now assume the standard maxwellian velocity distribution , eq .
( [ max ] ) . in sec .
[ affect ] we will discuss how uncertainties in the wimp distribution affect these signals .
the shape of the energy spectrum depends on both the wimp mass and the mass of the target nuclei .
this can be seen , following lewin and smith @xcite , by assuming a standard maxwellian velocity distribution , eq .
( [ max ] ) , and , initially , neglecting the earth s velocity and the galactic escape speed .
this allows the time averaged energy spectrum to be written as @xmath141 where @xmath142 is the event rate in the @xmath143 limit , and @xmath144 , the characteristic energy scale , is given by @xmath145 when the earth s velocity and the galactic escape speed are taken into account eq .
( [ drde0 ] ) is still a reasonable approximation to the event rate if @xmath146 and @xmath144 are both multiplied by constants of order unity . if @xmath147 , @xmath148 , while if @xmath149 , @xmath144 is independent of the wimp mass .
this indicates that the wimp mass can be determined from the energy spectrum , provided it is not significantly larger than the mass of the target nuclei @xcite . furthermore measuring consistent spectra for two different target nuclei
could in principle confirm the wimp origin of these spectra ( e.g. ref . ) .
this is sometimes referred to as the ` materials signal ' . due to the earth s orbit about the sun
, the net velocity of the lab with respect to the galactic rest frame varies annually .
the net speed is largest in the summer and hence there are more high speed wimps , and less low speed wimps , in the lab frame .
this produces an annual modulation of the event rate @xcite .
the differential event rate peaks in winter for small recoil energies and in summer for larger recoil energies @xcite .
the energy at which the annual modulation changes phase is often referred to as the ` crossing energy ' .
its value depends on the wimp and target masses , and hence could be used to determine the wimp mass @xcite .
since the earth s orbital speed is significantly smaller than the sun s circular speed the amplitude of the modulation is small and , to a first approximation , the differential event rate can , for the standard halo model , be written approximately as a taylor series : @xmath150 \,,\ ] ] where @xmath151 , @xmath152 year and @xmath153 days . for the standard halo model the fractional amplitude of the modulation
is approximately given by @xcite @xmath154 where @xmath155 , @xmath156 is the value of @xmath157 at which the sign of the modulation reverses and @xmath158 . for @xmath159 , in the extreme tail of the speed distribution , the shape of the modulation is non - sinusoidal .
for very small energies @xmath160 is negative ( i.e. the maximum occurs in december rather than june ) .
as @xmath120 is increased , @xmath161 initially decreases to zero at which point the phase of the annual modulation changes so that the maximum occurs in summer .
as @xmath120 is increased further the fractional amplitude continues increasing .
this is potentially misleading however , as the mean event rate , and hence the raw amplitude , becomes very small at large @xmath120 . after the phase change on increasing @xmath120 further the raw amplitude increases to a local maximum before decreasing again and tending to zero . for the shm , for measurable energies
, the amplitude of the modulation lies in the range @xmath162 .
our motion with respect to the galactic rest frame also produces a direction dependence of the event rate .
the wimp flux in the lab frame is peaked in the direction of motion of the sun ( towards the constellation cygnus ) , and hence the recoil spectrum is , for most energies , peaked in the direction opposite to this @xcite .
this directional signal is far larger than the annual modulation ; the event rate in the backward direction is several times larger than that in the forward direction @xcite .
a detector which can measure the recoil directions is required to detect this signal ( see ref . for an overview of the status of directional detection experiments ) .
the full direction dependence of the event rate is most compactly written in terms of the radon transform of the wimp velocity distribution @xcite @xmath163 where @xmath164 , @xmath165 is the recoil direction and @xmath166 is the 3-dimensional radon transform of the wimp velocity distribution @xmath167 geometrically the radon transform , @xmath168 , is the integral of the function @xmath0 on a plane orthogonal to the direction @xmath165 at a distance @xmath125 from the origin . see ref . for an alternative , but equivalent , expression . while the directional recoil rate depends on both of the angles which specify a given direction , the strongest signal is the differential of the event rate with respect to the angle between the recoil and the direction of solar motion , @xmath94 , @xcite @xmath169 for the standard maxwellian velocity distribution @xcite @xmath170 } \,,\ ] ] where @xmath171 is the component of the earth s velocity parallel to the direction of solar motion .
an ideal detector capable of measuring the nuclear recoil vectors ( including their senses , @xmath172 versus @xmath173 ) in 3-dimensions , with good angular resolution , could reject isotropy of wimp recoils with only of order 10 events @xcite .
most , but not all , backgrounds would produce an isotropic galactic recoil distribution ( due to the complicated motion of the earth with respect to the galactic rest frame ) . an anisotropic galactic recoil distribution would therefore provide strong , but not conclusive , evidence for a galactic origin of the recoils .
roughly 30 events would be required for an ideal detector to confirm that the peak recoil direction coincides with the inverse of the direction of solar motion , hence confirming the galactic origin of the recoil events @xcite .
in this section we will discuss how the uncertainties described in secs .
[ astro1 ] and [ astro2 ] affect the signals expected in direct detection experiments .
since the normalisation of the event rate is directly proportional to the product of the cross - section , @xmath174 , and the local density , @xmath48 , the uncertainty in @xmath48 leads directly to an uncertainty in measurements of , or constraints on , @xmath175 .
since the density is not expected to vary on mpc scales , the uncertainty in @xmath174 is the same for all direct detection experiments , and does not affect comparisons of e.g. the exclusion limits from different experiments .
it does however affect comparisons with collider and other constraints on @xmath174 .
furthermore it can significantly bias the determination of the wimp mass @xcite .
the characteristic energy scale , @xmath144 , given by eq .
( [ ereq ] ) , depends on the local circular speed .
therefore the uncertainty in the local circular speed , @xmath70 , leads to a @xmath176 uncertainty in the differential event rate , and hence exclusion limits @xcite .
the nature of the change in the exclusion limits is similar , but not identical , for different experiments @xcite .
it will also lead to a bias in determinations of the wimp mass @xcite .
this can be seen by differentiating the expression for the characteristic energy , eq .
( [ ereq ] ) , @xmath177 \frac{\delta v_{\rm c}}{v_{\rm c } } \,.\ ] ] since the differential event rate is given by an integral over the velocity distribution the differential event rate depends only weakly on the detailed form of the velocity distribution @xcite .
consequently the resulting uncertainty in exclusion limits @xcite and determinations of the wimp mass @xcite is usually fairly small .
there are , however , some exceptions to this statement .
firstly if the wimp is light and/or the experimental energy threshold is high compared with the characteristic energy , then only the high speed tail of the speed distribution is probed . in this case the uncertainties in its shape can have a significant effect on the expected energy spectrum and hence exclusions limits or allowed regions @xcite .
secondly if there is a dark disc there will be an additional population of low speed wimps .
if the dark disc density is sufficiently high and its velocity dispersion sufficiently small this will significantly change the energy spectrum and hence exclusion limits and mass determinations @xcite .
the size of these changes depends on the properties of the dark disc , which are currently uncertain @xcite . a wimp stream , with sufficiently high density ,
would add a sloping step to the differential event rate @xcite .
the annual modulation arises from the , relatively small , shift in the speed distribution in the lab frame over the course of the year .
it is therefore far more sensitive to the speed distribution than the time averaged energy spectrum .
both the amplitude and phase of the modulation , and hence the regions of parameter space compatible with an observed signal , can vary significantly ( e.g. refs . ) . the uncertainty in the value of @xmath178 can change the amplitude by a factor of order unity , while the uncertainty in the shape of the halo velocity distribution changes the amplitude by @xmath179 @xcite .
a high density dark disc could change the annual modulation signal significantly , producing a 2nd , low energy , maximum in the amplitude of the modulation and changing the phase significantly @xcite . for
a wimp stream the position and height of the step in the energy spectrum produced would vary annually ( e.g. ref . ) .
the detailed direction dependence of the event rate is sensitive to the velocity distribution , however the anisotropy is robust @xcite .
the number of events required to detect anisotropy or to demonstrate that the median inverse recoil direction coincides with the direction of solar motion only vary by of order @xmath180 @xcite . the features in the high speed tail can cause the median inverse direction of high energy recoils to deviate from the direction of solar motion @xcite .
a stream of wimps would produce recoils which are strongly peaked in the opposite direction @xcite
. in the long term , studying the median direction of high energy recoils could allow high speed features to be detected and the formation history of the milky way probed .
it is possible to carry out an ` astrophysics independent ' comparison of data from multiple experiments .
this effectively involves making comparisons of the energy spectra , so that the integral over the velocity distribution , @xmath181 defined in eq .
( [ gvmin ] ) , cancels .
for instance , with data from two different experiments using two different targets , the wimp mass can in principle be determined , without any assumptions about @xmath0 , by taking moments of the energy spectra @xcite .
however this method leads to a systematic underestimate of the wimp mass if it is comparable to or larger than the mass of the target nuclei .
ref . , see also ref .
, showed how the differential event rates in two experiments , ` 1 ' and ` 2 ' with target nuclei with mass numbers @xmath182 and @xmath183 are related . if the first experiment is sensitive to recoil energies in the range @xmath184 $ ] it probes @xmath181 for @xmath185 , where @xmath125 is related to the recoil energy , wimp mass and target nuclei mass by eq .
( [ vmin ] ) . in experiment 2
these values of @xmath125 correspond to recoils in the range @xmath186 = \frac{\mu_{a2 \chi}^2 m_{a1}}{\mu_{a1 \chi}^2 m_{a2 } } [ e_{1}^{\rm low } , e_{1}^{\rm high } ] \,.\ ] ] for recoil energies in this range the differential event rates in the two experiments are then related by @xmath187 this relation depends on the ( unknown ) wimp mass , therefore the comparison has to be made separately for a range of @xmath188 values .
this method has recently been extended to include annual modulation data @xcite .
this approach has proved particularly useful @xcite in comparing the event rate excesses and annual modulations found in the cogent @xcite , cresst @xcite and dama @xcite experiments with exclusions limits from cdms @xcite and xenon @xcite . ref
. found that the dama , cogent and cresst - ii data are compatible which each other , but not cdms and xenon , if the velocity distribution is very anisotropic ( which leads to a large modulation fraction ) .
the astrophysics independent methods are invaluable for assessing the compatibility of data from different experiments .
however the wimp mass is not a priori known , and the goal of wimp direct detection experiments is not just to detect wimps , but to also measure their mass and cross - section . while there are uncertainties in the velocity distribution it is not completely unknown , and therefore does not need to be completely removed from the analysis .
strigari and trotta @xcite have shown how the wimp properties can be constrained by a single future direct detection experiment when astronomical data ( such as the kinematics of halo stars ) are used to jointly constrain the wimp properties and the parameters of a model for the milky way halo .
they assumed an isotropic maxwellian speed distribution characterised by the peak speed , @xmath12 , and the escape speed , @xmath189 .
peter @xcite has examined how data sets from multiple direct detection experiments could be used to jointly constrain a maxwellian parametrisation of the wimp speed distribution and the wimp parameters ( mass and cross - section ) .
pato et al .
@xcite have taken a similar approach using the parameterization in eq .
( [ k ] ) .
peter has recently extended this work by allowing the peak speed and circular speed to differ ( as expected if the density profile is not @xmath5 ) and also considering an empirical speed distribution consisting of a five or ten bin step function @xcite .
combining data from multiple experiments with different targets significantly increases the accuracy with which the wimp parameters can be measured @xcite .
however fixing the form of the speed distribution leads to biases in the wimp properties , if the true speed distribution differs significantly from the parameterization assumed @xcite . with a suitable parameterization o the wimp parameters and speed distribution
can be jointly probed @xcite .
the form of the optimal parameterization of the speed distribution is an open question .
direct detection event rate calculations often assume the standard halo model , with an isotropic maxwellian velocity distribution with dispersion @xmath190 and a local wimp density @xmath191 .
however it has long been realised @xcite that uncertainties in the speed distribution will affect the signals expected in experiments .
we have discussed the standard halo model and other approaches to halo modelling , and the assumptions behind them .
we then reviewed observational determinations of quantities that are relevant to direct detection experiments , namely the local dark matter density , the local circular speed ( which is related to the velocity dispersion by the jeans equations ) and the local escape speed . while the statistical errors on these quantities are often small the systematic errors , from uncertainties in the modelling of the milky way , can be significantly larger .
next we turned our attention to high resolution , dark matter only simulations of the formation of milky way - like halos in a cosmological context .
they find velocity distributions that deviate significantly from the standard maxwellian , and have features at high speeds .
the effect of baryonic physics on the dark matter distribution is not yet well understood .
some recent simulations have found that a dark disc may be formed , however its properties ( and even existence ) are highly uncertain .
simulations resolve the dark matter distribution on scales many orders of magnitude larger than those probed by direct detection experiments .
the latest results suggest that the ultra - local dark matter distribution is largely smooth , but some features may exist .
we then reviewed the resulting uncertainties in the energy , time and direction dependence of the energy spectrum , and hence constraints on , or measurements of , the wimp parameters .
the uncertainty in the local circular speed has a significant effect on the event rate and hence exclusion limits and determinations of the wimp mass .
the uncertainty in the local density leads directly to an uncertainty in measurements of , or constraints on , the cross - section .
only the high energy tail of the energy spectrum is particularly sensitive to the exact shape of the speed distribution .
the annual modulation is far more sensitive to the shape of the velocity distribution ; its amplitude and phase can change significantly .
the anisotropy of the recoil rate is robust to changes in the velocity distribution , however high speed features can change the peak direction of high energy recoils , and hence provide a way of probing the formation of the milky way .
finally we discussed techniques for handling the astrophysical uncertainties when analysing direct detection data .
astrophysics independent comparisons between different experiments can be made using the integral of the velocity distribution , @xmath181 .
this approach is extremely useful for assessing the compatibility of various experiments , but requires the wimp mass as input .
parameterising the wimp speed distribution , and using data from multiple experiments to jointly constrain the wimp mass and cross - section and speed distribution is a promising approach , but the optimal form for the parameterisation is not yet known .
the author is supported by stfc .
she is grateful to the organisers of , and participants in , the ` dark matter underground and in the heavens ' workshop ( dmuh11 ) at cern , where she had various useful discussions about the topics covered in this review .
she is also grateful to bradley kavanagh for useful comments on a draft of this article . | direct detection experiments are poised to detect dark matter in the form of weakly interacting massive particles ( wimps ) .
the signals expected in these experiments depend on the ultra - local wimp density and velocity distribution .
firstly we review methods for modelling the dark matter distribution .
we then discuss observational determinations of the local dark matter density , circular speed and escape speed and the results of numerical simulations of milky way - like dark matter halos . in each case
we highlight the uncertainties and assumptions made .
we then overview the resulting uncertainties in the signals expected in direct detection experiments , specifically the energy , time and direction dependence of the event rate .
finally we conclude by discussing techniques for handling the astrophysical uncertainties when interpreting data from direct detection experiments . |
heavy quark physics offers a unique opportunity to explore the flavour structure of the standard model .
the third family of quarks is the less known .
its up - type quark , the top , has not yet been observed , although we are getting more and more stringent limits on its mass from direct search and radiative correction analysis . on the other hand , the corresponding down - type quark , the @xmath1 ,
is being studied with increasing accuracy . both growing statistics and the operation of dedicated vertex detector for @xmath1 tagging at high energy machines already allows a good determination of @xmath1 couplings .
present data clearly point to the @xmath1 as the @xmath41/2 component of a weak left - handed isodoublet @xcite .
at lep , one measures the @xmath1 neutral current couplings through the @xmath5 width and forward - backward asymmetry on the @xmath2 peak .
in fact , one has ( z b|b ) ( v_b^2+a_b^2 ) a_fb^0(m_z ) = a_e a_b where a_f ( f = e , b ) and i neglected terms of the order @xmath6 @xcite . in principle , another determination of the quantity @xmath7 ( or , equivalently , of the ratio @xmath8 ) can be derived from the measurement of the @xmath1-polarization asymmetry on the @xmath2 peak . according to the standard model ,
@xmath1 quarks are produced almost completely longitudinally polarized at @xmath9 .
the angular dependence of the polarization asymmetry is given by @xcite p_b ( ) = - [ angle ] where @xmath10 is the production cross section for a @xmath1 quark with right-(left-)handed longitudinal polarization and , as above , i neglected mass effects of the order @xmath11 . since @xmath12 is quite smaller than @xmath7 , eq.([angle ] )
depends only weakly on the @xmath1 production angle @xmath13 .
the corresponding average mean polarization is [ pbc ] p_b =- a_b = - qcd corrections to @xmath14 have been computed in ref.@xcite .
chirality - violating mass effects in the production cross section for polarized @xmath1 quarks at one - loop are found to decrease the tree - level polarization @xmath15 by less than @xmath16 .
in analogy to the @xmath17 polarization measurement at lep , measuring @xmath14 would provide an alternative method to determine @xmath7 that has the advantage , with respect to the forward - backward asymmetry @xmath18 , not to be suppressed by the small quantity @xmath12 ( eq.(1 ) ) .
furthermore , as anticipated , @xmath14 is quite large in the standard model .
one has ( assuming @xmath19 ) p_b=-0.936 [ pb ] to be confronted with the @xmath17 polarization value @xmath20 .
unfortunately , contrary to the @xmath17 polarization asymmetries that have already provided a nice determination of @xmath21 and @xmath12 at lep @xcite , a measurement of @xmath14 is not as much straightforward .
this is mainly due to hadronization and fragmentation properties of @xmath1 quarks that largely dilute and obscure @xmath1 polarization effects in the observed final states .
moreover , there are rather large theoretical uncertainties in the predictions of such strong interaction effects , that make the disentangling of the polarization @xmath14 a quite difficult task . in this paper , i will review the present status of theoretical understanding about the possibility of detecting some effect of such a large value of @xmath14 in @xmath1 hadron final states on the @xmath2 peak .
i will concentrate on @xmath14 sensitivity of lepton energy spectra in the inclusive semileptonic @xmath1 hadron decay .
both charged - lepton and neutrino spectra will be considered .
theoretical ambiguities coming from our limited knowledge of the perturbative component of the @xmath1 fragmentation function will be especially stressed .
then , i will present some information on @xmath14 that can be extracted from the already available lep and petra data .
recent developments on the neutrino spectrum as a particularly efficient tool to measure a polarization effect at lep will be described as well .
the @xmath1 polarization influences in a sizable way the lepton distributions in @xmath1 semileptonic decays @xmath22 , where @xmath23 is an electron or a muon ( see also ref.@xcite ) . the branching ratio for these modes is relatively high ( about 20% ) . the inclusive differential distribution for semileptonic hadron decay
have been demonstrated to be equal to the parton - model one up to corrections of order @xmath24 ( that in this case amount to a few percent ) @xcite . assuming a free quark , with no effects from hadronization and qcd corrections , the normalized decay distribution in the charged - lepton energy is , in the @xmath1 rest frame , @xcite & = & \{3 - 2y_+_c ( ) + & + & p _ [ 1 - 2y_-_c ( ) ] } [ cm ] where f(_c)=1 - 8_c+8_c^3-_c^4 - 12_c^2 _ c. [ fep ] in eq .
( [ cm ] ) , @xmath25 ( @xmath26 ) , @xmath27 is the @xmath28 ( or @xmath29 ) energy in the @xmath1 rest frame , @xmath30 is the angle between the lepton and the @xmath1 spin quantization axis , and @xmath31 . in the laboratory system , the @xmath1
quark moves with velocity @xmath32 .
if we quantize the @xmath1 spin along the direction of @xmath33 , the lepton - energy decay distribution in the laboratory frame can be obtained by [ boost ] = _ y^-^y^+ dy , where y = y _ , x&=&x _ , = _ = ( -1 ) , + y^-&= & , y^+=\ { , ( 1-_c ) } , and @xmath34 is the charged - lepton energy in the laboratory system . in order to reproduce more realistic experimental situations
, one can impose some cut @xmath35 on the minimum allowed lepton transverse momentum with respect to the @xmath33 axis by simply multiplying the integrand in eq.([boost ] ) by @xmath36 , where @xmath37 . in the lep case ( @xmath38 )
, one gets from eqs.([cm ] ) and ( [ boost ] ) & = & \ { 5 + 5x_-4x_^2 + p ( 1+x_-8x_^2 ) + & + & [ -22 + 5_c+(27 - 4_c)x_+ 5_cx_^2 - 5x_^3 + & + & p ( -8 - 17_c+(9 + 28_c)x_+(18 - 17_c)x_^2 - 19x_^3 ) ] + & + & ( ) [ 3-_c+ p ( 3+_c ) ] } , with @xmath39 defined in eq .
( [ fep ] ) . in fig . 1a ,
the results for @xmath40 for a nave @xmath1 polarization @xmath14 ( solid lines ) are compared with the unpolarized case ( dashed lines ) , assuming two different values ( 1.2 and 1.5 gev ) for the @xmath41-quark mass and @xmath425 gev . here ,
any @xmath1-fragmentation effect is neglected and the final lepton - energy fraction in the laboratory frame is given by @xmath43 .
the harder spectra correspond to @xmath441.2 gev , the softer ones to @xmath441.5 gev .
one can see that a negative value of the @xmath1 polarization makes the @xmath23-energy spectrum harder . [
cols="^,<,<,<,<",options="header " , ] .3 cm * table iii * : the ratio @xmath45 , defined by eq .
( [ rb ] ) , for different choices of @xmath46 and @xmath47 . by confronting tables ii and
iii , one can see that the ratio of moments tends to be quite more stable than the original moments also with respect to @xmath46 .
for instance , one has an error on @xmath48 of only about @xmath49 by varying @xmath46 in the range @xmath50 mev@xmath51 mev . at this point ,
the observed leptonic spectrum in the laboratory is obtained , in the collinear approximation , by convoluting @xmath52 with the polarization - dependent `` nave '' leptonic spectrum in eq .
( [ boost ] ) .
hence , one can now compute the ratios @xmath53 of the @xmath54-th moments of the leptonic spectrum at @xmath55 over the ones at petra energies versus the `` effective '' @xmath1 longitudinal polarization @xmath56 , defined as the fraction of @xmath14 that survives hadronization processes at lep . in fact , @xmath53 is given by @xmath45 times the polarization - dependent ratio of the moments , @xmath57 , of the `` nave '' lepton spectra at lep and petra [ rl ] r_^n()&=&= + & = & r_b^n .
one important point is that , at petra energies , @xmath1 quarks are produced , to a very good approximation , through @xmath58 exchange alone , that is with zero longitudinal polarization .
thus the low - energy data offer a way to calibrate on unpolarized @xmath1 hadron samples . in fig . 2 ,
the value predicted for @xmath53 with @xmath59 ( mean - value ratio ) , is shown as a function of the `` effective '' @xmath1 polarization @xmath56 .
the band corresponds to @xmath46 varying between 150 gev ( upper curves ) and 350 gev ( lower curves ) in @xmath45 .
a change @xmath60 in the polarization from the unpolarized case ( @xmath61 ) corresponds to a @xmath62 variation in @xmath63 .
the study of higher order moments is expecially useful for consistency checks on the measured value of @xmath56 ( ref.@xcite ) .
the sensitivity to the exact value of @xmath64 is in general reduced in the ratio , with respect to the original moments . in fig . 2 ,
a comparison with available experimental data @xcite is also shown .
the dashed area corresponds to the measured value @xmath65 , that has been determined assuming @xmath66 in the lep data . with this assumption , one gets @xmath67 ( neglecting @xmath68 effects in the convolution of @xmath52 with the nave leptonic spectrum ) and , as a consequence , @xmath69 .
experimental data are already accurate enough to exclude large polarization effects and point to a value of the `` effective '' @xmath1 polarization centred around 0 and compatible with the theoretical prediction of about @xmath70 .
higher statistics on @xmath1 events at lep will soon allow to study polarization effects in a purely @xmath71 sample , that is foreseen to retain most of the initial @xmath1-quark polarization , an order of magnitude larger than in the previous inclusive analysis . in refs.@xcite and @xcite ,
some studies of exclusive decays of polarized @xmath71 are presented .
of course , the ambiguities in the knowledge of the @xmath1-quark fragmentation also apply to polarization measurements based on semileptonic @xmath1-baryon decays .
a relevant question is whether , with a reasonable statistics on semileptonic @xmath71 decays ( some data are already available @xcite ) , lep data will no longer need to be normalized with lower - energy data .
one could study the ratios @xmath72 of the lepton moments for the @xmath71 sample over those for the @xmath73-meson sample , both measured at lep . in this case , one gets [ rl2 ] r_^n()__b&= & = + & = & .
hence , only moments of the nave charged - lepton spectra are left in the ratio and @xmath1-fragmentation effects are mostly cancelled .
the main problem with this method is that one should actually calibrate on the spectrum from unpolarized @xmath71 s ( and not @xmath73 s ) .
without such a normalization , a possible observed difference between the @xmath71 and the inclusive spectra could be blamed on the model dependent features of the exclusive @xmath71 decay .
the order of magnitude of these effects have been studied in ref.@xcite and found to be comparable to polarization effects ( fig .
3 ) . hence
, a reliable model for @xmath71 semileptonic decays is needed before polarization effects can really be established in this way .
recently , a very interesting and effective method to get rid of fragmentation effects in the @xmath71 semileptonic sample was suggested in ref.@xcite .
a new variable y= given by the ratio of the mean electron and neutrino laboratory energies in the inclusive semileptonic @xmath71 decay , has been introduced .
this variable depends mainly on the @xmath1 rest - frame lepton angular distributions , while the dependence on the @xmath1 boost in the laboratory ( and , hence , on @xmath1 fragmentation effects ) drops out .
indeed , since at lep @xmath74 , the mean lepton energy in the laboratory is given by e = e + p ( e + p ) where @xmath75 is the longitudinal momentum in the @xmath1 direction of flight , @xmath76 , and primed quantities refer to the @xmath1 rest - frame . as a consequence , one
has y = the average energies and longitudinal momenta in the @xmath1 rest - frame can be easily computed by means of eqs.([cm ] ) and ( [ cmnu ] ) .
one finds @xcite e_&=&(--6_c^2 + 10_c^3 -+-9_c^2_c -3_c^3_c ) + p_&=&- p
( --4_c^2 + 4_c^3 + --3_c^2_c -3_c^3_c ) + e_&=&(-3_c-2_c^2 + 6_c^3 -+-6_c^2_c ) + p_&=&- p ( -+_c+-2_c^3 + -+2_c^2_c ) while the average energies are independent of polarization , the average longitudinal momenta differ from 0 only for a polarized quark . in fig . 4 ,
@xmath77 is plotted versus @xmath78 .
besides being free of fragmentation ambiguities , @xmath77 has the advantage to be much more sensitive than @xmath79 to the @xmath1 polarization ( , for @xmath80 , @xmath77 goes from 7/6 at @xmath66 up to 2 at @xmath81 ) .
this is due to the opposite behaviour of @xmath82 with respect to @xmath83 when varying @xmath78 , as well as to the larger sensitivity to @xmath78 of neutrino spectra ( table i ) .
also , the @xmath77 sensitivity to @xmath84 is rather mild .
of course , neutrino spectra can be measured with less accuracy than electron and muon spectra .
indeed , a missing energy has to be determined in each event , whose experimental error ( @xmath85gev ) is limited essentially by the resolution on the accompanying hadronic jets @xcite .
this reflects in a @xmath77 resolution estimated as _
y ~ ( with @xmath54 number of measured events ) corresponding to an error of about @xmath86 on @xmath77 with present statistics .
this implies an accuracy of about @xmath86 on @xmath78 ( fig . 4 ) . in ref.@xcite ,
also other variables are considered , that have the advantage to eliminate most of systematic errors .
these are the ratio @xmath87 and difference @xmath88 of the @xmath77 s corresponding to the baryon and meson samples r_y & & = 1-k_r p + d_y&&y_baryon - y_meson = -k_d p + where , for @xmath89 , one can predict k_r=0.660.02 k_d=0.750.03 neither qcd corrections nor the adopted kinematical approximations do contribute to @xmath90 and @xmath91 more than 0.02 . hence , any measured deviation of @xmath87 and @xmath88 from 1 and 0 respectively will give a polarization measurement with very good efficiency .
in this paper , i have reviewed different strategies for detecting @xmath1 polarization effects through lepton energy spectra in the inclusive semileptonic decays of @xmath1 hadrons .
i have shown that , if one concentrate on electron and muon spectra in the inclusive @xmath1 hadrons decay , one has to deal with large theoretical uncertainties coming from the only - partial knowledge of the @xmath1 fragmentation function . a strategy to control these ambiguities that makes use of both lep and lower - energy petra data has been outlined . by comparing theoretical predictions with experimental data , one can already exclude large polarization effects in the lep inclusive leptonic spectrum .
much more can be done with large statistics on the @xmath1 baryon sample , by restricting the polarization study to @xmath1 baryons .
in fact , the corresponding @xmath78 effect should be about an order of magnitude larger than in the inclusive hadronic spectrum , since ( contrary to @xmath1 baryons ) @xmath1 meson , which make up most of the hadronic sample , are expected not to retain any original spin information . in principle , when restricting to @xmath71 samples , a realistic modelling of exclusive decays is required .
it has been shown that , if one can study also neutrino spectra with good accuracy , the new variable @xmath92 is almost free of theoretical uncertainties and optimizes the sensitivity to @xmath78 , due also to the very good sensitivity of neutrino spectra . * figure captions * + + a ) charged - lepton energy spectrum in the laboratory frame for a nave @xmath1 polarization @xmath14 ( solid lines ) compared to the unpolarized case @xmath66 ( dashed lines ) , assuming different values for @xmath84 .
@xmath1-fragmentation effects are not included ( @xmath95 ) .
+ b ) same as in fig .
1a for the neutrino energy spectrum .
+ prediction for @xmath53 with @xmath59 ( average - value ratio ) versus the `` effective '' @xmath1 polarization @xmath56 .
the band corresponds to @xmath46 varying between 150 gev ( upper curves ) and 350 gev ( lower curves ) in the prediction of @xmath45 .
the dashed area derives from the measured value of @xmath48 @xmath65 .
+ mimicking model - dependent effects by varying the effective quark masses .
the @xmath23 energy spectra in the laboratory frame for a nave @xmath1 polarization @xmath14 ( solid lines ) are compared with the unpolarized case @xmath66 ( dashed lines ) , assuming four different sets of values for @xmath96 : | i review the present state of knowledge on lepton energy spectra in the inclusive semileptonic decay of beauty hadrons ( @xmath0 ) as a means to measure @xmath1 polarization effects on the @xmath2 peak .
charged - lepton as well as neutrino spectra are considered .
ps .
# 1#2#3phys . lett . * # 1 * ( 19#2 ) # 3 # 1#2#3z . phys . * # 1 * ( 19#2 ) # 3 # 1#2#3phys .
rev .
lett . *
# 1 * ( 19#2 ) # 3 # 1#2#3phys . rep . * # 1 * ( 19#2 ) # 3 # 1#2#3phys . rev .
* # 1 * ( 19#2 ) # 3 # 1#2#3nucl . phys . * # 1 * ( 19#2 ) # 3 # 1#2#3acta phys . pol .
* # 1 * ( 19#2 ) # 3 # 1#2#3acta phys . austr . * # 1 * ( 19#2 ) # 3 # 1#2#3j .
comp .
phys . *
# 1 * ( 19#2 ) # 3 # 1#2#3prog .
. phys . *
# 1 * ( 19#2 ) # 3 # 1#2#3int .
j. mod .
phys . *
# 1 * ( 19#2 ) # 3 0.0 cm 0.0 cm 15.2 cm 21.5 cm 0.7 cm -1.0 cm 2.0 cm # 1@xmath3 * preprint n.1009 * dipartimento di fisica + march 17 , 1994 universit di roma `` la sapienza '' + i.n.f.n .
- sezione di roma + * lepton spectra and the b polarization + at lep * c barbara mele + + _ i.n.f.n .
, sezione di roma , italy and _ + _ dipartimento di fisica , universit `` la sapienza '' , _ + _ p.le aldo moro 2 , i-00185 rome , italy _ + |
the majority of the mass in the universe is known to consist of dark matter ( dm ) of unknown composition .
identifying the nature of this dark matter is one of the outstanding problems in physics and astrophysics .
leading candidates for this dark matter are weakly interacting massive particles ( wimps ) , a generic class of particles that includes the lightest supersymmetric particle .
these particles undergo weak interactions and their expected masses range from 1 gev to 10 tev .
many wimps , if present in thermal equilibrium in the early universe , annihilate with one another , leaving behind a relic density found to be roughly the correct value . furthermore , recent interest in low mass wimps lead us to mention asymmetric dark matter models , which naturally predict light wimps @xcite .
thirty years ago , refs . @xcite
first proposed the idea of detecting weakly interacting particles , including neutrinos and wimps , via coherent scattering with nuclei .
soon after @xcite computed detection rates in the context of a galactic halo of wimps .
this work also showed that the count rate in wimp direct detection experiments will experience an annual modulation @xcite as a result of the motion of the earth around the sun .
then development of ultra - pure ge detectors permitted the first limits on wimps @xcite . since that time
, a multitude of experimental efforts to detect wimps has been underway , with some of them currently claiming detection .
the basic goal of direct detection experiments is to measure the energy deposited when weakly interacting particles scatter off of nuclei in the detector , depositing small amounts of energy , e.g. 1 - 10 kev , in the nucleus . a recent review of the basic calculations of dark matter detection , with an emphasis on annual modulation ,
may be found in @xcite .
numerous collaborations worldwide have been searching for wimps using a variety of techniques to detect the nuclear recoil . in this paper
we elaborate on a novel mechanism for direct detection of wimps using explosives @xcite .
the small amount of energy deposited in the nucleus by the wimp scattering event can be enough to trigger an explosion .
the registration of such an explosion then indicates that a wimp / nucleon scattering event took place . in our search for appropriate explosive materials
, we realized a key limitation , which we named
greg s rule . "
everything on the surface of earth , including the conventional chemical explosives , has been constantly bombarded by ionizing particles coming from trace amounts of naturally occurring radioactive materials and cosmic radiation . since
conventional explosives can be stored in large quantities for extended periods of time ( without blowing up ) , we may conclude that all the conventional explosives that are currently being used in commercial or military applications can not be used in dm detection applications .
this does not imply that there are no explosives that can be detonated by a single highly ionizing particle .
if one were to synthesize such a material it would be highly unstable and would mysteriously explode .
we need to be contrarians " and test such unsafe " explosives , which were discovered but rejected in prior r&d .
luckily there are two directions to pursue .
first , the chemical explosive , nitrogen triodine ( ni@xmath2 ) , has been studied and can be ignited by a single highly ionizing particle ( e.g. an @xmath3-particle ) @xcite .
future work on using ni@xmath2 for dm detectors will be interesting . in this paper
we instead study the second approach , nanothermites .
thermites have been used for more than 100 years to obtain bursts of very high temperatures in small volumes , typically a few @xmath4 .
thermites are two component explosives , consisting of a metal and either an oxide or a halide .
these two components are stable when kept separated from one another ; but when they are brought together they undergo a rapid exothermic reaction an explosion .
the classical examples are @xmath5 one advantage of thermites is the impressive number of elements , which can be used .
classic implementation of thermites uses micron scale ( 1 to 10 microns ) granulation , but in recent years nano - sized granules of high explosives have been increasingly used @xcite .
these nano - thermites make interesting dark matter detectors . when a wimp strikes the metal layer , the metal may heat up sufficiently to overcome the chemical energy barrier between the metal and metal - oxide .
an explosion results .
nanoexplosive dark matter detectors have several advantages : 1 .
they can operate at room temperature ; 2 .
low energy threshold of 0.5 kev , allowing for study of low mass @xmath6 gev wimps ; 3 .
flexibility of materials : one may choose from a variety of elements with high atomic mass ( _ e.g. _ tl or ta ) to maximize the spin - independent scattering rate . given a variety of materials one can also extract information about the mass and cross section of the wimps ; 4 .
one can also select materials with high nuclear spin to maximize spin - dependent interaction rate ; 5 .
signal is amplified by the chain reaction of explosions ; 6 .
excellent background rejection due to physical granularity of the detector .
because the cells containing the nanoparticles are less than a micron in size , the detector has the resolution to differentiate between wimp nuclear recoils , which only interact with one cell of our detectors , and other backgrounds ( such as @xmath3-particles , @xmath7-particles and @xmath8-rays ) which travel through many cells .
thus , if the background has enough energy to cause the ignition of one cell , then it would ignite multiple cells . in the section @xmath9 , the typical ranges (
@xmath10 @xmath11 m ) of @xmath3 and @xmath7 particles are shown .
7 . depending on the specifics of the detector design , the possibility of directional sensitivity with nanometer tracking
; this possibility will be studied in future papers .
to allow for specific calculation we study oxide - based nano - thermites , which consists of metal spheres with a radius of 5 nm embedded in an oxide .
motivated by their optical , magnetic and electronic applications , metal nanoparticles have been synthesized using both liquid and gas phase methods @xcite@xcite . in situations where the metal nanoparticles are susceptible to oxidation
, the nanoparticles can be coated by a thin layer of an inert metal @xcite . to form a nano - thermite the metal particles
must be mixed by an appropriate gel of oxide @xcite .
alternatively , the oxide can be replaced by an appropriate halide @xcite .
enough energy deposit in the metal sphere heats it up to the point where there is an explosion beginning at the interface of the two materials at the edge of the nanoparticle . as a specific design ,
we imagine constructing a cell " which consists of @xmath12 metal nanoparticles embedded in an oxide .
a full detector will need many of these cells ; e.g. to obtain 1 kg of target material ( the metal ) there will be @xmath13 cells . a wimp hitting
the target will cause only one of these cells to explode .
m is embedded into an insulator , which thermally decouples the cells from each other .
each cell contains more than a few million metal nanoparticles embedded into a metal - oxide .
two different images are depicted at the bottom of the figure : ( a ) shows the design model used for all calculations , and ( b ) represents a more realistic depiction of the nano - thermite detector .
the dissimilarity between both images is the addition of a passivation layer in image ( b ) .
a passivation layer is a metal - oxide coating placed around the nanoparticle in order to prevent oxygen molecules interacting with the metal .
an oxidized metal will not react chemically with a metal - oxide , since it is no longer favorable to gain oxygen atoms .
thus , an oxidized metal will not produce a thermite reaction .
the passivation layer covering the metal nanoparticle would be required in the synthesis of the detector ; since it would prevent oxidation of the metal nanoparticle during construction of the detector ( i.e. before embedding the nanoparticle into the cell ) . as well , in some differing implementations , the metal - oxide of the cell could be comprised of mixed nano - wires @xcite , which would produce a larger temperature increase due to a higher effective thermal resistance between the oxide and the metal .
image ( a ) represents a simplified design model , which enabled analytic results in later sections .
] more precisely , when a wimp elastically scatters with a metal nucleus and deposits energy to the metal , then that energy is converted into a temperature increase .
if the temperature increase is big enough to overcome the potential barrier of the thermite reaction , then the metal will react with the surrounding oxidizer exothermically . in the design using metal nanoparticles ,
after the first thermite reaction of one nanoparticle occurs , the exothermic heat produced by the thermite will heat up the other metal nanoparticles within the 0.5 @xmath11 m cell ; thus creating a chain reaction which amplifies the signal to a measurable effect .
utilizing eq [ thermite ] as an example , the amplification factor for the signal is on the order of @xmath14-@xmath15 .
the detection of the cell explosions could be made by sensitive microphones or spectroscopic studies of the debris .
figure [ detectorfinal ] shows a schematic representation of the nano - thermite detector studied in this paper . on top
, the first picture of figure [ detectorfinal ] shows an array of cells embedded into an insulating material .
the insulator is used to thermally decouple the cells ; so that the reaction within a cell does not cause the explosion of neighboring cells .
the length of each cell is taken to be @xmath16 m .
the spatial scale of the cells enable us to distinguish background from wimp / nucleus collisions .
backgrounds composed of @xmath3 , @xmath7 and @xmath8 particles will traverse multiple cells ; whereas a recoiled ion from a wimp / nucleus collision will only interact with a single cell .
the middle picture in the figure is a magnified view of an individual cell .
inside each cell there will be more than a few million nanoparticles .
the nanoparticles , represented by the white circles , are embedded into the metal - oxide , shown as the black background .
finally , the bottom pictures of figure [ detectorfinal ] depicts an enlarged section of the cell surrounding a single nanoparticle of radius @xmath17 nm .
there are two pictures at the bottom .
image ( a ) shows the simplified model used to make all the calculations in the sections @xmath18 and @xmath19 .
in contrast , image ( b ) depicts a more realistic design for the nano - thermite detector .
a thin passivation layer is placed around the metal to prevent oxidation of the metal during the construction of the detector ( i.e. before embedding the metal nanoparticle into the cell ) .
the passivation layer is a metal - oxide coating placed around the nanoparticle in order to prevent oxygen molecules interacting with the metal .
an oxidized metal will not react chemically with a metal - oxide , since it is no longer favorable to gain oxygen atoms .
thus , an oxidized metal will not produce a thermite reaction . however , the passivating barrier is lowered if the metal nanoparticle or the passivation layer melts due to the temperature increase . in a realistic scenario ,
the synthesis of the nanoparticle embedded into the oxide would require a passivation layer .
it should be noted that the addition of an extra layer between the metal and the oxide of the cell would produce an additional thermal resistance at the interfaces .
this thermal resistance would cause the metal to hold in heat ; and thus , increase the temperature increase yield after a wimp / nucleus collision , when compared to the results presented in this paper . as well , in some differing implementations , the metal - oxide of the cell could be comprised of mixed nano - wires @xcite , which would produce a larger temperature increase due to a higher effective thermal resistance between the oxide and the metal .
as explained in the section @xmath18 , the temperature increase is calculated utilizing the design model of image ( a ) ( i.e. no passivation layer ) and zero thermal resistance between the oxide and the metal nanoparticle .
thus , our calculations are conservative and underestimate the temperature increase sourced by an elastic collision between a wimp and a metal nucleus .
more generally , many other detector designs may be possible , such as two parallel layers of the two components .
this latter design would allow determination of the direction from which the wimp came , as only wimps headed first into the metal ( rather than first into the oxide ) would initiate an explosion . the goal of this paper is to study the ignition of the explosion when a wimp hits the metal nanoparticle . a parallel paper @xcite studies the nano boom dark matter detectors more generally , including methods of detection and readout of the explosion ; alternate explosives other than thermites ; and other aspects of the problem . in this paper
we begin by reviewing the relevant particle and astrophysics of direct detection , and then turn to the viability of a nanothermite detector for wimps . for our calculations
we consider wimp masses of @xmath20 gev .
wimp direct detection experiments seek to measure the energy deposited when a wimp interacts with a nucleus in a detector .
if a wimp of mass @xmath21 scatters elastically from a nucleus of mass @xmath22 , it will deposit a recoil energy @xmath23 where @xmath24 is the reduced mass of the wimp - nucleus system , @xmath25 is the speed of the wimp relative to the nucleus , and @xmath26 is the scattering angle in the center of mass frame . the typical energy transferred to the nucleus in a scattering event is from 0.1 to 50 kev , depending on the wimp mass and the detector material .
typical count rates in detectors are less than 1 count per kg of detector per day
. reviews of the dark matter scattering process and direct detection can be found in refs .
@xcite . over the past twenty five years a variety of designs
have been developed to detect wimps .
they include detectors that measure scintillation ; ionization ; and dilution - refrigerator based calorimeters which measure the total energy deposed by means of a phonon spectrum .
current detector masses range in size up to 100 kg
. the upcoming generation of detectors will reach one tonne .
a major concern in all wimp detectors is backgrounds . to eliminate spurious events from cr ,
the detectors must be placed deep underground ( @xmath0 2,000 m of water equivalent ) . yet
radioactive backgrounds remain and must be eliminated .
thus the experimental determination of annual and/or diurnal modulation is a crucial test of the wimp origin of any events observed in the detector , as most backgrounds should not exhibit the same time dependence .
* particle physics : wimp / nucleus cross sections : * for a supersymmetric ( susy ) neutralino and many other wimp candidates , the dominant wimp - quark couplings in direct detection experiments are the scalar and axial - vector couplings , which give rise to spin - independent ( si ) and spin - dependent ( sd ) cross - sections for elastic scattering of a wimp with a nucleus , respectively .
si scattering is typically taken to be @xmath27 where @xmath28 is the atomic mass of the nucleus , @xmath29 is the wimp - proton reduced mass and @xmath30 is the si scattering cross section of wimps with protons . for large momentum transfer , this relation is multiplied by a form factor correction to account for the sensitivity to the spatial structure of the nucleus .
since the si cross - section grows rapidly with nuclear mass , direct detection experiments often use heavy nuclei to increase their sensitivity to wimp scattering .
spin - dependent ( sd ) wimp - nucleus interactions depend on the spin of the nucleus .
most nuclei have equal numbers of neutrons and protons so that there is no sd contribution ; specific nuclei must be chosen in experiments to search for nonzero sd couplings .
sd scattering is often of lesser significance than si scattering in direct detection experiments for the heavy elements used in most detectors due to the extra @xmath31 coherence factor in the cross section .
* astrophysics : velocity structure of the galactic halo : * the velocity distribution @xmath32 of dark matter particles in the galactic halo is crucial to their signals in dark matter detectors ( as first stressed by @xcite ) .
the dark matter halo in the local neighbourhood is likely to be composed mainly of a smooth , well mixed ( virialised ) component with an average density @xmath33 gev/@xmath4 .
the simplest model of this smooth component is the standard halo model ( shm ) , a spherically symmetric nonrotating isothermal sphere with an isotropic , maxwellian velocity distribution characterized by an rms velocity dispersion @xmath34 km / sec ; the distribution is truncated at escape velocity @xmath35 550 km / sec in the galactic rest frame , or @xmath35 750 km / sec in the laboratory frame ( the frame of the earth ) where the measurements are made .
the resultant count rates in direct detection experiments due to the shm were first discussed in @xcite .
a key issue for dark matter detectors using nano - explosives is the question of whether or not a wimp deposits enough energy in the detector to initiate an explosion .
the wimp velocity plays a key role , as can be seen in eq .
( [ eq : energy_deposit ] ) .
we will study two different wimp velocities .
first , we take the typical wimp speed , which is roughly 300 km / sec in the lab frame . however , it is possible that , although the typical wimp is not sufficient to set off an explosion , there are plenty of high velocity wimps on the tail of the distribution that do cause explosions .
in fact many existing dark matter detectors ( such as lux , xenon , and cdms ) rely on this tail in obtaining results for low mass wimps near the threshold of their sensitivities .
thus , as our second case , we will consider fast wimps with speeds of 700 km / sec in the lab frame .
herein , we study both cases for oxide - based nano - thermites ; and aim to extend the study both cases for halide - based nano - thermites in future work .
a related question is the value of the energy recoil , which depends on the scattering angle ( see eq .
( [ eq : energy_deposit ] ) ) .
the maximum recoil energy takes place for forward scattering with an angle of @xmath36 , i.e. @xmath37 in some cases a more typical value of the energy recoil is half of the maximum , but an accurate calculation including form factors for the detailed interactions with the nucleons inside the nucleus would be required and is the subject of future work . for our case of 300 km / sec wimp speed ,
we assume a characteristic energy of 1/2 the maximum energy recoil ; whereas for the fast wimps with 700 km / sec , we assume the maximum energy recoil . in the future
it would be interesting to integrate over the entire wimp velocity distribution , with results dependent on the detector material of choice as well as wimp mass .
* current experimental status for direct detection : * in the past decade , a host of direct detection experiments using a variety of different detector materials and designs have reported unexplained nuclear recoil signals which could be due to wimps .
detection of annual modulation has now been claimed by the dama and , more recently , cogent experiments .
the italian dark matter experiment , or dama @xcite , consists of 250 kg of radio pure nai scintillator situated in the gran sasso tunnel underneath the apennine mountains near rome , and became the first direct detection experiment to observe a positive signal .
the group now has accumulated 1 ton - yr of data over the past decade and finds an 8.9 @xmath38 annual modulation with the correct phase and spectrum to be consistent with a dark matter signal .
recently cogent @xcite , consisting of germanium , also claim to see annual modulation of the signal with the correct phase to be consistent with wimps , and together with a third cresst - ii @xcite experiment , could be seeing @xmath3910 gev wimps .
the cdms experiment also has seen a few unexplained events in their silicon detectors at low energies @xcite , that might be compatible with low mass wimps .
however , other experiments , notably cdms - ge @xcite , supercdms @xcite , xenon @xcite and lux @xcite , have null results that conflict with these positive signals and may rule them out .
many direct detection experiments are either currently running or gearing up to do so , and we can expect more data soon .
the coupp @xcite@xcite , picasso @xcite , simple @xcite , and pico @xcite ( a new collaboration between the coupp and picasso teams ) experiments are the most similar to the nano explosives proposed here .
picasso and simple use superheated droplet detectors and coupp uses bubble chambers . these detectors go through a phase transition when hit by a wimp ; easily visible bubbles form in the bubble chambers .
these detectors operate at room temperature and are sensitive to low mass wimps , down to about 15 gev .
the new pico collaboration aims for a 3 kev energy threshold , allowing it to study wimps of even lower masses ; their limitations are due to gamma - ray backgrounds .
one of our goals is to explore new materials and designs that allow the detector to have even lower threshold and sensitivity to even lighter wimps .
because our targets are nm in size , we have the resolution to differentiate between wimp nuclear recoils , which only interact with one cell of our detectors , and other backgrounds ( such as @xmath3s and @xmath8s ) which travel through many cells and cause multiple ignitions . in the past few years
the cross - sections that have been reached by detectors have improved by three orders of magnitude ; over the next few years another two orders of magnitude should be reached .
the next generation of detectors being built will be one tonne in mass or directional .
a review of the theory and experimental status of dark matter detection can be found in @xcite .
one of the goals of using nano explosive dm detectors is to design low threshold detectors that can test the light mass @xmath110 gev hypothesis .
for a given energy deposit by a wimp ( given in eq .
( [ eq : energy_deposit ] ) , the amount that the nanoparticle heats up is determined by @xmath40 where the heat capacity of the nanoparticle @xmath41 depends on the material as well as on the size of the nanoparticle .
it is instructive to see specific examples of eq [ [ eq.heat_capacity ] ] in order to get an idea of the amount of temperatures roughly expected to be produced inside the nanoparticle .
the specific heat per volume of al is @xmath42 kev / nm@xmath43 . the heat capacity for an aluminum nanoparticle of radius @xmath44 nm is @xmath45 kev / k .
thus , the expected temperature increase for an aluminum nanoparticle with an energy of @xmath46 kev deposited by a wimp interacting with an aluminum nucleus is @xmath47 k. the ignition temperature for a micron - size @xmath48 thermite necessary for the reaction to begin is roughly @xmath49 k , which implies that a nanoparticle with radius ( @xmath50 nm ) could produce high enough temperatures for the thermite reaction to begin and produce an explosion .
the low values of the heat capacity for metal nanoparticles allows for a very large temperature increase .
this is one of the main reasons for pursuing nano - thermites as a possible wimp detector . in this paper
, we aim to improve our previous calculation with eq [ [ eq.heat_capacity ] ] in order to establish if the nanoparticle could work as a wimp detector or not .
there are two issues that need to be considered .
the first complication arises from the fact that the metal nanoparticle will not retain the temperature increase for an infinite amount of time and will dissipate heat following the dynamics of the heat transfer equation .
the question of whether the thermite reaction will begin or not needs to be addressed by studying the heat transfer equation in order to establish the amount of time the nanoparticle stays heated .
then the arrhenius equation approximates the reaction rate given a specific chemical reaction , size of the nanoparticle and temperature .
this reaction rate is multiplied by the timescale determined from the heat transfer equation to estimate the probability of a nanoparticle exploding .
the arrhenius equation allows us to estimate the quantum efficiency of detonation , the probability that a detonation would occur following a nucleus recoil event .
the second complication considered is the distance traveled by the recoiling nucleus .
it is expected that sometimes the recoiling nucleus will escape the metal nanoparticle after interacting with the wimp .
the escaping nucleus will not depose all of its energy into the nanoparticle . the range at which the recoiling nucleus stops needs to be understood in order to approximate correctly the amount of energy deposed into the metal nanoparticle .
calculations to address these two complications ; utilizing the heat transfer equation , the reaction rate and stopping distance for the recoiling nucleus in order to establish the viability of nano - thermites as wimp detectors ; will be done in later sections below .
we may ask what temperature increase is required to initiate an explosion . here
we describe the basic idea , and continue in detail in the next section .
we treat the system as a phase transition with a barrier that must be overcome in order for the thermite reaction to take place .
the ability to use nano - thermites for dark matter detection is pulled in two contrary directions . on the one hand ,
the material must be chosen so that it does not spontaneously explode due to thermal fluctuations .
we may require no spontaneous explosion for at least one year ; this requirement determines the barrier height for the chemical reaction
. the required barrier height can be quite significant and is particularly restrictive for a detector operating at room temperature . yet , on the other hand , we would like the most sensitive possible detectors to incoming wimps .
ideally the materials with the smallest required temperature increase ( due to a wimp hit ) would detect the most wimps .
we ve seen that typical wimp interactions deposit 0.1 - 50 kev s of energy .
given the barrier height required to avoid spontaneous combustion , temperature increases of @xmath01000 k are required for the thermite reaction to take place at room temperature .
we will see that heavy wimps with masses above 100 gev ( or for some materials above 1 tev ) are able to deposit enough energy ( @xmath010 kev ) to cause ignition at room temperature for the detector geometry considered in this paper ( other more favorable geometries are considered in the companion paper @xcite .
it is also possible to operate at cryogenic temperatures , such as 77 k using liquid nitrogen or 4.2 k using liquid helium as coolants . at these lower temperatures , the thermal fluctuations that produce spontaneous ignition
are less effective , and we may choose a material with a lower barrier height for the phase transition without spontaneous detonation . in this case the detector can react to lower temperature increases @xmath39 50 k , corresponding to lower energy thresholds , for some materials as low as @xmath39 0.5 kev . some of the explosives that are designed to operate at 4.2 k may not be stable at room temperature . to realize such explosives we need to develop cryogenic methods of mixing metal nanoparticles and oxides . when operating cryogenically , the nano explosive wimp detector can detect @xmath110 gev wimps better than any previous dark matter detector .
we can see the advantages of using nano - thermites , rather than larger micron - sized ones : 1 .
only when the objects are nano - sized is their specific heat small enough to allow operating at room temperature .
smaller detector elements have smaller @xmath41 and thus , for the same energy deposited by a wimp , larger temperature increase ( see eq.([eq.heat_capacity ] ) ) . for nano - sized thermites
the temperature increase due to wimp interactions is large enough to cause an explosion ; whereas for micron - sized thermites the temperature increase would be too low . 2
. low energy threshold of 0.5 kev , allowing for study of low mass @xmath6gev wimps , can be achieved when : 1 .
operating at cryogenic temperatures ; 2 . using chemically active metals with low melting temperatures such as : gallium , rubidium , caesium , indium , tin , lead or bismuth ; 3 .
employing f - based ( fluorine - based ) nano - thermites
( e.g. al @xmath51 wf@xmath52 ) ; 4 . considering more advance models other than nanoparticles embedded in an oxide . +
the possibility of reaching lower energy thresholds through the implementation of cryogenic temperatures will be further discussed in this paper .
a detailed discussion for the design concepts ( b)-(d ) is beyond the scope of this paper and will be discussed in forthcoming papers .
3 . only when the detector elements have a size smaller than the track range of background can one use new methods of background rejection .
the track lengths of recoiling nuclei are @xmath53 nm , which is much smaller than the typical range of an @xmath3 particle ( approximately @xmath54 m ) emitted by radioactive decay . in nanothermites , a wimp will make one and only one cell explode , while other background particles will cause many cells to explode .
in contrast , currently employed detectors with a physical granularity greater than approximately @xmath54 m will not be able to make this differentiation .
4 . directional detectors are possible , not with spherical nanoparticles , but with asymmetric detector designs that may be the study of future work . here
the goal is to obtain the direction that the wimp came from ; for the case of forward scattering this is the same as the direction of the nuclear recoil .
directionality would prove wimp detection with much less statistics , with only @xmath39 100 wimps required ; in addition one would learn about the structure of the dark matter halo . for directional sensitivity , the detector resolution must be smaller than the size of the track range of recoiling nuclei in order to measure the particles track . if the resolution is micron - sized , while the track length is nanometers , then of course the track will be impossible to follow .
thus nano - scale detectors in principle have the capability to obtain nanometer track resolution .
the spherical nanoparticles studied in this paper do not have directional sensitivity but other designs may .
we take the rate of the thermite reaction @xmath55 at temperature @xmath56 to be represented by the arrhenius equation @xmath57 where @xmath58 is the number of interacting sites where a metal nucleus could chemically interact with the oxide , and @xmath28 is the arrhenius prefactor which is unique for each reaction but can be approximated by the vibrational frequency at the interface of the reactants .
the vibrational frequency of a crystal is roughly @xmath59 hz .
a survey of published values for the arrhenius prefactor @xmath28 experimentally measured for solid decomposition reactions showed a slight predominace of @xmath28 values between @xmath60 hz - @xmath59 hz@xcite . in this paper , we will approximate @xmath61 hz , but wish to note that differences in the arrhenius prefactor of order @xmath62 will not change our future results of the desired ignition and activation temperatures dramatically ( less than an order of magnitude ) due to the dominance of the exponential term .
the arrhenius equation eq [ eq : arrhenius ] will allow us to calculate the probability of ignition of the nano - thermite following a transient temperature increase .
now we turn to computing two important temperatures : one characterizing the barrier height of the phase transition and the other , the ignition temperature needed for a thermite reaction to take place .
we will find that each of these temperatures must exceed a minimal value : the temperature characterizing the barrier height must be high enough to prevent spontaneous explosion even when no wimp has hit the detector , and the ignition temperature @xmath63 sets a minimum value needed in order for an explosion due to a wimp interaction to take place .
these two conditions are important considerations when choosing a metal and oxide to make up the oxide - based thermite we will use in dm detectors . _ the first condition : _ the barrier height is characterized by the activation energy " + @xmath64 ( where @xmath65 is the boltzmann constant ) and its corresponding activation temperature " @xmath66 . as mentioned above , the barrier height is determined by requiring the thermite to be stable at room temperature @xmath67 k to thermal fluctuations for at least one year , i.e. , we require @xmath68 . even though the latter requirement is a conservative estimate
, we will later show that a nano - thermite dark matter detector could meet such a constraining stipulation .
in adapting eq.([eq : arrhenius ] ) to the requirement @xmath68 and computing the prefactor @xmath58 , we must add up all metal nuclei that could chemically interact with the metal - oxide and possibly produce an explosion within the entire detector .
as mentioned previously , each cell has a radius @xmath69 nm . since the typical size for the lattice constant @xmath70 , which measures the separation between nuclei in the metal , is a few angstroms ( @xmath71 ) , the number of sites per nanoparticle of radius @xmath72 nm is @xmath73 thus , the total number of interacting sites @xmath74 found in a detector of total mass @xmath75 is given by @xmath76 where @xmath77 and @xmath78 are the density and volume of a single nanoparticle . to obtain a numerical value we will take @xmath79 kg . to be as conservative as possible
, we will take the lowest metal density for all the elements we consider : aluminum .
then we find @xmath80 and set @xmath81 in eq.([eq : arrhenius ] ) .
now we can impose the condition of having a stable detector at room temperature that will not have any thermite reaction in the absence of a dark matter interaction for a running time of @xmath46 yr : @xmath82 thus , in searching through all possible thermite elements for possible use as dm detectors , we should choose those with thermite activation temperature @xmath66 greater than @xmath83 k. we note that because of additional methods of background rejections such as annual modulation , we could select a detector to be less stable to thermal fluctuations ( e.g. @xmath84 ) .
nevertheless , our previous result is very robust because of the exponential nature of the arrhenius equation ( i.e. the same calculation of the activation temperature for any varying conditions under the assumed simple nanoparticle / oxide model will not vary much from this result eq@xmath85 $ ] ) .
_ the second condition : _ now we can proceed to find the ignition temperature @xmath63 needed for a signal to be seen in our detector .
we want to find the minimal temperature increase required by the energy deposited in a wimp interaction that can lead to an explosion .
when a wimp hits a nucleus inside a metal nanoparticle with radius @xmath86 nm , the nucleus typically traverses the entire nanoparticle ( or even somewhat farther ) .
we may take the entire nanoparticle to be heated by some temperature @xmath87 .
however , this value of the temperature does not last more than nanoseconds . in the next section
we will solve the heat transfer equation to estimate the diffusion timescale of the heat out of the nanoparticle .
shortly after a characteristic timescale known as the conduction time " @xmath88 , the metal nanoparticle is no longer hot enough to induce ignition .
thus the appropriate timescale with which to multiply the rate in eq.([eq : arrhenius ] ) is this conduction time . to successfully have an exploding nanoparticle operating at some temperature @xmath87 , we thus require @xmath89 .
we will see below that the conduction time is given by @xmath90 , where @xmath91 is the radius of the metal nanoparticle and @xmath3 is the thermal diffusivity of the metal .
the biggest metal thermal diffusivity studied will be on the order of @xmath92 m@xmath93s@xmath94 ; thus the shortest time - scale considered will be @xmath95 s. however , if the passivation layer is introduced between the metal nanoparticle and oxide , then the conduction time may be order of magnitudes longer . in our detector model , we are interested in the explosion of a single metal nanoparticle embedded in an oxide .
thus , we take @xmath58 to be the number of sites on a single nanoparticle , @xmath96 .
the ignition temperature @xmath63 is now given by considering the following inequality : @xmath97 the last inequality follows from taking the lowest value of the activation energy allowed by eq.([eq : ta ] ) , @xmath98 k. thus , in order to have a detector that can ( a ) run for a full year without any spontaneous thermite reactions and ( b ) detect signals when a dark matter particle interacts with the nanoparticle , the ideal thermite reaction for the detector is one with an activation temperature @xmath99 k and an ignition temperature @xmath100 k. the value for the ignition temperature needed for the nano - thermite to ignite is different from values quoted in literature due to the difference in time and size scale . in order to find the ignition temperature of a given thermite
, experimentalist usually heat the bulk of the materials for prolonged periods of times when compared to the conduction time ( on the order of seconds - minutes ) . as an example , if we substitute @xmath101 , then the resulting ignition temperature @xmath102 is closer to experimental results for thermite ignition temperatures : @xmath103 k. for this reason , the minimum temperature needed for a single nano - particle to ignite due to a recoiling metal nucleus under our assumed simple model is higher than expected .
note that the ignition temperature is also a function of the activation energy , so that an increase in the activation energy increases the ignition temperature .
as well , the activation and ignition temperatures were calculated after alleviating the condition of having no signals sourced by thermal fluctuations for a year ( @xmath104 ) to 1 per day ( 1 per hour ) : @xmath105 k ( @xmath106 k ) and @xmath107 k ( @xmath108 k ) .
this calculation was done in order to gain perspective under our assumed simple model into how the activation and ignition temperatures change as a function of the stability timescale of the detector .
it should also be mentioned that the ignition temperature might be higher than the melting temperature of the metal and/or oxide .
the change in phase of the metal would change the previous calculation by increasing the amount of sites able to interact chemically ; and thus , lowering the ignition temperature . in order to be consistent and conservative in our calculation
, we will adopt @xmath109 k as our ignition temperature in all future comparisons . nevertheless , it is possible to have a different detector design that lowers the ignition temperature needed without sacrificing the stability of the nanothermite detector to thermal fluctuations .
our previous calculations have been done under the assumption that the explosion of one nanoparticle of radius @xmath110 nm initiates the chain reaction throughout the cell , which is interpreted as a signal . however , it is possible to conceive a detector in which two or more smaller metal nanoparticles ( e.g. @xmath111 nm ) need to ignite in order to initiate the chain reaction throughout the cell .
this multi - particle ignition mechanism suppresses activation of the nano - thermite cell due to thermal fluctuations and permits a lower ignition temperature ( i.e. energy threshold ) .
the specific design aspects for the multi - particle ignition detector is currently being pursued , but is beyond the scope of this paper and will be discussed in future work .
there are many different implementations of the nano - thermite detector that are worth pursuing in more detail . in this paper
we concentrate on the simple model of a metal nano - particle embedded in an oxide .
the minimum ignition temperature adopted for future comparison will be the conservative estimate of @xmath112 k. now that we have calculated the ignition temperature needed , in the next section we will study which metals have the necessary thermal and physical properties to reach the required ignition temperature @xmath113 and explode when struck by a wimp .
we will consider a variety of wimp masses , @xmath20 gev .
we proceed now to calculate the temperature increase given to a nanoparticle by a wimp collision with a nucleus in the metal . while eq.[[eq.heat_capacity ] ] gives a rough idea of the temperature increase , here we will compute this quantity more carefully .
a metal nucleus recoiling from a wimp interaction moves a certain distance before stopping .
the amount of energy per length lost by the metal nucleus as it traverses through the metal nanoparticle and/or oxide is given by the stopping power ( @xmath114 ) .
consequently , the range ( stopping distance ) is given by @xmath115 where @xmath116 is the initial energy of the metal nucleus and @xmath117 is the total stopping power given by lindhard ( 1961 ) theory @xcite when considering a nucleus of type @xmath118 moving through a medium of type @xmath119 .
lindhard theory is valid for heavy ions with a few kev of energy . a different theory , bethe - bloch , needs to be used when considering backgrounds like @xmath3 or @xmath7 particles , since they have larger energies ( usually a couple of mevs ) and a small charge ( @xmath120 or @xmath121 ) .
the stopping power of backgrounds will be discussed in a later section .
the total stopping power for a slowly moving heavy ion can be separated into the electronic @xmath122 and nuclear @xmath123 components , such that @xmath124 .
utilizing lindhard s equations @xcite , we can approximate both stopping powers as : @xmath125 in the equations for the nuclear and electronic stopping power , the subscripts @xmath126 correspond to the parameters of the scattered nucleus and the medium respectively , @xmath127 is the atomic number , @xmath128 is the atomic number density of the medium and @xmath22 is the atomic mass measured in atomic mass units . in this case
@xmath129 pertains to a metal nucleus moving through the metal nanoparticle and @xmath130 is the stopping power of a metal nucleus moving through the outside medium .
we will neglect any geometric factors and assume that all of the energy of the recoiling nucleus will be deposited into the nanoparticle if @xmath131 . if @xmath132 then only a fraction of the recoiling energy will be deposited into the nanoparticle . in the latter case ,
when the ion escapes the metal nanoparticle , the energy deposited into the nanoparticle @xmath133 can be found by solving for the final energy @xmath134 in the following equation : @xmath135 the effective amount of energy deposited into the nanoparticle , @xmath136 , is given by @xmath137 it is this quantity that determines the temperature increase of the metal nanoparticle .
we take the entire nanoparticle to be heated uniformly by this amount of energy .
the initial temperature increase of the nanoparticle is then given by a modified version of eq.([eq.heat_capacity ] ) , @xmath138 let us assume that the detector is operating at some uniform background temperature @xmath139 .
for example , it might be at room temperature @xmath140 .
then the entire nanoparticle is initially heated by the wimp interaction to @xmath141 in time , the temperature increase of the nanoparticle due to the wimp interaction dissipates .
we will now solve the heat transfer equation in order to follow the evolution of the temperature profile @xmath142 .
this allows us to calculate the temperature near a single nanoparticle as a function of time in order to establish the characteristic time scale for how long the metal particle remains heated , and give us its temperature at the time of dissipation .
the goal is to determine whether the temperature found at the interface is greater than the ignition temperature assumed for our model @xmath143 k , in which case the nano - thermite in question could plausibly work as a wimp detectors .
the heat transfer equation is solved for a single sphere of radius @xmath91 embedded into a semi - infinite medium with different thermal properties .
the heat diffusion of a sphere to a surrounding medium has been analytically calculated by a. brown ( 1965 ) @xcite .
our calculation assumes zero thermal resistance at the interface .
thus , the calculated temperature increase may be much lower than in the real system , since a finite resistance will contain the heat inside the nanoparticle .
in the following calculation , we will study the temperature increase @xmath144 profile as a function of radial distance ( @xmath145 ) and time ( @xmath146 ) .
specifically , the heat transfer equation @xmath147 is solved given the thermal conductivity ( @xmath148 ) and thermal diffusivity ( @xmath149 ) of the sphere and medium .
the subscript 1 " pertains to the sphere , whereas the subscript 2 " is given to the medium parameters .
the following initial conditions ( ic ) and boundary conditions ( bc ) are chosen in order to solve the heat transfer equation : @xmath150 the boundary condition found in the last line implies that there is no thermal resistance and no heat source at the interface . following ( brown 1965)@xcite
, we find that the temperature is given by @xmath151 where @xmath152 it can be seen from the solution above that there is a natural time - scale @xmath88 that arises from solving the heat eaquation ; @xmath153 .
this time - scale , which will be referred to as conduction time , gives a modest estimate as to how long does the metal nanoparticle stay heated .
figure [ [ temphist2 ] ] shows how the temperature diffuses from the nanoparticle to the medium at times @xmath154 .
thus , one can approximate the amount of time that the nanoparticle stays at a high temperature by @xmath88 .
the temperature found at the interface @xmath155 measured at a time @xmath156 is the relevant temperature value that needs to be compared to the ignition temperature @xmath87 in order to determine if the thermite reaction will ignite .
the temperature increase at the interface ( r=@xmath91 ) is given by : @xmath157 so that @xmath158 in the previous equations for @xmath159 , we have explicitly made the substitution @xmath160 and definition @xmath161 .
after a wimp / nucleus collision with maximum energy @xmath162 and speed of @xmath163 km / s .
the different plots represent @xmath159 at times @xmath164 , @xmath165 and @xmath166 as a function of the distance from the center of the metal nanoparticle .
the nanoparticle significantly cools shortly after the conduction time , @xmath167 .
this feature is general to all materials considered .
the rapid cooling at times greater than @xmath88 is due to the exponential term in eq [ [ delttt ] ] .
thus , @xmath88 serves as a very good estimate for the total time the nanoparticle is heated . ]
as can be easily appreciated , there are many physical parameters that can influence the temperature . to reduce the number of parameters we need to study
, we will make the following simplifying assumptions .
we will use the semi - empirical mass formula to approximately relate the atomic number to the atomic mass : @xmath168 thus we eliminate @xmath169 as a free parameter in the calculation of the of the stopping power in eq ( [ stopping]-[stopping1 ] ) , and consequently the effective energy in eq ( [ effectiveenergy ] ) . the thermal diffusivity will be simplified by taking advantage of the relation @xmath170 and the dulong - petit law which states that the heat capacity of a solid in crystalline form is given by @xmath171 @xmath172 thus , the thermal diffusivity is approximated by @xmath173 it should be noted that the dulong - petit law overestimates the heat capacity for light atoms bonded strongly to each other at room temperature such as beryllium , and for most solids kept at cryogenic temperatures ( i.e. 4.2 k or 77 k ) . in both cases , the overestimation of the heat capacity will give a smaller thermal diffusivity , which as a consequence gives a smaller @xmath159 in eq @xmath174 $ ] .
thus , use of the dulong - petit is a conservative assumption that leads to a calculated temperature increase smaller than reality . taking into account eq [ z ] and eq [ eq : nnano ] , the number of parameters is reduced to five : density ( @xmath175 ) , thermal conductivity ( @xmath176 ) and atomic mass ( @xmath22 ) of the metal , the wimp velocity ( @xmath25 ) , and the wimp mass ( @xmath177 ) . now
that we can calculate the temperature increase as a function of the 5 chosen parameters ( @xmath178 ) , we can study which metal would produce a high enough temperature to overcome the ignition temperature needed .
note , that the calculation for the temperature increase is independent of the assumed ignition temperature @xmath63 .
thus , we can select the optimal metals from our calculations of the temperature increase , and perform future experiments to confirm the viability of the nano - thermite as a dark matter detector .
the equation for the temperature increase gives the temperature profile for a sphere of a given material surrounded by a medium of another material .
specifically , eq [ @xmath179 will be used in order to calculate the temperature increase expected for a metal sphere embedded into a specific oxide ; copper ( ii)-oxide .
thermite reactions work by putting two metals of very different reactivity together and letting the more reactive ( in a chemical sense ) steal the oxygen from the least reactive .
specifically , we choose copper ( ii ) oxide ( cuo ) as the material for the medium due to the low reactivity of copper .
then any metal more chemically reactive than copper will create a thermite reaction when exposed to cuo at a temperature greater than the ignition temperature @xmath87 .
the fact that copper is one of the least reactive metals allows us to consider a large family of metals to use as nanoparticles .
another very promising metal - oxide is tungsten trioxide ( wo@xmath2 ) . like copper ,
tungsten is also a very low reactive metal .
the study of which metal - oxide would work best is beyond the scope of this paper and warrants further examination .
we wish to calculate the temperature increase caused by a wimp / metal - nucleus elastic collision at the interface of the metal nanoparticle with copper ( ii ) oxide for a couple of metal targets in order to identify potential metals that could produce enough temperature to create a thermite reaction .
four different metals are chosen as test metals : aluminum , ytterbium , thallium and tantalum .
figure [ [ contour ] ] shows the reason for this choice of metals .
the first argument is the fact that their densities range from one of the lightest elements in the periodic table , aluminum , to one of the densest active metals , tantalum . as a function of atomic mass @xmath22 and thermal conductivity @xmath176 of the metal used ,
given the densities : @xmath180 g/@xmath4 .
the chosen densities correspond to the metals : aluminum , ytterbium , thallium and tantalum , respectively .
the red dots on each contour plot shows where each metal lies .
the reason for showing these graphs , even though they span an unreal parameter space of density , thermal conductivity and atomic mass , is to show the general trend of the temperature output as a function of @xmath181 .
the hope is that the reader can familiarize himself / herself with this trend and , if interested , possibly explore a new set of metals as potential candidates for a nano - thermite wimp detector.,title="fig : " ] as a function of atomic mass @xmath22 and thermal conductivity @xmath176 of the metal used , given the densities : @xmath180 g/@xmath4 .
the chosen densities correspond to the metals : aluminum , ytterbium , thallium and tantalum , respectively .
the red dots on each contour plot shows where each metal lies .
the reason for showing these graphs , even though they span an unreal parameter space of density , thermal conductivity and atomic mass , is to show the general trend of the temperature output as a function of @xmath181 .
the hope is that the reader can familiarize himself / herself with this trend and , if interested , possibly explore a new set of metals as potential candidates for a nano - thermite wimp detector.,title="fig : " ] as a function of atomic mass @xmath22 and thermal conductivity @xmath176 of the metal used , given the densities : @xmath180 g/@xmath4 .
the chosen densities correspond to the metals : aluminum , ytterbium , thallium and tantalum , respectively .
the red dots on each contour plot shows where each metal lies .
the reason for showing these graphs , even though they span an unreal parameter space of density , thermal conductivity and atomic mass , is to show the general trend of the temperature output as a function of @xmath181 .
the hope is that the reader can familiarize himself / herself with this trend and , if interested , possibly explore a new set of metals as potential candidates for a nano - thermite wimp detector.,title="fig : " ] as a function of atomic mass @xmath22 and thermal conductivity @xmath176 of the metal used , given the densities : @xmath180 g/@xmath4 .
the chosen densities correspond to the metals : aluminum , ytterbium , thallium and tantalum , respectively .
the red dots on each contour plot shows where each metal lies .
the reason for showing these graphs , even though they span an unreal parameter space of density , thermal conductivity and atomic mass , is to show the general trend of the temperature output as a function of @xmath181 .
the hope is that the reader can familiarize himself / herself with this trend and , if interested , possibly explore a new set of metals as potential candidates for a nano - thermite wimp detector.,title="fig : " ] in order to study which metals would work best for the nanoboom detector under different limits of the wimp / nucleus interaction , the analysis is divided into two distinct sections .
the two differing cases are : i ) typical recoil energy , @xmath182 is deposited to the metal nucleus at the mean speed for galactic wimps in the lab frame ( @xmath183 km / s ) , and ii ) a very energetic wimp with a velocity close to the escape velocity ( @xmath163 km / s ) deposits the maximum energy , @xmath184 to the metal nucleus . for each case , the temperature increase at the interface is calculated for each of the four metals and assuming a wimp mass of either @xmath185 .
given a mass for the wimp and a specific consideration of the recoil energy , either case ( i ) or ( ii ) , then the temperature increase @xmath159 calculated for the metal nanoparticle is dependent on three further parameters : the density ( @xmath186 and thermal conductivity @xmath187 ) of the metal and the mass of the metal nucleus @xmath188 the plot shown in figure [ [ contour ] ] gives @xmath159 assuming a wimp mass of @xmath189 gev , a recoil energy @xmath190 with speed @xmath183 km / s and four different densities : @xmath191 @xmath192 which are the densities of aluminum , ytterbium , thallium and tantalum respectively .
the reason for choosing ytterbium , thallium and tantalum is due to their relatively large mass ( @xmath193 amu , @xmath194 amu and @xmath195 amu ) , which gives a higher temperature increase .
figure [ [ contour ] ] shows that for a @xmath196 gev wimp @xmath159 increases for metals with a higher thermal conductivity and mass .
the fact that a large thermal conductivity produces a higher temperature is seen clearly in the boundary condition .
looking at equation [ refbc ] one can see that if @xmath197 then @xmath198 , which is the boundary condition found for an insulator .
the difference in thermal conductivity between the metal and the oxide induces a phonon spectrum mismatch at the interface , which creates an effective thermal resistance .
this implies that materials with a very large thermal conductivity compared to cuo will retain the heat inside , and consequently have a higher temperature at the interface .
the recoil energy grows as the difference between the mass of the metal nucleus and wimp mass decrease , because the term @xmath199 in the equation for @xmath162 is maximized for @xmath200 .
thus , the temperature output produced by the wimp / metal - nucleus interaction will increase as the mass of the metal increases for very heavy wimps ( @xmath201 gev ) and gets closer to the wimp mass .
in contrast , the temperature increase for low mass wimps ( i.e. @xmath202 gev ) will be greatest for metals with low atomic mass ( e.g. aluminum ) , closest to the mass of the wimp .
aluminum is a low atomic mass metal with a high thermal conductivity , which are helpful characteristics for the study of low mass wimps .
furthermore , al@xmath203 is an attractive target for the study of spin - dependent wimp / nucleus interactions .
as well , aluminum is a popular metal fuel in thermite reactions .
aluminum based thermite reactions have been heavily studied within the scientific community and used in industry for a long time . by considering aluminum as one of our metal targets ,
we gain perspective as to what type of thermite reaction should be studied in order to make the nanoboom detector work .
.this table gives the temperature increase for the four chosen metals when interacting with wimps of differing mass with a recoil energy given by one of the two considered cases .
the temperature is given in kelvins .
the subscript @xmath204 ( @xmath205 ) of @xmath159 specifies if the recoil energy considered was from case i ( case ii ) . for comparison ,
the superscript * " indicates a temperature increase higher than the melting temperature of its respective metal .
the total temperature at which the metal heats up is given by adding room temperature to the resulting temperature increase : @xmath206 k. the total temperature @xmath207 is then compared to @xmath208 k in order to establish which metals would work as nano - thermite detectors at room temperature .
only thallium and tantalum could work at room temperature .
nevertheless , aluminum and ytterbium are possible metals for a nano - thermite detector operating at cryogenic temperatures . [
cols="^,^,^,^,^",options="header " , ] the table shows the results for the total temperature increase found for the chosen four metals when interacting with wimps of differing mass with a recoil energy given by one of the two cases considered .
the relevant parameters of copper ( ii ) oxide needed to calculate the temperature increase are : density ( @xmath209 g/@xmath4 ) , thermal conductivity ( @xmath210 w/(mk ) ) and heat capacity ( @xmath211 j/(kg))@xcite @xcite .
the subscript @xmath204 or @xmath205 of @xmath159 specifies if the recoil energy considered was a typical elastic collision with @xmath212 at a speed of @xmath204 km / s , or if the collision was very energetic with @xmath213 at a speed close to the escape velocity , @xmath163 km / s , respectively . for comparison ,
the superscript * " indicates a temperature increase higher than the melting temperature of its respective metal .
the total temperature at which the metal heats up is given by adding room temperature to the resulting temperature increase : @xmath206 k. in order to establish which metals would work as nano - thermite wimp detectors , we then compare the total temperature @xmath207 to the ignition temperature @xmath208 k. the results show that ytterbium has the highest temperature increase of the four metals when considering case i and wimps of mass @xmath214 gev .
ytterbium increases to a temperature of @xmath215 k whenever a typical collision happens between a wimp of mass @xmath214 gev ( i.e. case i ) .
ytterbium is the best metal out of the studied four in case i for a wimp of mass @xmath214 gev , because the term @xmath216 in the equation for @xmath162 eq [ [ eq : emax ] ] is maximized for ytterbium .
note that tantalum and thallium have a higher temperature when considering case ii and a wimp mass @xmath217 gev , because they are much denser metals than ytterbium .
thus , the stopping distance of a metal nucleus moving through thallium and tantalum would be shorter than in ytterbium .
this implies that thallium and tantalum recoiling nuclei would deposit a higher fraction of its total energy into the nanoparticle . +
the calculations show that a tantalum metal nanoparticle at room temperature could get hotter than @xmath218 k when considering highly energetic collisions ( @xmath219 @xmath220 gev and @xmath163 km / s ) .
thus , metal nanoparticles composed of tantalum could serve as a wimp detector operating at room temperature for @xmath217 gev .
similarly , a nano - thermite composed of thallium metal nanoparticles could detect wimps with mass @xmath221 gev .
in contrast , the table also shows that aluminum has a consistent temperature increase for a larger range of wimp mass .
specifically , if aluminum were to be a metal with an ignition temperature @xmath222 k , then aluminum nanoparticles could work to measure light wimps .
aluminum has a higher temperature increase for low mass wimps compared to ytterbium , thallium and tantalum , because it has the highest thermal conductivity and lowest mass difference to a 10 gev wimp . according to eq [ @xmath179 , case ii wimp / metal elastic collision can produce @xmath223 k for wimps with mass @xmath224 gev .
it should be noted that a detector composed of al / cuo and with an ignition temperature of @xmath225 k has an energy threshold of about 2kev . such a detector can be made stable if it operates at cryogenic temperatures .
operating the detector at cryogenic temperatures could increase its sensitivity in measuring low mass wimps .
we will consider configuring the detector at two different cryogenic temperatures : @xmath226 k. lowering the temperature at which the detector works allows us to consider thermite reaction with more favorable activation and ignition temperatures .
as well , a cryogenic detector opens the possibility of using halide - based thermites or other decomposition reactions .
herein , we discuss the possibility of an oxide - based nano - thermite dark matter detector working at cryogenic temperatures .
thus , following the same argument used to get equation eq@xmath85 $ ] and eq@xmath227 $ ] , we would need a much less increase in temperature in order to start the thermite reaction .
considering temperatures of 77 k and 4.2 k as the original temperature for the detector , then the ignition temperature will be @xmath228 k and @xmath229 k respectively .
this is specially helpful for detectors of wimp mass 10 gev , which had a lower temperature increase .
this variant on the original design is very promising , since it affects equally any material used as the metal and allows us to work with a larger family of thermite reactions .
it is important to note that a cryogenic detector with a very low activation temperature would make the detector unstable at room temperature .
thus , the materials would also need to be synthesized at cryogenic temperatures .
neverthless , a thermite detector working at cryogenic temperatures will be stable even when considering very low activation temperatures ; on the order of @xmath230 k and @xmath231 k for detector temperatures of @xmath232 k ( liquid nitrogen ) and @xmath233 k ( liquid helium ) respectively . specifically , an al / cuo nano - thermite detector with an ignition temperature of @xmath234 k permits an energy threshold of 0.5 kev .
only experiments would determine the actual ignition temperature of an al / cuo nanothermite .
fortunately , the ignition temperature of thermites is proportional to the size of the nanoparticle ( i.e. a smaller nanoparticle has a lower ignition temperature ) , and other similar metals / oxide to al / cuo could be used in order to achieve a ignition temperature close to 50k@xcite .
the latter consideration is beyond the scope of this paper .
any wimp detector is vulnerable to background coming from the cosmic rays or radiation from its components .
cosmic rays can be minimized by putting the detector deep inside a mine underground .
nevertheless , background will still be present due to radiation decay inside the detector , which are mostly due to natural impurities of the materials .
the energy deposed per unit length of the background onto the detector can be calculated by using bethe - bloch theory .
the bethe - bloch equation , in contrast to the lindhard equation , is used for relativistic ionizing particles like alphas and betas . the stopping power of alphas and betas
can be found utilizing the astar and estar programs [ @xcite,@xcite ] .
since the stopping power is linearly proportional to the density of the medium , we calculate the stopping power for alphas and betas moving in the densest material ( platinum ) common to both programs .
excellent background rejection can be achieved because of the nanoscale granularity of the detector .
single charged particles have a very long range .
alpha particles , for example , have a much longer range in the detector than do recoiling nuclei , which can be used to get rid of background due to alphas . the energy loss of particles moving through the detector cells is @xmath235 . for helium @xmath236 = 4 ( near the end of the range the ion will pick up electrons and de / dx increases , but for the most part one can take @xmath237 for helium ) .
z is much higher for nuclear recoils due to wimp interactions , causing faster energy loss ; thus the nuclei stop within a single cell .
the wimp makes only one cell explode : the chain reaction initiated by one exploding metal nanoparticle is restricted to nanoparticles within only one cell , which is thermally isolated from neighboring cells by an insulating material . on the other hand ,
the range of the @xmath3-particles , usually on the order of @xmath238 m or greater , is longer than size of one cell , and therefore make about 20 cells explode at once .
thus the nanoscale granularity is key for background rejection .
alpha particles produced in a radioactive decay usually have energies around 5 mev .
utilizing astar , the stopping power of an alpha particle moving through platinum with an energy of 5 mev is @xmath239 kev/(10 nm ) .
alphas have a large enough stopping power to produce a signal in a nano - thermite detector composed of a highly dense metal and energy threshold @xmath240 kev .
fortunately , the range of alpha particles is much larger than that of heavy ions .
rejection of alpha particle signals can be made due to the nanometer granularity of our detector .
specifically , the typical range of a heavy ion is around 50 nm ; whereas , an alpha particle moving through a material would typically travel approximately @xmath241 m or more . as an illustrative example ,
the stopping distance for a 5 mev alpha particle moving through al@xmath242o@xmath2 ( sio@xmath242 ) is 13.76 @xmath11 m ( 20 @xmath11 m ) .
since each cell has a size of @xmath243 m , then a heavy ion would only explode 1 cell ; and an alpha particle would explode 20 or more cells .
signals produced by alpha particles can be rejected by eliminating all exploded cell clusters .
the energy range of beta particles is wide , ranging from a few kev to hundreds of mevs .
this occurs , because the energy of the radioactive decay is usually shared between an electron and a neutrino .
specifically , the biggest stopping power for a beta particle moving through platinum in the range of energies @xmath244 mev(found at energy @xmath245 mev ) is @xmath246 kev/(10 nm ) .
a nano - thermite detector will produce a signal if one or more nanoparticles ignite and initiate a chain reaction within the cell .
thus , if our detector has an energy threshold greater than @xmath247 kev , then beta particles will not produce a signal in our detector .
it should be noted that the range of beta particles is much longer than alpha particles .
thus , assuming that a beta particle can deposit enough energy to the nanoparticle to make an explosion and/or the detector has a lower energy threshold of @xmath248 kev ; then the same method of rejecting signals originated by alpha particles could be used to identify explosions caused by beta particles .
the signal produced by a beta particle ( if any ) would explode many cells , much more than alpha particles .
this result would hold true also if we considered other metals with a lighter density , since the quoted stopping power is from one of the densest metals in the periodic table .
another source of background are gammas , which are highly energetic photons .
even though gamma particles are electrically neutral , they can create photo electrons through the photoelectric effect and the compton effect .
either of those interactions will eject an electron at relativistic speeds , turning it into a beta particle that will ionize many more atoms .
typical energies for gamma particles produced in a radioactive decay range from a few hundred kev to 10 mev .
estar can be employed to learn the behavior of the electrons produced by the interaction of an atom with a gamma ray . by the same argument
found for beta particles above , gammas will not produce a signal in a nano - thermite wimp detector with energy threshold @xmath249 kev .
the nanometer granularity of the nano - thermite detector is such that background from beta and gamma particles do not produce any signal ( assuming an @xmath249 kev ) , and signals originated by alpha particles can be discarded due to the large number of cells exploded .
we have studied the ignition properties of nanoscale explosives as a novel type of dark matter detector .
other design concepts may be employed for the nanothermite dark matter detector , which could obtain lower energy thresholds and/or measure directionality of the recoiling nucleus sourced by a wimp / nucleus interaction .
we focused on two - component nanothermite explosives consisting of a metal and an oxide . as a specific example
, we considered metal nanoparticles of 5 nm radius embedded in a gel of oxide , with millions of these nanoparticles constituting one cell " isolated from other cells . a large number of cells adds up to a total of 1 kg detector mass . a wimp striking a metal nucleus in the nanoparticle , deposits energy that may be enough to initiate a reaction at the interface between the two layers .
we calculated the temperature increase of a metal nanoparticle due to a wimp interaction and compared it to the ignition temperature required for the nanoparticle to explode .
we computed the range of the nuclear recoil using the lindhard formula ; if the recoiling nucleus did not stop inside the nanoparticle , we considered only the fraction of the energy that was deposited inside the metal nanoparticle itself .
this energy fraction was then converted to a temperature increase .
we needed to know how long the nanoparticle remained hot in order to determine whether an explosion was set off .
this timescale was obtained from the heat transfer equation .
all assumptions made during the calculations were chosen in the spirit of being as conservative as possible .
we then compared the temperature increase to the ignition temperature required to set off a nanothermite explosion .
this ignition temperature varies for different thermite materials , and was computed by requiring two conditions to be met : ( i ) for each of the thermites we considered , there should be no spontaneous combustion of any of the metal nanoparticles for at least a time period of one year , and ( ii ) the temperature increase from a wimp interaction must be sufficiently high to overcome an activation barrier and allow the thermite reaction to proceed .
we searched through a variety of thermite materials to find those whose temperature increases from wimp interactions would exceed their ignition temperatures for an explosion .
we found aluminum , ytterbium , thallium and tantalum to be particularly suited to discover wimps via the explosion they would induce .
we note that our model assumed that both the metal and oxide interact as solids .
however , if the metal changes physical state into a gas due to a correspondingly high temperature increase , then the nano - thermite reaction rate may drastically increase .
excellent background rejection can be achieved because of the nanoscale granularity of the detector .
the wimp makes only one cell explode : the chain reaction initiated by one exploding metal nanoparticle is restricted to nanoparticles within only one cell , which is thermally isolated from neighboring cells by an insulating material .
the range of the @xmath3-particles on the other hand is longer than size of one cell , on the order of @xmath54 m , and therefore makes approximately 20 or more cells explode at once .
thus the nanoscale granularity is key for background rejection .
betas and gammas , on the other hand , rarely set off an explosion at all .
we found a number of thermites that would serve as efficient wimp detectors . using a single model
, we found that if the detector operates at room temperature , wimps with masses above 100 gev ( or for some materials above 1 tev ) could be detected ; they deposit enough energy ( @xmath010 kev ) to cause an explosion .
when operating cryogenically at liquid nitrogen or liquid helium temperatures , the nano explosive wimp detector can detect energy deposits as low as 0.5 kev , making the nano explosive detector sensitive to very light @xmath110 gev wimps .
even with the conservative model presented in this paper , our calculations suggest that oxide - based nano - thermites would work as a dark matter detector .
we look forward to experiments which will establish accurately the minimal energy deposition by a recoiling nucleus necessary for a nano - thermite combustion .
+ k.f .
and a. l. acknowledge the support of the doe under grant doe - fg02 - 95er40899 and the michigan center for theoretical physics .
, k.f . , and
acknowledge the mcubed grant via the university of michigan for support .
s. nussinov , phys .
b 165 , 55- ( 1985 ) + r.s .
chivukula and t.p .
walker , nuclear.phys .
b329(2 ) , 445 - 63 , ( 1990 ) . + d.e .
kaplan , phys .
l. 68 , 741 - 48,1992 .
+ d.e.kaplan , m.a .
luty and k.m .
zurek , asymmetric dark matter " , phys .
d79 , 2009 .
+ h.an , s.l.chen , r.n .
mohapatra , s. nussinov and y.zhang,energy dependence of direct detection cross section for asymmetric mirror dark matter " , phys.rev.d . 82,023533 ( 2010 ) .
+ h. davoudiasl , r.n .
mohapatra , on relating the genesis of cosmic baryons and dark matter " , arxiv:1203.1247 [ hep.phys ] , 2012 .
a. drukier and l. stodolsky , `` principles and applications of a neutral current detector for neutrino physics and astronomy '' , phys .
d * 30 * , 2295 ( 1984 ) .
m. w. goodman and e. witten , `` detectability of certain dark - matter candidates '' , phys .
d * 31 * , 3059 ( 1985 ) .
a. k. drukier , k. freese , d. n. spergel , detecting cold dark matter candidates , phys .
d33 , 3495 - 3508 ( 1986 ) k. freese , j. a. frieman and a. gould , `` signal modulation in cold dark matter detection '' , phys .
d * 37 * , 3388 ( 1988 ) .
s. p. ahlen , f. t. avignone , r. l. brodzinski , a. k. drukier , g. gelmini and d. n. spergel , `` limits on cold dark matter candidates from an ultralow background germanium spectrometer '' , phys .
b * 195 * , 603 ( 1987 ) .
k. freese , m. lisanti , and c.m .
savage , `` annual modulation of dark matter : a review '' , rev.mod.phys .
* 85 * ( 2013 ) 1561 - 1581 .
g. h. anderson ,
@xmath3-particles as detonators , nature 109 .
2745 , 749 ( 1922 ) ; h. h. poole , on the detonating action of @xmath3-particles " , scientific proceedings of the royal dublin society 17 , no .
11 , 93 ( 1922 ) ; p. brousseau , h. e. dorsett and m. d. cliff , and c. anderson , detonation properties of explosives containing nanometric aluminum powder " , proceedings of the 12th international detonation symposium .
( 2002 ) ; k. t. sullivan , n. w. piekiel , s. chowdhury , c. wu , m. r. zachariah , and c. e. johnson , ignition and combustion characteristics of nanoscale al / agio3 : a potential energetic biocidal system " , combustion science and technology 183 , 285 ( 2010 ) .
w. lee perry , b. c. tappan , b. l. reardon , v. e. sanders , and s. f. son , energy release characteristics of the nanoscale aluminum - tungsten oxide hydrate metastable intermolecular composite , journal of applied physics 101 , 064313 ( 2007 ) ; t. foley , a. pacheco , j. malchi , r. yetter , and k. higa , kelvin `` development of nanothermite composites with variable electrostatic discharge ignition thresholds '' , propellants , explosives , pyrotechnics 32 , 431 ( 2007 ) . c. j. oconner , j. a. sims , a. kambhar , v. l. kolesnichenco , w. l. zhou , and j. a. wiemann , magnetic properties of feptx / au and coptx / au core - shell nanoparticles " , j. magn . matter .
226 - 230 1915 ( 2001 ) .
j. r. primack , d. seckel and b. sadoulet , `` detection of cosmic dark matter '' , ann .
nucl . part .
sci . * 38 * , 751 ( 1988 ) .
p. f. smith and j. d. lewin , `` dark matter detection '' , phys .
rept . * 187 * , 203 ( 1990 ) .
j. d. lewin and p. f. smith , `` review of mathematics , numerical factors , and corrections for dark matter experiments based on elastic nuclear recoil '' , astropart .
* 6 * , 87 ( 1996 ) .
g. jungman , m. kamionkowski and k. griest , `` supersymmetric dark matter '' , phys .
rept . * 267 * , 195 ( 1996 ) [ arxiv : hep - ph/9506380 ] .
g. bertone , d. hooper and j. silk , `` particle dark matter : evidence , candidates and constraints '' , phys .
* 405 * , 279 ( 2005 ) .
r. bernabei , p. belli , f. cappella et al . ,
new results from dama / libra " , eur .
j. c67 , 39 - 49 ( 2010 ) . c. e. aalseth , p. s. barbeau , j. colaresi , j. i. collar , j. diaz leon , j. e. fast , n. fields , t. w. hossbach et al . ,
search for an annual modulation in a p - type point contact germanium dark matter detector " , phys .
107 , 141301 ( 2011 ) .
g. angloher _
_ , results from 730 kg days of the cresst - ii dark matter search " , arxiv:1109.0702 [ astro-ph.co ] .
r. agnese _ et al .
_ [ cdms collaboration ] , `` silicon detector dark matter results from the final exposure of cdms ii '' , arxiv:1304.4279 [ hep - ex ] .
z. ahmed _ et al .
_ [ cdms - ii collaboration ] , `` results from a low - energy analysis of the cdms ii germanium data '' , phys .
* 106 * , 131302 ( 2011 ) .
r. agnese _ et al .
_ [ supercdms collaboration ] ,
search for low - mass wimps with supercdms " ,
arxiv:1402.7137 [ hep - ex ] . j. angle _ et al . _ [ xenon10 collaboration ] , `` a search for light dark matter in xenon10 data '' , phys . rev .
lett . * 107 * , 051301 ( 2011 ) .
d. s. akerib _ et al . _
[ lux collaboration ] , `` first results from the lux dark matter experiment at the sanford underground research facility '' , arxiv:1310.8214 [ astro-ph.co ] , 2013 .
e. behnke _
[ coupp collaboration ] , direct measurement of the bubble nucleation energy threshold in a cf@xmath2i bubble chamber " , arxiv:1304.6001 [ hep - ex ] , 2014 .
e. behnke _
[ coupp collaboration ] , improved limits on spin - dependent wimp - proton interactions from a two liter cf@xmath2i bubble chamber",phys .
* 106 * , 021303 , 2010 . s. archambault _
[ picasso collaboration ] ,
constraints on low - mass wimp interactions on 19f from picasso " , arxiv:1202.1240 [ hep - ex ] , 2012 .
m. felizardo _
[ simple collaboration ] , final analysis and results of the phase ii simple dark matter search `` , phys .
lett . * 108 * , 201302 , 2012 .
r. neilson , ' ' pico-2l : a bubble chamber to search for light wimps",ucla dark matter , 2014 .
a. k. galwey and m. e. brown , application of the arrhenius equation to solid state kinetics : can this be justified ? " , thermochim .
acta 386 ( 2002 ) 91 - 98 .
a. k. galwey , magnitudes of arrhenius parameters for decomposition reactions of solids " , thermochim .
acta 242 ( 1994 ) 259 . j. lindhard and m. scharff , energy dissipation by ions in the kev region " , phys . rev . *
124 * , 128 ( 1961 ) .
a. brown , diffusion of heat from a sphere to a surrounding medium " , aust .
j. phys . , 1965 ,
* 18 * , 483 - 9 . a.l
. edwards , a compilation of thermal property data for computer heat - conduction calculations " , ucrl-50589 , university of california lawrence radiation laboratory , 24 february 1969 .
j. leitner @xmath250 , heat capacity of cuo in the temperature range of 298.15 - 1300 k " , thermochimica acta 348 , 49 - 51 ( 2000 ) .
d. g. piercey and t. m. klap@xmath251tke , nanoscale aluminum - metal oxide ( thermite ) reactions for application in energetic materials " , cejem , * 2010 * , 7(2 ) , 115 - 129 .
berger , j.s .
coursey , m.a .
zucker and j. chang , database astar , obtained from http://www.nist.gov/pml/data/star/index.cfm .
berger , j.s .
coursey , m.a . zucker and j. chang , database estar , obtained from http://www.nist.gov/pml/data/star/index.cfm . | we present nanoscale explosives as a novel type of dark matter detector and study the ignition properties .
when a weakly interacting massive particle wimp from the galactic halo elastically scatters off of a nucleus in the detector , the small amount of energy deposited can trigger an explosion . for specificity , this paper focuses on a type of two - component explosive known as a nanothermite , consisting of a metal and an oxide in close proximity .
when the two components interact they undergo a rapid exothermic reaction an explosion . as a specific example , we consider metal nanoparticles of 5 nm radius embedded in an oxide . one cell contains more than a few million nanoparticles , and a large number of cells adds up to a total of 1 kg detector mass .
a wimp interacts with a metal nucleus of the nanoparticles , depositing enough energy to initiate a reaction at the interface between the two layers .
when one nanoparticle explodes it initiates a chain reaction throughout the cell .
a number of possible thermite materials are studied .
excellent background rejection can be achieved because of the nanoscale granularity of the detector : whereas a wimp will cause a single cell to explode , backgrounds will instead set off multiple cells .
if the detector operates at room temperature , we find that wimps with masses above 100 gev ( or for some materials above 1 tev ) could be detected ; they deposit enough energy ( @xmath010 kev ) to cause an explosion .
when operating cryogenically at liquid nitrogen or liquid helium temperatures , the nano explosive wimp detector can detect energy deposits as low as 0.5 kev , making the nano explosive detector more sensitive to very light @xmath110 gev wimps , better than other dark matter detectors . |
the rotation curves of high surface brightness ( hsb ) spiral galaxies rise fairly steeply to reach an extended , approximately flat part , well within the optical disk ( bosma 1978 , 1981a , b ; rubin , ford , & thonnard 1978 , 1980 ) . the discovery that the rotation curves of these galaxies are more or less flat out to one or two holmberg radii has been one of the key pieces of evidence for the existence of dark matter outside the optical disk ( see also van albada et al . 1985 ) . within the optical disk ,
the observed rotation curves can in most cases be explained by the stellar components alone ( kalnajs 1983 ; kent 1986 ) .
the rotation curves of so - called low surface brightness ( lsb ) galaxies have been studied only recently ( de blok , mcgaugh , & van der hulst 1996 , hereafter bmh ; see also pickering et al .
these rotation curves , derived from observations , were found to rise more slowly than those of hsb galaxies of the same luminosity , if the radii are measured in kpc . at the outermost measured point
, they were often still rising .
mcgaugh & de blok ( 1998 ) noted that , with radii expressed in disk scale lengths , the rotation curve shapes of lsb and hsb galaxies become more similar , but not necessarily identical .
based on mass modeling of these rotation curves , de blok & mcgaugh ( 1997 , hereafter bm ) concluded that lsb galaxies were dominated by dark matter and that the contribution of the stellar disk to the rotation curve , even if scaled to its maximum possible value , could not explain the observed rotation curve in the inner parts .
the rotation curves of lsb galaxies have received a great deal of attention , because they provide additional constraints on theories of galaxy formation and evolution , and dark halo structure ( e.g. , dalcanton , spergel , & summers 1997 ; mihos , mcgaugh , & de blok 1997 ; hernandez & gilmore 1998 ; kravtsov et al
. 1998 ; mcgaugh & de blok 1998 ) .
unfortunately , most of the galaxies studied in bmh and bm are only poorly resolved , making their results sensitive to the effects of beam smearing . for five of the galaxies in their sample ,
high resolution h@xmath0 rotation curves , which have been obtained in order to eliminate beam smearing and to investigate the rotation curve shapes in the inner parts of lsb galaxies , are presented in this letter .
the galaxies presented here were selected from the sample of lsb galaxies of bmh .
only galaxies were chosen that satisfied the criteria given in bm to define their useful rotation curves .
an overview of the properties of the galaxies is given in table [ tabprops ] , which lists the name of the galaxy ( 1 ) , the adopted distance in mpc , for @xmath1 km s@xmath2 mpc@xmath2 ( 2 ) , the central surface brightness in mag arcsec@xmath3 ( 3 ) , the disk scale length in kpc ( 4 ) , the inclination angle ( 5 ) , the position angle ( 6 ) , the absolute magnitude ( 7 ) and the systemic velocity ( 8) .
the observations were carried out at palomar observatory with the @xmath4 hale telescope telescope is operated in a joint agreement among the california institute of technology , the jet propulsion laboratory and cornell university ] , on november 20 1998 .
the fwhm velocity resolution was 54 km s@xmath2 , the pixel size in the spatial direction was @xmath5 .
each galaxy spectrum consisted of a single 1800s exposure .
the slit was oriented along the major axis , at the position angle derived by bmh ( see table [ tabprops ] ) . despite their low surface brightnesses
, all galaxies showed up clearly on the slit - viewing monitor .
the slit could therefore accurately be aligned with the center of the galaxy by eye .
the data were reduced using standard procedures in iraf and the resulting h@xmath0 position - velocity diagrams are presented in fig .
[ figure1 ] .
table 1 + properties of the galaxies@xmath6 lccccccc name & @xmath7 & @xmath8 & @xmath9 & @xmath10 & p.a . &
@xmath11 & v@xmath12 & ( mpc ) & ( mag/@xmath13 ) & ( kpc ) & ( @xmath14 ) & ( @xmath14 ) & ( mag ) & ( km s@xmath2 ) ( 1 ) & ( 2 ) & ( 3 ) & ( 4 ) & ( 5 ) & ( 6 ) & ( 7 ) & ( 8) f563-v2 & 61 & 22.1 & 2.1 & 29 & 148 & @xmath15 & @xmath16 + f568 - 1 & 85 & 23.8 & 5.3 & 26 & 13 & @xmath17 & @xmath18 + f568 - 3 & 77 & 23.1 & 4.0 & 40 & 169 & @xmath19 & @xmath20 + f568-v1 & 80 & 23.3 & 3.2 & 40 & 136 & @xmath21 & @xmath22 + f574 - 1 & 96 & 23.3 & 4.3 & 65 & 90@xmath23 & @xmath24 & @xmath25 + -6pt -2pt -2pt
to derive the rotation curves , we started by making gaussian fits to the line profiles at each position along the major axis to obtain the radial velocities . these fits and their errors
are overlayed on the h@xmath0 position - velocity diagram in fig .
[ figure1 ] .
the positions of the galaxy centers were determined from the peak of the continuum light along the slit .
all galaxies were suffi- ciently bright to allow to determine the position of the center along the slit with an accuracy of better than @xmath26 .
= 18.5 cm -0.0 cm to obtain a larger radial coverage of the rotation curves , the h@xmath0 data were combined with the data presented in bmh . to this end
, the derived h@xmath0 velocities were plotted on the position - velocity diagrams .
both sets of data were found to agree well with each other if the effects of beam smearing on the data are taken into account ( cf .
swaters 1999 ) .
therefore , we have used the data to determine rotation velocities beyond radii where we found h@xmath0 emission . note that in most cases the extends only little beyond the h@xmath0 .
next , the rotation curves derived for the approaching and the receding sides were combined .
the h@xmath0 points were sampled every @xmath27 , the points every @xmath28 ( approximately two points per beam ) . the errors on the rotation velocities were estimated from the differences between the two sides and the uncertainties in the derived velocities .
we will refer to these rotation curves as the high resolution rotation curves ( hrc ) .
the derived hrcs are shown in fig .
[ figure2 ] together with the rotation curves presented in bm .
probably because bm did not correct for beam smearing , the rotation curves systematically underestimate the inner slopes of the rotation curves , especially for f568-v1 and f574 - 1 .
both these galaxies have a central depression in the distribution , as can be seen in the maps presented in bmh .
the spatial smearing of from larger radii into the observed central depression leads to an apparent solid body - like rotation curve , in particular for the highly inclined galaxy f574 - 1 .
for the mass modeling presented here we have used the same parameters for the thickness of the gaseous and stellar disks as bm have done .
the stellar disk was assumed to have a vertical sech@xmath29 distribution , with a scale height @xmath30 .
@xmath31-band light profiles , presented in de blok , van der hulst , & bothun ( 1995 ) and bmh , were used to calculate the contribution of the stellar disk to the rotation curve .
the was assumed to reside in an infinitely thin disk .
the only difference with the mass models presented in bm is that we have decomposed the light profile of f568 - 1 into a disk and a central component , and fitted these components to the rotation curve separately . in the other galaxies
no significant central component is present . for the dark matter component
a pseudo - isothermal halo was used , following bm , which has a rotation curve given by : @xmath32,\ ] ] where @xmath33 is the halo core radius and @xmath34 is the central density .
one of the major uncertainties in fitting mass models to rotation curves , in absence of an independent measurement of the stellar mass - to - light ratio @xmath35 , is the uncertainty in the contribution of the stellar disk to the rotation curve .
however , lower and upper limits on @xmath35 , and hence on the dark matter content , can be obtained by assuming that the contribution of the stellar disk to the rotation curve is either minimal or maximal . in the maximum disk mass models ,
the contribution of the stellar disk to the rotation curve is scaled to explain most of the inner parts of the rotation curve .
the resulting rotation curve fits are shown in the top panel of fig .
[ figure3 ] .
what immediately strikes the eye is that , in contrast with the findings of bm , the -17pt table 2 + mass model parameters lccccccc & & & & & & & & lsbc & & @xmath35 & @xmath33 & @xmath34 & & @xmath33 & @xmath34 name & & m@xmath36 & kpc & @xmath37 & & kpc & @xmath38 f563-v2@xmath6 & & 5.4 & & & & 0.94 & 0.283 f568 - 1@xmath39 & & 17.2 & & & & 1.5 & 0.181 f568 - 3 & & 1.5 & 3.0 & 0.027 & & 2.5 & 0.48 f568-v1 & & 9.3 & 6.7 & 0.005 & & 1.2 & 0.188 f574 - 1 & & 3.7 & 3.4 & 0.003 & & 1.5 & 0.92 inner parts of the rotation curves can be explained almost entirely by the contribution of the stellar disk in all of these lsb galaxies , with the exception of f568 - 3 .
the dark halo parameters ( see table [ tabfits ] ) are ill - defined for most galaxies in our sample , because most of the hrcs do not extend to large radii .
nonetheless , it is clear that in these maximum disk fits the contribution of the dark halo will only become important outside the optical disk , as is also the case for hsb galaxies .
the required stellar mass - to - light ratios for the maximum disk fits ( listed in table [ tabfits ] ) , may be high , up to 17 in the @xmath31-band .
most of these are well outside the range of what current population synthesis models predict ( e.g. , worthey 1994 ) . if these high values of @xmath35 are to be explained solely by a stellar population , the stellar content and the processes of star formation in lsb galaxies need to be very different from those in hsb galaxies .
alternatively , these high mass - to - light ratios may indicate the presence of an additional baryonic component that is associated with the disk , as has been suggested by e.g. pfenniger , combes , & martinet ( 1994 ) . on the other hand , the fact that stellar disk can be scaled to explain the observed rotation curve may simply reflect the possibility that the luminous and dark mass have a similar distribution within the optical galaxy .
the other extreme for the contribution of the stellar disk to the rotation curve is to assume that its contribution is negligible .
the dark halo parameters for this minimum disk fit are listed in table [ tabfits ] .
high central densities of dark matter and small core radii are required to explain the observed steep rise in the hrcs . from fig .
[ figure3 ] it is clear that the minimum disk mass models fit the rotation curves equally well as the maximum disk models . in fact , a good fit can be obtained with any mass - to - light ratio lower than the maximum disk mass - to - light ratio , demonstrating that the degeneracy that exists in the mass modeling for hsb galaxies ( e.g. van albada et al .
1985 ) also exists for lsb galaxies .
irrespective of the contribution of the stellar disk to the rotation curve , the similarity between the shapes of the observed rotation curves and those of the stellar disks implies that the total mass density and the luminous mass density are coupled within the region of the optical disk .
the hrcs derived for the lsb galaxies rise steeply in the inner parts , and reach a flat part beyond about two disk scale lengths , as is found for hsb galaxies . in fig .
[ figure4 ] , the rotation curves for lsb galaxies ( dotted lines ) are compared with those of three typical late - type hsb galaxies from begeman ( 1987 ) , ngc 2403 , ngc 3198 and ngc 6503 .
all the galaxies in fig .
[ figure4 ] have no bulges , or only weak ones . in the top panel of fig .
[ figure4 ] the radii are given in kpc .
in these units , the rotation curves of lsb galaxies rise more slowly than those of hsb galaxies , indicating that these galaxies not only have lower central surface brightnesses , but also lower central mass densities , as was found by de blok & mcgaugh ( 1996 ) as well .
a different picture emerges in the bottom panel of fig .
[ figure4 ] , in which the rotation curves are scaled by their optical disk scale lengths and normalized to the velocity at two disk scale lengths .
the normalization probably does not introduce systematic effects because the maximum difference in absolute magnitude between the galaxies in fig .
[ figure4 ] is only 1.5 mag .
with radii expressed in disk scale lengths , the rotation curves of the lsb galaxies presented here and those of hsb galaxies have almost identical shapes .
this is consistent with the concept of a ` universal rotation curve ' ( persic , salucci , & stel 1996 , see also rubin et al . 1985 ) .
the similarity between the lsb and hsb rotation curve shapes suggests that rotation curve shapes are linked to the distribution of light in the stellar disks , independent of central disk surface brightness .
although such a link is most easily understood if the stellar disks are close to maximal , independent of surface brightness , the required high mass - to - light ratios seem to favor a picture in which lsb galaxies are dominated by dark matter within the optical disk , and hsb galaxies more by the stellar disk .
the relative contribution of the stellar disk to the rotation curve may change continuously from lsb to hsb , or perhaps lsb and hsb galaxies constitute discrete galaxy families , as has been suggested by tully & verheijen ( 1997 ) .
we thank roelof bottema for valuable discussions , and erwin de blok for making available the data and for useful comments .
rs thanks ipac for its hospitality during his visits which were funded in part by a grant to bfm as part of the nasa long - term space astrophysics program .
bosma , a. 1978 , phd thesis , rijksuniversiteit groningen bosma , a. 1981a , , 86 , 1791 bosma , a. 1981b , , 86 , 1825 begeman , k. 1987 , phd thesis , rijksuniversiteit groningen dalcanton , j. j. , spergel , d. n. , & summers , f. j. 1997 , , 482 , 659 de blok , w. j. g. , & mcgaugh , s. s. 1996 , , 469 , l89 de blok , w. j. g. , & mcgaugh , s. s. 1997 , , 290 , 533 ( bm ) de blok , w. j. g. , van der hulst , j. m. , & bothun , g. d. 1995 , , 274 , 235 de blok , w. j. g. , mcgaugh , s. s. , & van der hulst , j. m. 1996 , , 283 , 18 ( bmh ) hernandez , x. , & gilmore , g. 1998 , , 294 , 595 kalnajs , a. j. 1983 , in internal kinematics of galaxies , iau symp .
e. athanassoula ( dordrecht : reidel ) , 87 kent , s. m. 1986 , , 91 , 1301 kravtsov , a. v. , klypin , a. a. , bullock , j. s. , & primack , j. r. , 502 , 48 mcgaugh , s. s. , & de blok , w. j. g. 1998 , , 499 , 41 mihos , j.c . ,
mcgaugh , s. s. , & de blok , w. j. g. 1997 , , 477 , l79 persic , m. , salucci , p. , & stel , f. 1996 , , 281 , 27 pickering , t. e. , impey , c. d. , van gorkom , j. h. , & bothun , g. d. 1997 , , 114 , 1858 pfenniger , d. , combes , f. , & martinet , l. 1994 , , 285 , 79 rubin , v. c. , ford , w. k. , & thonnard , n. 1978 , , 225 , l107 rubin , v. c. , ford , w. k. , & thonnard , n. 1980 , , 238 , 471 rubin , v. c. , burstein , d. , ford , jr . , w. k. , & thonnard , n. 1985 , , 289 , 81 swaters , r.a . , 1999 , phd thesis , rijksuniversiteit groningen tully , r. b. , & verheijen , m. a. w. 1997 , , 484 , 145 van albada , t. s. , bahcall , j. n. , begeman , k. , & sancisi , r. 1985 , , 295 , 305 worthey , g. 1994 , , 95 , 107 | high resolution h@xmath0 rotation curves are presented for five low surface brightness galaxies .
these h@xmath0 rotation curves have shapes different from those previously derived from observations , probably because of the higher spatial resolution of the h@xmath0 observations .
the h@xmath0 rotation curves rise more steeply in the inner parts than the rotation curves and reach a flat part beyond about two disk scale lengths .
with radii expressed in optical disk scale lengths , the rotation curves of the low surface brightness galaxies presented here and those of hsb galaxies have almost identical shapes .
mass modeling shows that the contribution of the stellar component to the rotation curves may be scaled to explain most of the inner parts of the rotation curves , albeit with high stellar mass - to - light ratios . on the other hand ,
well fitting mass models can also be obtained with lower contributions of the stellar disk .
these observations suggest that the luminous mass density and the total mass density are coupled in the inner parts of these galaxies . |
there are geological evidence that earth had to face multiple episodes of global ( or quasi - global ) glaciation , at the end of the archean 2.45 - 2.22 gyrs ago and during the neoproterozoic era 710 and 650 myears ago @xcite .
it is widely believed that the carbonate@xmath0silicate cycle @xcite was the main agent to trigger deglaciations on past earth .
in particular , on a completely frozen planet , also called hard snowball , the weathering of co@xmath1 is stopped ; continuous volcanic outgassing builds up atmospheric co@xmath1 , which warms up the climate until liquid water is produced in the equatorial regions . by extension
, this mechanism may be crucial to stabilize the climate of earth - like exoplanets .
it is even central for the definition of the classical habitable zone @xcite , which assumes that planets can build up co@xmath1 atmospheres ( as massive as wanted ) suitable for the stability of surface liquid water .
planets near the outer edge of the habitable zone may suffer , as earth did , episodes of total glaciation .
recent work by @xcite has even shown that , in case of planets lacking land vascular plants , temperate climates may not be stable , because at high co@xmath1 partial pressure , the increased weathering rate does not allow temperate stable solutions .
this effect , enhanced for planets weakly irradiated by their star , may drive systematically earth - like exoplanets toward episodes of glaciation . can earth - like planets in the habitable zone of their star always escape from episodes of glaciation ? in this work ,
we use 3d - global climate model simulations of snowball planets to study the ability of the increased co@xmath1 greenhouse effect resulting from the carbonate - silicate cycle to drive them out of glaciation . as
the co@xmath1 outgassed by volcanoes accumulate in the atmosphere , the temperature of condensation of co@xmath1 can exceed the surface temperature of the poles .
this lead to the trapping of extra outgassed co@xmath1 , forming permanent co@xmath1 polar ice caps , and therefore seriously limits the efficiency of the carbonate@xmath0silicate cycle .
this possibility has already been suggested by @xcite and @xcite , but has never been explored yet .
we propose in this paper a detailed study of this scenario .
we use in this paper the 3-dimensions lmd generic global climate model to study the deglaciation of earth - like planets orbiting circularly around a sun - like star ( sun spectrum and luminosity ) in response to increasing amount of atmospheric co@xmath1 ( from 0.01 to 3 bars ) , and for orbital distances ranging from 1.10 to 1.45 astronomical units ( au ) .
detailed information on the model can be found in appendix [ appen_lmd_model ] .
our simulations were designed to represent completely frozen earth - like planet characteristics .
these include the radius ( 6370 km ) , the gravity field ( 9.81 m s@xmath4 ) , the obliquity ( 23.5@xmath8 ) and the rotation speed ( 7.28@xmath1310@xmath14 rad s@xmath15 ) .
the roles of obliquity , planetary mass and rotation rate on the ability of planets to escape glaciation episodes are discussed in sections [ null_obliquity_section ] and [ sensitivity ] .
eventually , most of the simulations were performed for a uniformly flat topography .
the effect of topography is discussed in section [ topography ] .
all the simulations performed in this study were forced initially in a cold and dry state , which assumes : 1 . a uniform and complete ice cover .
2 . atmospheric temperatures arbitrarily fixed to 230 kelvins everywhere .
no water vapor , no clouds .
depending on co@xmath1 partial pressure , obliquity , or parameterization of clouds , the simulations evolve in different steady state climate regimes that are discussed in the next section .
[ cols= " < , < " , ] \1 . melting , freezing , condensation , evaporation , sublimation and precipitation of h@xmath1o physical processes are all included in the model . in the atmosphere
, water vapor can condense into water ice particles clouds . at the surface
, we fix the h@xmath1o ice albedo at 0.6 and we use an emissivity of 1 . \2 . in our model , co@xmath1 can condense to form co@xmath1 ice clouds and surface frost if the temperature drops below the saturation temperature .
atmospheric co@xmath1 ice particles are sedimented and thus can accumulate at the surface .
the co@xmath1 ice layer formed at the surface can sublimate and recycle the co@xmath1 in the atmosphere .
the co@xmath1 ice on the surface contributes to the surface albedo calculation : if the co@xmath1 ice layer exceeds a threshold value of 1 mm thickness , then the local surface albedo is set immediately to the albedo of co@xmath1 ice ( 0.6 in this work ) . on mars ,
the albedo of co@xmath1 ice can substantially vary @xcite with insolation and presence of dust . without dust ,
the albedo can become very high so that 0.6 is probably a lower estimate .
for co@xmath1 ice , we use an emissivity of 0.9 .
the radiative effect of co@xmath1 ice clouds is discussed in details in section [ co2_cloud_section ] physical parameters used for both co@xmath1 and h@xmath1o ices are summarized in table [ param_ices ] .
what happens once co@xmath1 ice caps have reached their maximum size , as calculated in section [ max_reservoir_co2 ] ? using calculations , we show in the following lines that , depending on few parameters , the co@xmath1 ice caps when full may ( or may not ) be stable . the total mass of co@xmath1 available at the surface and in the atmosphere is : @xmath16 with a the area of the co@xmath1 ice caps , s the total area of the surface ( i.e. 4@xmath17r@xmath18 ) and @xmath19 the thickness of the co@xmath1 ice caps averaged over a. when co@xmath1 ice caps are full , @xmath20 .
let s assume that the system initially with full co@xmath1 ice caps is sightly perturbated from @xmath21 to @xmath22 . by conservation of co@xmath1 mass ,
the new mass of co@xmath1 ice caps is @xmath23 $ ] = @xmath24\footnote{hereafter , ' [ ' and ' ] ' are used to bracket variables . }
$ ] .
we know first from gcm simulations that the area of the ice caps a is not a monotonic function of @xmath33 ( see figure [ condens_plots ] ) .
instead , as @xmath33 grows , @xmath34 should be positive and then at some point become negative .
figure [ condens_plots]b ( 23.5@xmath35 obliquity , inactive co@xmath1 clouds ) tells us that the maximum of a(@xmath33 ) should roughly lie at 1 bar in this specific configuration .
it is possible to derive an analytical - complicated yet - expression of @xmath36 using the relation that served to compute figure [ 2d_basal_melting ] .
it tells us that , whatever the configuration , @xmath36 is in fact always negative , because as pco@xmath1 increases , surface temperature also increases , which limits the thickness of co@xmath1 ice caps shows that the mean thickness of co@xmath1 ice caps is proportional to the maximum thickness calculated by basal melting . ] ( see equation [ conductivity ] ) . for the experiment described in previous section ( 1.30 au
, 23.5@xmath35 obliquity , inactive co@xmath1 clouds , co@xmath1 partial pressure of 1 bar ) , we assume @xmath37 and we have therefore ( with @xmath38 , see section [ co2_rad_extent ] ) : @xmath39 from our calculations , this condition is valid for internal heat flux lower than 80 mw m@xmath4 .
it means that , for internal heat flux lower than 80 mw m@xmath4 : ( 1 ) co@xmath1 ice caps when full would not be stable , and ( 2 ) the maximum amount of co@xmath1 that can be trapped in the @xmath40surface , atmosphere@xmath41 system is attained for pco@xmath1 lower than 1 bar . for internal heat flux
higher than 80 mw m@xmath4 , because a significant fraction of co@xmath1 is in the gaseous phase : ( 1 ) co@xmath1 ice caps when full would be stable ( until @xmath34 term becomes significant ) , and ( 2 ) the maximum amount of co@xmath1 trapped in the @xmath40surface , atmosphere@xmath41 system occurs for pco@xmath1 larger than 1 bar . ,
b. , forget , f. , wordsworth , r. , leconte , j. , millour , e. , codron , f. , and spiga , a. ( 2013 ) . exploring the faint young sun problem and the possible climates of the archean earth with a 3-d gcm .
, 118:414431 . , f. , musat , i. , bony , s. , braconnot , p. , codron , f. , dufresne , j .- l . ,
fairhead , l. , filiberti , m .- a .
, friedlingstein , p. , grandpeix , j .- y . , krinner , g. , levan , p. , li , z .- x . , and lott , f. ( 2006 ) . .
, 27:787813 . , l. s. , gordon , i. e. , babikov , y. , barbe , a. , benner , d. c. , bernath , p. f. , birk , m. , bizzocchi , l. , boudon , v. , brown , l. r. , campargue , a. , chance , k. , coudert , l. h. , devi , v. m. , drouin , b. j. , fayt , a. , flaud , j .-
m . , gamache , r. r. , harrison , j. j. , hartmann , j. m. , hill , c. , hodges , j. t. , jacquemart , d. , jolly , a. , lamouroux , j. , le roi , r. j. , li , g. , long , d. a. , lyulin , o. m. , mackie , c. j. , massie , s. t. , mikhailenko , s. , mller , h. s. p. , naumenko , o. v. , nikitin , a. v. , orphal , j. , perevalov , v. i. , perrin , a. , polovtseva , e. r. , richard , c. , smith , m. a. h. , starikova , e. , sung , k. , tashkun , s. a. , tennyson , j. , toon , g. c. , tyuterev , v. g. , and wagner , g. ( 2013 ) . .
, 130:455 .
toon , o. b. , mckay , c. p. , ackerman , t. p. , and santhanam , k. ( 1989 ) .
rapid calculation of radiative heating rates and photodissociation rates in inhomogeneous multiple scattering atmospheres .
, 94:16,28716,301 . | it is widely believed that the carbonate@xmath0silicate cycle @xcite is the main agent - through volcanism - to trigger deglaciations by co@xmath1 greenhouse warming on earth and on earth - like planets when they get in frozen state .
here we use a 3d global climate model ( gcm ) to simulate the ability of planets initially completely frozen to escape from glaciation episodes by accumulating enough gaseous co@xmath1 .
the model includes co@xmath1 condensation and sublimation processes and the water cycle .
we find that planets with earth - like characteristics ( size , mass , obliquity , rotation rate , ... ) orbiting a sun - like star may never be able to escape from a glaciation era , if their orbital distance is greater than @xmath2 1.27 astronomical units ( flux @xmath3 847 w m@xmath4 , s@xmath5 @xmath3 0.62 ) , because co@xmath1 would condense at the poles here the cold traps forming permanent co@xmath1 ice caps .
this limits the amount of co@xmath1 in the atmosphere and thus its greenhouse effect .
furthermore , our results indicate that for ( 1 ) high rotation rates ( p@xmath624 h ) , ( 2 ) low obliquity ( [email protected]@xmath8 ) , ( 3 ) low background gas partial pressures ( @xmath91bar ) , and ( 4 ) high water ice albedo ( h@xmath1o [email protected] ) , this critical limit could occur for significantly lower equivalent distance ( or higher insolation ) . for each possible configuration
, we show that the amount of co@xmath1 that can be trapped in the polar caps depends on the efficiency of co@xmath1 ice to flow laterally as well as its graviational stability relative to subsurface water ice .
the flow of co@xmath1 ice from polar regions to the equator is mostly controlled by the bottom temperature , and hence by the internal heat flux of the planet .
we find that a frozen earth - like planet located at 1.30 au of a sun - like star could store as much as 1.5/4.5/15 bars of dry ice at the poles , for internal heat fluxes of 100/30/10 mw m@xmath4 .
but these amounts are in fact lower limits . for planets with a significant water ice cover ,
we show that co@xmath1 ice deposits should be gravitationnally unstable .
they get burried beneath the water ice cover in very short timescales of 10@xmath11 - 10@xmath12 yrs , mainly controlled by the viscosity of water ice . for water ice cover exceeding about 300 meters ,
we show that the co@xmath1 would be permanently sequestred underneath the water ice cover , in the form of co@xmath1 liquids , co@xmath1 clathrate hydrates and/or dissolved in subglacial water reservoirs ( if any ) .
this would considerably increase the amount of co@xmath1 trapped and further reduce the probability of deglaciation .
our results may have implications for the search for life - suitable extrasolar planets orbiting in the habitable zone of sun - like stars . |
with the successful launch of the _ fermi _ gamma - ray space telescope , we have a new opportunity to study the @xmath0-ray emission from different types of extragalactic sources not only blazars , but also radio galaxies and possibly other classes of active galactic nuclei ( agn ) with much improved sensitivity than previously available ( abdo et al .
2010a ) . during the initial all - sky survey performed during the first four months after its launch , the _ fermi _ large area telescope ( lat ) detected only two radio galaxies at high significance ( @xmath13 ) , namely ngc1275 ( abdo et al . 2009a
; hereafter paper - i ) and cena ( abdo et al .
2009b , 2009c ) .
more recently , the detection of mev / gev emission from yet another famous radio galaxy m87 , an established tev source , was reported based on ten - months of all - sky survey _ fermi_-lat data ( abdo et al .
2009d ) . yet
the detection of ngc1275 was particularly noteworthy because this source , unlike cena or m87 , was previously undetected in @xmath0-rays , neither by _
cgro_/egret during its @xmath1410 years of operation , nor by ground - based cherenkov telescopes . the @xmath0-ray flux of ngc1275 detected by the _ fermi_-lat was about seven times higher than the one implied by the @xmath15 upper limit reported by egret , namely @xmath16ph@xmath7s@xmath8 ( reimer et al .
we note that _ cos b _ data taken between 1975 and 1979 ( strong et al . 1982 ; mayer - hasselwander et al .
1982 ) showed a @xmath0-ray excess coincident with the position of this galaxy , although the evidence for the claimed high - energy source to be _ uniquely _ related to ngc1275 is ambiguous .
ngc1275 is a giant elliptical galaxy located at the center of the perseus cluster , luminosity distance @xmath17 mpc , scale @xmath18kpc arcmin@xmath8 ( for flat cosmology with @xmath19kms@xmath8mpc@xmath8 and @xmath20 ) . ] .
this cluster is the brightest cluster of galaxies in the x - ray band ( e.g. , bhringer et al .
1993 ; fabian et al .
2003 , 2006 ) , and as such it has been the focus of several extensive research programs over many years and across the entire available electromagnetic spectrum .
when observed at radio wavelengths , ngc1275 hosts the exceptionally bright radio fri radio galaxy perseusa = 3c84 ( e.g. , vermeulen et al . 1994 , taylor et al .
1996 , walker et al .
2000 , asada et al .
although high - energy @xmath0-rays may in general be produced within the intergalactic / interstellar medium of the perseus cluster , in paper - i , we argued that the inner radio jet of 3c84 was the most likely source of the observed @xmath0-ray photons because of the variability of the mev / gev flux on year / decade timescales implied by the egret and early _
fermi_-lat observations . specifically , these measurements implied the @xmath0-ray emission region size in ngc1275 has a radius , @xmath21pc .
note , however , that no significant variability was detected within the four - month long _
fermi_-lat dataset analyzed in paper - i . also , the lat error circle determined from these initial data was too large to exclude possible contributions from other galaxies to the observed @xmath0-ray flux ( specifically , ngc 1273 , 1274 , 1277 , 1278 and 1279 , were still within the previously determined @xmath2 error circle with @xmath22arcmin ) . meanwhile , follow - up observations by the veritas cherenkov telescope ( weekes et al .
2002 ) put strong constraints on the very high - energy ( vhe ) @xmath0-ray emission from ngc1275 above 100 gev .
in particular , no vhe @xmath0-ray emission from ngc1275 was detected by veritas , with a @xmath23 confidence level upper limit of @xmath24 of the crab nebula flux , corresponding to @xmath25 of the power - law extrapolation of the mev / gev flux observed during the first four months of the _ fermi_-lat observations ( assuming the photon index , @xmath26 ; acciari et al .
this naturally indicates a deviation from the pure power - law spectrum in the vhe regime , possibly compatible with the presence of an exponential cutoff around or below photon energies @xmath27gev ( acciari et al .
the magic cherenkov telescope also recently measured upper limits for the vhe @xmath0-ray emission of ngc1275 , namely @xmath28ph@xmath7s@xmath8 for the photon indices ranging from @xmath29 up to @xmath30 ( aleksi et al .
thus , the implied deviation in the @xmath0-ray spectrum of 3c84 from a simple power - law form , as well as a possibility for a short - timescale ( @xmath31 month ) variability of the perseusa @xmath0-ray flux , may now be finally addressed and re - examined by _
fermi_-lat , due to the much improved photon statistic ( especially above @xmath32gev ) after the one - year - all - sky survey
. obviously , such deep studies of ngc1275 in the mev / gev photon energy range are of the major importance for understanding the whole class of @xmath0-ray emitting radio galaxies in general .
firmly motivated , we performed a detailed investigation of ngc1275 in @xmath0-rays based on the accumulation of one - year of _ fermi_-lat all - sky survey data .
in particular , we aimed to address the following problems : ( i ) presence of short - timescale @xmath0-ray flux variability , ( ii ) positional coincidence of the @xmath0-ray emitting center with the active nucleus of ngc1275 , and ( iii ) spectral curvature and spectral evolution of ngc1275 in the mev / gev photon energy range . in @xmath332 , we describe the _ fermi_-lat @xmath0-ray observations and data reduction procedure .
the results of the analysis are given in @xmath333 , and the discussion and conclusions are presented in @xmath334 .
the lat instrument onboard _ fermi _ is described in detail in atwood et al .
( 2009 ) and references therein .
compared to earlier @xmath0-ray missions , the lat has a large effective area ( @xmath14@xmath34@xmath35 on axis at @xmath36gev for the event class considered here ) , wide energy coverage ( from @xmath37mev to @xmath38gev ) , and improved angular resolution .
the @xmath39 containment angles of the reconstructed incoming photon direction are approximated as @xmath40 below @xmath32gev , giving @xmath41 at @xmath42mev and @xmath43 at @xmath32gev ( update of values presented in atwood et al .
2009 ) . above @xmath32gev ,
the improvement of angular resolution becomes relatively gentle , such that @xmath44 at @xmath42gev . during the first year of operations , most of the telescope s time was dedicated to observing in `` survey mode '' where _ fermi _ points away from the earth and nominally rocks the spacecraft axis north and south from the orbital plane to enable monitoring of the entire sky on a time scale shorter than a day .
the whole sky is surveyed every @xmath45hours ( or 2 orbits ) .
the total live time included is 280.3 days ( 24.21 ms ) .
this corresponds to an absolute efficiency of 75.0 @xmath46 .
most of the inefficiency is due to time lost during passages through the south atlantic anomaly and to readout dead time .
the observations used here comprises all scientific data obtained between 4 august 2008 and 13 august 2009 .
this time interval runs from mission elapsed time ( met ) 239557417 to 271844550 .
we have applied the zenith angle cut of @xmath47 to eliminate photons from the earth s limb .
the same zenith cut is also accounted for in the exposure calculation using the lat science tool gtltcube .
we use the `` diffuse '' class events ( atwood et al .
2009 ) , which are those reconstructed events having the highest probability of being photons . in the analysis presented here
, we set the lower energy bound to a value of @xmath42mev .
science tools version v9r15p2 and irfs ( instrumental response functions ) p6_v3 ( a model of the spatial distribution of photon events calibrated post - launch ) were used throughout this paper .
the early portion of the data here coincides with the early lat observations of ngc1275 presented in paper - i ( 4 august 2008 and 5 december 2008 ) .
figure1 shows the lat @xmath2 @xmath0-ray localization error circles placed on the _ chandra _ acis - s x - ray image of ngc1275/perseus region ( obsid 4952 ; exposure @xmath48ks ) .
image is constructed between @xmath49 and @xmath50kev .
the positional center of the @xmath0-ray emission ( ra@xmath51 , dec@xmath52 ) is taken from the 11-month catalog ( 1fgl j0319.7 + 4130 ; abdo et al .
2010b ) , and is only @xmath1arcmin from the ngc1275 nucleus ( ra@xmath53 , dec@xmath54 ) , with @xmath2 radius @xmath3arcmin ( fig .
1 ) . since the localization error has been substantially improved compared to the one quoted in paper - i ( @xmath55arcmin , based on the 3-month bright source list ; abdo et al.2009c ) , all the other field galaxies mentioned in paper i within the 3-month lat error ( ngc 1273 , 1274 , 1277 , 1278 , and 1279 ) are now outside the 11-month lat error circle . indeed
, even the nearest galaxy , ngc 1274 , is 2.6 arcmin apart from the positional center of the @xmath0-ray emission with 95 @xmath46 radius of 1.5 arcmin . with the good positional association described above ,
we thus believe the @xmath0-ray source is most likely identified as the nucleus of ngc 1275 , but still needing further confirmation based on correlated variability with observations at other wavelengths .
we also checked the projection of the @xmath0-ray images as presented in paper - i .
the counts distributions of ngc1275 in different energy bands are consistent with the distributions for a point source , indicating that the diffuse extended component does not contaminate the ngc1275/perseus @xmath0-ray source , even with much improved photon statistics in the one - year dataset .
since the angular resolution of the lat improves at high energies ( atwood et al .
2009 and see above ) , it is interesting to compare the reconstructed incoming photon directions with the position of the ngc1275 nucleus around the higher energy photons that can be detected with the _ fermi_-lat
. figure2 shows the angular separation of @xmath56gev photons from the nucleus of ngc1275 as a function of photon energy .
the @xmath39 and @xmath2 angular resolutions of _
fermi_-lat are shown as dotted and dashed lines , respectively .
the angular displacement of the highest energy photon detected during one - year - all - sky survey ( @xmath57gev ; a double circle with a number `` i '' in figure2 ) from the nucleus is only @xmath12arcmin , which is well within the psf of the lat at this energy ( @xmath58arcmin ) .
note that the chance probability for detecting galactic and/or extragalactic @xmath0-ray background photons with energies at least @xmath59gev within @xmath60 is less than @xmath61 , hence we conclude that ngc1275 is the most likely source of the discussed photon .
the other three highest energy photons ( @xmath62gev ) are marked in figure2 as `` ii '' , `` iii '' and `` iv '' , and are all well within the @xmath2 psf of the _ fermi_-lat .
to study the average spectrum of ngc1275 during the one - year observation , we use the standard maximum - likelihood spectral estimator provided with the lat science tools gtlike .
this fits the data to a source model , along with the models for the uniform extragalactic and structured galactic backgrounds .
we use a recent galactic diffuse model , gll_iem_v02.fit , with the normalization free to vary in the fit .
the response function used is p6_v3_diffuse .
careful choice of the source region is important especially for relatively faint sources . following the detailed study of changing the region of interest ( roi ) radius from @xmath63 to @xmath64 in paper - i , we set @xmath65 in the following analysis to minimize the contamination from the galactic diffuse emission .
only one point source , corresponding to src_a in paper - i , was found in the roi .
the coordinates are ra@xmath66 , dec@xmath67 , with @xmath68arcmin .
we modeled the source for subsequent spectral analysis ( also see the _
fermi_-lat bright @xmath0-ray source list ; abdo et al .
2009c ) . we
first model the continuum @xmath0-ray emission of ngc1275 with a single power - law .
the extragalactic background is assumed to have a power - law spectrum as well , with the spectral index and the normalization free to vary in the fit . from an unbinned gtlike analysis ,
the best fit power - law parameters for ngc1275 are : @xmath69 or @xmath70 with only statistical errors taken into account .
systematic errors for the lat are still under investigation , but the estimated systematic uncertainty on the flux is 10 @xmath46 at 100 mev , 5 @xmath46 at 500 mev , and 20 @xmath46 at 10 gev , respectively ( abdo et al .
the results are consistent with those presented in paper - i , even though the uncertainties are now much smaller due to the improved photon statistics .
the predicted photon counts from ngc1275 in the roi are @xmath71 and the test statistic ( defined as @xmath72 , where @xmath73 and @xmath74 are the likelihood when the source is included or not , respectively ) is @xmath75 above @xmath76mev , corresponding to a @xmath77 detection . for the galactic diffuse background , the normalization is @xmath78 and @xmath79 .
the power - law photon index of the extragalactic background is @xmath80 with @xmath81 .
figure3 shows the lat spectrum of ngc1275 obtained by separately running gtlike for 10 energy bands ; @xmath82mev , @xmath83mev , @xmath84mev , @xmath85mev@xmath86gev , @xmath87gev , @xmath88gev , @xmath89gev , @xmath90gev , @xmath91gev , and @xmath92gev , where the dashed line shows the best fit power - law function for the ngc1275 data given in equation ( 1 ) . for the highest energy bin ( @xmath92gev )
, we plot a 2 @xmath93 upper limit since this bin includes only one photon ( see figure .
2 ) and is ignored in the subsequent statistical analysis . note an indication of a deviation of the model with respect to the data above @xmath94 gev .
indeed , a @xmath95 fit of the power - law model to the data fitting to the resultant 10 band lat spectrum as described above , to give a convenient estimate of the goodness of the fit . ]
gives a relatively poor fit with @xmath96 for @xmath97 degrees of freedom ( d.o.f . ) , where its probability is @xmath98 .
instead , an alternative fit with a cutoff power - law function with the form @xmath99 $ ] gives a much better representation of the data with an improved @xmath100 for @xmath97d.o.f . , corresponding to @xmath101 .
we therefore retried the gtlike fit assuming a power - law with an exponential cutoff function , and obtained @xmath102gev with @xmath103 .
moreover , we apply a likelihood ratio test between a simple power - law and a cutoff power - law function .
the test statistic ( @xmath104 ) is twice the difference in these log - likelihoods , which gives @xmath104 = 2.9 for our case .
note that the probability distribution of the test statistic can be approximated by a @xmath95 distribution with 1 d.o.f .
, corresponding to different d.o.f . between two functions .
we obtain @xmath105 @xmath106 0.08 , which again indicates a deviation from a simple power - law function although this is currently inconclusive .
the best - fit cutoff power - law function is shown as dotted line in figure3 .
next , we investigated the @xmath0-ray flux variations of ngc1275 from 4 august 2008 to 13 august 2009 . to this end , we constructed spectra with a time resolution of 14 days and fit each spectrum with a power - law model just for simplicity .
the roi radius ( @xmath65 ) , the energy range ( @xmath107mev ) , and other screening conditions are the same as described above . since no variability is expected for the underlying background diffuse emission , we fix the best - fit parameters to the average values determined from the one - year integrated spectrum for the galactic and extragalactic background components .
we first fix the spectral index to the best fit value over the full interval , @xmath108 , to minimize uncertainties in the flux estimates . in this case ,
variability is highly significant with @xmath109 for @xmath110d.o.f .
, where @xmath111 .
indeed , the @xmath0-ray flux varies even within a few month timescale . due to the limited photon statistics , however , it is difficult to investigate any shorter timescale variability of ngc1275 .
we also investigate the spectral evolution of ngc1275 , with the @xmath0-ray photon index free to vary .
figure4 shows variations of the flux ( @xmath107mev : _ upper panel _ ) and photon index ( _ lower panel _ ) versus time .
a flaring event is seen around @xmath112days after 2008 august 4th ( epoch `` b '' in figure4 ) , corresponding to the epoch april may 2009 .
a doubling timescale for the flaring period can not be accurately measured , but @xmath1420 days .
a @xmath95 fit to a constant gives @xmath113 and @xmath114 for @xmath110d.o.f .
, corresponding to the probability @xmath115 and @xmath116 , respectively , for the flux and the photon index variations .
interestingly , the spectral variation is predominantly due to the difference between the pre- and post - flare period , denoted in figure4 as epochs `` a '' and `` c '' . in fact , if we test the variability _ separately _ for the pre- and post - flare periods ( a and c ) , the significance of the variability decreases substantially , such that a constant fit provides @xmath117 or @xmath118 for both the flux and the photon index changes . during the flare ( epoch b ) , the photon index changes from @xmath119 to @xmath120 .
note that , after the flare ( epoch c ) , the flux drops to its original level but the spectrum remains relatively flat , with @xmath121 , persisting for more than a few months .
it should also be noted that three out of the four highest energy photons ( denoted as i , ii , and iii in figure2 ) were indeed detected in the post - flare period , when the spectrum appears the flattest .
in the previous sections , we reported on the analysis of the @xmath0-ray emission from ngc1275 observed with _
fermi_-lat during its one - year - all - sky survey .
we showed that with the increased photon statistics , the positional center of the @xmath0-ray emission is now much closer to the ngc1275 nucleus as compared to that reported in paper - i .
in addition , we have shown that the average @xmath0-ray spectrum of ngc1275 reveals a significant deviation from a simple power - law above photon energies @xmath122gev .
that is , the observed _
fermi_-lat spectrum is best fitted by a power - law function ( @xmath123 2.1 ) with an exponential cutoff at the break photon energy @xmath102gev .
finally , we argued that significant flux and spectral changes of ngc1275 are detected with _
fermi_-lat on a timescale of a few months , although the possibility for even shorter variability remains uncertain .
we also reported the detection of an interesting spectral evolution , consisting of a _ persistent _ ( over more than a few months ) spectral hardening ( from @xmath124 to @xmath121 ) after the largest flaring event observed in april may 2009 . during this flat - spectrum / low - flux - level epoch the highest energy photon ( @xmath125gev )
was detected from the direction of ngc1275 .
all these new findings basically support the idea put forward in paper - i that the observed @xmath0-ray emission from the perseus system originates in the ( sub ) pc - scale radio jet of ngc1275 , and is therefore most likely analogous to high - energy emission observed in blazars .
in fact , as shown in paper - i , the overall @xmath126 spectral energy distribution ( sed ) of ngc1275 constructed with multi - frequency radio to @xmath0-ray data shows a close similarity to the `` two - bump '' seds of so - called low - frequency peaked bl lac objects ( hereafter lbls ) .
till now , only a few lbls have been detected at tev photon energies bl lac ( albert et al .
2007 ) , 3c66a ( acciari et al .
2009a ) , s5 0716 + 714 ( anderhub et al .
2009 ) , and also w comae from an ibl class ( acciari et al . 2008 ; 2009d ) and more similar discoveries are expected .
hence , ngc1275 itself has been suggested to be a potential tev source as well , thus motivating deep veritas ( acciari et al .
2009b ) and magic ( aleksi et al .
2010 ) observations .
these observations so far resulted only in upper limits to the vhe @xmath0-ray emission of the studied region .
one should note , however , that both the low- and high - energy peaks of ngc1275 ( in the @xmath126 representations ) are located at substantially lower frequencies than those of typical lbls .
indeed , the low - energy ( synchrotron ) emission components of 3c66a and bl lac peak around @xmath127hz , while it is around @xmath128hz in the case of ngc1275 ( paper - i ) . correspondingly , the high - energy emission component of 3c66a ( but not necessarily of bl lac ) peaks at higher photon energies than that observed in ngc1275 .
within such an interpretation , this is consistent with the softer mev / gev spectrum in ngc1275 ( @xmath129 ) compared to 3c66a ( @xmath130 ; abdo et al .
2009b ) as measured by the _ fermi_-lat .
note also that the x - ray spectrum of ngc1275 is clearly `` rising '' in the @xmath126 representation ( @xmath131 = [email protected] ; balmaverde et al .
2006 , ajello et al .
2009 ) , indicating that it is dominated by the low - energy portion of the ic emission , whilst the x - ray spectra of lbls are typically very flat , suggesting a transition between the synchrotron and ic components ( @xmath131 @xmath133 2 ; e.g. , ghisellini et al .
1998 ) . such observational differences may indicate that , unlike in the case of lbls , the high - energy spectrum of ngc1275 does not extend up to tev photon energies .
an interesting comparison can be made to another nearby radio galaxy detected by _
fermi_-lat , namely m87 which resembles ngc1275 in many respects , and is an established tev source ( abdo et al .
while m87 is closer to us than ngc1275 ( @xmath134mpc versus @xmath17mpc ) , the gev flux of m87 is much lower than that of ngc1275 ( @xmath135ph@xmath7s@xmath8 versus @xmath136ph@xmath7s@xmath8 ) , with similar lat measured spectra ( @xmath137 versus @xmath138 ) .
note also that both radio galaxies are located at the centers of rich clusters , that the synchrotron emission components in both sources peak in the far infrared ( @xmath139hz ) , and the estimated jet powers are similar , @xmath140 erg s@xmath8 ( owen et al .
2002 , dunn & fabian 2004 ) .
thus , it may be surprising that only one of these has so far been detected at tev photon energies .
the detailed analysis of the spectral evolution of ngc1275 within the _ fermi_-lat range reported in this paper may provide a viable explanation for such a behavior .
in particular , as already emphasized above , we found that the epochs characterized by the flattest gev continuum of this source , as well as the arrival times of the highest energy photons from the direction of ngc1275 , do not coincide with the epochs of the highest photon flux above @xmath42mev .
this is clearly illustrated in figure5 , which shows a correlation between @xmath141 and the photon indices emerging from the power - law fits . because no significant flux or spectral changes were observed during the pre- and post - flare epochs a ( @xmath142day ) and c ( @xmath143day ) ( see figure4 ) ,
the average fluxes and photon indices for these time periods are reported .
note that the average fluxes of these pre- and post - flare epochs are comparable , whilst the corresponding photon indices differ significantly by @xmath144 .
moreover , the observed spectral evolution during one year of the _ fermi_-lat exposure reveals a hysteresis - like character , more clearly seen for the flaring period ( epoch b , with a time bin of 2 weeks ) , followed by a gradual flattening in the subsequent decay phase .
further insights into the spectral evolution of ngc1275 within the _ fermi_-lat photon energy range are provided by figure6 , which shows the two seds for pre- and post - flare epochs a and c. here the dotted line corresponds to the best `` power - law with an exponential cutoff '' fit function determined from an average @xmath0-ray spectrum , as given in figure3 .
interestingly , the difference between the two seds consists of an excess at photon energies @xmath145gev , with the low - energy @xmath0-ray flux remaining essentially unchanged between the two epochs .
this implies that ( 1 ) the @xmath0-ray variability in ngc1275 ( and possibly other radio galaxies ) may be restricted to @xmath146gev photon energies , and that ( 2 ) the position of the peak in the high - energy spectral component ( in the @xmath147 representation ) may change substantially even within the same object with no accompanying significant flux changes .
note in this context that both the veritas and magic non - detections were obtained during the pre - flare epoch a ( @xmath148day ; see acciari et al .
2009b ; aleksi et al .
hence , the emerging conclusion is that _ it is not the total flux above @xmath42mev which should play a major role in triggering tev observations of steep - spectrum _
fermi_-lat sources , but instead it is the flux and photon index determined at higher photon energies ( @xmath146gev)_. on the theoretical side , this conclusion could be possibly justified by noting that after a new episode of injection of freshly accelerated electrons into the emission zone ( e.g. , a downstream region of a shock ) , higher energy electrons may lag behind the lower - energy electrons . in such a case , as discussed previously in the context of blazar modeling ( e.g. , kirk et al . 1998 ; sato et al .
2008 ) , an `` anti - clockwise '' hysteresis in the flux versus photon index plane may arise , similar to what we observed in ngc1275 at @xmath0-ray photon energies . here ,
we comment on the question if ngc1275 being a representative example of a low - power radio galaxy may be considered as a misaligned blazar , most likely of the lbl type . assuming a homogeneous jet model
, one should expect its jet doppler factor @xmath149 , for the typically expected jet viewing angle of @xmath150 and jet bulk lorentz factors @xmath151 .
indeed , modeling of the broad - band emission of lbls ( beamed counterparts of radio galaxies such as ngc1275 by assumption ) requires @xmath152 . thus ,
if the only difference between radio galaxies and blazars is due to the viewing angle , this should manifest in ( i ) different observed positions of the spectral peaks in the @xmath126 representation ( @xmath153 ) , ( ii ) different observed variability patterns ( @xmath154 , assuming the emission region is a moving source ) , and finally in ( iii ) different observed luminosities ( @xmath155 for a moving blob case , or @xmath156 for the steady jet ( see sikora et al .
1997 ) .
indeed , both the low- and high - energy @xmath126 peaks of ngc1275 are located at substantially lower frequencies than those of typical lbls .
taking the difference of beaming factors into account ( i.e. , @xmath157 ) , the broad - band sed of the ngc1275 would be similar to the seds of blazars such as bl lacertae or 3c66a .
note however that the position of the high - energy spectral peak may change substantially in a single object even for comparable flux levels , at least in ngc1275 , so the diagnostics related to the location of the spectral peaks may not be very conclusive .
variability as short as day timescales is often observed in lbls .
for example , during the historical flare of bl lacertae in 1997 , correlated @xmath0-ray and optical flares were observed , with the @xmath0-ray flux increasing by a factor of @xmath30 within a day ( bloom et al . 1997 ) .
similarly , daily variability has been discovered in both _
fermi_-lat and veritas observations of 3c66a ( reyes et al .
hence , in the simple unification scheme outlined above , we should expect ngc1275 to vary in @xmath0-rays on timescales of several days .
this , however , is difficult to test even with the excellent sensitivity of the _ fermi_-lat instrument , due to the limited photon statistics for weekly time bins .
the analysis presented in this paper gives instead only a robust upper limit @xmath158 a few months .
however , note that day - timescale variability has been detected at tev photon energy range for the m87 radio galaxy ( acciari et al .
thus , more frequent monitoring of ngc1275 by ground - based cherenkov telescopes would be valuable in this respect .
the observed photon flux of @xmath159ph@xmath7s@xmath8 implies an _ observed _ ( isotropic ) @xmath0-ray luminosity of ngc1275 , @xmath160ergs@xmath8 , for @xmath161 and @xmath162gev .
this is already comparable to the typical observed @xmath0-ray luminosities of lbls , which range between @xmath163ergs@xmath8 .
note in particular that in the framework of the simple unification scheme discussed above , the beamed analog of ngc1275 would then be characterized by the observed @xmath0-ray luminosity larger by a factor of @xmath164 ^ 4 \sim 3 \times 10 ^ 4 $ ] ( or at least @xmath165 ^ 3 \sim 3 \times 10 ^ 3 $ ] ) than this , i.e. @xmath166ergs@xmath8 . such luminosities are not expected for lbl - type blazars ( see abdo et al .
2010a ) . on the other hand ,
the observed @xmath0-ray luminosity of ngc1275 is not energetically problematic , since the total _ emitted _ @xmath0-ray power in this source seems rather moderate _ as long as the emitting plasma moves with highly relativistic bulk velocities _ , namely @xmath167ergs@xmath8 for @xmath168 , where @xmath169 is the solid angle defined by the jet opening angle @xmath170 , for which we assumed @xmath171 .
such a relatively small emitted power would constitute less than @xmath172 of the total kinetic power of the ngc1275 jet , estimated by dunn & fabian ( 2004 ) to be roughly @xmath173ergs@xmath8 .
yet the problem of an unexpectedly large observed @xmath0-ray luminosity of the beamed analog of ngc1275 remains , and poses a serious challenge to the simplest version of the agn unification scheme . in this context , a viable explanation for this problem would be to postulate that the high - energy emission observed from `` misaligned '' blazars such as ngc1275 ( or m87 ) is dominated not by a jet `` spine '' characterized by large bulk lorentz factors ( @xmath174 as is the case in bona - fide blazars ) , but by the slower jet boundary layers ( @xmath175 few ) as discussed by several authors ( e.g. , celotti et al .
2001 , stawarz & ostrowski 2002 , ghisellini et al .
alternatively , one may propose that the @xmath0-ray emission observed from radio galaxies is not produced within the `` proper '' blazar emission zone , but at larger distances from the active center characterized by slower bulk velocities ( say , @xmath176 few ) of the emitting plasma .
yet another possibility may be that the inner jets in ngc1275 ( and also in similar objects ) are in general intrinsically less relativistic than the ones in bona - fide blazars ; this would be consistent with the conclusions of lister & marscher ( 1997 ) , who argue that radio - loud agn with nuclear jets characterized by @xmath177 must be rather rare among the general population .
whichever scenario is correct , the _ fermi _ results seem to indicate that low - power radio galaxies are most likely not simple off - axis analogs of bl lac objects in terms of their @xmath0-ray properties .
the _ fermi_-lat collaboration acknowledges generous ongoing support from a number of agencies and institutes that have supported both the development and the operation of the lat as well as scientific data analysis .
these include the national aeronautics and space administration and the department of energy in the united states , the commissariat lenergie atomique and the centre national de la recherche scientifique / institut national de physique nuclaire et de physique des particules in france , the agenzia spaziale italiana and the istituto nazionale di fisica nucleare in italy , the ministry of education , culture , sports , science and technology ( mext ) , high energy accelerator research organization ( kek ) and japan aerospace exploration agency ( jaxa ) in japan , and the k. a. wallenberg foundation , the swedish research council and the swedish national space board in sweden .
abdo , a. , et al .
( _ fermi_-lat collaboration ) 2009a , , 699 , 31 ( paper - i ) abdo , a. , et al .
( _ fermi_-lat collaboration ) 2009b , , 700 , 597 ( lat bright agn sample ) abdo , a. , et al .
( _ fermi_-lat collaboration ) 2009c , , 183 , 46 ( lat bright source list ) abdo , a. , et al .
( _ fermi_-lat collaboration ) 2009d , , 707 , 55 ( m87 ) abdo , a. , et al .
( _ fermi_-lat collaboration ) 2010a , , submitted ( arxiv:1002.0150v1 ) abdo , a. , et al .
( _ fermi_-lat collaboration ) 2010b , , submitted ( arxiv:1002.2280v1 ) acciari , v. a. , et al .
2008 , , 684 , l73 acciari , v. a. , et al .
2009a , , 693 , l104 acciari , v. a. , et al .
2009b , , 706 , l275 acciari , v. a. , et al
. 2009c , science , 325 , 444 acciari , v. a. , et al . 2009d , , 707 , 612 ajello , m. , et al .
2009 , apj , 690 , 367 albert , j. , et al .
2007 , , 666 , l17 aleksi , j. , et al .
2010 , , 710 , 634 anderhub , h. , et al .
2009 , , 704 , l129 asada , k. , et al , 2006 , pasj , 58 , 261 atwood , w. b. , et al .
( _ fermi_-lat collaboration ) 2009 , , 697 , 1071 balmaverde , b. , capetti , a. , & grandi , p. 2006
, , 451 , 35 bloom .
s. d. , et al .
1997 , , 490 , l145 bhringer , h. , et al .
1993 , , 264 , l25 celotti , a. , ghisellini , g. , & chiaberge , m. 2001 , , 321 , l1 dunn , r. j. h. , & fabian , a. c. 2004 , , 355 , 862 fabian , a. c. , et al .
2003 , , 344 , l43 fabian , a. c. , et al .
2006 , , 366 , 417 ghisellini , g. , celotti , a. , fossati , g. , maraschi , l. , & comastri , a. 1998 , , 301 , 451 ghisellini , g. , tavecchio , f. , & chiaberge , m. 2005 , , 432 , 401 kirk , j. g. , & rieger , f. m. , & mastichiadis , a. 1998 , , 333 , 452 lister , m. l. , & marscher , a. p. 1997
, , 476 , 572 mayer - hasselwander , h. a. , et al .
1982 , , 105 , 164 owen , f. n. , eilek , j. a. , & kassim , n. e. 2000 , , 543 , 611 reimer , o. , pohl , m.,sreekumar , p. , & mattox , j. r. 2003 , , 588 , 155 reyes , l. c. , et al .
2009 , in proceedings of the 31st icrc ( astro - ph/0907.5175 ) sato , r. , et al .
2008 , , 680 , l9 sikora , m. , madejski , g. , moderski , r. , & poutanen , j. 1997 , , 484 , 108 stawarz , . , & ostrowski , m. 2002 , , 578 , 763 strong , a. w. , et al .
1982 , , 115 , 404 taylor , g. b. , & vermeulen , r. c. 1996 , , 457 , l69 vermeulen , r. c. , readhead , a. c. s. , & backer , d. c. 1994 , , 430 , l41 walker , r. c. , et al .
2000 , , 530 , 233 weekes , t. c. , et al .
2002 , astropart .
phys . , 17 , 22 lat @xmath0-ray localization error circles placed on the @xmath178 acis - s image constructed between 0.4 and 8.0 kev .
the positional center of the @xmath0-ray emission , marked as the white `` x '' , is only 0.46 arcmin from the position of the ngc 1275 nucleus , with 95@xmath46 radii of @xmath179 = 5.2 arcmin for paper - i and 1.5 arcmin ( more accurately , 1.56 @xmath180 1.38 arcmin ; abdo et al . 2010b ) for 11 months , respectively . ]
gev ) photons from the nucleus of ngc1275 as a function of photon energy .
the @xmath39 and @xmath2 angular resolutions of _
fermi_-lat are shown as dotted and dashed lines , respectively .
these psf profiles are derived for p6_v3_diffuse and have been averaged over the acceptance of the lat .
the highest energy photon detected during the one - year - all - sky survey is @xmath11gev ( _ double circle _ denoted as `` i '' ) , whose angular separation from ngc1275 is only @xmath12arcmin .
arrival times of the four highest energy photons ( `` i - iv '' ) , are indicated in figure4 . ]
100 mev spectrum of ngc1275 derived from one year accumulation of data ( _ filled circles _ ; this work ) , as compared with that determined from the initial 4-month dataset ( _ open squares _ ; paper i ) .
the dashed line shows the power - law function for the one - year data determined from the @xmath181 .
the dotted line represents the best - fit cutoff power - law with @xmath182gev as described in the text .
] -ray flux and spectral index over the period august 2008 august 2009 .
the time ( in days ) is measured from the start of the _ fermi _ observation , i.e. , 2008 august 4 , 15:43:37 ut . * upper panel : * changes in the @xmath107mev flux . *
lower panel : * changes in the power - law photon index .
we have divided the analyzed time window into epochs a ( before the flare ) , b ( during the flare ) , and c ( after the flare ) .
the arrival times of the highest energy photons i - iv ( figure2 ) are indicated as arrows .
background diffuse emission ( both galactic and extra - galactic ) is fixed at the best - fit parameters determined from an average spectral fit as given in the text , and only statistical errors are shown . ]
day ) and after the flare ( @xmath143day ) , respectively ( see figure4 ) .
label `` b '' denotes the spectral evolution during the flare with a time bin of 2 weeks .
note that the photon index changes significantly before and after the flare , while the flux levels are comparable . ]
-ray spectral energy distribution of ngc1275 before ( crosses ) and after the flare ( open circles ) , corresponding to epochs a and c in figure4 , respectively .
note that the excess high - energy @xmath0-ray emission only appears above @xmath145gev .
the flux upper limits in the tev range resulting from the veritas and magic observations correspond to the pre - flare epoch a. the dotted line represents the best - fit cutoff power - law fit as given in figure3 . ] | we report on a detailed investigation of the high - energy @xmath0-ray emission from ngc1275 , a well - known radio galaxy hosted by a giant elliptical located at the center of the nearby perseus cluster . with the increased photon statistics , the center of the @xmath0-ray
emitting region is now measured to be separated by only @xmath1arcmin from the nucleus of ngc1275 , well within the @xmath2 confidence error circle with radius @xmath3arcmin . early _
fermi_-lat observations revealed a significant decade - timescale brightening of ngc1275 at gev photon energies , with a flux about seven times higher than the one implied by the upper limit from previous egret observations . with the accumulation of one - year of _ fermi_-lat all - sky - survey exposure , we now detect flux and spectral variations of this source on month timescales , as reported in this paper .
the average @xmath4100 mev @xmath0-ray spectrum of ngc1275 shows a possible deviation from a simple power - law shape , indicating a spectral cut - off around an observed photon energy of @xmath5 gev , with an average flux of @xmath6ph@xmath7s@xmath8 and a power - law photon index , @xmath9 .
the largest @xmath0-ray flaring event was observed in april may 2009 and was accompanied by significant spectral variability above @xmath10 gev .
the @xmath0-ray activity of ngc1275 during this flare can be described by a hysteresis behavior in the flux versus photon index plane .
the highest energy photon associated with the @xmath0-ray source was detected at the very end of the observation , with the observed energy of @xmath11gev and an angular separation of about @xmath12arcmin from the nucleus . in this paper
we present the details of the _ fermi_-lat data analysis , and briefly discuss the implications of the observed @xmath0-ray spectral evolution of ngc1275 in the context of @xmath0-ray blazar sources in general . |
it has long been known that a galaxy s environment has an impact on its morphology : spirals dominate in the field , and ellipticals and s0s dominate in dense cluster environments ( hubble & humason 1931 ) .
this has been quantified in the density - morphology relation ( oemler 1974 ; dressler 1980 ) , but this relation alone does not determine whether the cluster environment affects the formation or the evolution of galaxies . according to the butcher - oemler effect ,
clusters at z @xmath0 0.2 are bluer and contain more spirals than nearby clusters , indicating that cluster environments may be correlated with the evolution of galaxies from spirals to earlier types ( butcher & oemler 1978 ) .
more recently , by examining a specific merging cluster , tran et al .
( 2005 ) presents a strong relationship between the butcher - oemler effect and infalling galaxies .
the environment - driven evolution of spiral galaxies into s0s is caused by mechanisms that physically remove gas and/or induce star formation until almost no gas remains .
galaxy - intracluster medium ( icm ) interactions are ones in which the gas in a galaxy interacts with the ambient intracluster hot gas .
one such process is ram pressure stripping ( rps ) , which removes the interstellar medium ( ism ) of a galaxy as it moves through the icm ( gunn & gott 1972 ) .
ram pressure could also compress the gas within a galaxy to cause a burst of star formation that would consume all of the remaining gas ( fujita & nagashima 1999 )
. a galaxy - icm interaction does not affect the stellar component of a galaxy .
the cluster potential can also strip both gas and stars or compress gas to cause an increased star formation rate ( byrd & valtonen 1990 ) .
these galaxy - cluster interactions affect both the stellar and gas components of galaxies .
there are also galaxy - galaxy interactions that can take place between galaxies .
these include mergers between galaxies with low relative velocities , and galaxy harassment : high - speed interactions between cluster galaxies ( hashimoto et al .
1998 ; bekki 1999 ; barnes & hernquist 1991 ; bekki 1998 ; moore et al .
these interactions can cause an increased star formation rate and will also affect both the stellar and gas component of galaxies . in our earlier work ( tonnesen , bryan & van gorkom 2007 ) , we studied cluster galaxies that had evolved in a cosmological simulation .
this simulation included dark matter and stellar particles , and followed gas through an adaptive mesh refinement ( amr ) code .
it included radiative cooling and star formation .
this work was unique because it allowed the cluster to form in a cosmological simulation , but had the resolution to follow individual galaxies ( 3 kpc ) .
we found that most galaxies that lost all their gas did so without losing stellar mass , indicating that ram pressure stripping was the most effective evolutionary mechanism .
this convinced us to look more closely at ram pressure stripping of galaxies . while a substantial body of literature exists on ram pressure stripping ( e.g. roediger & hensler 2005 ; roediger & brggen 2006 ; vollmer , hutchmeier & van driel 2005 ; see references in van gorkom 2004 ) ,
the effect of a multiphase ism has only begun to be explored in simulations .
schulz and struck ( 2001 ) included cooling and found that it caused a disk instability that led to an inflow of dense gas , and that including cooling allows the disk to remain cool and thin despite icm heating . stripping of a multiphase ism
was also studied at higher resolution by quilis , moore & bower ( 2000 ) .
they did not include radiative cooling , but instead randomly placed holes and overdensities in their ism disk in an attempt to model an inhomogeneous ism .
they find that stripping is much more complete as icm streams through the holes in their galaxy and prevents stripped material from falling back onto the galaxy as earlier simulations had shown .
more recently , kronberger et al ( 2008 ) have modeled cooling in the context of star formation .
they find increased star formation in the center of their galaxy and star formation in the stripped material ( although it does not escape the galactic halo ) .
observations have been inconclusive about the effect of stripping on dense molecular clouds . in a survey of virgo cluster galaxies ,
kenney & young ( 1986 , 1989 ) found that co remains in hi deficient galaxies , indicating that molecular clouds are not being stripped entirely . whether this is because molecular clouds reside in the center of galaxies where even atomic gas is rarely stripped , or because molecular clouds are too dense to be stripped is unclear .
the likelihood of molecular gas being stripped varies with each individual case .
clemens et al .
( 2000 , 2001 ) have examined in detail the interacting galaxy pair ngc 4490 and ngc 4485 .
they find evidence that in moving through the extended hi envelope surrounding the two galaxies , ngc 4485 has had atomic gas , molecular gas , and dust removed via ram pressure stripping .
crowl et al .
( 2005 ) have studied the virgo cluster galaxy ngc 4402 , and observe ram pressure stripping and dense cloud ablation .
they find that it is possible to strip the lower density ism without stripping dense clouds , but that the surviving clouds will eventually be ablated by the icm wind .
they also observe an hii region which has likely formed within the stripped gas , so molecular gas was either stripped or formed from stripped material .
however , in ngc 4848 , vollmer et al ( 2001 ) find that the displaced molecular gas they observe is formed from re - accreted lower density gas , and that the molecular gas was not stripped at all .
ngc 4438 has been observed to have both stripped and surviving molecular gas , determined by the clouds positions in the galaxy ( vollmer et al .
another galaxy in the virgo cluster , ngc 4522 , has no hii regions in the outer disk , but extraplanar hii regions ( kenney & koopmann 1999 ; kenney , van gorkom & vollmer 2004 ) . as discussed in kenney
et al ( 2004 ) , there are four possibilities for stripping of molecular clouds .
( 1 ) molecular clouds can be directly stripped from the disk , although there is probably an upper limit to the surface density for which this is possible .
( 2 ) molecular clouds , which are thought to have a lifetime of about @xmath1 years ( blitz & shu 1980 ; larson 2003 , hartmann 2003 ) , can also evolve via star formation into lower density gas , which can then be stripped directly .
( 3 ) another possibility is that the molecular clouds are not directly stripped out of the galaxy , but are instead ablated by the icm wind on a longer timescale ( nulsen 1982 ) .
( 4 ) finally , dense clouds may be coupled to the rest of the ism by magnetic fields or some other mechanism , so that the entire ism is stripped out together . in this paper
, we run a set of high resolution simulations ( 40 pc resolution , which is small enough to marginally resolve giant molecular clouds ) to understand how a multiphase ism could affect ram pressure stripping of a galaxy s gas .
it is important to recognize that we do not attempt to include all of the physics involved in the ism .
we use these simulations to understand how the density fluctuations that are observed in the multiphase ism of galaxies will affect gas loss .
we vary ram pressure strength to investigate the result of different wind strengths . using our simulations , we are able to look in detail at possibilities ( 1 ) and ( 3 ) .
we also consider whether including molecular clouds affects the large scale stripping properties such as the total amount of gas stripped and the radius to which gas is stripped . in an introduction to our code
, we provide the general characteristics of our simulation ( 2.1 - 3 ) .
we then discuss our suite of runs and the measurements we use to compare them ( 2.4 - 5 ) . in
3 we discuss how cooling affects a galaxy evolving in a static icm .
we then ( 4 ) compare our simulations . in 5 we briefly discuss the limitations and implications of our results .
finally , we conclude in 6 with a summary of our results and the overall picture of ram pressure stripping of a galaxy with a multiphase ism .
we have simulated a galaxy using the adaptive mesh refinement ( amr ) code _ enzo_. although this code can follow stellar and dark matter particles , we chose to use a fixed stellar and dark matter potential for ease of computation , although the self - gravity of the gas is self - consistently computed . to follow the gas
, we employ an adaptive mesh for solving the fluid equations including gravity ( bryan 1999 ; norman & bryan 1999 ; oshea et al . 2004 ) .
the code begins with a fixed , static grid and automatically adds refined grids as required in order to resolve important features in the flow as defined by enhancements in the gas density .
the simulation includes radiative cooling using the sarazin & white ( 1987 ) cooling curve .
we allow cooling to 8,000 k , which results in gas with neutral hydrogen temperatures without overcooling a large fraction of our gas . with this cutoff , we find that we still can form gas clouds with densities typical of molecular clouds , although it is clear that the internal structure of such clouds is not reproduced in detail ( see figure [ fig : mass_rho ] for the density mass distribution in the gas disks of our galaxies ) .
we discuss this choice in greater detail in [ sec - cool ] .
our box is 311 kpc on a side .
our coarsest resolution is 2.5 kpc .
we allow 6 levels of refinement for our runs that include radiative cooling , for a best resolution of 40 pc .
the runs without radiative cooling have 5 levels of refinement .
we refine our simulation using baryon mass , with our minimum overdensity for refinement set as 51.2 , which we found immediately refined the entire galactic disk to our best resolution .
we discuss resolution effects in more detail in
[ sec - res ] .
our galaxy is placed at ( 0.5,0.5,0.22 ) or ( 155.5,155.5,68.42 ) kpc , so that we can follow the stripped gas as far as possible .
the icm wind flows along the z - axis in the positive direction , with the lower z boundary set for inflow and higher z boundary set as outflow .
the x and y boundaries are set to reflecting .
we model a massive spiral galaxy with a flat rotation curve of 200 km @xmath2 .
it consists of a gas disk that is followed using adaptive mesh refinement ( including self - gravity of the gas a crucial ingredient required to form self - gravitating molecular clouds ) , as well as the static potentials of a stellar disk , a stellar bulge , and a dark matter halo .
we directly follow roediger & brggen ( 2006 ) in our modeling of the stellar and dark matter potential and gas disk .
briefly , we model the stellar disk using a plummer - kuzmin disk ( see miyamoto & nagai 1975 ) , the stellar bulge using a spherical hernquist bulge ( hernquist 1993 ) , and the dark matter halo using the spherical model of burkert ( 1995 ) .
this dark matter halo model is compatible with observed rotation curves ( burkert 1995 ; trachternach et al .
the equation for the analytic potential is in mori & burkert ( 2000 ) .
the gas is described as a softened exponential disk : @xmath3 given this gas density distribution in the disk , the gas temperature and pressure are calculated to maintain the disk in hydrostatic equilibrium with the surrounding icm in the z direction .
the gas disk s rotational velocity is set so that the combination of the centrifugal force and the pressure gradient of the disk balances the radial gravitational force .
we cut the gas disk smoothly by multiplying the gas density distribution by @xmath4 kpc@xmath5 kpc@xmath6 for 20 kpc @xmath7 r @xmath8 26(21 ) kpc .
see our galaxy parameters in tables [ tbl - const ] and [ tbl - gconst ] .
we use different initial gas disk radii in order to more easily compare disks when the icm wind hits the galaxy .
we discuss this in greater detail in
[ sec - nowind ] .
c | c variable & value + m@xmath9 & @xmath10 m@xmath11 + a@xmath9 & 4 kpc + b@xmath9 & 0.25 kpc + m@xmath12 & @xmath13 m@xmath11 + r@xmath12 & 0.4 kpc + r@xmath14 & 23 kpc + @xmath15 & @xmath16 g @xmath17 + c | c | c variable & value rc & value nc + m@xmath18 & @xmath13 m@xmath11 & @xmath13 m@xmath11 + a@xmath18 & 7 kpc & 6.5 kpc + b@xmath18 & 0.4 kpc & 0.4 kpc + our galaxies evolve in a pressurized icm .
because a galaxy moving through the icm can be more easily simulated by modeling a fixed galaxy within a moving icm , our galaxy remains in the same place in our simulated box ( at least the stars and dark matter do ) .
the galaxy initially evolves in a static icm , to examine the stability in the static icm and to allow cool , dense gas to form .
later , we trigger a constant icm inflow along the z - axis , which is always face - on to the galaxy . see the appendix for exactly when the wind hits each galaxy .
briefly , the wind is triggered so that it hits the galaxy while gas is still distributed out to a large radius , but enough time has passed so that high density gas clouds have formed .
the exact time was chosen to be when the galaxy in the corresponding rcnw case had gas collapsed to @xmath19 @xmath20 10@xmath21 g @xmath17 , typical of densities found in molecular clouds .
dynes @xmath22 , @xmath23 dynes @xmath22 , and @xmath24 dynes @xmath22 .
[ sec : icm ] for discussion . ]
we choose to study three ram pressure strengths using realistic values for the icm density and velocity .
the three ram pressure strengths we choose are @xmath25 dynes @xmath22 , @xmath23 dynes @xmath22 , and @xmath26 dynes @xmath22 ( see table [ tbl - icm ] ) .
we choose the maximum and minimum ram pressure strengths to match two of the values used by roediger and brggen ( 2006 ) , and we pick @xmath23 dynes @xmath22 because it is the highest ram pressure value at the virial radius of our cluster simulation ( tonnesen et al .
2007 ) . to choose the corresponding icm parameters , @xmath19 and @xmath27
, we use the results from our earlier cluster simulation to find the mean density , velocity , and temperature of the icm at the three ram pressures ( tonnesen , bryan & van gorkom 2007 ) .
this is shown in figure [ fig : sample_select ] , where for comparison we plot the values used by roediger & brggen ( 2006 ) .
as discussed earlier , we set the inflow boundary condition in order to model the icm wind .
however , because we need to allow our galaxies to cool and evolve before the icm wind begins to strip them , the galaxies are initialized in a static icm and then , after a multiphase ism has developed , the boundary values are set to generate a wind with the desired characteristics . to get the parameters for this initial icm we backtrack from the density , temperature and velocity of the icm wind using the rankine - hugoniot jump conditions , assuming a mach number of 3.5 .
we choose this mach number because it gives us a supersonic shock ( which leads to cleaner behavior at the inflow boundary ) , while allowing our initial static icm to have a similar density and temperature as the icm wind .
this procedure is important because we compare the evolution of our stripped galaxy to one that only cools in a static medium that has the same density and temperature as the flowing icm .
we use the same density and temperature as in the moving icm for the cases in which the icm always remains static , because the stripped galaxies spend more time in the simulation in the moving icm .
c | ccccccc runs & @xmath28 & t@xmath29 & p@xmath30 & p@xmath31 & v@xmath29 & cooling ? &
r@xmath32 + phrcw & @xmath33 & @xmath34 & @xmath35 & @xmath36 & 1413.5 & y & 25 + phncw & @xmath33 & @xmath34 & @xmath35 & @xmath36 & 1413.5 & n & 21 + phrcnw & @xmath33 & @xmath34 & @xmath35 & 0.0 & 0.0 & y & 25 + pmrcw & @xmath37 & @xmath38 & @xmath39 & @xmath23 & 1045.4 & y & 25 + pmncw & @xmath37 & @xmath38 & @xmath39 & @xmath23 & 1045.4 & n & 21 + pmrcnw & @xmath37 & @xmath38 & @xmath39 & 0.0 & 0.0 & y & 25 + plrcw & @xmath40 & @xmath41 & @xmath42 & @xmath26 & 1111.7 & y & 25 + plncw & @xmath40 & @xmath41 & @xmath42 & @xmath26 & 1111.7 & n & 21 + plrcnw & @xmath40 & @xmath41 & @xmath42 & 0.0 & 0.0 & y & 25 + we perform nine full runs because while attempting to understand how cooling affects galactic gas stripping , we also need to understand how cooling alone affects a galaxy .
we examine how three different ram pressure strengths affect galaxies that include and do not include radiative cooling .
we choose a naming convention for the simulations which indicates the parameters used for that simulation ( see also table [ tbl - icm ] ) .
the three ram pressure strengths also correspond to the three icm thermal pressures , and are denoted ph , pm , pl for high , medium , and low .
if cooling is turned on , the next identifier is rc , if not , nc . the six runs with an icm wind end with w. in addition to these six runs , we also study three runs of galaxies evolving with radiative cooling in a static icm with the same temperature and density as the moving icm .
all these runs end with nw , for no wind . for details on these nine simulations , see table [ tbl - icm ] , as well as the appendix . because we are interested primarily in the impact of radiative cooling on gas stripping , we only consider cases with a face - on wind .
these cases are likely to be the most affected by an icm wind because the gas density fluctuations in the disk , caused by radiative cooling , are most apparent in a face - on view .
we look closely at three types of measurements : 1 ) the total amount of gas in the galaxy , 2 ) the radius of the dense gas in the galaxy , and 3 ) the amount of gas mass at different densities .
we first measure the total amount of gas that remains in the galaxy . we measure this in two ways : the amount of gas mass in a cylinder centered on the galaxy with a height of 10 kpc and a radius of 27 kpc , and the total amount of bound gas mass in the entire box .
mass is bound if the thermal and kinetic energy are smaller than the gravitational energy from the static potential of the stars and dark matter .
we can compare these values to an analytic estimate of how much gas a galaxy should lose using two analytic mass loss estimates , corresponding to two stages of gas stripping .
we will follow roediger & brggen ( 2007 ) by naming the initial stage ram pressure pushing and the later stage continuous stripping .
we can predict the amount of gas loss from ram pressure pushing by measuring the amount of gas mass outside the analytic gunn & gott ( 1972 ) stripping radius , the radius at which the restoring force per unit area from the galaxy is equal to the ram pressure acting on the galaxy , or where @xmath43 .
we calculate the radius using the initial gas distribution and measure the mass outside this radius immediately before the icm wind hits the galaxy .
once initial gas stripping has occurred , there is a slower continuous stripping phase which is caused by the kelvin - helmholtz instability .
nulsen ( 1982 ) derived an equation for the mass loss rate of a spherical cloud , which roediger & brggen ( 2007 ) adapted for a disk : @xmath44 .
because we use a constant density and velocity icm wind , only the changing radius of the galaxy will change this rate of gas loss .
we measure the stripping radius in the simulation by finding the radius of gas with a density @xmath45 g @xmath17 ( or approximately a number density of @xmath46 @xmath17 ) . to mitigate the effect of asymmetry on this measurement
, we find the radius in 12 azimuthal slices around the disk and take the average . because cooling causes both overdensities and underdensities , we measure this radius by finding the distance to the furthest cell with gas above our threshold density .
as may be expected , this increases the radius we measure for the galaxies that include cooling , and has no effect on the radius of the galaxies without cooling .
we compare this to the analytic stripping radius .
we also measure the amount of gas at different densities .
the gas is split into three density regimes : ( 1 ) high density : @xmath47 g @xmath17 , ( 2 ) moderate density : @xmath48 g @xmath49 g @xmath17 , and ( 3 ) low density : @xmath50 g @xmath51 g @xmath17 .
we choose these density ranges in order to follow three different types of gas : gas with high densities seen in molecular clouds ( crutcher 1991 and references therein ) , gas with densities that are nearly that of molecular gas , and finally gas which is clearly not molecular , but more dense than the surrounding icm .
we use the gas that is within the cylinder centered on our galaxy for this measurement .
although the stability of this galaxy model is discussed in detail in roediger & hensler ( 2005 ) and roediger & brggen ( 2006 ) , it is important to reiterate that using _ enzo _ , the galaxies are stable if there is no radiative cooling and no icm wind .
although they maintain the exact same amount of gas throughout the simulation , there is a fluctuation in the radius with the epicyclic frequency for this galaxy .
this is most likely due to the rapid change in gas density , and therefore pressure , in the outer kpc ( between 20 - 21 kpc ) of the galaxy in the initial galactic profile . in figure
[ fig : ncnw ] , we show how the radius varies with time for two of the runs in which the galaxy sits in a static environment with icm pressure and without radiative cooling . in the simulations in which we include radiative cooling , at all icm pressures , gas with a density of @xmath48 g @xmath17 forms . at this point ,
gas with the density of molecular clouds has formed , and we consider the gas disk to be a multiphase ism . in all cases , the highest density gas ( @xmath52 10@xmath53 g @xmath17 ) only forms stably within the central 7 kpc of the disk . also , the mass fraction of gas with densities of molecular gas ( @xmath52 10@xmath54 g @xmath17 ) is 38% - 50% after evolving for @xmath0 1 gyr in a static icm with radiative cooling .
we closely examine three galaxies with radiative cooling in three different pressure icm s and no wind ( the rcnw cases in table [ tbl - icm ] ) .
we find that as in the galaxies without cooling there does seem to be a radial epicyclic oscillation .
however , in the higher icm pressure cases ( phrcnw and pmrcnw ) , low density and low pressure regions in the disk allow non - rotating icm gas to enter and interact with the rotating denser clouds . these clouds therefore slow and lose angular momentum , leading the disk to settle towards the center .
this drop in disk radius can be seen in the bottom panels of figure [ fig : radii ] .
it is notable that in the lowest pressure icm there is very little entrainment of the icm gas , and the radius of the gas is dominated by epicyclic oscillations .
we also find evidence of this process , and how it varies with icm pressure , by following the total amount of gas in these galaxies .
for example , phrcnw gains @xmath55 m@xmath56 over the 950 myr run , while plrcnw gains @xmath57 m@xmath56 , because of the different amounts of gas entrained through cooling of the disk . because the radius of the gas disk decreases as the gas cools ( see the bottom panels of figure [ fig : radii ] ) , we lower the gas radius of the galaxies _ without _ cooling to the radius that the radiatively cooled galaxies reach due to their entrainment of the icm ( as described in the previous paragraph ) .
therefore , although the radius of the galaxies with cooling begins at 25 kpc , the radius of the galaxies without cooling start at 21 kpc . despite this different initial radius , we set the radial gas scale length , a@xmath18 , so that the galaxies with and without cooling have the same amount of mass .
this is all done in order to ensure a fair comparison between the radiatively cooling and non - cooling runs .
in figures [ fig : rcshots ] and [ fig : ncshots ] we show projections along the y- and z- axis towards phrcw and phncw about 40 myr , 250 myr , 500 myr , and 750 myr after the wind has hit the galaxy .
the projections are shaded to reflect surface density of the gas with a logarithmic stretch , and are 103.7 kpc across .
there are a number of important differences in the morphology of the gas disks that are reflected in our quantitative comparisons .
first , the gas in the galaxy with radiative cooling ( phrcw ) forms small clouds with radii of about 100 pc near the center of the disk ( within 7 kpc ) that are of higher density than anywhere in the gas disk without cooling ( phncw ) .
only these central clouds attain molecular densities , even in the galaxy that is never affected by an icm wind .
further , at any radius of the disk in phrcw , there is both higher and lower density gas than in the disk of phncw .
+ + + + + + + + this density structure has an important impact on the way that gas is stripped from the disk .
first , it allows icm gas to create holes in the disk , which leads to ablation of material from a wide range of radii , rather than just at the outer edge of the disk , as in the no - cooling run ( phncw ) .
as we will show in more detail below , this allows the gas to be stripped more quickly ( see also quilis et al 2000 ) .
in addition , the gas that is stripped tends to stream directly behind the disk , rather than flaring to the side , and therefore creates a narrower wake .
we will examine the wake in more detail in a future paper .
the density fluctuations in the disk in the phrcw galaxy create a ring of higher density gas from about 10 kpc to 16 kpc .
this gas never collapses to the highest densities .
in fact , in the phrcnw galaxy , gas with @xmath58 g @xmath17 only persists for @xmath59100 myr in the middle of the simulation .
although the vast majority of gas in this ring is always of a low enough density for direct stripping to occur , allowing the simulation to run until this ring had fully fragmented might result in a different stripping scenario .
it would also allow gas more time to drift towards the center of the galactic potential , making it more difficult to strip .
however , because little molecular gas is found outside the center of galaxies in both these simulations and in observations ( e.g. leroy et al . 2008 ) , we do nt expect the ring structure to make much difference in how much gas can be stripped .
although in this paper we will not discuss the stripped gas in any detail , it is notable that the stripped gas in phrcw has more density substructure than that of phncw .
higher density gas leaves the disk in phrcw than in phncw ( although not gas with molecular densities ) , and stripped clumps survive for at least 100 kpc .
we begin our quantitative analysis of the impact of cooling by comparing the amount of gas mass that is lost in the cases with and without cooling ( figure [ fig : masses ] ) . in general
, the amount of gas mass lost by the galaxies that include cooling is about the same as the gas mass lost by the galaxies that do not include cooling .
however , the timescale for stripping is much shorter for the radiatively cooled galaxies .
these differences are illustrated in table [ tbl - strip ] .
note that the last column in this table is the time from when the wind first hits the galaxy to when the galaxy has lost 85% of the total amount of gas lost . in the highest ram pressure runs ( phncw and phrcw ) , after the wind has been hitting the galaxy for 800 myr , both the galaxies with and without cooling have lost similar amounts of mass : @xmath60 m@xmath56 and @xmath61 m@xmath56 in the cooling and no cooling cases , respectively .
we can compare these values to an analytic prediction for how much mass the galaxies should lose based on the amount of gas mass outside the analytic gunn & gott ( 1972 ) stripping radius measured immediately before the wind hits the galaxy .
we find that the predictions are similar , although smaller , than the measured values : the galaxy with cooling should lose @xmath62 m@xmath56 and the galaxy without cooling should lose @xmath63 m@xmath56 . in the case with cooling ,
almost the entirety ( 86% ) of this mass loss occurs within 150 myr , while the same amount of mass loss takes about 650 myr in the phncw case .
also , both of these galaxies have a fallback event ( gas that is initially stripped from the disk region , although still gravitationally bound , falling back onto the disk ) about 250 myr after the wind hits the galaxy , although in phncw , the fallback is more drastic and seems to be followed by a few minor fallback events .
although the phrcw case seems to stop losing gas despite the continued icm wind , losing less than @xmath64 m@xmath11 in the final 300 myr we follow the galaxy , the phncw case does not slow its gas loss by nearly as much , losing twice as much gas in the same amount of time .
this difference is fit well by the difference in radius between the two cases and the dependence of kelvin - helmholtz stripping on radius squared . in pmrcw , our run with the middle ram pressure value , after 900 myr of the wind hitting the galaxy , it has lost about @xmath65 m@xmath11 of gas mass .
in fact , it loses almost this entire amount of gas ( 95% ) within 340 myr of the wind first hitting the galaxy . on the other hand , in pmncw
, the galaxy has only lost @xmath66 m@xmath56 in the 900 myr since the icm wind has hit the galaxy . unlike the highest ram pressure case , the analytic prediction for the amount of gas loss is larger than the measured amount : @xmath67 m@xmath56 for pmrcw and @xmath68 m@xmath56 for pmncw .
again , the flatness of the pmncw gas mass curve and the steeper grade of the pmrcw gas mass curve may be attributed to kelvin - helmholtz stripping effecting only the galaxy without cooling .
also , there does not seem to be any steep drop in the pmncw galaxy s gas that is the hallmark of ram pressure pushing .
even more than in the highest ram pressure case , the galaxy with cooling seems to be only instantaneously stripped , without much longer - term kelvin - helmholtz stripping .
the lowest ram pressure runs ( plrcw and plncw ) are very similar to the middle ram pressure runs , except that only @xmath69 and @xmath70 of gas are lost in the galaxies with and without cooling , respectively . like the middle ram pressure case , the analytic prediction for the amount of gas loss is more than the amount measured , @xmath71 ( plrcw ) and @xmath72 ( plncw ) m@xmath11 .
even when the ram pressure is only @xmath26 dynes @xmath22 , the galaxy with radiative cooling appears to lose gas through ram pressure pushing .
although similar amounts of gas mass are lost whether or not the galaxy is allowed to radiatively cool , there are a few trends .
first , in the middle and low ram pressure cases , the amount of gas lost by the galaxy with radiative cooling is more than that lost by the galaxy without radiative cooling .
although this qualitatively agrees with the analytic predictions for the amount of gas mass loss indicating that the radiatively cooled galaxy will lose more gas , the difference in measured mass loss between the rc and nc cases is greater than the difference in the analytic prediction .
thus , it may be that for lower ram pressures , a multiphase ism is easier to strip . also , galaxies that radiatively cool lose their gas more quickly .
there is slightly more mass in total outside the stripping radius in the radiatively cooling galaxies , and there is more mass at higher density .
a plausible reason for the quick stripping of gas mass , then , is that a lumpy gas distribution is more quickly and easily stripped than gas with a smoother profile .
later continuous stripping seems to mainly affect the galaxies without cooling , perhaps because their larger cohesive gas disk is more affected by the kelvin - helmholtz instability .
c | ccccc runs & m@xmath73 & m@xmath74 & r@xmath75 & r@xmath74 & t@xmath76 + phrcw & @xmath77 & @xmath78 & 11 & 11.46 & 148 + phncw & @xmath79 & @xmath80 & 7 & 11.46 & 625 + pmrcw & @xmath81 & @xmath67 & 11 & 16.8 & 288 + pmncw & @xmath82 & @xmath68 & 19.5 & 16.8 & 810 + plrcw & @xmath83 & @xmath84 & 15 & 18.22 & 316 + plncw & @xmath85 & @xmath72 & 20 ? & 18.22 & 754 + in general , the galaxies that include radiative cooling are stripped to a smaller radius than predicted using the gunn & gott ( 1972 ) analytic equation , and end at a smaller radius than stripped galaxies without cooling and at a smaller radius than galaxies with radiative cooling that are not affected by an icm wind ( figure [ fig : radii ] ) .
again , see table [ tbl - strip ] for the stripping radius values . in the highest ram pressure cases ,
the analytic stripping radius is 11.5 kpc , very similar to the stripping radius of the galaxy in phncw .
the oscillations in the radius correspond to fallback episodes in figure [ fig : masses ] .
the final radius of the phrcw case is about 7 kpc .
the final radius of the phrcnw galaxy is just above 11 kpc .
this means that including cooling without a wind causes galactic gas to spiral towards the center of the disk , most likely because the entrainment of nonrotating icm into the disk creates drag and angular momentum loss .
the entrainment of gas into the disk means that the average gas density of the resulting disk would be higher than before , making it more difficult to strip .
in fact , comparing the amount of gas at different densities in phrcw and phncw about 15 myr before the icm wind hits the galaxy , we find that phrcw has more high density gas outside 10 kpc and a higher fraction of the gas in the disk region is high density than in phncw .
therefore , the holes in the disk must allow for easier stripping down to smaller radii .
the holes may also create more drag against surviving gas , allowing it to spiral towards the center of the disk more quickly . in the middle ram pressure cases ,
the analytic stripping radius is 16.8 kpc .
the final radius of the galaxy in pmncw is between 19 and 20 kpc .
however , the initial radius drop is to about 17 kpc , while a static galaxy without cooling , only undergoing epicyclic variations , has a smallest radius of about 19 kpc .
the final average radius of the galaxy in pmrcnw is steady at 15 kpc , although it is very asymmetric .
the galaxy in pmrcw is stripped down to a radius of 11 kpc .
again , a mixture of stripping and cooling results in the smallest radius . in the lowest ram pressure cases ,
the analytic stripping radius is 18.2 kpc .
it is difficult to say whether the galaxy in plncw is stripped to a smaller radius because of the large oscillation in radius over time .
plrcw is stripped to about 15 kpc . in this case
more than in the previous two , it is clear that it must be the holes punched by the icm wind in the gas disk that allow the gas to be stripped to a small galactic radius .
neither plncw nor plrcnw have much change in the radius of the gas disk , so it must be a combination of the icm wind and the clumpiness of the gas in the disk that result in the lowest gas radius . by examining the stripping radius of gas in these galaxies , we find that allowing radiative cooling results in a smaller radius when hit by an icm wind . again , the most likely explanation is that the clumpiness of the disk gas allows for stripping to a smaller radius
. as gas in the galaxy cools , it becomes more dense , eventually attaining densities greater than 1000 @xmath17 , certainly in the range of molecular gas .
although with a resolution of 40 pc , we only marginally resolve giant molecular clouds , we can begin to answer whether ram pressure has any effect on the densest gas in a galaxy ( and hence on star formation ) .
when comparing the three cases with radiative cooling and no icm wind ( bottom panels in figure [ fig : mass_rho ] ) , it is clear that a higher pressure icm results in a faster formation of high density gas and the formation of more high density gas .
however , as we will see , including an icm wind can either strip the surrounding lower density gas that would accrete onto clouds , or create more high density gas when the wind pressure is low .
we find that with the highest ram pressure of @xmath36 dynes @xmath22 , the amount of higher density gas ( above @xmath86 g @xmath17 ) is less than if the galaxy were to radiatively cool without an icm wind ( see figure [ fig : mass_rho ] ) .
this is true even when ( as in figure [ fig : mass_rhoh ] ) , we shift the phrcw line showing how much gas has cooled as a function of time by 200 myr in order to account for the fact that the galaxy in phrcw has been in the higher pressure icm for almost 200 myr less than the galaxy in phrcnw . when the wind hits the galaxy , less dense gas throughout the disk is accelerated out of the galaxy , leaving less nearby lower density gas that would later either be accreted onto nearby clouds or cool and collapse into new clouds .
however , gas with densities above @xmath86 g @xmath17 is never seen outside of the disk ( always within 1 kpc of the disk plane ) , so molecular clouds are not being stripped in their entirety .
examining the snapshots indicates that higher density clouds do leave the disk , leading us to hypothesize that dense clouds can be ablated until they are of just low enough column density to be stripped from the disk .
the scenario seems less clear in the medium ram pressure case ( @xmath23 dynes @xmath22 ) .
although the amount of higher density gas increases more slowly in pmrcw than in pmrcnw , both end their runs with similar amounts of high density gas ( see figure [ fig : mass_rhoh ] ) . in the lowest ram pressure case ( @xmath26 dynes @xmath22 ) , the galaxy that is hit by the icm wind has more of the highest density gas ( above @xmath48 g @xmath17 ) than the galaxy that evolves in the post - shock icm . in this case , the added ram pressure from the wind appears to compress the gas , leading to more higher density gas . in summary ,
molecular clouds do not get stripped , even if the ram pressure is @xmath36 dynes @xmath22 .
they do not grow within a high ram pressure icm wind , however , because lower density gas is accelerated out of even the inner disk ( inside the stripping radius of 7 kpc ) .
however , with low ram pressure , @xmath26 dynes @xmath22 , more high density gas is formed , likely because the ram pressure , while not enough to strip much gas from the disk , is enough to compress gas to a higher density .
a galaxy is considered hi deficient if its deficiency value is 0.3 or greater .
deficiency is defined as def = @xmath7 log x @xmath87@xmath88 - log x@xmath89 ( giovanelli & haynes 1985 ) , where x@xmath89 is the observed hi mass of the cluster galaxy and @xmath7 log x @xmath87@xmath88 is averaged over a sample of field galaxies of morphological type t and optical diameter d. given this definition , the highest ram pressure case is almost but not quite hi deficient .
however , gas is truncated to less than half the original gas radius in the largest ram pressure case , and is down to less than 75% in the the lowest ram pressure case .
if we were to roughly classify these galaxies using the scheme of koopmann & kenney ( 2004 ) , assuming that hii regions will follow our gas distribution , all three of our runs result in truncated disks .
we can also compare the amount of high density gas we find in our galaxies to the amount of molecular gas observed using co. kenney & young ( 1986 , 1989 ) found that hi deficient galaxies in virgo have undepleted co , and nakanishi et al .
( 2006 ) found that the molecular fraction in three galaxies near the center of virgo is high .
we find that in our high ram pressure run , the fraction of high density ( i.e. molecular ) gas actually remains similar to the case with no wind . on the other hand , in the low ram pressure case
, the fraction of high density gas does increase as the icm wind compresses the ism .
this case better matches the observations of high molecular gas fractions ( nakanishi et al .
it is possible that galaxies will move through low ram pressure environments on their way to higher ram pressure , so the initial compression followed by strong stripping could result in hi depleted disks with unaffected , or increased , amounts of molecular gas .
however , we do note that these results are somewhat sensitive to the resolution and details of the cooling ( discussed in more detail below ) .
we do not include star formation in our simulations , but we do have an enhanced amount of high density gas in our lowest ram pressure simulation , while our highest ram pressure simulation has less high density gas in comparison to their respective no - wind comparison runs .
if we assume that higher density gas corresponds to a higher rate of star formation , the lowest ram pressure case would result in an enhanced star formation rate , while the highest ram pressure case would result in a lower star formation rate .
placing this into a cluster environment , we would expect to find enhanced star formation at or outside the virial radius in the lower density icm , and post - star - forming galaxies closer to the cluster center . in coma ,
poggianti et al ( 2004 ) found that post - starburst or post - star - forming galaxies were found at the edge of icm substructure , which seems to follow our claim that starbursts will occur outside of the densest icm , while high ram pressure will quench star formation .
we can also use morphological cues to determine how closely these simulations match observations .
a perfect example of a galaxy showing both the strength and weakness of our simulations that include a multiphase medium is the virgo spiral ngc 4402 ( crowl et al 2005 ) .
the galaxy has upturned edges in it s dust and gas distribution , similar to the earliest projection of our galaxy without cooling .
however , similar to our galaxy with radiative cooling , there is evidence that two dense clouds have survived longer than the lower density ism and are being ablated by the wind .
a molecular cloud may survive between a few and a few tens of myr ( blitz & shu 1980 ; larson 2003 ; hartmann 2003 ) . because we do not include star formation
, our dense clouds of gas do not evolve into less dense pockets of gas that could be stripped . in the highest ram pressure case
, holes are punched around the disk ( as close as 6 kpc to the disk center ) between 6 - 11 myr after the icm wind first hits the galaxy , which indicates that as long as gas stays low density for 10 myr it can be stripped by the wind .
if our dense regions of gas could turn into very low density regions through star formation ( @xmath19 @xmath90 10@xmath91 g @xmath17 ) , our galaxies that include radiative cooling could be entirely stripped of gas in the highest ram pressure case .
in the lowest ram pressure case , no gas appears to be stripped from the central region of the gas disk .
our simulation does not include magnetic fields or cosmic rays .
in general in the milky way , the magnetic field is measured to be about 2 - 3 microgauss ( men et al 2008 ) . a field of this strength would not have enough magnetic pressure ( @xmath92 dynes @xmath22 ) to significantly lessen the direct impact of ram pressure on the galactic gas .
however , there are measurements of the magnetic field in molecular clouds ( crutcher 1991 and references therein ) of up to @xmath93 microgauss , which can certainly withstand ram pressures of order 10@xmath94 dynes @xmath22 .
these magnetic fields could help protect molecular clouds from being ablated by the icm wind , so could help them survive .
also , a magnetic field could attach a molecular cloud to less dense gas , either anchoring the surrounding gas or helping to drag out the molecular cloud . using this simulation
we can not conjecture on the likely impact of magnetic fields on the total amount of gas stripped .
we also do not include cosmic rays in our simulation , which would add energy to the galactic gas and increase the vertical distribution ( e.g. joung & mac low 2006 ) .
it is also possible that the increase in gas pressure would lower the amount of entrainment into the disk from the lower pressure icm , which would in turn lessen the number of very low density regions and make the disk more cohesive .
this would make stripping more difficult because fewer holes could be punched through the disk by the icm wind .
however , the energy input from cosmic rays would slow high density gas formation , possibly allowing for more extended stripping if the gas is not locked in very dense clouds . as with magnetic fields
, the addition of cosmic rays into our simulation does not have an easily predicted effect . as has been discussed in detail in previous works ( e.g. tonnesen et al .
2007 and roediger & brggen 2006 ) , resolution does not have a large effect on our runs without radiative cooling .
the runs without radiative cooling that we discuss in this paper have a cell size of 80 pc , and we ran a comparison run with a cell size of 40 pc to check this .
we found very similar results in the amount of gas and the radius of the disk .
in addition , our results are similar to those in roediger & brggen ( 2006 ) , where even lower resolution was used . however , including radiative cooling leads to results that are very dependent on resolution .
the runs discussed in this paper have a cell size of 40 pc , which leads to gas collapsing to molecular densities .
for example , the surface density in the center of the disk immediately before the wind hits the galaxy is 200 m@xmath56 kpc@xmath95 , with values above 10 m@xmath56 kpc@xmath95 for much of the inner disk ( this surface density corresponds to the transition between hi and molecular gas observationally found in bigiel et al 2008 ) . to examine the impact of resolution
, we have run a comparison simulation with the same parameters as phrcw , but with a cell size of 80 pc .
we find that gas density does increase , but the highest density gas never forms , and there is very little gas in the middle density range . because the lack of cloud formation results in a smoother gas distribution in the galaxy disk , the icm wind does not have the same effect on this lower resolution disk as it does in the higher resolution cases .
figure [ fig : resapp ] shows that although the low resolution disk with cooling still gets stripped slightly more quickly than the no cooling case , it does not get stripped as quickly as the higher resolution case . this occurs because the corresponding low density regions in the disk do not form , so the icm wind does not punch holes throughout the disk .
the low resolution run also has much less fallback than phncw , but because very dense clouds are not formed ( see figure [ fig : resapp_rho ] ) , continuous stripping occurs and gas loss does not cease after a few hundred myr .
we also ran a pair of comparison runs between a galaxy with a cell size of 40 pc and one with a cell size of 20 pc . because of the computing time needed to run the high resolution simulation , we only had output for 160 myr of evolution .
however , within this time , the maximum surface density of the higher resolution run is more than double that of the regular resolution run .
further , after 160 myr the amount of gas with densities above @xmath48 g @xmath17 is already more than 25% of the total galaxy gas mass and growing at a rate of 40 m@xmath56 yr@xmath96 , while gas with those densities has not yet formed in the regular resolution run ( at 160 myr , @xmath97 @xmath20 10@xmath98 g @xmath17 ) . in fact , in the highest resolution run , @xmath64 m@xmath11 of gas with densities above @xmath99 g @xmath17 ( @xmath100 @xmath17 ) has already formed by 160 myr into the simulation , with no sign of slowing down much higher gas density formation .
a visual inspection shows that the cloud radius is set by the minimum cell size in the simulation .
we conclude that there are physical processes that we do not include in the simulation , such as star formation , magnetic fields , cosmic rays , and feedback from supernovae , that would all result in lower density gas than is found in our simulations , either by adding energy to the ism or locking the ism into place and hindering collapse . in this case ,
higher resolution is not necessarily better , because our omissions may result in unphysical gas densities . in the regular resolution run ,
the larger cells balance , to a certain extent , the smaller - scale physics we omit . for discussion .
] as mentioned in [ sec - sim ]
, we include radiative cooling using the sarazin & white ( 1987 ) cooling curve .
we allow cooling to 8,000 k , as this results in gas with neutral hydrogen temperatures without overcooling a large fraction of our gas . with this cutoff , we find that we form clouds with densities and sizes typical of molecular clouds , although we make no attempt to reproduce the internal structure of the clouds . for a more detailed discussion of the density distribution in the ism for similar simulations , but without a wind , see tasker & bryan ( 2006 , 2008 ) . to examine the impact of our adopted cooling floor
, we perform a run with no wind and cooling to 300 k which shows that after 800 myr the density distribution of disk gas was nearly identical to the case with cooling to 8000 k , although the case with cooling to 300 k reached the distribution more quickly and leveled out .
the highest density gas is less than a factor of three more dense in the 300 k case than in the 8000 k case . to better compare how an icm wind would affect a galaxy with a lower cooling floor , we ran a simulation which is identical to phrcw , but with radiative cooling to 300 k. we find a few interesting results , shown in figure [ fig : coolcomp ] . first
, more gas is lost than in either phrcw or phncw .
gas is initially lost more slowly than in the other two cases , but does not slow down with fallback as in phncw .
also , the final radius of the gas disk is similar to phrcw , but it reaches this radius more slowly . finally , the amount of gas in each of the three density ranges is similar to that in phrcw , but the two lower density regimes have a little less gas .
the highest density gas is the same in both runs throughout the simulations .
we also show ( figure [ fig : cool_image ] ) an image of each of the two cooling cases 150 myr after the wind has hit the galaxy .
all of the differences in these two cases are explained by the fact that the galaxy with the lower radiative cooling floor fragments more than our standard cooled galaxy . because more of the gas is in dense knots , the initial onslaught of the wind does not strip gas as quickly .
further , because gas that began in the outer regions of the disk survives , we see epicyclic variation in the radius , similar to our other simulations , resulting in an increase in the radius .
this is unlike phrcw , in which all the gas that began in the outskirts is quickly stripped . because the clouds in the 300 k cooling case move to large radii
, there is more time for them to be ablated as they drift back towards the center of the galaxy .
this explains why gas takes longer to be stripped , why more gas is stripped , and why the amount of mid - density gas decreases in this galaxy . in the inner region of the disk ,
the larger amount of fragmentation means that more gas is stripped at early times from the central regions of the disk , because the gas disk is not smooth and can not deflect the icm wind .
in fact , we do not see gas fallback , as in the other high ram pressure cases , presumably because the wind pushes gas at all radii . the longer period of gas stripping results both from ablation of the dense clouds in the center of the disk that are more exposed to the wind because of the larger amount of fragmentation , and because the dense clouds at large radii can be ablated more easily than their counterparts in the 8000 k case , which are closer to the galactic center .
it is unclear what level of fragmentation is more realistic in observed galaxies .
because of the lower jeans length in the lower temperature floor runs , we resolve the resulting fragmentation less well and so prefer the runs with a higher minimum temperature .
in addition , it is not clear if the lower minimum temperature runs are physically more realistic as we do not include some effects such as small - scale turbulence , cosmic rays and magnetic fields , all of which may provide a source of effective pressure in low temperature regions . while our simulations represent a sizable step forward in realism , it is clear that more work is required .
we have run a set of detailed galaxy simulations including radiative cooling to understand how a multiphase ism interacts with an oncoming icm wind .
we find : \1 .
the total amount of gas mass loss is similar when comparing galaxies with and without radiative cooling .
however , the time over which the gas is stripped is shorter in the galaxies with radiative cooling .
the continued stripping of galaxies without radiative cooling could be kelvin - helmholtz stripping acting on the larger disk .
the stripping radius is significantly smaller for the galaxy simulations with radiative cooling .
this is even true when accounting for the angular momentum loss in the ism gas due to the interaction between the ism and icm gas ( see section [ sec - nowind ] ) .
the morphology of the stripped disks are considerably different , with the cooling case producing a multiphase medium which results in many holes in the disk , while the case without cooling produces a coherent disk .
these holes allow icm gas to stream through the galaxy disk , affecting the evolution of the dense clouds which remain .
the amount of high density gas is affected differently depending on the amount of ram pressure .
when the ram pressure is strong , there is less gas at all densities , including the highest density .
this is because the low density gas is stripped , even from the inner regions of the galaxy , so there is less gas to feed molecular clouds . with a weaker ram pressure
, there is more of the highest density gas , although there is also less of the lower density gas . a weaker wind strips less of the lower density gas , and increases the pressure on disk gas to create more high density gas .
this leads us to a picture in which radiative cooling results in both overdensities and underdensities at all radii in the galactic gas disk .
the underdensities are clearly stripped quickly throughout the disk , and the overdense regions that then have the wind streaming around them through the holes in the disk are stripped more rapidly than without a multiphase medium .
however , if a dense cloud is able to survive this initial onslaught , it loses angular momentum through its interaction with the non - rotating icm wind and therefore settles toward a lower radius .
this is how the runs with cooling and a wind ( the rcw cases ) are stripped quickly and to a small radius while not losing much more gas than the galaxies without radiative cooling .
also , the galaxies with radiative cooling seem to stop losing gas while those without cooling continue to lose gas through the entire simulation .
this may be because the kelvin - helmholtz instability is inefficient on small dense clouds but can affect the large cohesive disk in the no - cooling runs . allowing cooling to a lower temperature supports our picture that dense clouds can not be directly stripped but can be ablated over time . although these simulations are arguably the most realistic stripping simulations performed to date , we have discussed in detail the limitations involved in our work in [ sec - discussion ] .
we acknowledge support from nsf grants ast-05 - 07161 , ast-05 - 47823 , and ast-06 - 06959 , as well as computational resources from the national center for supercomputing applications .
we thank jacqueline van gorkom , jeff kenney , hugh crowl , airee chung , tomer tal , and anne abramson for useful discussions , as well as elizabeth tasker for invaluable help setting up the initial conditions .
we would also like to thank our referee for editing and advice that greatly improved our paper .
barnes , j. e. , & hernquist , l. e. 1991 , , 370 , l65 bekki , k. 1999 , , 510 , l15 bekki , k. 1998 , , 502 , l133 bigiel , f. , leroy , a. , walter , f. , brinks , e. , de blok , w. j. g. , madore , b. , z & thornley , m. d. 2008 , , 136 , 2846 blitz , l. , & shu , f. h. 1980 , , 238 , 148 bryan , g.l .
phys . and eng . 1999 , 1:2 , p. burkert , a. 1995 , , 447 , l25 butcher , h. , & oemler , a. , jr .
1978 , , 219 , 18 byrd , g. , & valtonen , m. 1990 , , 350 , 89 clemens , m. , alexander , p. , & green , d. 2001 , , 276 , 451 clemens , m. s. , alexander , p. , & green , d. a. 2000 , , 312 , 236 crowl , h. h. , kenney , j. d. p. , van gorkom , j. h. , & vollmer , b. 2005 , , 130 , 65 crutcher , r. m. 1991 , fragmentation of molecular clouds and star formation , 147 , 61 dressler , a. 1980 , , 236 , 351 fujita , y. , & nagashima , m. 1999 , , 516 , 619 giovanelli , r. , & haynes , m. p. 1985 , , 292 , 404 gunn , j. e. , & gott , j. r. i. 1972 , , 176 , 1 joung , m. k. r. , & mac low , m .-
2006 , , 653 , 1266 hartmann , l. 2003 , , 585 , 398 hashimoto , y. , oemler , a. j. , lin , h. , & tucker , d. l. 1998 , , 499 , 589 hernquist , l. 1993 , , 86 , 389 hubble , e. , & humason , m. l. 1931 , , 74 , 43 kenney , j. d. , & young , j. s. 1986 , , 301 , l13 kenney , j. d. p. , & young , j. s. 1989 , , 344 , 171 kenney , j. d. p. , & koopmann , r. a. 1999 , , 117 , 181 kenney , j. d. p. , van gorkom , j. h. , & vollmer , b. 2004 , , 127 , 3361 kronberger , t. , kapferer , w. , ferrari , c. , unterguggenberger , s. , & schindler , s. 2008 , , 481 , 337 larson , r. b. 2003 , reports of progress in physics , 66 , 1651 leroy , a. k. , walter , f. , brinks , e. , bigiel , f. , de blok , w. j. g. , madore , b. , & thornley , m. d. 2008 , arxiv:0810.2556 men , h. , ferrire , k. , & han , j. l. 2008 , , 486 , 819 miyamoto , m. , & nagai , r. 1975 , , 27 , 533 moore , b. , katz , n. , lake , g. , dressler , a. , & oemler , a. 1996 , , 379 , 613 mori , m. , & burkert , a. 2000 , , 538 , 559 nakanishi , h. , et al .
2006 , , 651 , 804 norman , m. l. & bryan , g. l. 1999 , assl vol . 240 : numerical astrophysics , 19 nulsen , p. e. j. 1982 , , 198 , 1007 oshea , b. w. , bryan , g. , bordner , j. , norman , m. l. , abel , t. , harkness , r. , & kritsuk , a. 2004 , arxiv astrophysics e - prints , astro - ph/0403044 oemler , a. j. 1974 , , 194 , 1 poggianti , b. m. , bridges , t. j. , komiyama , y. , yagi , m. , carter , d. , mobasher , b. , okamura , s. , & kashikawa , n. 2004 , , 601 , 197 quilis , v. , moore , b. , & bower , r. 2000 , science , 288 , 1617 roediger , e. , & hensler , g. 2005 , , 433 , 875 roediger , e. , & brggen , m. 2007 , , 380 , 1399 roediger , e. , & brggen , m. 2006 , , 369 , 567 sarazin , c. & white , 1987 , apj , 320 , 32 schulz , s. , & struck , c. 2001 , , 328 , 185 tasker , e. j. , & bryan , g. l. 2006 , , 641 , 878 tasker , e. j. , & bryan , g. l. 2008 , , 673 , 810 tonnesen , s. , bryan , g. l. , & van gorkom , j. h. 2007 , , 671 , 1434 trachternach , c. , de blok , w. j. g. , walter , f. , brinks , e. , & kennicutt , r. c. 2008 , , 136 , 2720 tran , k .- v .
h. , van dokkum , p. , illingworth , g. d. , kelson , d. , gonzalez , a. , & franx , m. 2005 , , 619 , 134 van gorkom , j. h. 2004 , clusters of galaxies : probes of cosmological structure and galaxy evolution , 305 vollmer , b. , braine , j. , balkowski , c. , cayatte , v. , & duschl , w. j. 2001 , , 374 , 824 vollmer , b. , huchtmeier , w. , & van driel , w. 2005 , , 439 , 921 vollmer , b. , braine , j. , combes , f. , & sofue , y. 2005 , , 441 , 473
this galaxy begins with a gas radius of 21 kpc .
the surrounding icm has a density and temperature listed in table [ tbl - icm ] .
after about 67 myr , an icm wind with a mach number of 3.5 and the density and temperature listed in table [ tbl - icm ] hits the galaxy with a ram pressure of @xmath36 dynes @xmath17 . for about 50 myr
very little gas is stripped from the galaxy . at that point , @xmath101 m@xmath56
is stripped from the disk in just over 50 myr , although it remains gravitationally bound in the halo for another 150 myr . during this time , some of the gas that was stripped to more than 5 kpc above the disk falls back onto the galactic disk .
once the gas that was initially stripped from the galaxy through ram pressure pushing has left the halo , gas that leaves the disk is unbound . the total amount of gas mass loss in 900 myr is @xmath61 m@xmath56 .
although there are strong fluctuations , the radius of remaining gas settles to about 11 kpc .
although all the gas in this galaxy is low density , by making finer density cuts we find that the gas with the highest densities , in this case between @xmath102 and @xmath103 g @xmath17 , stays nearly constant , with 80% of the stripped gas coming from lower density gas .
this galaxy begins with a gas radius of 25 kpc ( as do all of the galaxies in simulations that include radiative cooling ) . throughout this run
the surrounding icm has the density and temperature of the post - shock icm when there is a wind in the high ram pressure cases .
we chose the hotter icm because of it s longer cooling time .
initially the radius seems to fluctuate with the epicyclic period ( @xmath104 myr ) , however , cooling creates both over densities and underdensities in the disk .
this allows some of the non - rotating icm to fall into holes in the disk and to slow the disk s rotation , leading to a drop in the disk radius through the loss of angular momentum .
this gas infall is also seen by considering the total amount of gas in the disk region : the galaxy gains @xmath55 m@xmath56 over the 950 myr run .
high density gas forms about 100 myr into the galaxy s evolution , with the highest density gas , above @xmath48 g @xmath17 , forming closer to 150 myr . throughout the run
the galaxy gains gas with densities above @xmath86 g @xmath17 , loses less dense gas as it turns into more dense gas .
although we would need to extend this run to be sure the radius would stay constant , the radius settles between 11 and 12 kpc in the final 200 myr of the run . as mentioned above , because this galaxy includes radiative cooling , it begins with a radius of 25 kpc .
the surrounding density and temperature are the same as run phncw , shown in table [ tbl - icm ] .
after about 190 myr , the same icm wind hits this galaxy as hits the galaxy in phncw .
very little gas is lost for a few tens of myrs , at which point @xmath105 m@xmath56 is lost from the galaxy in under 150 myr .
@xmath106 m@xmath56 is immediately gravitationally unbound from the galaxy , and almost all of the remaining @xmath101 m@xmath56 leaves the halo within 400 myr . a small amount of this gas falls back onto the galactic plane .
after the gas immediately removed from the galactic disk by ram pressure pushing is no longer gravitationally bound , about 550 myr after the wind first hits the galaxy , gas loss precedes at about 1 m@xmath56 yr@xmath96 for the final 250 myr we continue the run .
the total gas mass loss is about @xmath107 m@xmath56 . as in the phrcnw run ,
the radius of the disk falls before the icm wind hits the galaxy .
once the wind hits the galaxy the gas is stripped to a radius below 9 kpc within 250 myr .
the radius after 800 myr of ram pressure stripping is at or below 7 kpc . as in phrcnw
, the galaxy loses lower density gas ( below @xmath86 g @xmath17 ) throughout the run .
although the galaxy does increase all it s higher density gas , gas with densities between @xmath48 and @xmath86 g @xmath17 remains constant after the icm wind has been hitting the galaxy for 350 myr .
also , the highest density gas does not appear until 350 myr into the simulation , about 150 myr after the icm wind hits the galaxy .
this galaxy begins with a radius of 21 kpc .
the surrounding icm density and temperature is listed in table [ tbl - icm ] .
after the galaxy has evolved for about 80 myr , an icm wind hits the galaxy with a ram pressure of @xmath23 dynes @xmath22 , and density and temperature listed in table [ tbl - icm ] . the total amount of gas loss in the more than 900 myr in which the icm wind hits the galaxy is less than @xmath66 m@xmath56 . the mass loss rate for the first 100 myr is less than 0.5 m@xmath56 yr@xmath96 , and increases until in the final 100 myr it is a little over 2 m@xmath56 yr@xmath96 .
the radius fluctuates , but seems to settle at about 19 kpc .
in this case , all the gas that is stripped is less dense than @xmath102 g @xmath17 .
this galaxy begins with a radius of 21 kpc .
this surrounding icm density and temperature are the middle post - shock values listed in table [ tbl - icm ] .
again the radius seems to vary due to a combination of epicyclic variation and mixing from non - rotating icm gas .
less icm gas is accreted in this case than in phrcnw , @xmath57 m@xmath56 , which can be expected since the icm pressure is lower . as in the higher icm pressure case , phrcnw , throughout the run the galaxy gains gas with densities above @xmath86 g @xmath17 , and loses gas less dense than @xmath86 g @xmath17 . however , there is less gas with densities greater than @xmath48 g @xmath17 in this run than in phrcnw , and it does not form until about 250 myr .
the radius remains at about 15.5 kpc for the final 400 myr of this run .
this galaxy begins with a radius of 25 kpc .
the surrounding icm density and temperature values are the same as in pmncw .
after the galaxy has evolved for about 275 myr , the icm wind hits the galaxy with a ram pressure of @xmath23 dynes @xmath22 . for 100 myr little gas
is lost , and then in less than 350 myr after the wind first hits the galaxy the majority of the gas is stripped , @xmath108 m@xmath56 of a total loss of @xmath65 m@xmath56 .
the radius falls to 21 kpc before the icm wind hits the galaxy , and then begins to fall again about 230 myr after the wind first hits the galaxy .
the final radius , just below 11 kpc , is reached in slightly less than 900 myr after the wind hits the galaxy and is constant for the final 200 myr of the simulation .
the galaxy loses lower density gas ( below @xmath86 g @xmath17 ) throughout the run . although the galaxy does increase it s higher density gas , gas with densities between
@xmath48 and @xmath86 g @xmath17 remains constant after the icm wind has been hitting the galaxy for 700 myr .
the highest density gas , above @xmath48 g @xmath17 , is first formed 450 myr into the simulation , or 175 myr after the higher pressure icm hits the galaxy .
this galaxy begins with a radius of 21 kpc .
the surrounding icm density and temperature are listed in the bottom pre - shock rows of table [ tbl - icm ] .
after the galaxy has evolved for about 77 myr , the icm wind hits the galaxy with the post - shock values in table [ tbl - icm ] and a ram pressure of @xmath26 dynes @xmath22 .
the total gas loss is @xmath109 m@xmath56 .
the gas loss is spread throughout the entire time the wind hits the galaxy , with the fastest gas loss rate less than 2 m@xmath56 yr@xmath96 .
the gas loss rate does not constantly ascend or descend . the radius of the disk oscillates with a period of about 370 myr , the epicyclic period of a disk with a radius of 18 kpc , the gunn & gott stripping radius for this case .
the oscillations are too large , however , to make any claims about a stripping radius . once again , in this case , all the gas that is stripped is less dense than @xmath102 g @xmath17 .
this galaxy begins with a radius of 25 kpc .
the surrounding icm density and temperature are listed in the bottom post - shock rows of table [ tbl - icm ] . in this case radial variation
seems to be mostly due to epicycles , with mixing perhaps having the effect of shifting the oscillations to lower radii .
this galaxy gains less than @xmath110 m@xmath56 in the entire 1.2 myr of the run . as in both of the other rcnw cases , throughout the run
the galaxy gains gas with densities above @xmath86 g @xmath17 , and loses gas less dense than @xmath103 g @xmath17
. however , there is less gas with densities greater than @xmath48 g @xmath17 in this run than in the runs with higher icm pressure , although it does form at about 300 myr , nearly the same time as in pmrcnw .
this galaxy begins with a radius of 25 kpc .
the surrounding icm density and temperature are the same pre - shock values as in plncw .
after the galaxy has evolved for 271 myr , the icm wind hits the galaxy with a ram pressure of @xmath26 dynes @xmath22 . as in pmrcw , there is little change in gas mass for the first 100 myr after the wind hits the galaxy , and then the galaxy loses @xmath111 m@xmath56 , of total gas mass loss of @xmath69 m@xmath56 , in less than 350 myr . even with this small gas mass loss rate from the disk of less than 2 m@xmath56 yr@xmath96 , some small ( less than @xmath112 m@xmath56 ) amount of the stripped gas remains in the halo until 600 myr after the wind has hit the galaxy .
after the initial gas loss , the removal rate is small and continues to approach zero . again
the radius of the galaxy decreases in a smooth curve and is hit by the wind before it would rise in an epicyclic pattern .
once the galaxy is stripped , the radius decreases to between 15 and 14 kpc , and seems to hold steady there for the last 600 myr of the simulation .
the galaxy loses lower density gas ( below @xmath86 g @xmath17 ) throughout the run .
the galaxy increases it s higher density gas ( denser than @xmath86 g @xmath17 ) throughout the run .
the highest density gas , above @xmath48 g @xmath17 , is first formed 500 myr into the simulation , or 225 myr after the higher pressure icm hits the galaxy . | cluster galaxies moving through the intracluster medium ( icm ) are expected to lose some of their interstellar medium ( ism ) through ism - icm interactions .
we perform high resolution ( 40 pc ) three - dimensional hydrodynamical simulations of a galaxy undergoing ram pressure stripping including radiative cooling in order to investigate stripping of a multiphase medium .
the clumpy , multiphase ism is self - consistently produced by the inclusion of radiative cooling , and spans six orders of magnitude in gas density .
we find no large variations in the amount of gas lost whether or not cooling is involved , although the gas in the multiphase galaxy is stripped more quickly and to a smaller radius .
we also see significant differences in the morphology of the stripped disks .
this occurs because the multiphase medium naturally includes high density clouds set inside regions of lower density .
we find that the lower density gas is stripped quickly from any radius of the galaxy , and the higher density gas can then be ablated .
if high density clouds survive , through interaction with the icm they lose enough angular momentum to drift towards the center of the galaxy where they are no longer stripped .
finally , we find that low ram pressure values compress gas into high density clouds that could lead to enhanced star formation , while high ram pressure leads to a smaller amount of high - density gas . |
the spectrum of cosmic rays extends more than five orders of magnitude beyond the highest energy at which it has been possible so far to observe the primaries directly with experiments on balloons or spacecraft . in this
energy range the spectrum has at least two features , the knee between @xmath0 and @xmath1 ev and the ankle between @xmath2 and @xmath3 ev .
the knee may be associated with a transition between different classes of galactic cosmic rays ( or with a feature of propagation of galactic cosmic rays ) and the ankle with a transition from galactic to extra - galactic origin of the particles but these are only plausible conjectures . in order to understand the implications of these features for the origin of the high - energy particles it is necessary to measure the relative contribution of the different groups of nuclei .
this must be done with large ground - based air - shower experiments in order to achieve sufficient exposure to collect a large sample of high - energy events . because of the indirect nature of air shower experiments and the problems of fluctuations superimposed on a steeply falling energy spectrum , progress toward the goal of measuring the primary composition at high energy has been slow and difficult .
new experiments with better resolution , coupled with a better understanding of the hadronic interaction models used to interpret the data , show promise for improving the situation . the current status is summarized very nicely in the review of kalmykov and khristiansen @xcite . concerning the knee region ,
they conclude that measurements of fluctuations in the muon to electron ratio suggest a gradual change toward a larger fraction of heavy nuclei in the energy range from @xmath0 to @xmath4 ev .
they consider two quite different models which share this feature .
one is a diffusion model @xcite in which a single underlying composition and source spectrum are modified by propagation in the galactic magnetic field .
the other is a model @xcite in which there are two different classes of sources .
the diffusion model is a modern realization of the classic suggestion of peters @xcite and zatsepin @xcite .
its characteristic feature is a steepening of the spectrum at a certain value of magnetic rigidity so that there are relatively more heavy particles at high energy when the classification is by total energy per nucleus .
to fit the data the break has to occur at a rigidity of @xmath5 gv .
the two - component model is motivated by the observation @xcite that the proton spectrum as measured by jacee @xcite appears to steepen around 100 tev or somewhat below while there is no steepening of helium or heavier nuclei at the corresponding rigidity .
the models are further stimulated by the fact that 100 tev is the maximum energy expected for shock acceleration by supernova blast waves expanding into the typical interstellar medium @xcite . in this kind of model
higher energy particles come from acceleration by supernovae whose environments are richer in heavy nuclei , so the high - energy component contains fewer protons and relatively more heavy nuclei . the specific version of this model described in ref .
@xcite has less than 10% protons already for @xmath6 tev .
according to kalmykov and khristiansen the muon / electron measurements rule out such a low fraction of protons below the knee , presumably because without the light nuclei one would see smaller fluctuations in the muon / electron ratio than observed .
most likely , the real situation combines features of both diffusion and different types of sources .
in addition , one would expect supernovae in the same general class to have somewhat different environments ( magnetic fields and composition ) and thus to exhibit a gradation of cutoffs and compositions of the cosmic - ray spectra they produce .
nevertheless it is interesting to look at the mass fractions in these specific models just to illustrate how well experiments will need to determine the mass composition in order to distinguish among models .
if the composition is 64% protons and helium and 36% heavier nuclei below the knee ( `` normal composition '' ) and all components have a differential spectral index of -2.7 , then the corresponding fractions above the knee ( e.g. around @xmath4 ev ) would be 44% and 56% if the spectral index changes by -0.3 to -3.0 .
the corresponding numbers given in ref .
@xcite for the hall diffusion model of ptuskin et al .
@xcite are 55/45 below the knee and 28/72 at @xmath4 ev . the light / heavy ratio is similar to this at high energy in the two - source model of ref .
@xcite , but the light component is reduced before the knee as well .
an analysis of the fly s eye data @xcite on mean depth of shower maximum as a function of primary energy leads to the conclusion that the primaries are mostly heavy nuclei at the low end of the energy range of this experiment ( @xmath4 ev ) , with an increasing fraction of protons as energy increases .
it was noted @xcite that this transition corresponds in energy with the `` ankle '' feature of the spectrum .
if it is correct , this interpretation is circumstantial evidence for a transition to extragalactic origin for the highest energy cosmic rays , and this extragalactic component is primarily light nuclei ( protons and helium ) . in their review ,
kalmykov and khristiansen reach a similar conclusion , although the evidence for a change of compositon is less strong than indicated by the figure of ref .
yakutsk data are consistent with those of the fly s eye group @xcite . in fact
various calculations give a wide range of results for depth of maximum in the fly s eye energy range . after discussing some of the input to the models
, i will return to this question in the last section .
comparison to fly s eye data , with typical energies of @xmath2 ev and above , requires extrapolation of hadronic interaction models more than two orders of magnitude in center of mass energy beyond the highest accelerator energies ( @xmath7 tev ) .
in fact the required extrapolation is much greater than this because the showers involve nuclei as well as single hadrons ( both as targets and as projectiles ) and because showers depend essentially on how the energy is divided when particles interact ( fragmentation region ) rather than on multiplicity ( central region ) .
physical models are required to relate the various processes and to extrapolate to high energy .
thus also in the knee region there can be significant model - dependence in air shower simulations . in preparation for this meeting knapp et al .
@xcite have made a very instructive comparison of some results of different event generators , several of which represent theoretical models of hadronic interactions with similar underlying physical assumptions .
the models ( codes ) that i consider here are : 1 .
dual parton model @xcite ( dpmjet @xcite ) .
quark - gluon string model @xcite ( qgsjet @xcite ) .
venus @xcite .
minijet model @xcite ( sibyll @xcite ) .
all of these codes use strings of hadrons coupled to constituent partons of the incident hadrons to represent particle production in high - energy interactions .
the common feature of all these models that i want to draw attention to concerns interaction of a single hadron with a target nucleus and , in particular , the propagation of the projectile hadron and its fragments through the target nucleus . in all the models listed above the valence partons in the projectile only radiate one pair of strings of secondary hadrons . in cases with
more than one wounded nucleon in the target the extra strings are connected with sea - quarks in the projectile .
this means that the inelasticity in hadron - nucleus collisions is not much larger than that in corresponding hadron - nucleon collisions .
this is illustrated by comparison of figs .
14 and 35 in the report of knapp et al .
@xcite .
one can contrast this , for example , with the knp model @xcite , which was favored by the fly s eye group in its comparison @xcite of measurements of depth of maximum between @xmath4 and @xmath3 ev with simulations .
cronin ( private communication ) has emphasized that other codes , such as sibyll and mocca @xcite , give a rather different impression of the interpretation of the measurments than that shown in fig . 2 of ref .
@xcite . in that figure
the fly s eye data are compared to mean depth of maximum vs. energy for proton showers and for iron showers as simulated with the knp model .
this choice was made because , of the several models compared with the same data in an earlier paper @xcite , the knp model gave the most rapid shower development and hence the best agreement with the observed depth of maximum .
one feature of the knp model that led to relatively rapid shower development is the treatment of propagation of the projectile nucleon inside the target nucleus . in this model
the projectile nucleon is assumed to lose energy in the same way each time it interacts with a target nucleon in the nucleus .
this leads to collisions that are significantly more inelastic .
inelasticity is the fraction of energy in a hadron - nucleus interaction _ not _ carried off by the fragment of the projectile hadron . because there is no way to distinguish a `` fragment '' hadron from a `` produced '' hadron of the same type , this is only a theoretical definition ( but still a useful one for comparison of theoretical models ) .
the following definition of inelasticity , @xmath8 , in high - energy proton - nucleus collisions in terms of momentum - weighted integrals of inclusive cross sections is one that could in principle be realized experimentally : @xmath9 here @xmath10 is the cross section for production of a proton or neutron and @xmath11 the cross section for producing an antinucleon .
the definition becomes more accurate as energy increases and the target fragments become relatively less important .
an experimental definition is more difficult for pions , but what counts is the inclusive cross section in the fragmentation region ( e.g. for @xmath12 ) .
hfner and klar @xcite compared a simple model with data on p - nucleus interactions in the 100 gev region @xcite . within the context of a glauber multiple scattering treatment of hadron - nucleus collisions the mean number of wounded nucleons in the target
is given by @xmath13 and the probability , @xmath14 of having exactly @xmath15 wounded nucleons is calculated from the nuclear profile functions @xcite in the standard way @xcite .
hfner and klar write the energy of the projectile fragment nucleon after encounters with exactly @xmath15 wounded nucleons on average as @xmath16 the total inelasticity is then @xmath17 in this relation @xmath18 is the average inelasticity for proton - proton collisions .
within the string models listed above one expects @xmath19 .
comparison of this simple formulation to the data in ref .
@xcite gives @xmath20 for @xmath21 .
the knp model has instead,@xmath22 and hence a larger value of inelasticity .
figure 1 shows a plot of inelasticity as a function of @xmath23 in ev for the two sets of parameters calculated from eq .
[ inelasticity ] .
the solid line corresponds to \{@xmath24 for all @xmath15 } ; the dashed line to \{@xmath25}. in both cases the inelasticity increases with energy .
this is a consequence of the increasing cross section , which , according eq . [ nwound ] , drives an increase in the number of wounded nucleons .
the same cross sections @xcite and hence the same distributions of wounded nucleons have been used for both curves in fig .
the difference is in the assumption for @xmath26 .
the upper curve represents the knp model , and the lower curve is a good approximation to the inelasticity in the full sibyll model .
this model is based on the idea that the increase in cross section is driven by the production of minijets @xcite .
the emphasis is on the fragmentation region and on collisions of hadrons with light nuclei .
depth of shower maximum and tev muons are the principal intended applications . with one significant exception , all the ideas of the dual parton model are incorporated .
the program uses lund techniques @xcite and is tailored for efficient operation up to at least @xmath27 ev .
the exception is that in sibyll the soft part of the eikonal function is taken as a constant , fitted to low energy data .
the non - jet part of every event is represented by exchange of a single pair of strings . in the other string - type models listed above
, non - jet contributions to events can involve exchange of multiple pairs of soft strings . to the extent that the parameters in the two approaches are tuned to reproduce the same increasing cross section ,
the resulting multiplicity should be similar in one case it comes from short pairs of multiple soft strings and in the other from minijets .
since the fragmentation of the minijet pair is treated as a loop of string , the multiplicity should be similar . in principle , the soft part of the eikonal should also be energy - dependent , and multiple exchanges of both hard and soft kinds should occur .
the question of sensitivity of the parameters of the fits to data when both multiple soft and multiple hard exchanges are allowed is an interesting one . for sibyll at a lab energy of 2 tev
we find @xcite @xmath28 the overall inelasticity from sibyll for proton - nitrogen collisions is shown by the inverted triangles in fig .
it is very close to the simple formula [ inelasticity ] with the partial inelasticities from ref .
@xcite .
because the inelasticities tend to converge as energy decreases , it is not clear with what certainty the comparison of ref .
@xcite to the data restricts the behavior of the inelasticity and even whether the knp treatment is completely ruled out .
this question needs futher investigation .
interpretation of measurements of the muon to electron ratio in air showers deep in the atmosphere depend on monte carlo simulations .
differences in existing codes at present may make it difficult to use this feature alone to discriminate among light / heavy ratios of the kind illustrated in 1.1 above ( see for example fig .
71 of ref .
obvious remedies are to understand and resolve differences among the simulation codes and to measure more components of the showers , as is being done by kascade . in
the work referred to in ref .
@xcite , the model - dependence is addressed by considering fluctuations in the muon to electron ratio as well as the mean values . in comparing production of @xmath29 gev muons in cascades simulated with different interaction models it may be useful to tabulate the distribution of interactions of various types in the cascade .
this diagnostic relates the energy - dependence of low energy muons in showers to fundamental features of the interaction models . generally it will be possible to express the distribution of hadronic interactions in a proton - induced air shower by an expression of the form @xcite @xmath30 where @xmath31 , @xmath32 and @xmath33 is the primary energy .
the first two terms on the right side of eq . 1 represent the interactions of nucleons in the cascade , and the approximate expression for @xmath34 is exact for a flat inelasticity distribution and neglecting production of @xmath35 pairs . the 3rd and 4th terms represent interactions of pions and kaons respectively . an approximation for the number of low - energy muons in a shower initiated by a proton when @xmath36 is @xcite @xmath37 ( plus a similar term for kaons ) . here
gev is the critical energy for pions to decay rather than interact in the atmosphere .
the quantity @xmath39 is approximately the rapidity density in the central region for production of charged pions in interactions of pions with nuclei of the atmosphere evaluated for pion interaction energy of @xmath40 . in a superposition approximation ,
the a - dependence of the muon content of a shower will be @xmath41 for @xmath42 .
the accuracy of these approximations needs to be checked with simulations , but the relations should be a useful guide to investigating the model - dependence of the ratio of low energy muons to electrons in air showers . the akeno group have published an analysis of low - energy muons in large showers at agasa @xcite .
they fit the muon dependence on primary energy to a power @xmath43 with @xmath44 . here
@xmath45 is the scintillator density at @xmath46 m from the shower core , which is assumed to be a good measure of the primary energy , and @xmath47 is the corresponding muon density .
a change in composition of the degree suggested by fly s eye would require a value of @xmath48 larger by roughly 10% .
( the analysis involves a comparison of eqs . [ nmu ] and [ nmua ] above . ) because of possible model - dependence of the parameters @xmath49 ( and @xmath50 ) there is a corresponding model - dependence of of the @xmath51 ratio which needs further investigation .
in this section i return to the question of the interpretation of the depth of shower maximum as measured by fly s eye .
figure 2 shows the fly s eye data @xcite and comparisons to various model calculations .
the upper three lines in fig .
2 are for proton - induced showers calculated in three models : sibyll dotted @xcite ; qgs dashed @xcite ; knp highest solid line .
the representation of the knp model @xcite used for the fly s eye calculation is described in refs .
comparison of these three curves gives an indication of the extent of the uncertainty in interpretation of the depth of maximum measurements arising from uncertainties in the models , such as different inelasticities and different cross sections and interaction lengths .
in addition , an important point is that in the fly s eye analysis of bird et al .
@xcite the simulations have been corrected for effects of detector acceptance and other systematics .
the lower pair of solid lines in fig . 2 ( taken from ref .
@xcite ) shows the knp model for protons and for iron primaries _ after _
these corrections have been applied .
there are several compensating effects , but the net shift of about @xmath52 to @xmath53 g/@xmath54 is shown by comparison to the unshifted knp curve for protons . the recent work of the utah group @xcite represents a quite different approach to the calculation of depth of shower maximum .
they have analyzed the data in the context of a chou - yang model @xcite .
this interaction model features a statistical picture of particle production in the central region according to which the inclusive cross section is of the form @xmath55 the parameters @xmath56 ( normalization ) , @xmath57 ( mean transverse momentum ) and @xmath58 ( partition temperature ) are adjusted to fit collider data in the central region from @xmath59 gev to @xmath60 gev and extrapolated to higher energies as described in ref .
the mean inelasticity is the integral of eq .
[ chouyang ] , which is approximately @xmath61 the striking feature of this model is that , in order to fit the collider data , the inelasticity must _ decrease _ as energy increases from @xmath62 at low energy to @xmath63 for @xmath64 interactions in the fly s eye energy range @xcite . to compensate and obtain a reasonably shallow depth of maximum
, they use a naive ( knp ) assumption for the inelasticity as well as a rapidly increasing cross section .
since both effects increase with energy , the mean inelasticity is decreased more at high energy . in other words ,
assumptions which reduce the depth of maximum at one energy also tend to decrease the elongation rate ( logarithmic slope of @xmath65 ) . in view of the model dependence of depth of maximum and the significant systematic corrections involved in comparing to the data , it is important to look at the distribution of depth of maximum as well as its mean value .
this is illustrated in fig .
3 @xcite , which shows the contribution of protons and iron separately to the overall distribution for the group of events with @xmath66 ev .
the calculations were made with the knp model .
it appears that in this energy range comparable numbers of light and heavy nuclei are needed to fit the data .
* acknowledgments*. i am grateful to bob fletcher , george frichter , paolo lipari , todor stanev and lou voyvodic for useful discussions and collaboration on this work , and i thank paolo lipari for helpful comments on the paper .
this research is supported in part by the u.s .
department of energy under grant no .
de - fg02 - 91er40626 .
kalmykov & g.b .
khristiansen , j. phys .
g 21 ( 1995 ) 1279 .
ptuskin et al . , astron .
astrophys . 268
( 1993 ) 726 .
biermann , astron .
astrophys . 271
( 1993 ) 649 .
todor stanev , p.l .
biermann , t.k .
gaisser , astron .
astrophys . 274
( 1993 ) 902 .
biermann , t.k .
gaisser & todor stanev , phys .
d51 ( 1995 ) 3450 .
b. peters , il nuovo cimento ( serie x ) 22 ( 1961 ) 800 .
zatsepin , n.n .
gorunov & l.g .
dedenko , izv .
nauk ussr ser .
( 1962 ) 685 .
simon swordy , proc .
xxiii int . cosmic ray conference ( calgary , ed .
leahy , r.b .
hicks & d. venkatesan , world scientific , 1993 ) p. 243 .
k. asakamori et al .
( jacee collaboration ) proc .
xxiii int .
cosmic ray conf .
( calgary , 1993 ) pp .
21 and 25 .
lagage & c.j .
cesarsky , astron . astrophys . 118 ( 1983 ) 223 and 125 ( 1983 ) 249 .
bird et al .
rev . letters 71 ( 1993 ) 3401 .
dyakonov et al . ,
. cosmic ray conf .
( calgary ) 4 ( 1993 ) 303 .
j. knapp , d. heck , s.s .
ostapchenko & g. schatz , `` comparison of hadronic interaction models used in air shower simulations and of their influence on shower development and observables '' ( draft , july 31 , 1996 ) .
see also papers by j. knapp and d. heck in these proceedings .
a. capella , u. sukhatme , c .- i .
tan & j. tran thanh van , physics reports 236 ( 1994 ) 225 .
j. ranft , phys .
d51 ( 1995 ) 64 and g. battistoni , c. forti & j. ranft , astroparticle physics 3 ( 1995 ) 157 .
kaidalov , k.a .
ter - martirosyan & yu .
m. shabelsky , yad .
43 ( 1986 ) 1282 .
kalmykov & s.s .
ostapchenko , phys . at .
56 ( 1993 ) 346
. k. werner , physics reports 232 ( 1993 ) 87 .
l. durand and h. pi , phys .
58 ( 1987 ) 303 ; phys .
d38 ( 1988 ) 78 .
gaisser & todor stanev , phys .
letters 219 ( 1989 ) 375 .
fletcher et al .
d50 ( 1994 ) 5710 .
kopeliovich , n.n .
nikolaev & i.k .
potashnikova , phys .
d39 ( 1989 ) 769 .
a.m. hillas , proc .
. cosmic ray conf .
( rome ) vol . 1 ( 1995 ) 270 .
gaisser et al . , phys .
d47 ( 1993 ) 1919 .
j. hfner & a. klar , phys .
letters 145b ( 1984 ) 167 . d.s .
barton et al . , phys .
d27 ( 1983 ) 2580 . c. de marzo et al .
d26 ( 1982 ) 1019 . r.j .
glauber & g. matthiae , nucl .
b21 ( 1970 ) 135 .
d. cline , f. halzen and j. luthe , phys .
lettters 31 ( 1973 ) 491 .
gaisser and f. halzen , phys .
letters 54 ( 1987 ) 1754 .
g. pancheri and y. srivastava , phys .
b 159 ( 1985 ) 69 .
h. bengtsson and t. sjstrand , comput .
46 ( 1987 ) 43 .
george frichter et al . ,
( in preparation ) .
gaisser _ cosmic rays and particle physics _ ( cambridge university press , 1990 .
n. hayashida et al . , j. phys .
g 21 ( 1995 ) 1101 .
see also the talk of m. nagano at this conference .
serap tilav , ph.d .
thesis , university of delaware ( 1991 ) .
dai et al . , ap.j . (
1996 ) to be published .
chou and c.n .
yang , phys .
d32 ( 1985 ) 1692 .
gaisser et al .
, comments on astrophysics 17 ( 1993 ) 103 . | several models of minimum - bias hadronic interactions at ultra - high energy that have been used for calculations of air showers share essential common features . in this talk
i review these common elements and discuss some consequences .
i concentrate on properties of hadron - nucleus interactions , and i use mean depth of shower maximum as a function of primary energy to illustrate my main points .
i will contrast these models with models that use a more naive treatment of hadronic interactions in nuclei but which have been successfully used to interpret measurements of depth of shower maximum .
= 16.5 cm = -.5 cm = 0.1 cm = -1 cm |
the achievement of producing bose einstein condenstates with ultracold dilute gases of atoms has seen a wealth of theoretical and experimental activity . one enticing prospect of bose - einstein condensates is that they may allow for a better understanding of the interface between classical and quantum mechnics , through the possibility of macroscopic schrdinger cat states @xcite and macroscopic quantum tunneling @xcite .
another intriguing field of study is the chemistry of bose - einstein condensates , where the atomic constituents may form molecules through feshbach resonances @xcite or photoassociation @xcite . a novel feature of a molecular bose
einstein condensate is that the atomic and molecular states can exist as a superposition @xcite , providing a chemical analogue of a schrdinger cat state . in cases where the molecules are heteronuclear
, the presence of a permanent electric dipole moment also opens the possibility for manipulating the condensate through electrostatic forces @xcite .
since systems of bose einstein condensates exist at ultracold temperatures , it is to be expected that significant insights into their behaviour can be obtained from studying their ground - state properties . from a general theoretical perspective
there has been substantial progress in the understanding of quantum ( i.e. ground - state ) phases in many - body quantum systems , due largely to a cross fertilisation of ideas between the condensed matter theory and the quantum information theory communities .
much of this study has explored the relationship between entanglement and quantum criticality @xcite . however
other characterisations of quantum criticality have been sought too @xcite . recently
the notion of wavefunction overlaps ( also known as the _ fidelity _ ) , which is again common in quantum information theory , has been applied to the study of quantum phase transitions @xcite .
an advantage of this approach is its universality , as it can be applied to any system independent of the choice of decomposition into subsystems . with the above points in mind here we analyse
a simple , yet non - trivial , three - mode model describing a heteronuclear molecular condensate .
two modes are associated with two distinguishable atomic constituents , which can combine to form a molecule represented by the third mode .
besides the interaction describing the interconversion of atoms and molecules , the hamiltonian contains terms which are linear in the mode number operators ( corresponding to external fields ) and terms which are second - order in the mode number operators ( corresponding to scattering interactions between atoms and molecules ) .
we mainly concern ourselves with the ground - state properties of the model , with the aim of identifying the ground - state phases .
we avoid taking the thermodynamic limit and restrict our analysis to finite systems , for reasons which will be discussed later .
this in turn presents challenges in rigourously identifying quantum phases , since for finite systems there are no singularities in physical quantities such as the ground state energy and its derivatives .
however several recent works have addressed the issues of quantum phases in finite systems @xcite .
we mention that traditional techniques of renormalisation group methods are not applicable to the model under consideration , due to the low number of degrees of freedom .
neither is the concept of symmetry breaking , as the model does not admit global symmetries , nor long - range order , as the model is in essence zero - dimensional .
we start our analysis with a semi - classical treatment , following the approach of @xcite .
since the model with which we are dealing is integrable , the semi - classical many - body system can be reduced to a problem with a single degree of freedom .
we study the phase space of this system , in particular determining the fixed points . it is found that for certain coupling parameters bifurcations of the fixed points occur , and we can determine a parameter space diagram which classifies the fixed points .
an unexpected result is that the boundaries between the regions in parameter space are extremely sensitive on whether the number of constituent atoms is equal or not .
specifically , when the number of constituent atoms is equal ( i.e. the _ atomic imbalance _ is zero ) there is a spontaneous appearance of additional boundaries in the parameter space , some of which can be identified with bifurcations of the global minimum of the classical hamiltonian .
we next investigate the extent to which the classical behaviour influences the ground - state properties of the quantum system .
our first goal in the full quantum analysis is to derive an exact bethe ansatz solution for the model .
we use the bethe ansatz solution to map the spectrum of the hamiltonian into that of a one - body schrdinger equation in one - dimensional .
an advantage of this method is that it allows for an analysis of the finite system , following the ideas of @xcite , as the mapping to the one - body schrdinger equation is not dependent on taking the thermodynamic limit of the original many - body system .
the results of the analysis of the associated schrdinger equation are in general agreement with the results obtained from the semi - classical treatment , supporting the picture of an additional phase boundary when the atomic imbalance is zero . however , due to quantum fluctuations , the emergence of the phase boundary is smooth rather than spontaneous .
this property is apparent from a study of ground - state expectation values and quantum dynamics . in order to simply characterise the ground - state phases for the finite system ,
we finally define the notion of a _ quantum phase pre - transition _ in terms of wavefunction overlaps .
specifically , a quantum phase pre - transition is identified with each coupling for which the incremental ground - state wavefunction overlap is a local minimum .
we numerically calculate these for several cases and discuss these results in relation to the semi - classical and quantum analyses which have been described above .
the results confirm the emergence of a quantum phase boundary in the limit of zero atomic imbalance .
we consider a general three - mode hamiltonian describing a heteronuclear molecular bose
einstein condensate with two distinct species of atoms , labelled by @xmath0 and @xmath1 , which can combine to produce a molecule labelled by @xmath2 .
we introduce canonical creation and annihilation operators @xmath3 satisfying the usual commutation relations @xmath4=i$ ] etc . , which represent the three degrees of freedom in the model .
the hamiltonian reads @xcite @xmath5 the parameters @xmath6 describe s - wave scattering , @xmath7 are external potentials and @xmath8 is the amplitude for interconversion of atoms and molecules .
we remark that in the limit @xmath9 , equation ( [ ham ] ) is the hamiltonian studied in @xcite in the context of quantum optics . in the latter stages of the manuscript
we will study this limiting case in some detail .
the hamiltonian acts on the fock space spanned by the ( unnormalised ) vectors @xmath10 where @xmath11 is the fock vacuum .
we then have @xmath12 etc .
, where @xmath13 , @xmath14 and @xmath15 .
the hamiltonian commutes with @xmath16 and the total atom number @xmath17 . we refer to @xmath18 as the atomic imbalance and introduce @xmath19 $ ] as the fractional atomic imbalance .
as there are three degrees of freedom and three conserved operators , the system is integrable .
this fact will allow us to analyse the model in some depth .
below we begin with a semi - classical analogue of the model , and determine the fixed points of the system .
we now regard ( [ ham2 ] ) as a classical hamiltonian and investigate the fixed points of the system .
the first step is to derive hamilton s equations of motion yielding @xmath33 the fixed points of the system are determined by the condition @xmath34 due to periodicity of the solutions , below we restrict to @xmath35 .
it is necessary to treat the cases of @xmath36 and @xmath37 separately , and without loss of generality we assume @xmath38 .
define the functions @xmath39 note that the domain of @xmath40 is @xmath41 $ ] , and @xmath40 is divergent at @xmath42 and @xmath43 . for @xmath36 , we then have the following classification of solutions for ( [ fixed ] ) : * @xmath44 , and @xmath26 is a solution of @xmath45 which can admit one , two or three solutions . *
@xmath46 , and @xmath26 is a solution of @xmath47 which can admit one , two or three solutions . + a graphical representation of possible types of solutions for @xmath48 is given in fig .
[ curve2 ] . from the equations ( [ sol2 ] , [ sol1 ] ) we can determine there
are fixed point bifurcations for certain choices of the coupling parameters .
these bifurcations allow us to divide the coupling parameter space into different regions . to construct this diagram
, we observe that bifurcations occur when @xmath49 is the tangent line to @xmath50 ; i.e. for values of @xmath51 such that @xmath52 for some @xmath53 .
this requirement determines the boundaries in parameter space , which are depicted in fig .
[ curvekdiff ] next we consider the case @xmath37 for which the function @xmath40 has substantially different properties .
setting @xmath54 into ( [ g ] ) , we find that @xmath40 reduces to @xmath55 here we observe that @xmath40 is divergent at @xmath43 , but finite at @xmath56 .
this property affects the types of solutions for ( [ fixed ] ) . specifically , we now have the following classifications of solutions for @xmath37 : * @xmath44 , and @xmath26 is a solution of @xmath57 which can admit zero , one or two solutions .
* @xmath46 , and @xmath26 is a solution of @xmath58 which can admit zero , one or two solutions . *
@xmath56 and @xmath27 is a solution of @xmath59 which can admit zero , one or two solutions . a graphical representation of possible types of solutions for @xmath48 is given in fig .
[ curve2 ] . because @xmath60 is finite , for this case there can be either zero , one or two solutions . as in the @xmath36 case
, we can determine the region boundaries in parameter space from equations ( [ boundarya],[boundaryb ] ) . moreover , because of the existence of solutions of the form given by ( [ ssol3 ] ) for @xmath37 , which do not have an analogue for @xmath36 , we see the appearance of new boundaries given by the conditions @xmath61 for all values of @xmath62 .
the boundaries in parameter space are depicted in fig .
[ parametro1 ] . to help visualise the classical dynamics ,
it is useful to plot the level curves of the hamiltonian ( [ ham2 ] ) .
since the fixed point bifurcations change the topology of the level curves , qualitative differences can be observed between each of the regions .
the results are shown respectively in fig .
[ levelcurve2 ] for @xmath63 and fig .
[ levelcurve1 ] for @xmath37 , where for clarity we show @xmath64 $ ] . [ cols="^,^,^,^ " , ] we now formally define a quantum phase pre - transition in terms of ground - state wavefunction overlaps .
let @xmath65 denote a generic hamiltonian depending on a coupling parameter @xmath66 .
assuming the ground state of the system is non - degenerate , let @xmath67 denote the unique normalised ground state .
for fixed small @xmath68 we define the function @xmath69 by @xmath70 which is symmetric in @xmath68 , bounded between 0 and 1 , and satisfies @xmath71 .
generically , @xmath72 is a decreasing function of @xmath68 .
[ gsoverlaps1 ] shows the behaviour of the wavefunction overlaps for the hamiltonian ( [ hamres ] ) with @xmath73 , and different values of @xmath68 .
it is clear that there is a distinct dip in the quantity @xmath74 near the threshold coupling @xmath75 .
the choice for @xmath68 affects the magnitude of the minimum , which can be made arbitrarily small . in fig .
[ gsoverlaps1 ] when @xmath76 , representing a coupling change of about @xmath77 on either side of the threshold coupling , the ground states are essentially orthogonal .
however the value of @xmath62 at which the minimum occurs is largely independent of @xmath68 . for a given @xmath68
we say that there is a quantum phase pre - transition at @xmath78 if @xmath72 , treated as a single - variable function of @xmath66 , is locally minimal at @xmath78 .
we have computed the wavefunction overlaps for several cases with both zero and non - zero atomic imbalance . fig .
[ gsoverlaps2 ] ( a ) shows the behaviour of @xmath74 with @xmath79 and varying @xmath29 for both @xmath37 and @xmath80 .
it is clear the minimum value of @xmath74 , which determines the quantum phase pre - transition , is at @xmath81 .
the distinction between the predicted threshold coupling and the observed pre - transition coupling is that the pre - transition coupling also occurs for @xmath36 , although for fixed @xmath29 the value of the minimum is lower for @xmath37 compared to @xmath80 . in all instances the value of the mimimum decreases with increasing @xmath29 .
[ gsoverlaps2 ] ( b ) shows similar results for fixed @xmath73 and varying @xmath82 . in this latter case
we see that the occurences of the mimima , determining the pre - transition couplings , fit well with the predicted boundary of threshold couplings given by @xmath83 ( cf .
[ parametro1 ] ) .
however we again observe pre - transitions couplings for @xmath80 which were not predicted in sect .
[ sca ] . in the above cases
the minimum of @xmath74 is substantially more pronounced for zero imbalance compared to non - zero imbalance .
we can see that although quantum phase pre - transitions as defined exist for @xmath36 , the distiguishability of the two phases is more reliable in the limit as @xmath32 goes to zero . in the analyses of sect .
[ sca ] and sect .
[ sem ] qualitative differences are only found precisely when @xmath37 .
we interpret these results as the emergence of quantum phase boundaries at @xmath37 .
we have studied the ground - state phases of a three - mode model describing a heteronuclear molecular bose
einstein condensate , through a variety of techniques . using a semi - classical analysis
we were able to determine threshold couplings associated with fixed point bifurcations in phase space .
we then derived the exact bethe ansatz solution for the system and discussed how the spectrum of the hamiltonian maps into that of an associated one - body schrdinger equation .
it was shown that in the particular subcase of no scattering interactions the threshold coupling for the global minimum in phase space for zero atomic imbalance was consistent with the existence of a bifurcation of the potential of the schrdinger equation .
for the non - zero imbalance case where the semi - classical results do not predict any threshold coupling , it was found that there was no bifurcation of the schrdinger potential .
these results suggested that the ground - state properties of the model are sensitive to whether the atomic imbalance was zero or non - zero .
however due to the finite nature of the system , such a transition is not associated with a discontinuity that is defined in the thermodynamic limit .
we then introduced the notion of a quantum phase pre - transition for finite systems , defined in terms of ground - state wavefunction overlaps .
applying this idea to the model under consideration we argued that a quantum phase boundary emerges in the limit as the atomic imbalance goes to zero . for future work
it would be useful to investigate the ground - state entanglement properties of the hamiltonian .
not only would this be of interest in the study of the behaviour of the entanglement at the quantum phase boundary as @xmath32 approaches zero , but the model is also a simple example of a strictly tripartite system which may offer insights into the role of three - way entanglement in the description of quantum phases .
e.c.m . thanks capes - coordenao de aperfeioamento de pessoal de nvel superior for financial support . a.t . and a.f .
thank fapergs - fundao de amparo pesquisa do estado do rio grande do sul for financial support .
a.f also acknowledges support from pronex under contract cnpq 66.2002/1998 - 99 .
and n.o . are funded by the australian research council through the discovery projects dp0557949 and dp0663773 .
cirac , m. lewenstein , k. molmer , and p. zoller , phys .
a * 57 * , ( 1998 ) 1208 .
m. ueda and a. j. leggett , phys .
* 80 * , ( 1998 ) 1576 .
s. b. papp and c. e. wieman , cond - mat/0607667 .
b. damski , l. santos , e. tiemann , m. lewenstein , s. kotochigova , p. julienne , and p. zoller , phys .
lett . * 90 * , ( 2003 ) 110401 .
e. a. donley , n. r. claussen , s. t. thompson , and c. e. wieman , nature * 417 * , ( 2002 ) 529 .
d. wang , j. qi , m. f. stone , o. nikolayeva , h. wang , b. hattaway , s. d. gensemer , p. l. gould , e. e. eyler , and w. c. stwalley , phys .
93 * , ( 2004 ) 243005 .
t. j. osborne and m. a. nielsen , phys .
a * 66 * , ( 2002 ) 032110 .
a. osterloh , l. amico , g. falci , and r. fazio , nature * 416 * , ( 2002 ) 608 .
g. vidal , j. i. latorre , e. rico , and a. kitaev , phys .
lett . * 90 * , ( 2003 ) 227902 .
r. somma , h. barnum , e. knill , g. ortiz , and l. viola , int . j. mod
b * 20 * , ( 2006 ) 2760 .
a. hamma , quant - ph/0602091 .
p. zanardi and n. paunkovi , phys .
e * 74 * , ( 2006 ) 031123 .
zhou and j. p. barjaktarevic , _ fidelity and quantum phase transitions _ , preprint .
f. iachello and n. v. zamfir , phys .
lett . * 92 * , ( 2004 ) 212501 . c. dunning , k. e. hibberd , and j. links , _ on quantum phase crossovers in finite systems _ , j. stat .
, to appear
. j. m. arias , j. dukelsky , j. e. garca - ramos , and j. vidal , nucl - th/0609001 .
a. leviatan , nucl - th/0603011 .
g. n. santos , a. p. tonel , a. foerster , and j. links , phys .
a * 73 * , ( 2006 ) 023609 .
j. links , h .- q .
zhou , r. h. mckenzie , and m. d. gould , j. phys . a : math
. gen . * 36 * , ( 2003 ) r63 d. f. walls and r. barakat , phys .
a * 1 * , ( 1970 ) 446 .
d. f. walls and c. t. tindle , j. phys . a : math
. gen . * 5 * , ( 1972 ) 534 .
a. g. ushveridze , _ quasi - exactly solvable models in quantum mechanics _ ( institute of physics publishing , bristol and philadelphia , 1994 ) . v. v. ulyanov and o. b. zaslavskii , phys
* 216 * ( 1992 ) 179 .
m. a. nielsen and i. l. chuang , _ quantum computation and quantum information _
( cambridge university press , 1999 ) . | we study a three - mode hamiltonian modelling a heteronuclear molecular bose
einstein condensate .
two modes are associated with two distinguishable atomic constituents , which can combine to form a molecule represented by the third mode . beginning with a semi - classical analogue of the model , we conduct an analysis to determine the phase space fixed points of the system .
bifurcations of the fixed points naturally separate the coupling parameter space into different regions .
two distinct scenarios are found , dependent on whether the imbalance between the number operators for the atomic modes is zero or non - zero .
this result suggests the ground - state properties of the model exhibit an unusual sensitivity on the atomic imbalance .
we then test this finding for the quantum mechanical model .
specifically we use bethe ansatz methods , ground - state expectation values , the character of the quantum dynamics , and ground - state wavefunction overlaps to clarify the nature of the ground - state phases .
the character of the transition is smoothed due to quantum fluctuations , but we may nonetheless identify the emergence of a quantum phase boundary in the limit of zero atomic imbalance .
pacs : 02.30.ik , 03.65.sq , 03.75.nt |
the general model for detection is that the response is triggered by the transfer of an electron from a donor ( d ) to an acceptor ( a ) within the receptor site by the presence of an olfactant molecule that provides a bridge between the two . in the absence of the odorant ,
the distance between d and a is too great and electron transfer is inefficient .
placing an olfactant ( b ) between the two allows the electron transfer to occur either as a single coherent scattering event from d to a , or as a sequence of two incoherent hops , first from d to b then from b to a. this simple model is consistent with the standard `` swipe - card '' model for odor detection since implicit in this is that b must fit into some sort of pocket and be in a correct alignment between d and a in order for the charge transfer process to occur .
an interesting extension to this paradigm is that molecular shape may not be the sole deciding factor in scent recognition .
indeed , it has been observed recently that fruit flies ( _ d . melanogaster _ ) can navely discriminate between several isotopomers that have deuteriums substituted for hydrogen@xcite .
these flies can also be trained to associate a specific odorant with an electric shock resulting in the specific avoidance of the conditioned odorant@xcite .
flies trained in this manner can discriminate deuterated from hydrogenated isotopomers , indicating an ability to perceive differences in these odorants .
the ability to discriminate between the h - acp and the d8-acp has been shown to require the orco ( or83b ) olfactory co - receptor @xcite .
mutant flies that lack the gene are broadly anosmic and fail to discriminate between isotopomers , indicating the behavioral discrimination of isotopomers relies on olfactory perception .
in addition , there is a curious observation that fruit flies trained to discriminate deuterated olefins also discriminate non - deuterated olfactants with strong ir peaks in the 2300 @xmath0 range.@xcite
lastly , using the negatively - reinforced learning paradigm , drosophila has been shown capable of generalizing a trained odorant to another that shares a similar vibrational spectrum .
for example , when deuterated d5-benzaldehyde is paired with electric shocks , the flies will subsequently avoid fully - deuterated d17-octanol ( vs. undeuterated h - octanol ) , suggesting that the c - d vibrational mode is a salient feature of odorant perception in this species .
a controversial and speculative explanation of these observations is that there is an spectroscopic component to olfaction .
@xcite this model supposes that the bridge facilitates inelastic electron scattering and as such the olfactory response can be predicted by comparing infra - red ( ir ) spectra of various olfactants .
isotopic substitution of h for d shifts the ch stretching frequency from the 2850 - 3100 @xmath0 range into the 2300 @xmath0 range .
few molecules absorb in this ir region and there is little or no biological need or evolutionary pressure that we know for detecting deuterated compounds .
needless to say , this theory has been met with considerable skepticism since it runs contrary to more commonly held model in which the geometric shape and chemical nature of the olfactant are the primary components of olfactory reception .
@xcite moreover , many of the claims in ref
. and ref . were show to be inconsistent with psychophysical tests performed on human subjects by keller and vosshall .
@xcite for example , in this study human subjects could not discriminate by smell between deuterated and non - deuterated forms of acetophenone . on the other hand ,
the molecular mechanism for olfaction in humans might be different since humans have far fewer active odor receptor genes than other mammals and insects .
while vertebrate olfactory receptors are g - protein coupled , insect ors appear to have a deferent structure and act as ligand - gated cation channels .
the or83b odorant receptor gene is a broadly expressed receptor in drosophila and is remarkably conserved amongst the insect species .
information concerning the molecular structure at the olfactant binding site remains illusive , may involve odorants binding to cu(ii ) and zn(ii ) ions bound to the protein loop , activating the g protein through a structural a `` shuttle - cock '' mechanism that leads to cascade of events that eventually leads to some neural activity .
@xcite since very little is known about the molecular level details of the binding of the olfactant to a receptor site , our goal here is to piece apart the possible contributions to the transfer rate , determine whether or not a given bridge molecule can have an inelastic component to the rate , and give at best an indication of the relative strengths of the inelastic components .
however , for a given chemical olfactant , it is unlikely that discrimination between isotopomers can be a purely electronic effect since isotopomers have identical electronic structures within the born - oppenheimer approximation .
it is unlikely that the discrimination could be differences in mass since this would account for at most a few percent in terms of the diffusivity .
for example , in comparing the diffusion constant of acetophenone to its fully deuterated isotopomer , one has at best a 3% difference .
also , at ordinary physiological temperatures , there is no significant vibrational contribution to the heat capacity coming from the relevant c - h or c - d stretching or bending modes at 3100 @xmath0 and 2300 @xmath0 , respectively .
we assume that at the heart of the process , a charge is transferred from a donor to an acceptor within the receptor site itself . in the absence of an odorant ,
the distance between the donor and acceptor is too great to allow efficient electron transfer . when the odorant is present , it acts like a bridge between the donor and acceptor allowing for more efficient electron tunneling between the two .
@xcite aromatic compounds are typically better electron donors ( _ i.e. _ hole receptors ) due to their electron - rich @xmath1 -system . with this in mind
, we shall assume that the bridge / olfactant acts as an intermediate for hole transfer between a donor site and an acceptor site .
energy conservation requires that the energy transferred to vibrational motion is equal to the tunneling gap . for a donor - bridge - acceptor ( dba )
system in the non - adiabatic limit of electron transfer theory , the golden rule rate is given by @xmath2 there are two contributing factors to the transfer rate .
first , @xmath3 is the effective coupling between donor and acceptor which depends upon the electronic coupling between the donor and acceptor and @xmath4 is the thermally averaged franck - condon weighted density of nuclear vibrational states between the donor and acceptor and @xmath5 is the energy gap between donor and acceptor . before describing our approach ,
let us briefly review the model by lambe and jakelvic for molecular vibrational spectroscopy via inelastic electron tunneling in a tunnel junction system .
@xcite in a tunneling device , where there is a barrier potential @xmath6 separating two metallic leads , the wkb approximation give the electronic part of the tunneling matrix element as @xmath7 is the spacing between leads ( i.e. the spatial thickness of the barrier ) , @xmath8 is the total electronic energy , @xmath9 is the kinetic energy associated with motion perpendicular to the barrier , and @xmath10 is small perturbing potential due to the presence of an impurity molecule in the tunneling region . in a classic paper by scalapino and marcus@xcite , they considered the the case where a molecule with a permanent dipole moment @xmath11 is located close to one of the electrodes so that its image dipole must be included .
this leads to an interaction potential between the passing electron and dipole of the form @xmath12 where @xmath13 is the distance from the molecule and the electrode and @xmath11 is the molecular dipole operator . for the case of a single molecule in the tunneling region the current couples to the the dipole oscillations of the molecule leading to an inelastic contribution that contains the vibrational transition moment .
lambe and jakelvic also show that even in when the molecules in the tunneling gap lack permanent dipole moments , the current can couple to the polarizibility of the molecules leading to raman contributions to the inelastic current . here
, again one considers the interaction between the impurity molecule , the passing electron , and the nearest image dipole to show that @xmath14 where @xmath15 is the polarizability of the molecule . in this case , the scattering of the electron induces raman - like transitions within the impurity molecule . both cases lead to golden rule expressions for the inelastic contributions to the current - voltage curves in the presence of molecular impurities in the system .
et al._@xcite follow a similar lines and arrive at estimates for whether or not an inelastic tunneling component would be possible to observe in a model parameterized by physiological considerations .
however we argue that such inelastic barrier tunneling models are not suitable for the case at hand where we have electron transfer between localized states on the donor , bridge , and acceptor .
while connection between the electron transfer rate and the zero - bias molecular conduction was discussed by nitzan @xcite in comparing the landauer formula to the marcus rate , the relation is established when the orbitals of the bridging molecule are in contact with orbitals of the donor and acceptor leads .
this is not the case in the model described above . since the electron transfer rates are likely far slower than the structural responses they trigger , the central goal in this paper is to determine those internal vibrational modes of the bridge species that are important in accommodating the inelastic scattering rather than providing actual transition rates .
bridge mediated charge transfer is a broadly studied topic@xcite , and we start by assuming that we have three relevant sets of diabatic states denoted by @xmath16 , @xmath17 , and @xmath18 corresponding to the initial , intermediate , and final quantum states of the system . in @xmath19 , the charge is localized in a donor orbital and the bridge is in its neutral electronic state . in @xmath20 ,
the bridge is in a singly oxidized state , lastly in @xmath21 , the bridge is again in a neutral state and the charge is localized in an orbital on the acceptor . within a diabatic picture , we have the following hamiltonian : @xmath22\end{aligned}\ ] ] where each term on the diagonal represents the electronic + nuclear hamiltonians for each state .
@xmath23 and @xmath24 are the electronic interactions between bridge and donor or acceptor states respectively .
we assume that the the through - space coupling between the donor and acceptor are negligible compared to the other couplings in the system .
@xcite let us re - cast this as a reduced two state problem using the feshbach method so that the transition from the donor to acceptor states is via resonant scattering involving the bridge ( olfactant ) molecule.@xcite the results in an effective schrdinger equation @xmath25 where @xmath26 is the unperturbed system lacking the bridge and @xmath27 is an _ effective _ coupling matrix element between the donor and acceptor due to the presence of the bridging molecule which depends upon the scattering energy , @xmath8 .
@xcite this operator has a series of poles located at the eigen - energies of @xmath28 and contains both resonant and non - resonant components . at the heart of eq .
[ veff ] is the green s function for the evolution of a nuclei on the potential energy surface of the ionized species .
we can imagine this in the dynamical sense : upon ionization , the nuclei in @xmath29 experience a sudden change in their electronic environment corresponding to the charge transfer from the donor to the bridge .
this creates a vibrational wave function on the potential energy surface of the ionized species centered about the ground - state nuclear geometry .
we can write this as @xmath30 where @xmath31 is correlation function for the propagation a vibrational wave packet on the born - oppenheimer potential , @xmath32 , corresponding to this new electronic configuration , then projected onto the manifold of vibrational states of the final electronic configuration , where as above we partition the electronic contribution from the nuclear dynamics , except that our time evolution occurs on the @xmath32 potential corresponding to the bridging state .
this gives the correlation function as @xmath33 where we have assumed the initial wave packet to the be the ground - state vibrational wave function and the sum is over final vibrational states @xmath34 . upon promotion to the ionized state , the nuclei receive an impulsive force along the direction of the gradient of @xmath32 .
this is certainly the case for a classical oscillator displaced from its equilibrium position @xmath35 to @xmath36 where the force acting on the particle is @xmath37 .
the oscillator then follows an elliptical trajectory in phase - space . secondly ,
since the wave function is simply the displaced harmonic oscillator ground - state wave function and the well is harmonic , the shifted state evolves as a glauber coherent state in phase space without spreading or contracting .
thus , we approximate the time correlation function as @xmath38 where @xmath39 is the momentum imparted along normal coordinate @xmath40 , @xmath41 is the normal mode frequency , and @xmath42 is the transition dipole moment for the @xmath40th vibrational transition .
we can now evaluate eq .
[ cot ] by taking the fourier transform of the correlation function .
this gives the effective potential as a series of @xmath43-functions @xmath44 and we recall that the delta - function carries units of inverse - energy .
energy conservation requires that total momentum imparted to all the oscillators be such that @xmath45 in mass - scaled units . from above
, we assumed that the momentum transferred to the mode is proportional to the energy gradient of the ionized species along that direction evaluated at the equilibrium position of the neutral .
this implies that to a first approximation the inelastic scattering of an electron via the bridge species excites the vibrational modes of the bridge that are also infra - red ( ir ) active .
this is a central component of turin s spectroscopic theory of olfaction.@xcite however , here we see that _ only those modes that are directed along @xmath46 will be excited by the impulsive scattering process . _
( [ cot2 ] ) gives us a direct way to rapidly screen whether or not a given odorant is expected to exhibit an isotope effect using quantum chemical means .
starting from the equilibrium position of the neutral , one first determines the vibrational normal modes and frequencies of a given olfactant .
this gives the ir response of the molecule .
we then determine the energy gradient of the ionized species at this geometry and project this onto the normal mode coordinates , which are normalized linear combinations of the cartesian displacement coordinates for each atom in the molecule . since @xmath5 remains an unknown in our model , we require that this energy be distributed amongst the normal coordinates in proportion to their projection onto the energy gradient .
thus , even if a mode has a strong ir response and satisfies energy conservation , unless that mode is directly excited by the impulsive scattering process , then it will not contribute to the overall transition rate .
moreover , in order for cd stretches to play any role in the transfer rate , we have to make the assumption that @xmath47 . as a test case , we consider acetophenone ( acp ) and its deuterated isotopomers . in d3-acp ,
the hydrogens on the methyl group are replaced , in d5-acp the hydrogens on the phenyl ring are replaced , and in d8-acp all hydrogens are replaced with deuteriums . in all cases , we first perform geometry optimizations of the neutral species followed by calculations of vibrational frequencies using the b3lyp density functional with 6 - 31g(d ) basis set in vacuum using the nwchem quantum chemistry package.@xcite the same functional and basis set were then used to calculate the nuclear gradients and frequencies of the radical cation at the equilibrium geometry of the neutral .
the gradients are then projected on to the normal mode eigenvectors giving the projection of the force onto a given normal mode with frequency giving us @xmath48 .
the normal mode calculations also give us the infra - red response , @xmath49 .
since we know little about the actual binding site , we can only assume that the transfer integrals @xmath50 and @xmath51 are independent of the vibrational modes and are the same for each isotopomer .
figure [ acp - grad ] shows the results of our calculations for the deuterated and non - deuterated forms of acp along with the predicted infra - red responses .
this approach is numerically reliable and produces accurate potentials and gradients for this system without undue effort .
moreover , this general procedure can be applied to a broad set of olfactants .
briefly , the peaks around 3100 @xmath0 correspond to the c - h stretching modes of the molecule . in the fully deuterated d8-acp ,
this peak is shifted to 2300 @xmath0 , consistent with the typical isotope shift of a c - d stretching mode .
furthermore , the ir spectra for all but the d8-acp is void of peaks between 1800 - 3000 @xmath0 .
only the fully deuterated d8-acp has any appreciable spectral density in this region .
while there is little to note in comparing the intensities for the modes below 1800 @xmath0 , the contribution to the gradient coming from the c - h and c - d stretching modes is remarkably strong .
most notable , the peak at 2300 @xmath0 is more or less the sum of contributions from the ring and methyl c - d stretching modes .
based upon the relative intensities of both the ir and energy gradients around 2300 @xmath0 , the electron transfer model suggests that _
d. melanogaster _ should readily discriminate h - acp and d8-acp . in comparing the predicted ir spectra
, there is almost no ir oscillator strength in the c - d stretching modes for d3- and d5-acp while d8-acp exhibits a fairly strong signal .
however , in comparing the gradients , the d5-acp gradient has considerably smaller projection onto the c - d stretches modes and the d3-acp projection is almost vanishingly small . based upon gradients ,
one predicts that there should be an ability to differentiate between h - acp and d8-acp .
moreover , the flies may be able to discriminate between d3-acp and d5-acp even though both have very weak ir signals at 2300@xmath0 .
a crucial test of our model would be to identify an olfactant that can not be discriminated from its deuterated isotopomers .
for example , in removing an electron from ethylene , the c = c bond - length in @xmath52 will be longer than in @xmath53 ( due to reduction in bond - order ) but the c - h bond - lengths will be more or less unchanged .
this suggests that the energy gradient in our model will be directed primarily along the c = c bond .
quantum chemical calculations of the sorts described above on ethylene and two of its isotopomers , @xmath54 and @xmath55 , indicate the forces along the cd stretching coordinates these are far weaker in magnitude than along the ch stretching coordinates .
more over they are about 40-fold weaker than the corresponding forces in d8acp but are only half of those in d3acp .
however , the ir spectrum of ethylene does show an obvious isotope shift of the c - h stretching modes into the 2300@xmath0 region .
consequently , our model would predict that while drosophila can discriminate between hacp and d3acp it is doubtful that they could discriminate between ethylene and its isotopomers within the limits of the error bars of the t - maze experiments .
previously , d5-acp and d8-acp odorants were found to be significantly more aversive to drosophila than h - acp @xcite .
we verified this result , examining the nave avoidance of drosophila @xmath56 to 0.3% h , d3 , d5 , and d8-acp using an olfactory t - maze@xcite ( figure [ gr2]a ) .
the avoidance response to h - acp and d3-acp were indistinguishable ( f3,80 = 11.724 , bonferonni - dunn _ post hoc _ @xmath57 ) ; the responses to d5 and d8 were also not significantly different from each other ( p = 0.842 ) . however , the avoidance response to d5-acp was significantly greater than the response to d3-acp ( @xmath58 ) , suggesting deuteriums on the benzene ring provided a different odorant quality than the methyl deuteriums , leading to an increased saliency for the d5-acp and d8-acp . to test this hypothesis , we used negatively - reinforced olfactory learning @xcite . surprisingly , the @xmath56 flies are capable of discriminating between d5-acp and d3-acp after training , indicating a significant difference in perceptual quality between these odorants ( figure [ gr2]b ; @xmath59 , @xmath60 ) .
the @xmath56 flies are also capable of discriminating between h - acp and d3-acp after training ( figure 3c ; @xmath61 ; @xmath62 ) .
the conditioned avoidance in this experiment was reduced in comparison to the conditioned avoidance of d5-acp vs. d3-acp , indicating that d3-acp is more similar in odorant quality to h - acp than to d5-acp .
the robust discrimination between these very closely related odorants is consistent with our model that predicts differences in the excitation of vibrational modes of the methyl and deuterium c - d bonds by the impulsive scatting .
it is clearly evident that _
d. melanogaster _ can be trained to distinguish deuterated isotopomers of various olfactants . whether or not this is the result of a spectroscopic detection mechanism is an another question .
our model predicted that drosophila could be trained to discriminate between d8-acp and d5-acp , between d5-acp and d3-acp , and between d3acp and hacp based upon comparison of the gradients at 2300@xmath0 .
this prediction was borne out in the training experiments .
moreover , discrimination between d8-acp and d5-acp would not have been predicted based upon _ just _ the ir spectrum .
while deuteration does shift the c - h stretching mode into the otherwise empty 2300 @xmath0 region , our model suggests that _ ir response alone does not determine whether or not the flies can discriminate isotopomers .
_ moreover , this effect appears to be limited to a single spectral region and does not involve the entire ir spectrum.@xcite lacking detailed molecular level knowledge of the receptor sites , we speculate the tunneling gap between donor and acceptor sites is around 2300 @xmath0 and that the narrow spectral region about @xmath5 is giving rise to the observed ability for drosophela to discriminate between deuterated isotopomers by scent .
it is possible that the tunneling gap may be tuned to the nitrile ( -c@xmath63n ) stretching mode which does lie in this spectral region and may serve a role in detecting a variety of naturally occurring nitrile containing odorants .
a test of our model would be to identify an olfactant that has ir active c - h stretching modes but upon ionization distorts along modes that do not involve the c - h stretching modes .
for example , it is known that upon ionization of ethylene , the c = c bond - length in @xmath64 is shorter than in @xmath65 and the c - h bond - lengths are unchanged .
this suggests that the energy gradient in our model will be directed primarily along the c = c bond and have very little projection onto the c - h stretching modes .
however , the ir spectrum does show an isotope shift of the ch stretching modes into the 2300@xmath0 region .
this is evident in our computed spectra shown in fig .
[ ethylene ] .
our model then predicts that _ drosophila _ should _ not _ be able to easily discriminate between ethylene and any of its isotopomers .
the model also suggests that the isotope effect may not generalize between different classes of organic olfactants .
for example , here we focused upon aromatic olfactants which are good hole acceptors since they have electron - rich @xmath1-systems .
however , the electron transfer dynamics through aliphatic system would not likely involve transfer of a positive charge and may well involve different olfactory receptors all together .
erb was supported by the national science foundation ( che-1011894 ) and robert a. welch foundation ( e-1334 ) .
the work by gr was supported by darpa ( n66001 - 11 - 1 - 4119 ) .
erb jointly conceived the study with gr , developed the theoretical model and prepared the manuscript .
ac performed the quantum chemical calculations .
gr and am designed and performed the experiments .
the authors have no competing financial interests .
28ifxundefined [ 1 ] ifx#1 ifnum [ 1 ] # 1firstoftwo secondoftwo ifx [ 1 ] # 1firstoftwo secondoftwo `` `` # 1''''@noop [ 0]secondoftwosanitize@url [ 0 ]
+ 12$12 & 12#1212_12%12@startlink[1]@endlink[0]@bib@innerbibempty link:\doibase 10.1073/pnas.1012293108 [ * * , ( ) ] , @noop * * , ( ) link:\doibase 10.1016/j.neuron.2004.08.019 [ * * , ( ) ] http://dx.doi.org/10.1038/35046000 [ * * , ( ) ] link:\doibase 10.1093/chemse/21.6.773 [ * * , ( ) ] , link:\doibase 10.1006/jtbi.2001.2504 [ * * , ( ) ] link:\doibase 10.1073/pnas.1103992108 [ * * , ( ) ] , http://www.ncbi.nlm.nih.gov/pubmed/15034588 [ * * , ( ) ] link:\doibase 10.1073/pnas.262792899 [ * * , ( ) ] , @noop _
( , ) link:\doibase 10.1103/physrev.165.821 [ * * , ( ) ] @noop * * , ( ) link:\doibase 10.1103/physrevlett.98.038101 [ * * , ( ) ] link:\doibase 10.1021/jp003884h [ * * , ( ) ] , http://link.aps.org/abstract/rmp/v65/p599 [ * * , ( ) ] link:\doibase 10.1063/1.1696792 [ * * , ( ) ] link:\doibase 10.1063/1.1742723 [ * * , ( ) ] @noop * * , ( ) link:\doibase 10.1063/1.474741 [ * * , ( ) ] link:\doibase 10.1103/physreve.81.027101 [ * * , ( ) ] link:\doibase 10.1063/1.480471 [ * * , ( ) ] link:\doibase 10.1039/c2cp40579b [ * * , ( ) ] @noop _ _
( , , ) @noop _ _ ( , ) link:\doibase 10.1021/cr00005a007 [ * * , ( ) ] , link:\doibase 10.1002/9780470141656.ch3 [ * * , ( ) ] link:\doibase 10.1016/j.cpc.2010.04.018 [ * * , ( ) ] @noop * * , ( ) | the standard model for molecular recognition of an odorant is that receptor sites discriminate by molecular geometry as evidenced that two chiral molecules may smell very differently .
however , recent studies of isotopically labeled olfactants indicate that there may be a molecular vibration - sensing component to olfactory reception , specifically in the spectral region around 2300 @xmath0 . here
we present a donor - bridge - acceptor model for olfaction which attempts to explain this effect .
our model , based upon accurate quantum chemical calculations of the olfactant ( bridge ) in its neutral and ionized states , posits that internal modes of the olfactant are excited impulsively during hole transfer from a donor to acceptor site on the receptor , specifically those modes that are resonant with the tunneling gap . by projecting the impulsive force onto the internal modes
, we can determine which modes are excited at a given value of the donor - acceptor tunneling gap . only those modes resonant with the tunneling gap and are impulsively excited will give a significant contribution to the inelastic transfer rate . using acetophenone as a test case , our model and experiments on _ d. melanogaster
_ suggest that isotopomers of a given olfactant give rise to different odorant qualities .
these results support the notion that inelastic scattering effects may play a role in discriminating between isotopomers but that this is not a general spectroscopic effect . |
over the past years , observations and theory have been progressing significantly in understanding how the first large disk galaxies could form .
near - infrared ( rest - frame optical ) imaging from the hst reveals that , while @xmath140% of z@xmath11 - 3 most massive ( i.e. , @xmath2 ) galaxies are relatively compact and quiescent , @xmath160% of them are actively forming stars and show quite extended stellar structures ( @xcite ; see also @xcite ) . in a larger range of mass (
i.e. , @xmath110@xmath3 - 10@xmath4 ) , @xcite suggested that the stellar phase of [email protected] - 3.6 galaxies is probably distributed as triaxial ellipsoids rather than thick disks .
integral field spectroscopy now routinely provides us with spatially - resolved kinematics of the sub - population of z@xmath11 - 3 emission line galaxies @xcite . actively forming galaxies at these epochs reveal a relatively balanced mix of obvious ( major ) mergers , relatively compact dispersion - supported objects , and rotating extended structures , all being much more turbulent in their gaseous phase compared to local galaxies @xcite . on the theoretical side ,
cosmological simulations have predicted that streams of cold gas would occur in the halos in which these galaxies are expected to inhabit @xcite .
such streams would be responsible for bringing a continuous amount of high angular momentum gas in these halos over a relatively short timescale @xcite . because the amount of mass brought that way
is expected to be considerable , this could in principle pressurize the central forming disk and result in clump fragmentation @xcite , as ubiquitously observed at these epochs @xcite .
such `` clumpy galaxies '' are now more and more considered as an important step of galaxy formation .
however , it remains unclear whether the lifetime of such clumps is long enough to allow them to survive feedback effects and coalesce to form a central bulge @xcite .
it remains also unclear how _ thin _ disks can form from the high angular momentum gas which is expected to be accreted into the halos . finally , the smoking gun detection of cold gas accretion is still missing , although different options are being extensively discussed @xcite .
in contrast , it remains more uncertain how these z@xmath12 thick rotating disks evolved down to z@xmath10 , and how the local thin disks formed .
indeed , the hubble sequence of galaxies such as the milky way ( @xmath5 ) reveals a predominance ( 72% ) of rotationally - supported disks in the present - day universe @xcite . using the cosmological principle , their progenitors at [email protected] , i.e. ,
six billion years ago , can be selected and their morphology studied using the advanced camera for survey on - board the hubble space telescope @xcite .
using methodologies and rest - frame luminosities that mimic those used for local galaxies , the likeliest ancestor of the hubble sequence can then be established @xcite .
half the progenitors of spiral galaxies show peculiar morphologies that exclude them from this sequence @xcite .
most of them are starbursts , whose emergence at [email protected] reflects that of emission line galaxies @xcite . as part of the images survey ,
the large - scale spatially - resolved kinematics of a representative subsample of 63 emission line galaxies at z=[0.4 - 0.75 ] was obtained using the giraffe multi - ifu spectrograph @xcite .
most peculiar galaxies also reveal peculiar large - scale motions @xcite . comparing the morpho - kinematic properties of the subsample of the 33 galaxies lying into the cdfs @xcite with those of a grid of simple major merger models ,
convincing matches were found in about two - thirds of the cases @xcite .
this leads to at least one third of the progenitors of present - day spirals potentially involved in a major merger .
since major mergers can easily destroy thin rotating disks @xcite , this creates a critical tension between the large fraction of present - day disks and their survival within the @xmath0-cdm paradigm @xcite . further investigating the `` disk survival issue '' requires an accurate estimation of the evolution of the merger rate , i.e. , the fraction of galaxy involved in mergers per unit time .
several techniques were developed over the past years to try to quantify accurately the evolution of the merger rate , such as the pair technique ( e.g. , @xcite ) or the morphological technique , either through non - parametric ( e.g. , @xcite ) or visual ( e.g. , @xcite ) classifications .
alternatively , the amplitude of the two - point correlation function can also be used to infer a merger rate @xcite .
all these techniques are very dependent on the visibility timescale of the specific merging phase sampled by the observations , which requires to be estimated using additional numerical simulations and/or analytical arguments @xcite .
accounting for the exact distributions of merger mass ratios or orbits involved is also challenging @xcite .
as a result , observational estimates spread over almost one decade at z@xmath10 - 1 @xcite .
recent progresses combine numerical simulations with transfer radiation to derive a weighted average visibility timescale for each observational technique , using predictions for the distribution of mass ratios and other merger parameters from galaxy formation models @xcite .
taking these aspects into account , the discrepancy between the different merger rates derived in the literature is significantly reduced , while remaining differences are attributed to differences in range of mass ratio and sample selection @xcite .
this illustrates how crucial it is to carefully accounts for these aforementioned effects .
the present study puts this effort one step forward : we show that by combining spatially - resolved morphological and kinematic data as those provided by the images survey , one can date back the epoch where observed galaxy mergers were still in pair , and then derive the evolution of the fraction of mergers out to [email protected] , in a consistent way as a function of redshift .
such data also allow us to estimate relevant visibility timescales that are by construction consistent with the distribution of merger mass ratios and orbits of the galaxies in the observed sample .
the derived evolution of the merger rate out to [email protected] is found to be within a factor 2 - 3 of theoretical predictions from a state - of - the - art @xmath0-cdm semi - empirical model .
this paper is organized as follows .
2 presents the data used in this paper , which come from the images survey of [email protected] emission line intermediate - mass galaxies . in sect .
3 , we argue why large morpho - kinematic perturbances most likely result from major mergers . sect .
4 presents the major merger fractions and rates derived from these data .
a comparison with state - of - the - art semi - empirical model of galaxy formation is presented in sect .
finally , a discussion is given in sect .
6 and conclusions are drawn in sect .
7 . throughout the paper ,
magnitudes are given in the @xmath6 system , and a wmap-7 cosmology is used @xcite , with ( @xmath7 , @xmath8 , @xmath9 , @xmath10 , @xmath11)=(0.275 , 0.725 , 0.0458 , 0.702 , 0.816 ) .
we adopt the diet salpeter imf @xcite throughout this paper .
note that all imf - dependant quantities cited from other studies were converted to this imf .
in the frame of the images survey , a sample of 35 emission line galaxies at z=[0.4 - 0.75 ] , i.e. , about six billion years ago ( mean / median redshift of z=0.6/0.65 ) , were observed in the chandra deep field south ( cdfs ) @xcite .
this subsample of images is of particular interest since it benefits from homogeneous imaging and 3d data @xcite .
images targets were selected to have @xmath12 23.5 and @xmath13)\geq 15\aa$ ] to guarantee their detection by the multi - ifu spectrograph flames / giraffe at vlt @xcite .
these galaxies were selected to be intermediate - mass galaxies ( @xmath14 using a diet salpeter imf and simplified prescriptions to convert j - band luminosity and b - v colors into stellar mass , following @xcite ) .
this range of mass is of particular interest since most of the star formation at z@xmath151 occurred in such galaxies @xcite .
since the fraction of e / s0 did not raise between [email protected] and z=0 @xcite , z@xmath151 intermediate - mass galaxies are the likeliest progenitors of local spirals ( see also app .
b ) . from the full images sample ,
one galaxy was rejected because it turned out not to be a starburst galaxy @xcite , and another one because the hst / acs images were corrupted and no morphological analysis could be conducted .
the resulting sample is fully representative of the j - band luminosity function at z=0.5 - 1.0 @xcite . at these redshifts ,
the near - infrared luminosity functions of the global population and of the blue sub - population of galaxies are very similar @xcite . at first order ,
emission - line and blue galaxies sample the same population @xcite , which in principle guarantees that the images - cdfs sample is representative of the star forming population of galaxies at these epochs .
the total sfr of all the 33 remaining galaxies in the images - cdfs sample was estimated as the sum of the unobscured sfr measured from the rest - frame uv luminosity at 2800@xmath16 combined to the dust - reprocessed contribution , which was estimated from the ir luminosity as measured using 24 @xmath17 m spitzer fluxes ( see @xcite for details ) .
we find a mean sfr of @xmath112m@xmath18/yr , which compares well with other uv+ir measurements in larger samples at a similar range of mass and redshift ( e.g. , @xcite ) . in fig .
[ sfr ] , the location of the images - cdfs galaxies in the sfr - m@xmath19 plan is compared to the sequence of starbursting galaxies observed by @xcite at z=0.45 - 0.7 . as expected , the images - cdfs sample comprises luminous infrared galaxies , which populates the high - sfr tail of the distribution at sfr@xmath2014m@xmath18/yr @xcite , and less obscured ( uv ) starbursts , which populate the low - sfr tail of the distribution .
this figure reveals that most ( 91% ) images - cdfs galaxies lie well within the observed sfr - m@xmath19 sequence at these redshifts . the star formation rate density in the images - cdfs sample is found to be @xmath21 - 1.4 ( where @xmath22 is expressed in @xmath23 ) , also in good agreement with expectations from larger surveys that include ir data at 24 @xmath17 m @xcite.-1.4 at [email protected] for galaxies with m@xmath24m@xmath18 and @xmath25 - 1.9 for galaxies with m@xmath26m@xmath18 ( using a diet salpeter imf ) .
using a larger area survey , @xcite found that for the mass range m@xmath27m@xmath18 , i.e. , very close to the range of mass considered in the present study , @xmath28 ranges from -1.56@xmath29 at z=0.4 - 0.6 to [email protected] at z=0.5 - 0.8 , which is consistent with the value found in the images - cdfs sample .
] this confirms that the images - cdfs sample is representative of star - forming galaxies at [email protected] .
1-@xmath31 scatter from @xcite , as reported by @xcite.,width=340 ] gas masses were estimated from the total sfr by inverting the schmidt - kennicutt law @xcite .
all galaxies were found in a range of total baryonic mass @xmath3210.2 to 11 , or @xmath33 in the range 10 to 11 . the stellar - mass density in the images - cdfs sample
is found to be @xmath348.1 ( where @xmath35 is expressed in @xmath36 ) , which is in relatively good agreement with expectations from larger surveys of blue galaxies @xcite , within expected uncertainties .
detailed color morphology ( at a @xmath1500 pc resolution scale ) and large - scale kinematics ( at a @xmath17 kpc resolution scale ) were obtained thanks to the hubble space telescope , and 3d spectroscopy using the giraffe multi - ifu spectrograph , respectively .
galaxies were classified independently according to their morphology @xcite and kinematics @xcite , based on similarities with the properties of local galaxies .
the kinematical maps allowed us to classify the images - cdfs galaxies into three distinct kinematic classes ( see @xcite ) . rotating disks [ rds ] were those for which the velocity field [ vf ] showed an order gradient , a dynamical axis aligned with the morphological axis , and a velocity dispersion peak located at the dynamical center .
for these objects , it was required that the sigma peak location and amplitude can be reproduced by modeling the velocity dispersion map from the velocity gradient observed in the vf . at the giraffe spatial resolution
, it is indeed expected that most of the rotation curve gradient falls into on giraffe pixel , which results in an velocity dispersion peak at the dynamical center of the object .
conversely , galaxies with complex kinematics [ cks ] were those for which no ordered gradient was observed in the vf , or when a significant misalignment between the kinematical rotation axis and the morphological principle axis was detected .
perturbed rotators [ prs ] gathered objects with an order velocity gradient in the vf , but for which the velocity dispersion peak ( location and/or amplitude ) could not be reproduced from the vf , which indicates that this peak can not be attributed to rotation .
this classification was tested against quantified criteria and considering all uncertainties associated to kinematic measurements , which makes it objective and reproducible , as described extensively in @xcite .
the detailed color morphology of all the 33 galaxies in the images - cdfs sample was studied by @xcite . using a reproducible decision tree relying on parametric parameters and visual inspection of images and color maps ,
galaxy morphologies were classified into six different classes : compact systems , mergers , peculiar/ possible mergers , peculiar tadpoles , peculiar irregulars , and spiral disks .
in particular , it was required for a rotating disks to show redder colors towards the center as observed in local galaxies .
this classification was done by three independent classifiers using a decision tree , which limits the subjectivity and makes it reproducible .
the comparison between the morphological and kinematical classes show a very good agreement , with more than 80% of spiral disks having rotating disks , and more than 90% of peculiar galaxies having complex or perturbed kinematics .
since perturbations in the ( ionized ) gaseous phase generally correspond to perturbations in the stellar phase , the result of both the morphological and kinematical classifications can be robustly synthesized into a morpho - kinematic classification @xcite . with this classification ,
rotating spiral disks are those classified both as rotating disks from the kinematic point of view , and as spiral disks from the morphological point of view .
non - relaxed systems are those that show a complex kinematics and a peculiar morphology .
semi - relaxed systems are galaxies that show either some rotation from their kinematics ( being classified as either a rotating disk or either a perturbed rotation ) but with a peculiar morphology , or galaxies that do not show rotation in their kinematics ( being classified as having a complex kinematics ) but with a spiral disk morphology . according to this classification , 80% of [email protected] starburst galaxies
are non - relaxed or semi - relaxed systems @xcite , which corresponds to @xmath150% of all intermediate - mass , [email protected] galaxies , since 60% of intermediate - mass , [email protected] galaxies are indeed starbursts @xcite .
the hst / acs images and kinematic maps for a representative sub - sample of the images - cdfs can be found in fig .
[ vf ] ( see app .
stellar kinematic disturbances are indubitable evidence for an out - of - equilibrium dynamical state . in high - redshift galaxies , deriving spatially - resolved stellar kinematics remains out of reach , and one has to rely on gas kinematics as a best proxy .
this implicitly assumes that gas kinematics traces the gravitational potential , and that no other mechanisms are perturbing gaseous motions , such as , e.g. , outflows . in the images sample ,
large - scale outflows have been detected in only a handful of cases @xcite . using fors2 slit spectroscopy , @xcite further estimated that only 8% of intermediate - mass galaxies at [email protected] have significant outflows .
this suggests that gas kinematics is indeed a reliable tracer of the dynamical state of galaxies , at least on large spatial scales .
several dynamical processes are known to be a source of kinematic disturbances .
some of them are local in space , while others are global .
relative to the equilibrium state of rotation ( as expected in the intermediate mass range since more massive galaxies are expected to reach a different equilibrium state supported by dispersion ) , local disturbances are either due to minor mergers or internal instabilities such as clump formation , while global disturbances can be due to major mergers or cosmological gas accretion @xcite . in the images sample , kinematic perturbations are detected over a @xmath377 kpc spatial scale , which corresponds to the giraffe - ifu spatial resolution element ( i.e. , two spaxels ) .
they generally affect the galaxies across a significant fraction of their diameter , which , in principle , excludes _ de facto _ any mechanism resulting in local kinematic disturbances . in order to quantify this , we derived the jeans scale length @xmath38 @xcite in the images - cdfs , which is found to be [email protected] kpc .
@xmath38 increases to [email protected] kpc in the sub - sample of clumpy galaxies that show rotation @xcite .
thus , such jeans instabilities should remain largely undetected at the giraffe ifu spatial resolution . in the bottom of fig .
[ figgals ] one of the most clumpiest galaxy in the images - cdfs sample .
this galaxy shows at least seven clumps spread over its disks and is classified as ck .
however , the kinematic perturbations are mostly large - scale , with a misalignment between the morphological and kinematic axis as defined by the location of the maximal and minimal rotation velocities , and with a velocity dispersion peak that can not be reproduced by the rotation model inferred from the vf @xcite .
this confirms that even in the most disturbed clumpy galaxies , the kinematic perturbations can not be associated with the position of the clumps , i.e. , they are not local but global . in addition ,
clump instabilities are limited to a fraction of 20% of intermediate - mass galaxies at [email protected] that are effectively clumpy @xcite , which also rules out clump instabilities as a driver for the strongest kinematic perturbations . regarding minor mergers , such processes
are also expected to lead to local kinematic perturbations only .
an obvious case of minor merger with a mass ratio of @xmath11:18 was studied in detail by @xcite : only a local signature is found in the velocity dispersion map ( see top panel in fig . [ figgals ] ) .
even if we can not exclude that in very specific cases such local processes could result in large kinematic disturbances such as those observed in images galaxies , we can safely conclude that they can not be the main driver for the majority of galaxies , at least for the ck galaxies .
this does not mean that minor mergers are not ongoing in the images sample , but given the spatial resolution of giraffe - ifu data , they probably remain mostly undetected . regarding the impact of processes resulting in global disturbances , it is instructive to consider the range of the relative variations of the gravitational potential ( compared to the dynamical equilibrium state ) @xmath39 , in which galaxies can be considered as isolated and subject to secular evolution , as a function of mass and redshift @xcite .
we estimated @xmath39 for galaxies in the images - cdfs sample using rotation kinematic models @xcite .
for this , we derived the pixel - by - pixel relative difference between the observed vf and each galaxy model , and considered the sum @xmath40 of this map over all pixels , weighted by the observed s / n in each pixel .
the quantity 2.@xmath41 can then be used as a proxy for @xmath39 .
we also derived the dynamical timescale @xmath42 for each galaxy in the images - cdfs sample using as a proxy the ratio between the gas radius and the rotation velocity as determined by @xcite . in galaxies showing rotation ( rds+prs )
, we found that @[email protected] ( median value ) and @xmath42=60@xmath303 myr , while for ck galaxies , @[email protected] and @xmath42=60@xmath3024 myr .
uncertainties correspond to 1-@xmath31 error - bar , and were estimated using bootstrap resampling . in this regime ( i.e. , @xmath4310 myr and @xmath441 ) , the two dominant physical processes expected to drive gravitational perturbations in this range of mass and redshifts are secular evolution ( i.e. , instabilities occurring in a galaxy ) and the direct dynamical evolution of individual substructures due to some previously initiated perturbation , which can eventually dominate the evolution of the whole system @xcite .
large global perturbations ( i.e. , @xmath451 ) are typically triggered by major mergers .
the characteristic timescale for secular evolution can be estimated using the circular disk period i.e. , @xmath46 , with @xmath47 the disk radius and @xmath48 the circular velocity .
we used again gas radius and the rotation velocity as determined by @xcite as proxies for these two quantities .
the timescale for dynamical friction associated with the dynamical evolution of substructure was estimated following the simplified prescription given by @xcite : @xmath49 where @xmath50 , and @xmath0=1 + 1/@xmath51 .
this timescale quantifies how fast the dynamical evolution of substructures evolve on their own , once they have been triggered . for instance
, a merging galaxy will trigger instabilities in the main progenitors that will subsequently evolve on their own , on a timescale given by @xmath52 @xcite .
the dominant process at given amplitude of perturbation is the one associated with the shortest timescale . in fig .
[ figpot ] , we plotted the expected range of value of these two regimes at [email protected] in the images - cdfs sample .
secular evolution is found to be dominant at small relative variations of the gravitational potential @xmath39 , while the influence of ( major ) mergers dominate above @xmath530.3 .
more specifically , isolated galaxies should be found in the upper left region of fig .
[ figpot ] , close to the grey region ( as well as large local disks such as m31 ) , while on - going mergers and mergers remnants should be found at @xmath541 and close to the red region . comparing with these theoretical expectations , ck galaxies lie closer to the individual dynamical perturbation regime ( i.e. , the merger - driven regime ) ,
which is consistent with ( major ) merger - driven kinematic disturbances in these galaxies .
galaxies showing rotation also lie in this regime , although with a smaller range of gravitational perturbation , which is consistent with these galaxies being more ( temporally ) advanced merger remnants , as suggested by their other morpho - kinematic properties ( see @xcite and sect .
4.1 ) . on overall ,
[email protected] rotators ( rds+prs ) lie closer to the expected merger - driven regime , which probably reflects the fact that they are closer to dynamical equilibrium than ck galaxies , in which gravitational perturbations are about three times larger .
this suggests that any other known process ( gas accretion or secular evolution ) is very unlikely to be responsible for the high level of kinematic disturbance observed on large spatial scales , at least for the majority of the images - cdfs sample .
uncertainty range around the median timescale for secular perturbations , while the red region represents the expected 2-@xmath31 uncertainty range around the median timescale for individual perturbations to evolve on its own ( as approximated by eq .
( 5 ) of @xcite , see text ) .
the blue region is the 2-@xmath31 uncertainty range around the median value obtained for rotating objects at [email protected] , while the brown region represents the same for objects with a complex kinematics.[figpot],width=340 ] in this context , it was explored whether hydrodynamical major merger models for the 27 galaxies in the images - cdfs sample which are not relaxed morphogically and/or kinematically could generally account for their perturbations @xcite . morphological and kinematic maps from these simulations were produced and degraded to the spatial resolution of the observations .
such maps were then explored visually as a function of time and viewing angles to determine the best match to observations .
the so - determined best model was graded by three different examiners ( independent of the one who determined the best model ) using specific constraints @xcite .
secure major merger candidates ( 18/27 ) were identified as galaxies having a model that fitted observations with a good level of confidence ( i.e. , grades @xmath20 4/6 ) .
examples of such models are shown in app .
a. these candidates can be used as a basis to estimate the fraction of mergers and the merger rate at different epochs as detailed in sect . 4.1 .
interestingly , the modeling exercise lead to a median merger mass ratio @[email protected] in the images - cdfs sample , while theoretical expectations from variations of the gravitational potential give consistent predictions with @[email protected] in ck galaxies ( see above ) .
we refer the reader to @xcite for more details about the comparison between simulations and observations .
this comparison suggests that at least 33% of [email protected] galaxies ( 50% of galaxies not relaxed morpho - kinematically according to the morpho - kinematic classifications @xmath56 the ratio of secured models , which is 18/27=66% ) are involved in major mergers with a median mass ratio @xmath13 .
all the merger phases that are expected to lead to a significant sfr enhancement , i.e. , the pre - fusion , fusion / post - fusion , and relaxation phases @xcite , are homogeneously populated as a function of time by the images - cdfs galaxies , as evidenced by fig .
[ figtime ] . in this figure
, sfrs were homogenized by dividing the observed sfr by the estimated mass of gas in each galaxy and multiplying each derived number by the average mass of gas in the sample .
the time sequence reproduces quite well expectations from modeling the star formation during a typical merger ( see insert ) .
this shows that morpho - kinematic data such as those provided by the images survey are sensible to all the phases of a gas - rich major merger event . amongst the galaxies populating the relaxation phase , six of them are rotating ,
but they are clearly distinct of local rotating disks because of their high average star formation rate , and higher gas velocity dispersion @xcite .
the latter suggests that these galaxies are possibly observed during a post major merger relaxation phase ( @xcite ; see also fig .
[ figpot ] and sect .
most of the star formation activity is indeed concentrated at the disk outskirts , which suggests an ongoing inside - out growth of their disks @xcite .
the time sequence in fig . [ figtime ] empirically accounts for a cosmological distribution of merger parameters , since it is constructed using a representative sample .
it can thus be used to estimate accurately the evolution of the merger fraction and merger rate as follows .
theory usually gives the merger rate at the time when halos start merging , i.e. , where galaxies that inhabit these halos are still in pair .
the epoch when a given merging system was still a pair is dated using individual numerical models from @xcite .
an homogeneous and self - consistent determination of the merger rate as a function of lookback time can be estimated by simply counting the fraction of galaxies in each phase and dividing this number by the corresponding average visibility timescale provided by the modeling of images galaxies .
this method assumes that the visibility timescales do not evolve with redshift .
such an assumption is consistent with semi - analytic models , which predict almost no evolution of the merger visibility timescale as a function of lookback time , at least within uncertainties @xcite .
another possible caveat is the observed evolution of the galaxy gas fraction with redshift @xcite , which could potentially result in evolving timescales depending of the merging phase probed @xcite . however , hydrodynamical simulations of binary mergers @xcite predict that unequal mass ratio mergers ( which dominate the empirical distribution of merger mass ratios , see sect .
3.2 ) generally show no evolution of the pair visibility timescale with gas fraction . depending on whether or not the two progenitor candidates potentially involved in the merger could be identified visually in the hst / acs images ( e.g. , when two candidates for the progenitor nuclei could be identified ) , the secured major merger candidates were split into pre - fusion ( 8 gal . ) or post - fusion phases ( 10 gal . ) . amongst the objects that could not be secured as major merger candidates using numerical models , three were found to correspond to the pre - fusion phase , since they reveal distinct potential progenitors .
they were therefore added to the pre - fusion phase .
we attribute the fact that their numerical models did not succeed in reproducing their morpho - kinematical properties with sufficient precision to the limited exploration of the parameter space .
the pre- and post - fusion phases as observed at [email protected] correspond to galaxies that were on average in pairs ( and before the first passage ) at earlier epochs .
to date these epochs , we simply averaged the time elapsed since the beginning of the merger model for each of the 18 secured major merger candidates , and converted these numbers into a redshift .
all numbers were weighted by the respective grade given to the different models .
it was found that the pre - fusion phase corresponds on average to [email protected]@xmath57 ( i.e. , [email protected] gyr earlier than z=0.6 ) , while the post - fusion phase corresponds on average to [email protected]@xmath58 ( i.e. , [email protected] gyr earlier than z=0.6 ) .
uncertainties were estimated using bootstrap resampling .
the fraction of mergers in emission - line galaxies at z=0.4 - 0.75 for the pre - fusion phase is 0.5x11/33 = 16.7% ( with the factor 0.5 accounting for the fact that only one galaxy in the pair is part of the sample ) , while the fraction of mergers in emission - line galaxies for the post - fusion phase is directly given by 10/33 = 30.3% .
since 60% of intermediate - mass galaxies at [email protected] are emission - line galaxies @xcite , this translates into a total fraction of mergers of 10.0% and 18.1% for the pre- and post - fusion phases , respectively .
this assumes that major mergers produce a star formation activity that results in @xmath13)\geq 15\aa$ ] .
however , it has been shown that major mergers can lead to modest sfr enhancement over relatively short timescales @xcite .
this assumption can therefore lead to possible systematic effects that are considered separately ( see sect .
random uncertainties due to statistical fluctuations in the respective sub - samples ( see tab . [ rates ] ) are derived accurately using a beta distribution generator as an estimator of confidence intervals for the binomial statistics @xcite .
the 18 secured pre- and post - fusion major merger models were used to determine the fusion time for each galaxy .
the fusion time was defined as the interval of time between the beginning of the simulation ( i.e. , when the two progenitors are in pair and before the first passage ) and the time at which the two nuclei merge according to the best - fit model .
these individual timescales were averaged to mitigate uncertainties due to statistical fluctuations in the respective sub - samples . the averaged pre - fusion phase is found to be [email protected] gyr long , which gives an estimate of the pre - fusion visibility timescale in the images - cdfs sample .
the uncertainty was derived by bootstrap resampling of the grade - weighted distribution of individual timescales .
we assume the same visibility timescale for the post - fusion phase , since both appear to be similar in the images sample , according to fig .
[ figtime ] .
we compare this estimate with hydrodynamical simulations including radiation transfer @xcite .
we selected the g2 and g3 models as being the closest to the images - cdfs galaxies in terms of baryonic mass and gas fractions .
indeed , the most important parameters driving the result of a major merger are the baryonic mass ( as a proxy for the total mass inside the optical radius ) and the gas fraction @xcite .
for these two models , the timescale from the beginning of the interaction to the fusion of the two nuclei is found to be 1.24 and 2.44 gyr respectively , with an average @xmath11.84 gyr , in good agreement with our estimate of [email protected] gyr .
we also compared the pre - fusion visibility timescale with an average pair visibility timescale for large projection distances ( i.e. , @xmath59 kpc@xmath60 ) derived from similar simulations @xcite , which is found to be [email protected] gyr .
taking into account predictions of mass ratios from galaxy formation models slightly reduces the average value to 1.5 - 1.6 gyr @xcite .
the pre - fusion phase includes galaxies that are too close to be selected as pairs by usual criteria ( see also sect .
it is therefore expected that the pre - fusion visibility timescale in the images - cdfs sample is slightly larger than the average pair visibility timescale .
the six rotating spiral disks in the images - cdfs can be used to estimate the fraction of mergers at even earlier epochs ( see sect . 3.2 and 4.1 ) .
their mass doubling timescale @xmath61 is provided by the ratio between their total sfr and stellar mass @xcite .
note that this timescale is independent of the imf .
one of them shows a very large @xmath61 ( @xmath17 gyr ) and was therefore considered as an interloper and discarded in the following . assuming that these objects are merger remnants
, one can attribute most of their star formation activity to the formation of the disk ( see sect . 4.1 and 6.2 ) .
therefore , this timescale can be used to roughly estimate the average time elapsed from fusion to the observation epoch of these galaxies . on average , these galaxies were therefore observed as pairs at a lookback time 1.8 + 1.8 ( pre - fusion + post - fusion phases ) + 0.5x(@xmath62 - 1.8 ) = [email protected] gyr , with @[email protected] gyr ( where the error - bar was estimated using bootstrap resampling ) .
this corresponds to [email protected]@xmath63 .
the average relaxation timescale is found to be 5.5 - 1.8 - 1.8=1.9 gyr ( see fig .
[ figtime ] ) . amongst the nine non- or semi - relaxed galaxies in the images - cdfs sample that could not be securely identified as major mergers , six were found to be in the post - merger phase according to their best models .
they were all modeled quite at the end of the 3 gyr duration of the simulations , which suggests that the morpho - kinematics of these galaxies might not have been identified as resulting from a major merger due to the limited duration of the simulations . amongst them , four show evidence for rotation in their velocity fields and were classified as perturbed rotators , which is consistent with this interpretation . we therefore included them as part of the relaxation phase ( see red points in the relaxation phase in fig .
[ figtime ] ) .
this leads to a fraction of mergers at [email protected] of 20.0% ( see tab . [ rates ] ) .
the merger rate r [ gyr@xmath64 corresponding to the three merger phases considered above ( i.e. , the pre - fusion , post - fusion , and relaxation phases ) can be estimated using the ratio between the corresponding merger fraction and visibility timescale ( see tab .
[ rates ] ) .
error - bars due to random fluctuations in the sub - samples were estimated by propagating the random uncertainties on the fraction of mergers and on the visibility timescale .
the derived merger rates and their associated uncertainties are shown in fig .
[ figrate ] .
cccc merger phase & fraction of mergers ( % ) & merger rate ( % .@xmath65 ) & redshift + pre - fusion & @xmath66@xmath67 & @xmath68@xmath69 & @xmath70 + post - fusion & @xmath71@xmath72 & @xmath73@xmath74 & @xmath75 + relaxation & @xmath76@xmath77 & @xmath78@xmath79 & @xmath80 + for selection criteria similar to those of the images - cdfs sample , a fraction of galaxies in pair @xmath12 - 6% at [email protected] is generally found @xcite . in the present study ,
the fraction of galaxies involved in major mergers is found to be 10@xmath303% at [email protected] .
such a fraction is larger than the fraction of galaxies in pairs since the pair technique only picks up progenitors when they are approaching for the first time and do not consider progenitors between the first and second passage and/or when they are getting too close one from each other , while morpho - kinematic data are sensitive to all phases .
this is reflected in the visibility timescale associated with pair fractions , which is often assumed to be @xmath10.5 gyr from analytic arguments or independent simulations ( i.e. , which do not necessarily reflect the exact properties of the observations ; see also @xcite ) .
using such a timescale and a fraction of galaxies in pair @xmath12 - 6% indeed leads to a merger rate @xmath14 - 12% gyr@xmath81 , which is consistent with our [email protected] estimate of 5.5% .
when no lower limit to the distance between the two progenitors is imposed , the pair technique gives very similar results , with @xmath19% of galaxies in pair at [email protected] , and @xmath115% at z@xmath11 @xcite .
the morphological technique provides estimates of the fraction of morphologically - disturbed mergers of 7 - 11% at [email protected] , 11 - 17% at [email protected] , and 27@xmath3011% at [email protected] for similar mass ranges and
depending and the method used ( cas / gm system , see @xcite , or visual inspection , see @xcite ) . within uncertainties
, this is consistent with the fraction of mergers found using morpho - kinematic data ( see tab . [ rates ] ) .
we conclude that alternative techniques generally give consistent estimates , although with a resulting large range of values @xcite , which reflects systematic effects due to their limited sensitivity to the merging process and different sample selection effects ( see also @xcite ) .
a first possible cause of systematic effects is linked to the selection criteria of the images - cdfs sample , and in particular to the criterion @xmath13)\geq 15\aa$ ] .
indeed , we assumed that all quiescent ( @xmath13 ) < 15\aa$ ] ) galaxies are not major mergers , whereas a significant fraction of them ( 25% ) were found to have peculiar morphologies @xcite , which could be linked to less active mergers .
indeed , numerical simulations have revealed that major mergers do not always lead to strong enhancements of the star formation rate @xcite , and therefore these peculiar quiescent galaxies might also be undergoing major merger events .
this could potentially rise the fraction of mergers by 10.0% since quiescent galaxies represent 40% of intermediate - mass galaxies at [email protected] @xcite . since we do not have kinematic data for these objects
, this could potentially affect any of the three merger phase considered in this study .
we therefore considered that , on average , the merger fraction in each phase could be affected by a @xmath13.3% systematic effect upward ( see tab . [ rates ] ) .
the number of mergers in the pre - fusion phase could also be overestimated due to fly - by interlopers .
we estimated that four out of the eleven objects in this phase could be possible fly - byes and escape a subsequent merger ( j033217.62 - 274257.4 , j033219.32 - 274514.0 , j033220.48 - 275143.9 , j033238.60 - 274631.4 ) .
these objects correspond to cases where the two potential progenitors are well separated and photometric redshifts are consistent , but no reliable spectroscopic information was available to confirm whether or not the two galaxies were indeed gravitationally bound .
this can potentially have an impact on the fraction of mergers in the pre - fusion phase by 3.7% downward , and the corresponding merger rate by 2.0% ( see tab .
[ rates ] ) .
six galaxies in the images - cdfs are consistent with expectations from clumpy galaxies @xcite , i.e. , gas - rich toomre - unstable disks .
these six galaxies are found to be in the post - fusion phase ( 3 out of 6 ) .
since the numerical merger models do not consider small - scale perturbations such as clumps , these galaxies might correspond to toomre - unstable disks driven by internal instabilities , rather than major mergers .
the post - fusion fraction of merger and merger rate might therefore be affected by a 5.5% and 3.3% effect downward , respectively ( see tab .
[ rates ] ) .
the impact on the pre - fusion and relaxation phases is marginal ( one object ) compared to other uncertainties , so we did not consider it further . regarding the relaxation phase , we added six objects corresponding to the post - fusion phase but which were not safely identified as mergers .
this could be due to insufficiently long evolution times , so that the objects could not be safely identified , or to a too limited exploration of the parameter space .
considering these six galaxies as possible interlopers for the relaxation phase , this would impact the corresponding fraction of merger and merger rate by a 9.1% and 5.1% effect downward , respectively ( see tab .
[ rates ] ) .
systematic uncertainties are shown as a grey region in fig .
[ figrate ] .
instead of trying to predict galaxy properties all the way from initial conditions given by the cmb , semi - empirical models by - pass the unknown coupling between dark and baryonic matter using empirical relations , such as gas fraction or stellar mass functions as a function of redshifts .
such an approach allows us to overcome several issues encountered with semi - analytic models , which can lead to significantly underestimate the merger rate @xcite . the semi - empirical model used here @xcite
was shown to reproduce a number a local galaxy properties , such as their bulge - to - disk flux ratio distribution , tully - fisher relation , or luminosity function of luminous infrared galaxies @xcite .
we used the @xcite semi - empirical modelphopkins / site/ + mergercalc.html as an idl code . ] to predict the merger rate in a given range of baryonic mass [ @xmath82,@xmath83 , gas fraction [ @xmath84 , @xmath85 , and baryonic mass ratio between the two progenitors [ @xmath86,@xmath87 as a function of redshift @xmath88 .
indeed , @xcite emphasized that the most appropriate mass would be the tightly bound material that survives stripping to strongly perturb the primary , which is found to be the baryonic plus tightly bound dark matter mass within a few disk scale lengths .
the best observable proxy for this quantity is the baryonic mass @xcite , which was therefore used in this study . in most studies ,
the galaxy merger rate is estimated in a fixed range of stellar or baryonic mass , constant with redshift , i.e. , [ @xmath82,@xmath83=[@xmath89,@xmath90 . here ,
we need to estimate the proper range of baryonic mass [ @xmath82,@xmath83 at all @xmath88 in order to sample the progenitors and descendants of the images - cdfs galaxies . given [ @xmath82,@xmath83 , we determined the corresponding range of halo mass [ @xmath91,@xmath92 using the relationship between baryonic and halo masses @xmath93 as provided by the @xcite halo occupation model ( hod ) using the mass function from @xcite . using alternative mass functions change the merger rate by less than 1% .
we then estimate the average mass of the progenitor halos at @xmath94 following @xcite , i.e. , [ @xmath95,@xmath96 , which is then converted back to [ @xmath97,@xmath98 using the @xmath99-@xmath100 relationship . the hod model is used to estimate the galaxy merger rate at @xmath94 in the updated range of baryonic mass .
this ensures that the derived galaxy merger rate is integrated over the population of the progenitors of all galaxies within [ @xmath101,@xmath102 at @xmath103 only , and not over all the mass spectrum , which would overestimate the merger rate compared to estimates from observations . in practice , we set [ @xmath104 , @xmath105@xmath106[@xmath107,@xmath108 in order to sample the range [ @xmath109,@xmath110=[@xmath111,@xmath112 , which corresponds to the baryonic mass range of the images - cdfs sample ( see sect .
2 ) . we tested that following the mean progenitor mass instead of the average progenitor mass between the upper and lower limit of the mass range does not change the resulting merger rate significantly . to test the consistency of the adopted hod model with the images sample
, we checked that , according to the hod model , the range of baryonic mass sampled at z=0.6 corresponds to a range in @xmath33=[@xmath113,@xmath112 , which also matches the range of observed stellar mass in the images sample ( see sect .
2.1 ) . at z=0 ,
the hod model predicts that the z=0.6 progenitors as defined by their range of baryonic mass as above , will end up in a stellar mass range of @xmath33=[@xmath114,@xmath108 , corresponding to the baryonic mass range of @xmath115=[@xmath107,@xmath108 as quoted above .
the evolution in stellar mass as predicted by the hod model is therefore @xmath10.1 - 0.3 dex .
the median gas fraction in the progenitors of the images - cdfs sample were estimated using their observed sfrs @xcite .
we used this to parameterize the expected gas fraction in the images - cdfs progenitors and descendants , using @xmath116=\{15%,32%,50%,75%,90% } at @xmath88=\{0,0.6,0.72,1.12,1.55}. estimates at @xmath88=0 and 0.6 were derived using local hi measurements and inversion of the schmidt - kennicutt relation , respectively @xcite .
estimates at @xmath88=0.72 , 1.12 , and 1.55 were derived from the estimated median gas fractions in the pre - fusion , post - fusion , and relaxation phase , respectively @xcite .
the hod model predicts that the progenitors of [email protected] on - going mergers in the images - cdfs sample are galaxies with stellar mass in the range @xmath117 [ @xmath118,@xmath107 ] .
gas fractions of the order of @xmath190% in such a range of mass at [email protected] is consistent with observational estimates of the gas fraction in more [email protected] massive galaxies ( see discussion in sect .
we added a @xmath301-@xmath31 scatter of 0.15 dex , which appears to represent well the observed scatter at all @xmath88 @xcite . at given @xmath88 , the range of gas fraction [ @xmath84,@xmath85 over which the merger rate was integrated
is interpolated from the @xmath120 relationship as defined above , with an added @xmath302-@xmath31 scatter .
we explored the impact of a larger @xmath303-@xmath31 scatter with no qualitative change in the results .
finally , we set [ @xmath121,@xmath87=[@xmath122,@xmath123=[0.25,1.0 ] , as found in the images - cdfs sample @xcite .
figure [ figrate ] compares the derived estimates of the merger rate with state - of - the - art semi - empirical @xmath0-cdm models @xcite .
this theoretical model is found to be in remarkable agreement with the observed estimates without any particular fine - tuning . at the redshifts and masses probed by the images - cdfs sample , semi - empirical ( hod - based )
models can predict the merger rate within a factor @xmath11.5 @xcite .
hod models are in relatively good agreement one with each other because they rely on observational constraints ( e.g. , the galaxy stellar mass function ) that are well constrained at z@xmath151 and for stellar masses around @xmath124 @xcite .
as stated , above , changing the galaxy stellar mass function changes the resulting merger rate by less than @xmath11% .
the dominant source of random uncertainty is therefore due to variations from one ( hod ) model to another , which is illustrated in fig .
[ figrate ] as a red region .
this is the same order of magnitude that random uncertainties associated to the observed merger rate ( see tab .
[ rates ] ) .
this implies that random uncertainties limit the comparison between observations and theory at a level of a factor @xmath12 . in this comparison
, we adopted a diet salpeter imf for all observables at any epoch .
several studies suggest that the imf might become top - heavy under extreme starburst conditions @xcite .
this could result in an evolution of the imf with redshift , which would impact the estimation of stellar mass by up to @xmath1 -0.18 dex ( rising almost linearly with redshift from z=0 to [email protected] ) compared to estimates at z=0 @xcite .
this could result in a systematic shift of the merger rate with redshift .
several systematic effects associated with the stellar population models can also affect the determination of the stellar mass .
the most significant effect is the influence of the tp - agb phase , which omission in stellar population models as those used in the present study can lead to overestimate the stellar mass by 0.14 dex @xcite .
in addition , using simplified prescriptions for deriving the stellar - mass instead of full sed modelling @xcite , when used at high redshifts , can also potentially lead to overestimate stellar masses by 0.2 dex @xcite . in order to bracket the upper limit of the predicted merger rate at all redshifts
, we combined quadratically the systemic uncertainty associated with a possible evolution of the imf , and the effects of stellar population models , since they are independent .
this results in a total systematic uncertainty of -0.27 dex on the baryonic mass . in fig .
[ figrate ] , we plotted theoretical predictions with [ @xmath82,@xmath83 - 0.27 , using a dashed - red line . in doing so ,
[ @xmath84,@xmath85 was updated accordingly ( upward ) .
all these systematic uncertainties associated to the estimation of the stellar mass tend to decrease the galaxy merger rate at all redshifts .
however , the theoretical predictions remain consistent with the observational estimates within the corresponding range of observational systematic uncertainties .
quantitatively , this could decrease the integrated merger rate by @xmath120% .
we conclude that the major merger rate is determined from both theory and observations within a factor 2 - 3 , accounting for random and systematic uncertainties .
such an agreement between the merger rate determined using morpho - kinematic data and theoretical predictions without any particular fine - tuning , is remarkable .
this is an improvement of a factor @xmath13 compared to previous comparisons between observations and theory @xcite .
using morpho - kinematic data of [email protected] galaxies from the images - cdfs survey , we derived the evolution of the major merger rate out to [email protected] , i.e. , nine billion years ago . using comparisons of the spatially - resolved morpho - kinematic properties between observations and a grid of hydrodynamical simulations of galaxy mergers ,
we constrain their time evolution , which allowed us to accurately date back when they were still in pair . in this comparison ,
the estimate of the merger rate at [email protected] is much more uncertain since it relies on the relaxation phase during which it is more difficult and/or degenerate to interpret morpho - kinematic features as merger relics .
however , it makes sense to interpret such galaxies as post - mergers in the relaxation phase as detailed above , and fig .
[ figpot ] & [ figtime ] indeed suggest a plausible causal link between distant merging galaxies and such dynamically hot rotating disks .
we found a remarkable agreement between these new observational estimates and predictions from state - of - the - art semi - empirical @xmath0-cdm models within a factor 2 - 3 , accounting for both random and systematic uncertainties .
as a test of consistency , the @xmath0-cdm model predicts a large fraction of peculiar galaxies at z=[0.4 - 0.75 ] , which is found to be @xmath125 ( i.e. , the average merger rate at z=0.65 , @xmath10.06 gyr@xmath81 , see fig .
[ figrate ] ) @xmath126 ( i.e. , the total duration of the average major merger at z=0.65 , @xmath15.5 gyr , see fig .
[ figtime ] ) @xmath1 33% .
this corresponds exactly to the fraction of observed galaxies in which the peculiar morphologies and/or kinematics can be reproduced by a major merger model ( see sect .
the @xmath0-cdm model therefore predicts that @xmath133% of the progenitors of local disks were destroyed over the past 9 gyr during major ( i.e. , baryonic mass ratio @xmath201:4 ) merger events , as observed .
the occurrence of 33% of gas - rich major mergers in [email protected] galaxies is also consistent with an increase of the scatter of the tully - fisher relation at high redshifts @xcite , and the observed random walk in the specific angular momentum of [email protected] galaxies compared to local galaxies @xcite .
the agreement found between observations and theoretical predictions ( see fig .
[ figrate ] ) evidences that 50% ( 65% ) of local galaxies with stellar mass in the range 10@xmath127 - 10@xmath128m@xmath18 ( see sect .
5.1 ) have undergone a major merger ( defined here as baryonic mass ratio @xmath200.25 ) since z=1.5 ( z=2.0 ) . the @xcite model takes into account the specific physics involved in gas - rich major mergers , which was shown to reproduce the observed local relation between bulge - to - disk ratio and stellar mass @xcite .
this suggests that the time at which local galaxies have undergone their last major merger could be an important factor in determining how they acquired their internal structure ( b / t ratio and angular momentum ) .
depending on the gas fraction at fusion time , major mergers are well - known to result in ellipticals ( through dry mergers ) , but also in spirals ( through gas - dominated mergers ) .
indeed , hydrodynamical simulations of isolated merging disk galaxies have shown that with a fraction of gas at the fusion time larger than 50% , a significant disk can reform in the remnant @xcite .
note that the required fraction could be smaller with significant cosmological gas accretion @xcite , but we will adopt 50% as a conservative limit .
the median gas fraction at z=0.6 is found to be @xmath130% @xcite .
it is therefore likely that major mergers occurring between [email protected] and z=0 ( which involved 20% of local galaxies according to fig .
[ figrate ] ) were relatively dry , probably resulting in galaxies with significant bulges .
the major mergers sampled by the images - cdfs sample are those occurring between [email protected] and [email protected] ( see tab .
[ rates ] ) . fig .
[ figrate ] shows that 50% of local galaxies were involved in a major merger since z=1.5 , of which 30% occurred between z=0.6 and z=1.5 .
@xcite have extrapolated the gas fraction in the progenitors of these mergers , and found gas fractions well in excess of the required value of 50% to reform a disk ( @xmath190% , see sect .
several measurements of the gas fraction were attempted at [email protected] @xcite .
most measurements involve relatively massive galaxies ( @xmath1m@xmath129m@xmath18 ) and are limited to their half light radius .
however , the hod model predicts that the progenitors of the on - going z@xmath1 0.6 major mergers at z=1.5 should be in a lower range of stellar mass @xmath110@xmath130 - 10@xmath131m@xmath18 .
@xcite homogenized gas estimates from the literature to parameterize the evolution with redshift and stellar mass of the gas fraction within the full gas radius .
extrapolating their gas fraction - stellar mass relation at [email protected] in the range of mass 10@xmath130 - 10@xmath131m@xmath18 , one finds that gas fractions at a level of @xmath1 80 - 100% are probably close to average conditions ( see fig
. 7 of @xcite ) .
the progenitors of all the [email protected] on - going major mergers at z=[0.75 - 1.5 ] should therefore be rich enough in gas so that a significant disk reform after fusion , as also evidenced by observations of young dust - enshrouded disks at [email protected] @xcite .
we emphasize that we have considered here only the `` average '' results of major mergers since z@xmath12 and such statements have to be taken at the lowest order only , since one expects significant scatter in the b / t of a major merger remnant as a function of orbital parameters , gas fraction , and feedback @xcite .
we also emphasize that we considered here only a specific range of mass and that the impact of more modest mergers , i.e. , with baryonic mass ratio smaller than 0.25 was not considered .
results of semi - analytic models suggest that the likeliest channel for the formation of disk - dominated galaxies is a relatively quiet merger history ( e.g. , see discussion and references in @xcite ) .
of particular interest are local `` bulgeless galaxies '' ( which we define here as galaxies with b / [email protected] , as generally considered ) , which are often thought to have undergone their last major merger above z@xmath12 . the relatively large fraction of such galaxies
is often considered to be in tension with the fraction of major mergers as inferred from predictions of @xmath0-cdm models ( e.g. , @xcite ) .
it was argued in sect .
6.2 that most descendants of images - cdfs galaxies should lie within early - type spirals .
we now examine whether the fraction of mergers found in fig .
[ figrate ] is in tension with the observed fraction of local bulgeless galaxies .
@xcite used bulge+bar+disk decompositions on h - band images and found that two - third of local spirals with m@xmath19=10@xmath132 - 10@xmath133m@xmath18 are bulgeless .
their determination maximizes the fraction of bulgeless galaxies since they removed the contribution of the bar to the central light and followed the common assumption that bars are pure products of secular evolution and not merging ( see discussion in @xcite ) . since the fraction of spirals amongst local galaxies is @xmath170% ( see app .
b ) , this would translate into a maximal fraction of bulgeless galaxies in the local universe @xmath146% .
as @xcite argued , their sample appears to be representative of the b - band luminosity function at m@xmath13410@xmath132m@xmath18(see their fig .
2 ) . however , their fig .
3 evidences that their sample can be representative of stellar mass function only above m@xmath13410@xmath135m@xmath18 . in this range of mass , the maximal fraction of local bulgeless
is found to be 61% , or 43% of local galaxies . assuming that such galaxies can not be formed by mergers ,
this would therefore limit the fraction of major merger in local galaxies since z=2 to 57% .
we re - derived fig .
[ figrate ] for the range of mass m@xmath19=10@xmath135 - 10@xmath133m@xmath18 .
indeed , the hod model used in the present study predicts that on - going major merger in the [email protected] images - cdfs sample should end up in a stellar mass range of @xmath114-@xmath136m@xmath18 , which is shifted to smaller mass relative to the range @xmath137-@xmath138m@xmath18 in which the @xcite sample is representative . in this higher range of mass
, we found that the fraction of local galaxies that have undergone a major merger since z=2 decreases from 65% to 53% . within uncertainties , this matches quite well the maximal fraction of bulgeless galaxies found by @xcite in the same range of mass ( 57% ) .
we conclude that there is no tension between the fraction of mergers inferred from morpho - kinematic data and the fraction of local bulgeless galaxies . in other words , there is no need for disks to `` survive '' major mergers in the @xmath0-cdm paradigm , at least at z@xmath152 . _ the results of this paper point toward a re - building of half of local disks during the past 9.4 gyr after their last gas - rich major merger @xcite .
_ this does not imply that the evolution of the star formation and stellar mass densities are primarily driven by such processes over the same period @xcite .
these results are consistent with recent developments of the @xmath0-cdm model at earlier lookback times ( i.e. , [email protected] ) , according to which galaxy formation would be mainly driven by streams of cold gas , rather than by major mergers @xcite .
indeed , cosmological simulations predict that cold flows are strongly suppressed below z@xmath11 - 1.5 @xcite .
such an epoch might correspond to a transition between two different mechanisms driving galaxy formation , i.e. , cold gas streams and gas - rich major mergers , as suggested by the strong evolution of the fraction and properties of clumpy galaxies as a function of redshift @xcite .
in such simulations , halos with a relatively quiet merger history are usually selected as the best probes for the formation of local thin disks ( e.g. , @xcite ) .
the present results suggest that gas - rich major mergers could be be an equally plausible way of forming large thin disks .
further work will be needed to quantify what is the main reservoir of gas from which such thin disks can reform .
potential sources of gas include gas expelled during the merger and which falls back onto the remnant @xcite , gas coming from external gas accretion @xcite , and/or gas recycled through stellar feedback @xcite .
this emphasizes the need to further explore how such disks can ( re-)form in halos with a more recently active merger history @xcite .
interestingly , recent cosmological simulations conducted with state - of - the - art hydrodynamical codes reveal cold gas disks that have larger extent and specific angular momentum compared to disks formed in former sph cosmologicial simulations @xcite .
these simulations show that massive halos ( i.e. , @xmath139 ) have significantly larger sfrs at z=0 - 2 , compared to previous simulations @xcite .
using morpho - kinematic data of z=[0.4 - 0.75 ] galaxies from the images - cdfs survey , we derived the evolution of the major merger rate out to [email protected] , i.e. , nine billion years ago . using comparisons of the spatially - resolved morpho - kinematic properties between observations and a grid of hydrodynamical simulations of galaxy mergers ,
we constrain their time evolution , which allowed us to accurately date back when they were still in pair .
we then compared constrains from observations with predictions from semi - empirical models .
our main conclusions are as follows . 1 .
morpho - kinematic observations are found to be sensitive to all the phase of the merger process , from the pre - fusion phase in which galaxy nuclei are still separable by visual inspection , to the post - fusion and relaxation phases , in which the two progenitors merge and progressively reach equilibrium .
each of the three phases are found to last @xmath11.8 gyr .
the major merger rate ( defined as baryonic mass ratio @xmath200.25 ) inferred from morpho - kinematic observations is found to be evolving from 5.5%.gyr@xmath81 to 11.1%.gyr@xmath81 between z=0.72 to z=1.55 in the range of stellar mass 10@xmath140 - 10@xmath141m@xmath18 ( diet salpeter imf ) , or 10@xmath142 - 10@xmath141m@xmath18 in baryonic mass .
3 . a remarkable agreement is found between observational estimates and predictions from the @xcite semi - empirical model , which accounts for the specificity of gas - rich mergers at high redshifts .
the agreement is found to be at a level of a factor 2 - 3 without any particular fine - tuning of the model , accounting for both random and systematic uncertainties on the observational and theoretical sides , which significantly improves the accuracy of previous comparisons .
the @xmath0-cdm model predicts , in agreement with observations , that 33% of [email protected] intermediate - mass galaxies are on - going major mergers .
observational signatures of these mergers encompass morpho - kinematic disturbances in a large fraction of these galaxies at this redshift , and an increase of the scatter of scaling relations between mass , velocity and radius ( or angular momentum ) .
the @xmath0-cdm model predicts , in agreement with observations , that 20% / 50% / 65% of local galaxies in the range of stellar mass 10@xmath127 - 10@xmath128m@xmath18 were involved in a major merger between z=0.6 / 1.5 / z=2.0 and z=0 .
most of the major mergers occurring between z=0.6 and z=0 were probably relatively dry , resulting in galaxies with a prominent bulge , while most of the major mergers occurring between z=1.5 and z=0.6 probably involved large gas fractions , which should lead to the rebuilding of a significant disk .
theoretical models often suggest that `` bulgeless galaxies '' ( b / [email protected] ) result from relatively quiet merger histories .
we found no tension between the fraction of local bulgeless galaxies and the fraction of local galaxies that have undergone a major merger since z=2 , i.e. , there seems to be no `` disk survival issue '' but rather a `` disk rebuilding '' era , at least at [email protected] - 2 , when the influence of cosmological gas accretion predicted by theory and simulations would start declining in relatively massive haloes .
this work was partly supported by the `` laboratoire international associ '' origins .
we thank an anonymous referee for her / his careful reading of the paper and thorough comments .
we present here the i - band hst / acs images and kinematic maps for a sub - sample of the images - cdfs , see fig .
galaxies are the same that those used to illustrate the sfr vs. time sequence of fig .
[ figtime ] , in the same ( i.e. , temporal ) order , from top to bottom .
galaxies with no major merger model are rds .
we discuss here the evolution of the fraction of morphological types between [email protected] and z=0 in the range of stellar mass which encompasses the descendant of the images - cdfs galaxies .
we estimated the fraction of each morphological type in local galaxies as follows .
we first considered morphologically determined luminosity functions ( mdlf ) of local galaxies from @xcite .
they indeed constructed lfs for e / s0 , s0/a - sb , sbc - sd , and i m galaxies using @xmath11500 galaxies from the sdss using the r band luminosity .
we first converted their observed r band luminosity into stellar mass using prescriptions from @xcite , which gives a mean stellar mass / r - band luminosity ratio in the local volume of 3.05 within @xmath3030% of systematic uncertainty ( diet salpeter imf ) .
we then integrated each of the mdlf ( once converted into the proper imf ) from 10@xmath127 to 10@xmath128m@xmath18 , within which the descendants of the on - going [email protected] major mergers should end up , as predicted by the hod model ( see sect .
this results in the fractions of morphological types as listed in tab . [ types ] . deriving the fraction of morphological type that way
has the advantage of avoiding systematics associated with the representativity of the sample , and is limited only by the statistical scatter associated with the visual classification of @xcite and the systematic uncertainty associated to the mean stellar mass - to - light ratio in the local volume as determined by @xcite . as a consistency check
, we also derived the fraction of morphological types in the local universe using the catalogue of @xmath114000 visually classified galaxies in the sdss from @xcite .
we selected in their catalogue galaxies in the range of mass 10@xmath127 - 10@xmath128m@xmath18 , and with redshift @xmath143 0.05 .
according to their fig .
1 , this select a volume - limited sample in the stellar mass range of interest for our purposes .
@xcite emphasized that such a cut in @xmath88 will remove some of the brightest , most massive galaxies in the local volume , but these galaxies are all above the range of mass of interest , so this will not impact the morphological fractions in this range .
derived fractions are also listed in tab .
[ types ] .
it is interesting to compare them with those derived by integrating the mdlfs in the local volume , since both derivations rely on completely different selection effects .
we also listed results from @xcite , which selected a sample of the catalogue from @xcite , who visually classified a sample of more than 2000 galaxies from the sdss .
their sub - sample was selected to be volume - limited with m@xmath13410@xmath127m@xmath18 to specifically match the descendants of the galaxies sampled by the images survey .
ccccc morph .
type & fraction & fraction & fraction & fraction + & at z=0 & at z=0 & at z=0 & at [email protected] + & from n03 & from n10 & from ds10 & from ds10 + e / s0 & 28@xmath303% & 31% & 18@xmath305% & 17@xmath303% + all sp & & & & + s0/a - sd & 74@xmath303% & 68% & 72@xmath308% & 31@xmath307% + early - type sp & & & + s0/a - sb & 54@xmath304% & 45% & & + late - type sp & & & & + sbc - sd & 20@xmath301% & 23% & & + i m / pec & 2@xmath301% & @xmath1441% & 10@xmath303% & 52@xmath309% + the comparison between these three different determinations of the fraction of morphological types in the local universe _ in the same mass range _ reveals interesting features .
first , the fraction of spirals is found to be in remarkable agreement , around @xmath170% .
there are some variations within the fractions of early - type spirals , probably reflecting the uncertainty associated to eyeball classification between s0 and s0a galaxies .
the second interesting feature is the smaller fraction of e / s0 found by @xcite in conjunction with their higher fraction of peculiar galaxies compared to other catalogues .
this was attributed by @xcite to a number a factors that include ( 1 ) their systematic use of color maps to identify unusual color gradients , ( 2 ) the quantification of the misalignment between bulge and disk centroids , and ( 3 ) the compactness of galaxies , which were moved to the peculiar class if too compact to be decomposed into bulge and disk .
galaxies harboring such features were classified as peculiar by @xcite , but are probably in the e / s0 class in other catalogues that do not take into account such an information .
this suggests that @xmath110% of the e / s0 classified by @xcite or @xcite are actually perturbed and not fully relaxed .
such an interpretation is consistent with the fraction ( @xmath130% ) of early - type galaxies showing negative color gradients ( `` blue - cored '' early - types ) in the local universe in the sdss @xcite . at face value ,
the fraction of late - type spirals in tab .
[ types ] , i.e. , 20 - 23% might appear in conflict with the fraction of bulgeless spirals derived by @xcite , which is found to be @xmath160% ( see sect .
we recall that `` bulgeless '' means here b / [email protected] and that the correlation between b / t and morphological type has very large scatter ( see , e.g. , fig .
14 of @xcite ) . selecting galaxies with morphological types in the range sbc - sd in the sample of @xcite
leads to a fraction of late - type galaxies @xmath133% ( above m@xmath19=10@xmath135m@xmath18 ) . since spirals account for @xmath170% of local galaxies ( tab .
[ types ] ) , this translates into a fraction of @xmath123% of late - types amongst local galaxies , which is fully consistent with fractions listed in tab .
[ types ] .
we also listed in tab .
[ types ] the fraction of morphological types found by @xcite at [email protected] .
note that the fractions of peculiar galaxies , i.e. , those that do not fit into the hubble sequence , are at both redshifts in very good agreement with those found by @xcite .
@xcite used exactly the same classification method and strictly equivalent data to classify local and distant galaxies , which minimizes systematics when comparing the two epochs . when one uses the same method , it is found that the fraction of fully relaxed e / s0 does not evolve between [email protected] and z=0 .
compared to other catalogues , an evolution of the fraction of e / s0 of the order of @xmath110% is found , but this can be attributed to systematics associated to the different classification schemes used .
these 10% likely correspond to non - relaxed e / s0 , as discussed above .
agertz , o. , teyssier , r. & moore , b. 2009 , , 397 , 64 arnouts , s. et al .
2007 , , 476 , 137 athanassoula , e. 2008 , , 390 , l69 barnes , j. e. 1988 , , 331 , 699 barnes , j. e. 1992 , , 393 , 484 bell , e.f . ,
mcintosh , d.h . ,
katz , n. & weinberg , m.d .
2003 , , 149 , 289 barnes , j. e. 2002 , , 333 , 481 barnes , j. e. , & hernquist , l. 1996 , , 471 , 115 bell , e.f .
et al . 2005 , , 625 , 23 bell e. f. , phleps s. , somerville r. s. , wolf c. , borch a. , meisenheimer k. , 2006 , apj , 652 , 270 bell , e.f .
2007 , , 663 , 834 benson , a. j. , & bower , r. 2011 , , 410 , 2653 bournaud , f. , daddi , e. , elmegreen , b. g. , et al .
2008 , , 486 , 741 bournaud , f. & elmegreen , b.g .
2009 , , 694 , 158 bournaud , f. , elmegreen , b. g. , & martig , m. 2009b , , 707 , l1 bridge , c.r . ,
carlberg , r.g .
2010 , , 709 , 1067 brook , c. et al .
2012 , , 419 , 771 brooks , a. m. , governato , f. , quinn , t. , brook , c. b. , & wadsley , j. 2009 , , 694 , 396 bundy , k. et al .
2009 , , 697 , 1369 cameron , e. 2011 , pasa , 28 , 128 ceverino , d. , dekel , a. , & bournaud , f. 2010 , , 404 , 2151 cirasuolo , m. et al .
2007 , , 380 , 585 conselice , c.j . ,
bershady , m.a . ,
dickinson , m. & papovich , c. 2003 , , 126 , 1183 conselice , c.j .
, rajgor , s. & myers , r. 2008 , , 386 , 909 conselice , c.j . , yang , c. & bluck , a.f.l . 2009 , , 394 , 1956 contini , t. , garilli , b. , le fvre , o. , et al .
2012 , a&a , 539 , 91 contini , t. , epinat , b. , vergani , d. , et al . 2012b , the messenger , 147 , 32 covington , m. d. , kassin , s. a. , dutton , a. a. , et al .
2010 , , 710 , 279 cox , t.j . , jonsson , p. , primack , j.r .
& somerville , r.s .
2006 , , 373 , 1013 cox , t. j. , jonsson , p. , somerville , r. s. , primack , j. r. , & dekel , a. 2008 , , 384 , 386 daddi , e. et al .
2010 , , 713 , 686 dekel , a. , birnboim , y. , engel , g. , et al .
2009 , , 457 , 451 dekel , a. , sari , r. , & ceverino , d. 2009b , , 703 , 785 delgado - serrano , r. et al .
2010 , , 509 , a78 de ravel l. , et al .
, 2009 , a&a , 498 , 379 de ravel l. , et al . , 2011 ,
arxiv , arxiv:1104.5470 di matteo , p. et al .
2008 , , 492 , 31 dijkstra , m. , & loeb , a. 2009 , , 400 , 1109 dutton , a. a. , van den bosch , f. c. , & dekel , a. 2010 , , 405 , 1690 elmegreen , d. m. , elmegreen , b. g. , rubin , d. s. , & schaffer , m. a. 2005 , , 631 , 85 elmegreen , b. g. , bournaud , f. , & elmegreen , d. m. 2008 , , 688 , 67 elmegreen , b. g. , elmegreen , d. m. , fernandez , m. x. , & lemonias , j. j. 2009 , , 692 , 12 elmegreen , d. m. , elmegreen , b. g. , marcus , m. t. , et al . 2009b , , 701 , 306 epinat b. , et al . , 2009 , a&a , 504 , 789 epinat , b. , amram , p. , balkowski , c. , marcelin , m. 2010 , , 401 , 2113 epinat , b. , tasca , b. , amram , p. , et al .
2012 , a&a , 539 , 92 erb , d.k .
2006 , , 646 , 107 faucher - gigure , c .- a . , & kere , d. 2011 , , 412 , l118 flores , h. , hammer , f. , puech , m. , amram , p. & balkowski , c. 2006 , , 455 , 107 frster schreiber , n. m. , genzel , r. , lehnert , m. d. , et al .
2006 , , 645 , 1062 frster - schreiber , n.m .
2009 , , 706 , 1364 frster schreiber , n. m. , shapley , a. e. , erb , d. k. , et al .
2011 , , 731 , 65 frster schreiber , n. m. , shapley , a. e. , genzel , r. , et al .
2011b , arxiv:1104.0248 fuentes - carrera , i. et al .
2010 , , 513 , a43 fukugita , m. , nakamura , o. , okamura , s. , et al .
2007 , , 134 , 579 fumagalli , m. , prochaska , j. x. , kasen , d. , et al .
2011 , arxiv:1103.2130 gallazzi a. , bell e. f. , 2009 , apjs , 185 , 253 genel s. , et al . , 2010 ,
arxiv , arxiv:1011.0433 genzel , r. , newman , s. , jones , t. , et al .
2011 , , 733 , 101 giavalisco , m. , vanzella , e. , salimbeni , s. , et al . 2011 ,
arxiv:1106.1205 goerdt , t. , dekel , a. , sternberg , a. , et al .
2010 , , 407 , 613 governato f. , et al .
, 2009 , mnras , 398 , 312 hammer , f. et al .
1997 , , 481 , 49 .
u t hammer , f. et al .
2005 , , 430 , 115 hammer , f. et al .
2009 , , 507 , 1313 hammer , f. et al . 2009b , , 496 , 381 hernquist l. , 1993 , apj , 409 , 548 hopkins , p. f. , hernquist , l. , cox , t. j. , younger , j. d. , & besla , g. 2008 , , 688 , 757 hopkins , p.f .
2009 , , 715 , 202 hopkins , p.f .
et al . 2009b , , 397 , 802 hopkins , p.f . ,
cox , t.j .
, younger , j.d . & hernquist , l. 2009c , , 691 , 1168 hopkins , p.f . ,
bundy , k. , hernquist , l. , wuyts , s. & cox , t.j .
2010 , , 401 , 1131 hopkins , p.f .
2010b , , 724 , 915 hopkins , p.f . , younger , j.d . ,
hayward , c.c . , narayanan , d. & hernquist , l. 2010c , , 402 , 1693 hopkins , p. f. , keres , d. , murray , n. , quataert , e. , & hernquist , l. 2011 , arxiv:1111.6591 jogee , s. et al .
2009 , , 697 , 1971 kartaltepe et al .
2007 , , 172 , 320 kassin , s. a. , weiner , b. j. , faber , s. m. , et al .
2007 , , 660 , l35 keres , d. , katz , n. , fardal , m. , dav , r. & weinberg , d.h .
2009 , , 395 , 160 keres , d. , vogelsberger , m. , sijacki , d. , springel , v. , & hernquist , l. 2011 , arxiv:1109.4638 kimm , t. , devriendt , j. , slyz , a. , et al . 2011 ,
arxiv:1106.0538 kimm , t. , slyz , a. , devriendt , j. , & pichon , c. 2011b , , 413 , l51 kitzbichler , m.g .
& white , s.d.m .
2008 , , 391 , 1489 komatsu , e. et al .
2011 , , 192 , 18 krumholz m. r. , dekel a. , 2010 , mnras , 406 , 112 larson , r.b . & tinsley , b.m .
1978 , , 219 , 46 law , d. r. , steidel , c. c. , erb , d. k. , et al .
2009 , , 697 , 2057 law ,
d. r. , steidel , c. c. , shapley , a. e. , et al .
2012 , , 745 , 85 le fvre , o. et al .
2000 , , 311 , 565 le floch , e. , papovich , c. , dole , h. , et al .
2005 , , 632 , 169 le tiran , l. , lehnert , m. d. , di matteo , p. , nesvadba , n. p. h. , & van driel , w. 2011 , , 530 , l6 lemoine - busserolle , m. , bunker , a. , lamareille , f. , & kissler - patig , m. 2010 , , 401 , 1657 lemoine - busserolle , m. , & lamareille , f. 2010 , , 402 , 2291 lin , l. et al .
2008 , , 681 , 232 lpez - sanjuan c. , balcells m. , prez - gonzlez p. g. , barro g. , garca - dab c. e. , gallego j. , zamorano j. , 2009 , a&a , 501 , 505 lotz , j.m . , jonsson , p. , cox , t.j . & primack , j.r .
2008 , , 391 , 1137 lotz , j.m .
et al . 2008b , , 672 , 177 lotz , j. m. , jonsson , p. , cox , t. j. , & primack , j. r. 2010 , , 404 , 575 lotz , j.m . ,
jonsson , p. , cox , t.j .
& primack , j.r .
2010b , , 404 , 590 lotz , j. m. , jonsson , p. , cox , t. j. , et al .
2011 , , 742 , 103 martig , m. , & bournaud , f. 2010 , , 714 , l275 mignoli , m. , cimatti , a. , zamorani , g. , et al .
2005 , , 437 , 883 miller , s. h. , bundy , k. , sullivan , m. , ellis , r. s. , & treu , t. 2011 , , 741 , 115 monreal - ibero , a. , et al .
2010 , , 517 , a28 murray , n. , quataert , e. , & thompson , t. a. 2010 , , 709 , 191 nair , p. b. , & abraham , r. g. 2010 , , 186 , 427 nakamura , o. , fukugita , m. , yasuda , n. , et al .
2003 , , 125 , 1682 neichel , b. et al .
2008 , , 484 , 159 neistein , e. van den bosch , f.c . & dekel , a. 2006 , , 372 , 933 neistein , e. & dekel , a. 2008 , , 383 , 615 noeske , k. g. , weiner , b. j. , faber , s. m. , et al .
2007 , , 660 , l43 peirani , s. , hammer , f. , flores , h. , yang , y.b . & athanassoula , e. 2009 , , 496 , 51 perez - gonzalez , p.g .
2009 , , 675 , 234 pichon , c. , pogosyan , d. , kimm , t. , et al .
2011 , , 418 , 2493 powell , l. c. , slyz , a. , & devriendt , j. 2011 , , 414 , 3671 pozzetti , l. et al .
2007 , , 474 , 443 puech , m. , hammer , f. , flores , h. , stlin , g. , & marquart , t. 2006 , , 455 , 119 puech , m. , hammer , f. , lehnert , m.d . & flores , h. 2007 , , 466 , 83 puech , m. et al .
2007 , , 476 , 21 puech , m. et al .
2008 , , 484 , 173 puech , m. , hammer , f. , flores , h. , neichel , b. & yang , y.b .
2009 , , 493 , 899 puech , m. et al .
2010 , , 510 , a68 puech , m. 2010 , , 406 , 535 ravikumar , c.d .
2007 , , 465 , 1099 rawat , a. , hammer , f. , kembhavi , a.k . & flores , h. 2008 , , 681 , 1089 robaina , a. r. , bell , e. f. , skelton , r. e. , et al .
2009 , , 704 , 324 robaina a. r. , bell e. f. , van der wel a. , somerville r. s. , skelton r. e. , mcintosh d. h. , meisenheimer k. , wolf c. , 2010 , apj , 719 , 844 robertson , b. et al .
2006 , , 645 , 986 robertson , b. e. , & bullock , j. s. 2008 , , 685 , l27 rodrigues , m. , puech , m. , hammer , f. , flores , h. & rothberg , b. 2012 , , 421 , 2888 ryan r. e. , jr . , cohen s. h. , windhorst r. a. , silk j. , 2008 , apj , 678 , 751 shi y. , rieke g. , lotz j. , perez - gonzalez p. g. , 2009 , apj , 697 , 1764 springel , v. , & hernquist , l. 2005 , , 622 , l9 steidel , c. c. , erb , d. k. , shapley , a. e. , et al .
2010 , , 717 , 289 stewart , k.r . ,
bullock , j.s . ,
wechsler , r.h . ,
maller , a.h . &
zentner , a.r .
2008 , , 683 , 597 stewart , k.r . ,
bullock , j.s . ,
wechsler , r.h .
& maller , a.h .
2009 , , 702 , 307 stewart , k. r. , kaufmann , t. , bullock , j. s. , et al .
2011 , , 735 , l1 stewart , k. r. , kaufmann , t. , bullock , j. s. , et al . 2011b , , 738 , 39 suh , h. , jeong , h. , oh , k. , et al .
2010 , , 187 , 374 toomre , a. , & toomre , j. 1972 , , 178 , 623 toomre , a. 1977 , , 15 , 437 torrey , p. , vogelsberger , m. , sijacki , d. , springel , v. , & hernquist , l. 2011 , arxiv:1110.5635 toth , g. & ostriker , j.p .
1992 , , 389 , 5 van den bergh , s. 2002 , , 114 , 797 van de voort , f. , schaye , j. , booth , c. m. , haas , m. r. , & dalla vecchia , c. 2011 , , 414 , 2458 van dokkum , p.g . & van der marel , r.p .
2007 , , 655 , 30 van dokkum , p. g. , brammer , g. , fumagalli , m. , et al .
2011 , , 743 , 15 van starkenburg , l. , van der werf , p. p. , franx , m. , et al .
2008 , , 488 , 99 weidner , c. , kroupa , p. & bonnell , i.a.d .
2010 , , 401 , 275 weinzirl , t. , jogee , s. , khochfar , s. , burkert , a. , & kormendy , j. 2009 , , 696 , 411 weinzirl , t. , jogee , s. , conselice , c. j. , et al .
2011 , , 743 , 87 wright , s. a. , larkin , j. e. , law , d. r. , et al .
2009 , , 699 , 421 yang , y.b .
2008 , , 477 , 789 yang , y.b . , hammer , f. , flores , h. , puech , m. & rodrigues , m. 2009 , , 501 , 437 zheng , x. z. , bell , e. f. , papovich , c. , et al .
2007 , , 661 , l41 | about two - thirds of present - day , large galaxies are spirals such as the milky way or andromeda , but the way their thin rotating disks formed remains uncertain .
observations have revealed that half of their progenitors , six billion years ago , had peculiar morphologies and/or kinematics , which exclude them from the hubble sequence .
major mergers , i.e. , fusions between galaxies of similar mass , are found to be the likeliest driver for such strong peculiarities . however , thin disks are fragile and easily destroyed by such violent collisions , which creates a critical tension between the observed fraction of thin disks and their survival within the @xmath0-cdm paradigm . here
we show that the observed high occurrence of mergers amongst their progenitors is only apparent and is resolved when using morpho - kinematic observations which are sensitive to all the phases of the merging process .
this provides an original way of narrowing down observational estimates of the galaxy merger rate and leads to a perfect match with predictions by state - of - the - art @xmath0-cdm semi - empirical models with no particular fine - tuning needed .
these results imply that half of local thin disks do not survive but are actually rebuilt after a gas - rich major merger occurring in the past nine billion years , i.e. , two - thirds of the lifetime of the universe .
this emphasizes the need to study how thin disks can form in halos with a more active merger history than previously considered , and to investigate what is the origin of the gas reservoir from which local disks would reform . |
unifying quantum mechanics @xcite-@xcite with general relativity @xcite is a major open problem in physics .
after the unification of quantum mechanics and special relativity has been accomplished @xcite , in the resulting relativistic quantum field theories @xcite - @xcite , spacetime and quantum - mechanical degrees of freedom are still treated on a separate footing .
for example , the electromagnetic , weak and strong forces are described as being mediated by particles , namely photons , @xmath0 and @xmath1-bosons , and gluons .
an analogous consistent formulation of gravity has so far not succeeded . in order to overcome the schism between the theory of gravity on one side and quantum mechanics on the other
, one may therefore ask : is there an element in the structure of quantum mechanics that forces spacetime into existence ?
i.e. , is spacetime an emergent phenomenon , predicted by quantum mechanics ? in the present paper , we approach this question in a restricted and controlled setting , with emphasis on the notion of time .
namely , we study a scheme of quantum computation @xcite , so - called ` measurement - based ' or ` one - way ' quantum computation ( mbqc ) @xcite as a quantum - mechanical toy model for spacetime . in mbqc
, the process of computation is driven by local measurements on a suitably chosen entangled state .
` time ' in deterministic mbqc is a partial ordering of measurement events , which is compatible with the notion of time in the theory of relativity .
an external and continuously increasing clock time is not required .
the possible temporal orders are severely constrained by symmetries present in every mbqc .
the exploration of these symmetry - enforced constraints is a central theme in this paper .
a fundamental property of quantum mechanics is that measurement outcomes as opposed to measurement bases cannot be chosen .
they are in general random .
mbqc , which is driven by measurements ( no unitary evolution ever takes place in this computational scheme ) , were entirely useless if this randomness could not be prevented from creeping into the logical processing .
but it can : all that is needed is the adjustment of measurement bases according to measurement outcomes obtained at other qubit locations _ before_. one can turn this mechanism for adjusting measurement bases around , and say that it is the inherently probabilistic nature of quantum measurement which determines the ` before ' in mbqc . indeed , without even knowing what is being computed in a given mbqc , the requirement of preventing the randomness of quantum mechanical measurement from affecting the logical processing imposes stringent constraints on the possible temporal orders of measurement events .
it is because of this link between temporal order and quantum measurement that we propose mbqc as a quantum - mechanical toy model for spacetime . in the context of the present work
, it has previously been shown for the case the initial state in mbqc is a stabilizer state @xcite known in a suitable local basis , that the entire temporal order of measurements is fully specified if the set @xmath2 of first - measurable and the set @xmath3 of last - measurable qubits are known @xcite , @xcite .
but not every pair @xmath2 , @xmath3 leads to a possible temporal order . a condition on admissible pairs @xmath2 ,
@xmath3 has been given in @xcite .
on a different route , an mbqc - analogue of closed time - like curves ( ctcs ) has been identified @xcite , via post - selection models @xcite , @xcite for ctcs in quantum circuits .
certain solutions of the einstein equations permit closed time - like curves , and if mbqc is to serve as a toy model for a quantum spacetime it is a bonus that it accommodates this part of the phenomenology of general relativity .
the results of the present paper are three - fold .
first , we identify a gauge transformation leaving the classical output of all mbqcs invariant , and causing additional terms in the classical processing relations for measurement outcomes .
we show that the resulting generalized processing relations fully specify the resource state and measurement setting , up to equivalence . in the reverse direction ,
we provide a classification for all mbqc temporal relations compatible with a given resource stabilizer state and set of measurement bases .
specifically , we introduce a matroid capturing all information of the resource state and measurement bases relevant for temporal order , and show that the classical processing relations including temporal order are in one - to - one correspondence with the bases of this matroid .
second , we identify a further group of symmetry transformations generally changing mbqcs but preserving temporal order .
the possible temporal orders in mbqc thus arise as representations of this group .
these symmetry transformations turn out to be related to local complementation @xcite-@xcite .
third , after light cones and closed time - like curves , we identify event horizons as a third piece of gr phenomenology that has a counterpart in mbqc .
in this section we review some basic facts about measurement - based quantum computation , essential definitions for the discussion of mbqc temporal order , as well as previous work in this area .
the scheme of mbqc itself , with a proof of its computational universality , is not reviewed here since various articles on this subject exist in the literature @xcite , @xcite - @xcite also , we require familiarity with the stabilizer formalism @xcite . in mbqc ,
the process of computation is driven by local ( = 1-qubit ) measurements on an initial highly entangled state , generally taken to be a so - called cluster state .
the local measurements can only reduce entanglement , and therefore all entanglement needed for the computation must come from the initial state . for this reason , the initial state
is often called the ` resource state ' .
the computational power of a given mbqc strongly depends on the choice of the initial resource state .
for example , a local resource state has obviously no computational power .
other states may be used for a restricted class of computations . two - or - higher dimensional cluster states of unbounded size have the property that they enable universal quantum computation .
that is , _ any _ quantum computation can be realized on such a state , by suitable choice of the local measurement bases .
we now define graph states , to be used later on , and cluster states as a subclass thereof .
be @xmath4 a graph with vertex set @xmath5 and edge set @xmath6 , such that there is one qubit for each vertex @xmath7 .
then , the graph state @xmath8 is the unique ( up to global phase ) joint eigenstate , @xmath9 , for all @xmath7 , of the operators @xmath10 the cluster state @xmath11 is a graph state where the corresponding graph is a @xmath12-dimensional lattice @xmath13 . in the standard scheme @xcite , the measured observables are of the form @xmath14=\cos \varphi_a\ , \sigma_x^{(a ) } + ( -1)^{q_a}\sin \varphi_a \,\sigma_y^{(a)},\ ] ] with @xmath15 the qubit to which the measurement is applied , and @xmath16 depending on outcomes of ( earlier ) measurements on other qubit locations .
as noted above , the temporal order of measurement in mbqc is a consequence of the randomness of measurement outcomes . by adjusting measurement bases according to measurement outcomes obtained on other qubits ,
this randomness inherent in quantum mechanical measurement can be kept from creeping into the logical processing @xcite .
if the measurement outcome of qubit @xmath15 influences the choice of measurement basis for qubit @xmath17 , clearly , qubit @xmath15 must be measured before qubit @xmath17 can .
this is how a temporal order among the measurement events arises .
we now generalize the above scenario of measuring observables of form eq .
( [ obsxy ] ) on cluster states .
namely , we now consider general stabilizer states @xmath18 as resources , with @xmath19 and stabilizer group @xmath20 .
furthermore , we generalize the local observables whose measurement drives the computation from eq .
( [ obsxy ] ) to @xmath21 = \cos \varphi_a\ , \sigma_\phi^{(a ) } + ( -1)^{q_a } \sin \varphi_a\ , \sigma_{s\phi}^{(a)},\;\ ; \forall a \in \omega,\ ] ] with @xmath22 .
therein , the measurement angles @xmath23 are in the range @xmath24 , and @xmath25 may depend on measurement outcomes from several ( other ) qubits in @xmath26 .
[ measurement plane ] for every qubit @xmath27 , the measurement plane at @xmath15 is the ordered pair @xmath28 $ ] . we define a third pauli operator , @xmath29 . as we will see shortly , the pauli operators @xmath30 and @xmath31 are useful because of the relations @xmath32
\sigma_\phi^\dagger = o[q \oplus 1],\;\ ; \sigma_s o[q ] \sigma_s^\dagger = -o[q].\ ] ] the basic mechanism of accounting for an `` undesired '' measurement outcome is the following .
suppose on some qubit @xmath33 , instead of the `` desired '' post - measurement state @xmath34 the `` undesired '' post - measurement state @xmath35 has been obtained .
the goal is to get the computation back on track by only adjusting the subsequent measurements . to do that , we require a stabilizer operator @xmath36 with the following properties @xcite : ( 1 ) @xmath37 has support only on @xmath15 and the yet unmeasured qubits , and ( 2 ) @xmath38 . recall that @xmath39 for the eigenstates @xmath40 of the local measured observable @xmath41 $ ] , c.f .
( [ obs2 ] ) .
denote by @xmath42 and @xmath43 the past and future of @xmath15 , respectively .
then , @xmath44 therein , the first equality follows from @xmath36 , and the second from the above properties ( 1 ) and ( 2 ) .
since the overlaps between local states ( representing the local measurements ) with the resource state @xmath18 contain all information about the computation , we thus find that we can correct for `` undesired '' outcomes by ( a ) adjusting measurement bases of future measurements ( caused by tensor factors @xmath30 in @xmath45 ) and ( b ) re - interpretation of measurement outcomes ( caused by tensor factors @xmath31 in @xmath45 ) .
_ example : _ consider mbqc on a cluster state @xmath46 of three qubits on a line , each measured in the @xmath47$]-plane . that is , for all three qubits @xmath48 and @xmath49 .
( here and from now on , we use the shorthand @xmath50 , @xmath51 and @xmath52 . )
the stabilizer generators of @xmath46 are @xmath53 when the three cluster qubits are measured in the order @xmath54 , the corresponding quantum circuit is @xcite @xmath55{eqcirc}}\ ] ] using eq .
( [ corr1 ] ) , here we show that if qubits 1 , 2 , 3 are measured in the order @xmath54 , then the randomness of the measurement outcomes on qubits 1 and 2 can be corrected for .
first we consider the stabilizer operator @xmath56 .
if inserted in the state overlap of eq .
( [ corr1 ] ) , the measurement outcome of qubits 1 and 3 is flipped , as well as the measurement basis at qubit 2 .
since qubits 2 and 3 are yet unmeasured when qubit 1 is measured , this is a valid correction operation for qubit 1 ; hence the notation @xmath57 .
similarly , @xmath58 can be used as correction operation for qubit 2 .
@xmath59 flips the measurement outcome of qubit 2 and the measurement basis of qubit 3 .
since qubit 3 is yet unmeasured when qubit 2 is measured , this corresponds to a valid correction operation .
the above argument also works in reverse .
if the correction operations @xmath57 and @xmath59 are used , then @xmath60 follows .
@xmath57 implies that the measurement basis of qubit 2 depends on the measurement outcome at qubit 1 , hence @xmath61 .
note that @xmath62 does not yet follow !
@xmath57 does _ not _ affect the measurement basis at qubit 3 .
only the meaning of the eigenstates is interchanged , which by itself does not require qubit 3 to be measured after qubit 1 .
the interpretation of the measurement outcome may take place long after the measurement itself has taken place .
@xmath63 implies that the measurement basis of qubit 3 depends on the measurement outcome of qubit 2 , and hence @xmath64 . both relations taken together
yield @xmath60 . from the equivalence with the circuit of eq .
( [ eqcirc ] ) one would expect one bit of classical output .
indeed , if no correction operations need to be used , the eigenvalue measured at the output of the circuit corresponds to the eigenvalue @xmath65 measured on qubit 3 of the cluster .
now recall that @xmath57 , applied conditioned upon @xmath66 , flips @xmath65 .
therefore , with or without corrections , the eigenvalue measured in the circuit eq .
( [ eqcirc ] ) equals @xmath67 . or , in binary notation @xmath68
, the single bit of classical output takes the value @xmath69 . note that we have not made any use of @xmath70 in the above argument .
still , @xmath71 has a role to play , as we discuss in section [ gauge ] . to counteract the randomness of measurement outcomes , two measurement settings ( i.e. , bases ) per qubit suffice , which may be labeled by @xmath72 and @xmath73 , respectively , for each qubit @xmath15 .
the measurement settings may collectively be described by a binary vector @xmath74 , @xmath75_a = q_a$ ] for all @xmath76 , and the measurement outcomes by a binary vector @xmath77 , @xmath78_a = s_a$ ] , forall @xmath33 .
it turns out that the relation between measurement bases @xmath74 and measurement outcomes * s * is _ linear _ @xcite , @xmath79 with @xmath80 a binary matrix .
we call @xmath80 the influence matrix .
the set of all qubits @xmath17 whose measurement basis must be adjusted according to the measurement outcome on @xmath15 is denoted as the _ forward cone _ of a qubit @xmath15 .
similarly , the backward cone of a qubit @xmath17 is the set of all those qubits @xmath15 whose measurement outcome influence the measurement basis at @xmath17 . more formally , with eq .
( [ to1 ] ) , [ fbc ] for any @xmath33 the forward cone @xmath81 is given by @xmath82 for any @xmath83 the backward cone @xmath84 is given by @xmath85 we denote the characteristic vectors of @xmath86 and @xmath87 by @xmath88 and @xmath89 , respectively .
then , the influence matrix @xmath80 takes the form @xmath90 the influence matrix @xmath80 generates a temporal relation among the measurement events under transitivity .
we say @xmath91 ( @xmath15 precedes @xmath17 ) if @xmath92 . a priori , it is not forbidden that for two qubits @xmath93 , @xmath91 and @xmath94 .
however , such a computation could not be run deterministically in a world like ours where time progresses linearly .
an mbqc is _ deterministically runnable _ if the temporal relation `` @xmath95 '' between the measurement events is a strict partial order .
a strict partial order is a relation among the elements @xmath33 with the following properties
@xmath96 [ iodef ] for a given mbqc , the input set @xmath97 is the set of qubits whose backward cones are empty , @xmath98 .
the output set @xmath99 is the set of qubits whose forward cones are empty , @xmath100 .
that is , with respect to a given temporal relation among the measurement events , @xmath2 is the maximal set of qubits which can be measured first , and @xmath3 is the maximal set of qubits which can be measured last . regarding the computational output , for the purpose of this paper we are exclusively interested in mbqcs for which the computational result is a classical bit string .
that is , every qubit in @xmath26 is measured .
then , as a consequence of the randomness of individual measurement outcomes , the classical output @xmath101 of an mbqc is given by _ correlations _ among measurement outcomes .
again , the relation between classical output and measurement outcomes is linear , @xmath102 for a suitable binary matrix @xmath1 .
_ example : _ to illustrate the above notions , we briefly return to the three - qubit cluster state example of section [ oto ] . from the previous discussion we find that @xmath103 therefore , @xmath104 , @xmath105 , @xmath106 and @xmath107 , @xmath108 , @xmath109 .
hence , @xmath110 and @xmath111 .
also , @xmath112 and @xmath64 .
the latter two relations generate a third under transitivity , namely @xmath113 .
asymmetry and irreflexivity are obeyed in this example . regarding the single bit of output in this computation , the matrix @xmath1 of eq .
( [ to1b ] ) is @xmath114 .
_ mbqc temporal order as an ( almost ) emergent phenomenon . _ in @xcite
the following question is asked : `` given graph @xmath4 and the set @xmath115 of measurement planes for all vertices , can the temporal order of measurements in mbqc with a graph state @xmath8 be uniquely reconstructed from this information ? ''
the graph @xmath4 and the measurement planes @xmath115 are undirected objects .
thus , if the answer to this question was yes , then temporal order in mbqc were truly emergent .
however , it turns out that the pair @xmath116 does not specify the temporal order of measurements in mbqc uniquely ; there are in general a number of consistent temporal orders respecting the requirement that the randomness of measurement outcomes should not affect the logical processing .
one may then ask how constraining on temporal order this requirement actually is . to this question ,
the following answer is provided by @xcite : if in addition to @xmath4 and @xmath115 the set @xmath2 of first - measurable and the set @xmath3 of last - measurable qubits is known , then the complete temporal order ( if existing ) can be uniquely reconstructed from this information .
thus , mbqc temporal order is not emergent in the strict sense ; a seed @xmath117 must be provided in addition to @xmath4 and @xmath115 , and the complete temporal order then follows .
but not every pair @xmath2 , @xmath3 will lead to a consistent temporal order .
the question that now arises is which pairs @xmath117 do . for the case where the stabilizer resource state is a graph state and all qubits
are measured in the @xmath118$]-plane then the answer to this question is given in @xcite .
denote by @xmath119 the adjacency matrix of the graph @xmath4 describing the resource state @xmath8 , and by @xmath120 the submatrix of @xmath119 where the rows are restricted to @xmath121 and the columns are restricted to @xmath122 .
then , the pair @xmath117 leads to a partial order of measurement events in mbqc iff there exists a matrix @xmath80 such that @xmath123 , and @xmath80 is free of cycles ( that is @xmath124 , @xmath125 , @xmath126 , etc ) .
the resulting temporal order is generated by @xmath80 under transitivity .
[ cols= " < , < " , ] here we introduce the notion of `` gauge transformations '' acting on a given quantum computation .
these transformations exist for both the circuit model and mbqc . to obtain an intuition for the gauge transformations introduced here , it is instructive to first inspect them in the circuit model .
specifically , we consider a quantum circuit which consists of ( 1 ) the preparation of the quantum register in the initial state @xmath127 , ( 2 ) unitary evolution composed of , say , cnot gates and one - qubit rotations about the @xmath128- and @xmath1-axes , and ( 3 ) local measurements for readout .
such a circuit is displayed in fig .
[ gt ] above .
then , into every qubit line individually , we may insert an identity @xmath129 next to the input ; see fig .
the left @xmath130 is propagated backwards in time until absorbed by the input state @xmath131 .
the right @xmath130 is propagated forward in time , flipping rotation angles and readout - measurement outcomes in the passing .
this transformation is an _ equivalence transformation _ , since it is caused by the insertion of an identity gate into the circuit of fig .
the reason we call it a gauge transformation is that it changes reference frames for certain rotation angles . specifically , for the rotation gates next to each input qubit we can individually choose our convention for which rotation angles are called positive or negative , respectively .
once those signs are fixed on the input side , they are fixed throughout the circuit . translating the above discussed gauge transformations from the circuit model into mbqc
it is easily seen that the above relations eq .
( [ to1 ] ) and ( [ to1b ] ) are incomplete .
we find the more general relations @xmath132 with @xmath133 a choice of gauge . under a change of @xmath133 , the measurement bases in a particular mbqc change , but
the probability distribution for the classical output values remains unchanged .
of course , knowledge of the @xmath133-dependent extra parts in eq .
( [ to7a],[to7b ] ) is not necessary to run any given mbqc , since @xmath134 is always a valid choice .
however , the presence of the extra terms in the processing relations strengthens their interdependence with the resource state , which is the reason why we discuss them here .
we may now want to derive the generalized processing relations eq .
( [ to7a],[to7b ] ) directly in mbqc , without reference to the circuit model . before discussing the general case , we return to the specific example of the three - qubit cluster state in section [ oto ] .
_ example : _ consider the product @xmath135 of @xmath71 , @xmath136 in eq .
( [ 3cluster ] ) . when used in eq .
( [ corr1 ] ) , the effect of this stabilizer element is to flip the measurement bases of qubits 1 and 3 .
hence , the relation eq .
( [ trelex0 ] ) generalizes to @xmath137 this is precisely what one would expect from the insertion of @xmath138 next to the input of the equivalent quantum circuit in eq .
( [ eqcirc ] ) .
we now turn to the general case . via eq .
( [ corr1 ] ) , the stabilizer group @xmath20 acts on @xmath77 and @xmath74 . denote the post - measurement state of qubit @xmath33 by @xmath139 , with @xmath140 the measurement outcome and @xmath141 specifying the measurement basis . then , as in eq .
( [ corr1 ] ) , @xmath142 since @xmath143 and @xmath144 , for a stabilizer element @xmath145 the action of @xmath146 on @xmath77 , @xmath74 is @xmath147 again , nothing changes by the insertion of a stabilizer ( identity ) operator into the state overlap of eq .
( [ corr1 ] ) , and transformations @xmath146 are therefore equivalence transformations .
they can be used to constrain the possible temporal orders in mbqc , as we discuss explicitly in section [ gto ] . to state and prove our results on mbqc temporal order , we need to make a few more definitions .
first , let us emphasize that in this paper we allow closed time like curves ( ctc ) in the computation ; we are interested in classifying all transitive _ temporal relations _ consistent with a given resource stabilizer state and set of measurement planes
. therefore , we do not restrict to partial orders per se , and drop the conditions of irreflexivity and antisymmetry on the temporal relations .
it turns out that the possible temporal relations , and indeed the classical processing relations eq . ( [ to7a ] ) , ( [ to7b ] ) ,
can be parametrized by two subsets of @xmath26 , namely the computational output set @xmath148 and the gauge input set @xmath149 .
we define these sets next .
we say that the measurement outcome @xmath140 of qubit @xmath15 is _ corrected _ in a given mbqc , if by insertion of a suitable stabilizer operator @xmath150 in eq .
( [ gaugeid ] ) equivalence of @xmath151 with the reference outcome @xmath152 is established , at the cost of adjustment of measurement bases and/or re - interpretation of measurement outcomes on other qubits .
we observe that if for a given mbqc _ all _ measurement outcomes @xmath140 , @xmath33 , can be corrected then these measurement outcomes contain no information and no linear combination of them is worth outputting .
thus , in general there will be a set of qubits whose measurement outcomes are not corrected .
[ cos ] for a given mbqc , the computational output set @xmath153 is the set of qubits whose measurement outcomes are not corrected .
the correction operations for the qubits in @xmath154 , c.f .
( [ corr1 ] ) , are each implemented by correction operators @xmath155 . for each @xmath156
, @xmath157 has the property that it flips the measurement outcome @xmath140 at qubit @xmath15 , but does not flip the measurement outcome on any other qubit in @xmath158 .
in this way , it is ensured that the correction operation is for qubit @xmath15 individually .
[ corrop ] for an mbqc with a given stabilizer resource state @xmath18 , fixed set @xmath115 of measurement planes and computational output set @xmath148 , for each @xmath154 the corresponding correction operator @xmath155 is a pauli operator satisfying the conditions @xmath159 the correction operator @xmath157 can be used to correct an `` undesired '' measurement outcome at qubit @xmath15 , c.f .
( [ corr1 ] ) . if , for a given operator @xmath157 , @xmath160 then , by eq .
( [ corr1 ] ) , the measurement basis at qubit @xmath17 depends on the measurement outcome for qubit @xmath15 . in terms of the influence matrix @xmath80 ,
@xmath161_{ba } = 1,\;\;\;\;\forall a \in ( o_{\text{comp}})^c,\ , b\in \omega.\ ] ] note that in the first line of eq .
( [ corrcorr ] ) we allow @xmath162 only because we are admitting closed time - like curves in the present discussion .
@xmath162 means that the measurement basis for qubit @xmath15 depends on the outcome @xmath140 of the measurement of qubit @xmath15 .
this amounts to a closed time - like curve only involving qubit @xmath15 ( self - loop ) and is an obstacle to deterministic runnability . because the qubits @xmath163 have no correction operations , @xmath164 for all @xmath83 , and thus @xmath165 thus , @xmath148 and the correction operators @xmath166 completely determine the influence matrix @xmath80 and hence the temporal relation among the measurements .
conversely , if @xmath148 and @xmath166 are unknown , then the constraints eq .
( [ corrcorr ] ) pose self - consistency conditions on them .
we discuss these conditions further below .
as we have seen in the concrete three - qubit example above , for a given mbqc the relation between the choice @xmath74 of measurement bases and the measurement outcomes @xmath77 in general allows for an offset term @xmath167 , c.f .
( [ to7a ] ) .
thus , the measurement bases for a certain set of qubits can be freely chosen until the initially arbitrary @xmath133 becomes fixed .
this observation leads to [ gis]for a given mbqc with input set @xmath2 , the gauge input set @xmath168 is a set of qubits such that for each @xmath169 the parameter @xmath170 specifying the locally measured observable @xmath171 $ ] can be freely chosen .
analogously to the correction operations , there will be gauge operators which implement the ` corrections ' of measurement _ bases _ for the qubits in @xmath172 .
the definition of the gauge operators ensures that for all @xmath169 the corresponding @xmath170 can be changed individually without changing the others .
[ gaugeop ] for an mbqc with a given stabilizer resource state @xmath18 , fixed set @xmath115 of measurement planes , computational output set @xmath148 and gauge input set @xmath149 , for each @xmath169 , the corresponding gauge operator @xmath173 is a pauli operator satisfying the conditions @xmath174 like previously for the correction operations , eq .
( [ corrgauge ] ) poses self - consistency condition on the possible sets @xmath149 and @xmath175 .
[ indepl ] the correction operators @xmath157 of eq .
( [ corrcorr ] ) , @xmath154 , and the gauge operators @xmath176 of eq .
( [ corrgauge ] ) , @xmath169 , are independent .
that is , for all sets @xmath177 , @xmath178 with @xmath179 , @xmath180 .
_ proof of lemma [ indepl ] . _
( indirect ) assume that there exists a pair @xmath181 , with @xmath179 , such that @xmath182 .
then , for all @xmath183 , @xmath184 . now , with eq .
( [ corrcorr ] ) , @xmath185 , and @xmath186 for all @xmath187 . with eq .
( [ corrgauge ] ) , @xmath188 for all @xmath169 .
therefore , no other @xmath189 , @xmath190 can cancel a @xmath191-contribution from @xmath192 to @xmath193 .
hence , @xmath194 for all @xmath183 , and thus @xmath195 . by an analogous argument
, no @xmath189 can cancel the @xmath196-contribution of @xmath176 to @xmath193 , hence @xmath197 for all @xmath169 , and @xmath198 .
thus , @xmath199 . contradiction .
hence , the @xmath157 , @xmath176 are independent .
@xmath200 _ remark : _ for any given mbqc with fixed temporal relation , the computational output set @xmath148 is a subset of the output set @xmath3 , see eq .
( [ ooc ] ) . but
@xmath148 and @xmath3 are not necessarily equal .
likewise , @xmath168 by definition , but @xmath149 and @xmath2 may not be equal . to illustrate this point
, we consider the following two examples . _ example 1 .
_ consider the 3-qubit cluster state of section [ oto ] , with measurement planes @xmath201 $ ] for all three qubits .
we consider the correction operators @xmath202 , @xmath203 for qubits 1 and 2 , and gauge operator @xmath204 . @xmath205 and @xmath206 are then permitted by eqs .
( [ corrcorr ] ) and ( [ corrgauge ] ) , respectively . as was discussed previously for the above choice of correction operations , @xmath110 and @xmath111 .
thus , in the present example @xmath207 and @xmath208 .
_ example 2 .
_ consider a greenberger - horne - zeilinger state @xmath209 as resource state , with all three qubits measured in a basis in the @xmath118$]-plane .
we use the stabilizer elements @xmath210 and @xmath211 as correction operations for qubits 1 and 2 , and @xmath212 as gauge operator .
then , the choice @xmath206 and @xmath205 is admitted by eqs .
( [ corrcorr ] ) and ( [ corrgauge ] ) , respectively . on the other hand
, this is an example of a temporarily flat mqc , @xmath213 .
therefore , @xmath214 , and @xmath215 , @xmath216 .
[ sizes ] for any mbqc on a stabilizer state , @xmath217 .
we prove lemma [ sizes ] in section [ stabgnf ] .
a pair @xmath149 , @xmath148 is called extremal iff @xmath218 . as will become clear in the next section , extremal pairs
@xmath149 , @xmath148 are easier to handle than general pairs , and are not very restrictive ( c.f . theorem [ extr ] ) .
we still need to relate the classical output vector @xmath101 appearing in eq .
( [ to7b ] ) to the set @xmath148 . to this end , we make the following [ oco ] a classical output vector @xmath101 , with processing relations @xmath219 , is optimal iff the following conditions hold 1 .
_ maximality : _ upon left - multiplication by an invertible matrix , @xmath1 can be brought into a unique normal form @xmath220 where the column split is between @xmath221 , and 2 . _
determinism : _ for the matrix @xmath222 in eq .
( [ zn ] ) , @xmath223_{ij}=1 \longleftrightarrow k(j)|_i \in \ { \sigma_s^{(i ) } , \sigma_{s\phi}^{(i)}\}.\ ] ] in other words eq . informs us that the @xmath224th column of @xmath222 is simply the restriction of the support of @xmath225 to @xmath148 .
the reason for defining an ` optimal classical output ' besides a ` classical output ' is the following : one could , in principle , run an mbqc perfectly deterministically and then choose @xmath101 such that nothing is outputted at all , or all outputted bits , independent of the measurement angles chosen , are zero guaranteed or perfectly random guaranteed .
the above definition of an optimal classical output eliminates such choices .
maximality says that there is one bit of optimal output per qubit of @xmath148 .
the determinism condition can be understood from the correction procedure explained in eq .
( [ corr1 ] ) . for a qubit @xmath226
, we account for the undesired outcome @xmath227 by inserting @xmath225 into the state overlap @xmath228 .
now consider a qubit @xmath229 . by eq .
( [ zn ] ) , @xmath230 contributes to the output bit @xmath231 , the @xmath232-th bit of @xmath101 .
if @xmath233 , then the insertion of @xmath225 into the overlap flips @xmath230 , i.e. , @xmath234 .
this needs to be taken into account when reading out @xmath230 .
the linear combination @xmath235 remains unaffected by the correction for @xmath236 .
we have now made the definitions needed to state and prove our results on temporal order in mbqc . in section [ stabgnf ]
we establish a normal form for the stabilizer generator matrix of the resource state @xmath18 .
this normal form is the basis for our results on mbqc temporal order , which are stated in sections [ intdep ] and [ gto ] .
let us briefly review which pieces of information specify an mbqc .
a priori , there are four : the set of measurement angles , the set @xmath115 of measurement planes , the resource state @xmath18 and the classical processing relations eq . ( [ to7a],[to7b ] ) .
the measurement angles entirely drop out of all our considerations about temporal order .
next , we observe that when specifying the measurement planes and the resource state separately , we really specify too much .
starting from a given pair @xmath237 of resource state and set of measurement planes , for any local clifford unitary @xmath238 , the pair @xmath239 obtained by applying @xmath238 to both the measurement planes @xmath115 and the stabilizer state @xmath18 is again a valid pair , i.e. , it consists of a set of a stabilizer state and a set of measurement planes .
furthermore , it amounts to exactly the same computation as the original pair .
the pair @xmath237 is thus redundant . to remove this redundancy ,
we combine the measurement planes and the stabilizer state @xmath18 into the stabilizer generator matrix @xmath240 in the @xmath241-stabilizer basis , @xmath242 therein , the columns to the left ( right ) form the @xmath30 ( @xmath31- ) part of the stabilizer generator matrix .
@xmath240 comprises all information from the resource state @xmath18 and the set of measurement planes @xmath115 relevant for the discussion of mbqc .
we do not need to know the state and the measurement planes separately . now note that the correction operators @xmath157 , @xmath154 , and the gauge operators @xmath176 , @xmath243 are all elements of the stabilizer @xmath20 , and , by lemma [ indepl ] , are independent .
thus , they either form or can be completed to a set of generators for @xmath20 .
this observation leads us to the following [ nf ] for any mbqc on a stabilizer state @xmath18 with extremal @xmath149 , @xmath244 , the generator matrix @xmath245 of @xmath20 can be written in the normal form @xmath246 the matrices @xmath247 , @xmath248 , @xmath249 , @xmath250 are related to the matrices @xmath251 , @xmath252 , @xmath80 , @xmath1 governing the classical processing of measurement outcomes in mbqc , @xmath253 and @xmath254 , via @xmath255 by definition [ corrop ] , a correction operator @xmath157 exists for every @xmath256 . by eq .
( [ corrcorr ] ) these operators have no support in @xmath257 and are @xmath258 in number and must take the following form : @xmath259 , where the @xmath260 part is split as @xmath149 , @xmath261 while the @xmath31-part is split as @xmath148 , @xmath158 .
the gauge operators are , by definition , of the form : @xmath262 , where the column splits are as for the correction operators .
since there are @xmath263 generators for the stabilizer @xmath264 , of which @xmath258 are already accounted for , there can be at most @xmath265 independent gauge operators , i.e. , @xmath266 which proves lemma [ sizes ] . in the present setting , the pair @xmath149
, @xmath148 is extremal by assumption , thus these two sets of generators exhaust the stabilizer generators and we can write the stabilizer as @xmath267 for suitable matrices @xmath268 , @xmath269 , @xmath270 and @xmath271 .
we now need to identify these matrices . by definition , @xmath272 .
measurement outcomes on qubits @xmath163 are not corrected , hence @xmath273 for all @xmath163 , and @xmath274 follows from the definition of the output set @xmath3 .
then , the influence matrix @xmath80 takes the form @xmath275 where the column split is @xmath221 and the row split is @xmath276 .
now consider the correction operator @xmath157 for @xmath154 in the upper part of @xmath240 in eq .
( [ gn2 ] ) .
we already know from eq .
, that the @xmath260 part of @xmath157 is the forward cone of @xmath15
. therefore we must have @xmath277 . further comparing , with eq . which states that the restriction of the @xmath278 part of the correction operator @xmath157 to @xmath148 is the @xmath15th column of @xmath250 .
but this is precisely the @xmath15th row of @xmath269 , thus we infer that @xmath279
. next , consider row @xmath232 of the lower part of @xmath240 in eq .
( [ gn2 ] ) .
row @xmath232 is @xmath280 . the corresponding stabilizer operator @xmath176 , when inserted into the overlap @xmath281 as in eq .
( [ corr1 ] ) , flips the measurement basis at qubit @xmath232 and of qubits @xmath282 with @xmath283_l=1 $ ] .
it further flips the measurement outcomes at qubits @xmath284 with @xmath285_m=1 $ ] .
therefore , @xmath286 we thus arrive at the normal form eq .
( [ gnf ] ) .
let us return to our discussion from the beginning of section [ stabgnf ] , on which pieces of information are needed to describe an mbqc . at this stage ,
apart from the set of measurement angles which do not enter our discussion , we remain with two pieces of data specifying a given mbqc , namely @xmath240 and the processing relations eq . ( [ to7a ] ) , ( [ to7b ] ) .
but it does nt stop there . as the results of @xcite , @xcite , @xcite show , the resource state @xmath18 , the measurement planes @xmath115 and the influence matrix @xmath80being part of the classical processing relations eq .
( [ to7a ] ) , ( [ to7b])are not independent .
specifically , given the sets @xmath2 of first - measurable and @xmath3 of last - measurable qubits in addition to @xmath18 and @xmath115 , the temporal order ( generated by @xmath80 ) can be worked out completely . here
we prove a statement about the interdependence of @xmath240 and the mbqc classical processing relations which goes the opposite direction .
namely we show that the classical processing relations eq .
( [ to7a ] ) , ( [ to7b ] ) uniquely specify the stabilizer generator matrix @xmath240 , i.e. , the pair @xmath18 , @xmath115 up to equivalence ; see theorem [ onetoonetwo ] below .
thus , only two pieces of data are needed to specify an mbqc that satisfies the determinism constraints , namely the measurement angles and the classical processing relations for the measurement outcomes .
a further question is whether the temporal relations compatible with a resource state @xmath18 and set of measurement planes @xmath115 fit into a common framework . in this regard , we show that the classical processing relations ( containing the temporal order ) for mbqc with a fixed resource state and set of measurement planes , for extremal pairs @xmath287 , are in one - to - one correspondence with the bases of @xmath240 , c.f .
theorem [ onetoonethree ] .
we now present four theorems on the mutual dependence of the resource state and the classical processing relations .
[ onetooneone ] consider an mbqc on a stabilizer state @xmath18 , with fixed measurement planes and an extremal pair of gauge input set @xmath149 and computational output set @xmath148 .
then , the relations @xmath253 , and @xmath254 for an optimal output @xmath101 are unique . that is , once the resource state @xmath18
, the measurement planes and @xmath149 , @xmath148 are fixed , there is no freedom left to choose the classical processing relations .
they are uniquely determined by the former .
in particular , for fixed stabilizer state @xmath18 and measurement planes , @xmath288 , @xmath289 etc .
a corollary of theorem [ onetooneone ] is that given the measurement planes and an extremal pair @xmath149 , @xmath148 , the resource state @xmath18 uniquely determines the influence matrix @xmath80 .
one may ask how restrictive a condition the extremality of the pair @xmath149 , @xmath148 is . in this regard , note [ extr ] consider an mbqc on a fixed resource stabilizer state for fixed measurement planes , with an influence matrix @xmath80 and input and output sets @xmath290 , @xmath291 , such that no qubit @xmath292 can be individually gauged with respect to @xmath290 , @xmath291 .
then , there exists an extremal pair @xmath272 , @xmath274 such that @xmath293 .
the input set @xmath290 and the output set @xmath291 which appear in theorem [ extr ] are uniquely specified by @xmath80 through definition [ iodef ] .
@xmath294 is uniquely specified by the pair @xmath149 , @xmath148 through theorem [ onetooneone ] .
theorem [ extr ] states that all temporal relations for an mbqc , subject to the extra condition on the qubits which can be individually gauged , arise from extremal pairs @xmath149 , @xmath148 . by establishing theorem [ extr ] we trade the condition of the pairs @xmath149 , @xmath148 being extremal for the condition that no qubit in @xmath122 can be individually gauged .
the latter is a more meaningful condition .
suppose a qubit @xmath15 in @xmath122 could be individually gauged wrt @xmath2 , @xmath3
. then @xmath295 exists . for any @xmath17 with @xmath296
, @xmath297 is a valid correction operator for qubit @xmath17 , and @xmath298 .
hence , @xmath15 could be removed from all forward cones and thereby be made a qubit in @xmath2 . by imposing the extra condition in theorem [ extr ]
, we exclude temporal relations where certain qubits could be in the input set @xmath2 but are nt .
theorem [ onetooneone ] is mute on the question of which extremal pairs @xmath149 , @xmath148 are admissible .
theorem [ onetoonethree ] below describes how much freedom remains for the choice of the classical processing relations , given @xmath240 .
[ onetoonethree ] for mbqc with a fixed resource stabilizer state @xmath18 and fixed measurement planes , the classical processing relations for extremal @xmath149 , @xmath148 , as specified by the matrices @xmath251 , @xmath252 , @xmath80 , @xmath1 and the sets @xmath287 , are in one - to - one correspondence with the bases of the matroid @xmath240 .
after we have justified our restriction to extremal pairs of gauge input and computational output sets in theorem [ extr ] and have characterized the set of temporal relations compatible with a given resource state and set of measurement planes in theorem [ onetoonethree ] , we now return to theorem [ onetooneone ] , and show that a converse also holds .
[ onetoonetwo ] consider an mbqc on a stabilizer state @xmath18 , with classical processing relations @xmath253 , @xmath299 for an optimal classical output @xmath101 , such that @xmath300 .
then the classical processing relations uniquely specify the stabilizer generator matrix @xmath240 in the @xmath241-basis , i.e. the resource stabilizer state @xmath18 and set @xmath115 of measurement planes up to equivalence .
our proposal of mbqc as a toy model for spacetime is inspired by malament s theorem @xcite , stating that for a continuous spacetime manifold , the metric is determined by the temporal order of spacetime events up to a conformal factor .
our theorem [ onetoonetwo ] shows close resemblance with malament s theorem if we assert the following correspondence : @xmath301{corresp}}\ ] ] while the first line of the correspondence ( [ corresp ] ) appears intuitive , the second line really comes about by identifying the left - over pieces on either side .
nevertheless , it seems a match : a reel - valued parameter , the measurement angle at a given qubit is mapped to a real - valued parameter , the scale factor at the corrsponding spacetime point .
( note , though , that the domain of the measurement angle is a compact set while that of the scale factor is not ) .
a physical interpretation for this correspondence remains to be found .
the support for the conjectured correspondence ( [ corresp ] ) comes from its third line . by analogy with malament s theorem
, the temporal order of measurement events should , up to equivalence at least , determine the resource state @xmath18 and the set @xmath115 of measurement planes .
theorem [ onetoonetwo ] almost gives this result .
but the correspondence seems imperfect : it is not the temporal order alone contained in the influence matrix @xmath80which yields the pair @xmath18 , @xmath115 up to equivalence .
rather it is the entire classical processing relations eq .
( [ to7a ] ) , ( [ to7b ] ) , specified by the four matrices @xmath80 , @xmath251 , @xmath252 , @xmath1 .
but we can do better !
namely , it can be shown that the four matrices @xmath80 , @xmath251 , @xmath252 , @xmath1 can be assembled to an influence matrix @xmath302 of an mbqc on a slightly bigger resource state @xmath303 .
specifically , the support of @xmath303 is , in comparison to @xmath18 , enlarged at the temporal boundaries @xmath149 , @xmath148 . by this extension ,
all information contained in the classical processing relations becomes temporal information . to obtain @xmath302
we first choose a pair @xmath304 ( @xmath117 follow from @xmath80 ) , and extract from @xmath305 the corresponding matrices @xmath306 wrt to the normal form eq .
( [ gnf ] ) specified by @xmath307 .
we assemble the composite matrix @xmath308 which is a square matrix of size @xmath309 .
@xmath302 is the influence matrix for mbqc on a bigger resource state @xmath303 constructed from @xmath18 . for the support @xmath310 of
@xmath303 we require two additional sets of qubits , @xmath311 and @xmath312 , with @xmath313 , such that @xmath314 .
@xmath303 is obtained from @xmath18 by the following construction : @xmath315{psipsipr}}\ ] ] therein , the gates @xmath316{lambda_s}}=\lambda_{s}$ ] are @xmath31-controlled @xmath31-gates , i.e. , @xmath317 , @xmath318 , etc , and the extra qubits in @xmath311 and @xmath312 are initially prepared in the eigenstate of @xmath30 with eigenvalue 1 . with the definition eq .
( [ psiext ] ) of @xmath303 , the labelling of the blocks of rows and columns for the matrix on the r.h.s . of eq .
( [ tlcext ] ) is @xmath319 for the columns and @xmath320 for the rows .
we thus find that all information in the classical mbqc processing relations is temporal information for the computation on a slightly extended resource state .
theorem [ onetooneone ] is an immediate consequence of lemma [ nf ] .
assume that @xmath2 is valid input set and @xmath3 is a valid output set for a given mbqc .
then , the stabilizer generator matrix of the resource state can be written in the @xmath241-basis as @xmath321 for some matrices @xmath268 , @xmath269 , @xmath270 and @xmath271 .
the influence matrix @xmath80 can be obtained as @xmath322 with the column split between @xmath2 and @xmath122 , and the row split between @xmath323 and @xmath3 .
the matrices @xmath269 and @xmath324 do not necessarily have maximal row - rank . by row transformations of @xmath325
we extract the dependent rows , and obtain @xmath326 however , note that each row in the third set of rows in the above matrix can either be interpreted as a correction operator @xmath157 , with non - empty forward cone @xmath86 , for some @xmath327 , or a gauge operator @xmath176 for some @xmath328 .
the former is ruled out because every @xmath327 must have an empty forward cone .
the latter is ruled out by the assumption that no qubit outside the set @xmath2 can be individually gauged .
therefore that set of rows must identically vanish .
since @xmath240 has full row rank , so does the matrix @xmath329 appearing in eq .
( [ gn3b ] ) .
we may then choose a set @xmath330 such that the columns of @xmath329 indexed by @xmath331 form a maximal independent set .
we set @xmath332 .
then , by further row transformations which do not affect @xmath333 , the matrix in eq .
( [ gn3b ] ) can be converted to @xmath334 @xmath335 has full row rank by construction .
we can therefore find a set @xmath272 such that the columns of @xmath335 indexed by @xmath149 form a maximal independent set .
for any such set @xmath149 we can convert the matrix in eq .
( [ gn4 ] ) fully into the normal form eq .
( [ gnf ] ) without affecting @xmath80 .
for any of the above choices for @xmath272 and @xmath336 , the resulting influence matrix @xmath337 can be extracted as @xmath338 with the column split between @xmath2 and @xmath122 , and the row split between @xmath323 , @xmath331 and @xmath339 . by comparison of eqs .
( [ t ] ) and ( [ tpr ] ) we verify @xmath293 .
_ remark : _ comparing eq . with the normal form in eq .
, we can write the stabilizer matrix in a slightly varied form that can be useful later . @xmath340 where @xmath341 and where @xmath342 .
denote by @xmath343 the set of bases of @xmath245 , and by @xmath344 the set of extremal classical processing relations of form eq .
( [ to7a ] ) , ( [ to7b ] ) , specified by the triple @xmath345 . then , the mapping @xmath346 exists and is a bijection .
( 1 ) existence of @xmath347 : by the normal form eq . (
[ gnf ] ) of @xmath348 , for a given basis @xmath349 the sets @xmath149 and @xmath148 are extracted as follows .
a qubit @xmath232 is in @xmath149 if and only if the corresponding column of @xmath350 appears in @xmath349 .
a qubit @xmath15 is in @xmath148 if and only if the corresponding column of @xmath351 does not appear in @xmath349 .
knowing @xmath149 and @xmath148 , @xmath352 is uniquely determined via theorem [ onetooneone ] .
( 2 ) surjectivity of @xmath347 : by definition of `` extremal '' . ( 3 ) injectivity of @xmath347 : given @xmath149 and @xmath148 , @xmath353 is unique .
we divide the proof into the following steps : i ) first , with @xmath354 , the processing relations stem from an extremal pair @xmath149,@xmath148 .
we show that given an extremal pair @xmath149 , @xmath148 , the matrices @xmath247 , @xmath248 , @xmath355 and @xmath250 are uniquely determined by the classical processing relations eq .
( [ to7a],[to7b ] ) .
ii ) from these matrices we can derive the corresponding normal form of the resource state uniquely . let us denote this by @xmath356 . in iii ) and iv ) we show that the following diagram commutes , which establishes the equivalence of normal forms of all extremal pairs . @xmath357 in eq .
, @xmath358 is a transformation on the normal form @xmath356 , dependent on @xmath359 . because of their independence
, we consider the transformations @xmath360 and @xmath361 separately in iii ) and iv ) respectively .
extracting @xmath247 , @xmath248 , @xmath355 and @xmath250 : given any set of inputs @xmath2 and outputs @xmath3 by theorem [ extr ] , we know that there always exists an extremal pair @xmath362 , where @xmath363 . by eq .
( [ tshape ] ) , @xmath249 is uniquely specified by @xmath80 . by the assumption of the classical output being optimal
, @xmath1 can be brought into the unique normal form of eq .
( [ zn ] ) by left multiplication with an invertible matrix and ( permuting the columns if necessary ) , @xmath222 can be extracted .
the matrix @xmath251 may not appear in eq .
( [ to7a ] ) in its normal form eq .
( [ hrtzshape ] ) , nonetheless for an invertible matrix @xmath359 , the vector @xmath74 specifying the measurement bases is invariant under @xmath364 , @xmath365 . by definition of @xmath149 ,
every qubit in @xmath149 can be individually gauged with respect to @xmath149 , @xmath148 . therefore , ( up to row permutations ) , we can choose @xmath359 such that @xmath366 where the row split is between @xmath149 ( upper ) and @xmath367 ( lower ) . @xmath359 and @xmath368 are unique .
the classical output @xmath101 is invariant under the transformation @xmath369 .
however , since eqs .
( [ to7a ] ) and ( [ to7b ] ) refer to @xmath133 in the same basis , @xmath370 is now fixed : @xmath371
. then , @xmath372 , with the @xmath359 of eq .
( [ hnf ] ) .
assembling the normal form : using the unique matrices @xmath247 , @xmath248 , @xmath249 , @xmath250 , by lemma [ nf ] , we can now assemble the normal form eq .
( [ gnf ] ) of the stabilizer generator matrix for the resource state @xmath18 . by assumption of theorem [ onetoonetwo ] ,
the matrices @xmath251 , @xmath252 , @xmath80 , @xmath1 describe a valid computation , and the normal form eq .
( [ gnf ] ) derived from them must thus yield a valid description of a quantum state .
in particular , all pauli operators specified by the rows of @xmath373 in eq .
( [ gnf ] ) must commute .
( the rows are independent by design of the normal form . )
we have thus constructed a description of @xmath18 . since @xmath247 , @xmath248 , @xmath249 ,
@xmath250 are unique , so is @xmath18 .
we now proceed to construct the stabilizer @xmath20 from the processing relations when @xmath149 and @xmath148 are not specified . from the classical processing relation eq .
( [ to7a ] ) we can still extract the input set @xmath2 and the output set @xmath3 , by testing which rows and columns of @xmath80 identically vanish .
then , the possible choices for @xmath149 and @xmath148 are limited to @xmath272 and @xmath274 by definition .
equivalence under @xmath374 . in order to prove this we rely on the slightly variant version of the normal form of @xmath240 as shown in eq .
( [ gn5 ] ) which also includes additional detail about the correction operators .
as noted above , the different normal forms for @xmath251 can be interconverted by right - multiplication with an invertible matrix @xmath359 ( change of basis for @xmath133 ) , i.e. @xmath375 , where @xmath376 .
under such a transformation , the upper part of the normal form eq .
( [ gnf ] ) for @xmath240 remains unchanged , and the lower part is transformed @xmath377 .
invertible row transformations on @xmath240 leave @xmath18 unchanged , and @xmath18 is thus independent on the precise choice of @xmath272 .
equivalence under @xmath378 . proving that a different choice of @xmath148 does not change the stabilizer state is a little more complicated . we proceed in the following manner
. from theorem [ onetoonethree
] , we know that @xmath379 are the bases of a matroid . therefore for two distinct computational output sets @xmath148 and @xmath380 , there exists another computational output set @xmath381 , where @xmath382 and @xmath383 .
therefore , it suffices if we show that the stabilizer state does not change if we change the computational output set from @xmath148 to @xmath384 .
assume that the classical relation for the computational output set @xmath148 is given as @xmath385 where the column split of @xmath1 is between @xmath386 , @xmath331 , and @xmath148 .
the corresponding normal form ( with the column split in @xmath260-part @xmath387 ) is @xmath388 where @xmath389 .
suppose that we transform @xmath148 to @xmath390 , where @xmath391 and @xmath392 .
without loss of generality assume that @xmath232 is the last column of @xmath393 .
( it can not be an all zero column because , then it would not be possible for it to be in @xmath148 . )
let @xmath394 and @xmath395 .
then @xmath396 acting on @xmath1 achieves the transformation @xmath148 to @xmath384 .
@xmath397 we claim that this same transformation can be effected by row transformations of @xmath240 .
first let us focus on the middle set of rows in @xmath240 , namely the correction operators for @xmath331 .
then acting by @xmath398 gives us @xmath399 now if take the last row in eq .
, namely @xmath400 and add @xmath401 to the top set of rows in eq .
we obtain @xmath402 showing the equivalence of @xmath250 and @xmath403 .
the equivalence of @xmath248 and @xmath404 under @xmath359 can be shown in exactly the same fashion as for @xmath405 and @xmath406 .
this concludes the proof that the extremal classical relations completely determine the stabilizer state .
in this section we provide a different angle at theorem [ onetooneone ] , namely we show that the gauge transformations eq .
( [ g1 ] ) impose severe constraints on the possible temporal orders for a given stabilizer state and set of measurement planes .
the gauge transformations eq .
( [ g1 ] ) act on @xmath74 and @xmath77 , and can therefore have a non - trivial effect on the classical processing relation eq .
( [ to7a ] ) , @xmath407 .
we now study this effect . since the transformations eq . ( [ g1 ] ) are caused by the insertion of stabilizer operators into the state overlap in eq .
( [ corr1 ] ) , they do not change the physical situation .
therefore , the temporal relations before and after any such transformation must be equally valid , although not necessarily identical . by insertion of the stabilizer operator into the overlap @xmath408 , the stabilizer of @xmath18 and the the sets @xmath149 , @xmath148 do not change .
therefore , by theorem [ onetooneone ] , the matrices @xmath80 and @xmath251 do not change .
thus , besides @xmath74 and @xmath77 , all that can change in the relation eq .
( [ to7a ] ) under a transformation eq .
( [ g1 ] ) is @xmath133 .
the following two viewpoints are always equivalent : ( a ) the relation @xmath409 , under the action eq .
( [ g1 ] ) of a @xmath146 on @xmath410 is changed into an equivalent such relation @xmath411 , with @xmath412 .
( b ) the relation @xmath413 _ remains invariant _ under all transformations @xmath146 , acting on the triple @xmath414 .
we choose the latter viewpoint .
we now infer the action of the transformations @xmath146 on @xmath133 . without loss of generality
we assume that the relations eq .
( [ to7a ] ) are given with @xmath251 in its normal form eq .
( [ hrtzshape ] ) , @xmath415 furthermore , we assume that the pair @xmath149 , @xmath148 is extremal . since @xmath272 by definition , for each @xmath169 , @xmath170 only depends on @xmath133 but not on the measurement outcomes @xmath77 , @xmath416 .
now , the correction operators @xmath157 , @xmath154 derived from the normal form eq .
( [ gnf ] ) of @xmath240 , @xmath417 has no @xmath30-part .
therefore , the corresponding transformations @xmath146 do not flip @xmath170 , for all @xmath169 . in order to preserve the relation @xmath418
, they thus leave @xmath133 unchanged . now consider the other stabilizer generators , @xmath176 , @xmath169 , obeying the conditions eq .
( [ corrgauge ] ) . by construction
, @xmath419 flips @xmath170 but no other @xmath420 , for @xmath421 .
hence , to preserve the relations @xmath418 , it must also flip @xmath422 , but no other @xmath423 , @xmath424 . thus , for a stabilizer element @xmath145 , @xmath425 therein , we have assumed that the basis choice for @xmath133 is such that the matrix @xmath251 appearing in eq .
( [ to7a ] ) is of normal form eq .
( [ hrtzshape ] ) .
we have now fully specified the action of @xmath146 on the triple @xmath426 , c.f .
( [ g1 ] ) , ( [ g2 ] ) .
mbqcs satisfy the invariance condition @xmath427 it is evident that the requirement ( [ gk1 ] ) of invariance of the processing relations ( [ to7a ] ) under the gauge transformations poses constraints on the possible matrices @xmath80 and @xmath368 .
in fact , as we show below , given @xmath148 the matrices @xmath80 and @xmath368 are uniquely specified uniquely by the above invariance condition . to check the invariance condition eq .
( [ gk1 ] ) in a specific case , we return to our 3-qubit cluster state example of section [ oto ] .
we consider the effect of the transformations induced by generators @xmath428 , @xmath429 and @xmath430 on the processing relations eq .
( [ trelex ] ) . as noted earlier , @xmath206 .
then , with eqs .
( [ g1 ] ) and ( [ g2 ] ) , @xmath431 it is easily checked that the relation eq .
( [ trelex ] ) is invariant under the transformations @xmath432 , @xmath433 and @xmath434 of eq .
( [ gex3 ] ) . however , if the transformations are restricted to @xmath74 , @xmath77 , the relation eq .
( [ trelex ] ) is no longer invariant under the transformation induced by @xmath71 .
we now return to the general case and show that , given the set @xmath148 and the action eq .
( [ g1 ] ) , ( [ g2 ] ) of the gauge transformations on the triple @xmath435 , the invariance condition eq .
( [ gk1 ] ) uniquely specifies the classical processing relations eq .
( [ to7a ] ) for the adaption of measurement bases .
recall that we write the stabilizer generator matrix for @xmath18 in the @xmath241-basis as @xmath436 .
then , for the stabilizer generator @xmath437 corresponding to the @xmath15-th row of @xmath240 , with eq .
( [ g1 ] ) the action of the gauge transformation @xmath438 on @xmath77 , @xmath74 is @xmath439 with eq .
( [ g2 ] ) , the action of @xmath438 on @xmath133 is @xmath440 here , @xmath441 denotes @xmath442 restricted to @xmath149 .
then , the condition eq .
( [ gk1 ] ) for invariance of @xmath443 under @xmath438 becomes @xmath444 this condition must hold for all stabilizer generators @xmath445 simultaneously , hence @xmath446 by definition , the qubits in @xmath149 have empty backward cones , and the qubits in @xmath148 have empty forward cones , hence @xmath80 is of the form @xmath447 where the column split is @xmath221 and the row split is @xmath276 , c.f .
( [ tshape ] ) . by right - multiplication of relation eq .
( [ symcon ] ) with a suitable matrix , we transform @xmath448 into a matrix of form @xmath449 where the column split is between @xmath149 and @xmath261 . by definition of @xmath149 ,
such a transformation is always possible . under the same transformation , @xmath450 inserting the above into eq .
( [ symcon ] ) , we find that @xmath251 must be of normal form eq .
( [ hrtzshape ] ) , @xmath451 , and @xmath452 now , @xmath453 must be an invertible matrix .
this is the condition that , by definition of @xmath148 , every measurement outcome in @xmath158 is correctable . then , @xmath454 hence the relation @xmath407 is uniquely specified .
in addition to eq .
[ gk1 ] , we also require invariance of the classical output under the transformations eq .
( [ g1 ] ) , ( [ g2 ] ) , @xmath455 like eq .
( [ gk1 ] ) , eq .
( [ gk2 ] ) is a determinism constraint .
if for a single output bit @xmath456 the relation @xmath457 is not invariant under all gauge transformations eq .
( [ g1 ] ) , ( [ g2 ] ) , then the value of @xmath456 is guaranteed to be random , and thus useless as readout bit of a computation .
specifically , [ det ] assume an mbqc where the relation @xmath458 is invariant under the gauge transformations eq .
( [ g1 ] ) , ( [ g2 ] ) , but an output bit @xmath456 exists whose defining relation @xmath457 is not invariant under the action of @xmath146 for some @xmath459 .
then , the value of @xmath456 is completely random , independent of the choice of measurement angles .
_ proof of lemma [ det ] .
_ for simplicity , consider first the special case where @xmath459 acts trivially on @xmath133 , @xmath460 .
we may then write @xmath461 for an offset @xmath462 .
we call the string @xmath463 of measurement outcomes on @xmath464 even ( odd ) if it has even ( odd ) weight .
we denote the local post - measurement state on qubit @xmath15 by @xmath465 , where @xmath23 is the measurement angle , @xmath140 the measurement outcome and @xmath141 specifies the chosen measurement basis . under the transformation @xmath146 , @xmath466 , where , by assumption
, @xmath467 is odd .
now , the probability of outputting @xmath468 is @xmath469 thus , @xmath470 .
note that in transitioning from the third to the fourth line of eq .
( [ rand ] ) we have used the invariance property eq .
( [ gk1 ] ) , i.e. , the assumption that the adaption of measurement bases is deterministic . in the general case , @xmath471 .
we note that we can choose any gauge fixing @xmath133 , and thus @xmath472 . by an argument analogous to the above we then find @xmath473 . @xmath200
it has been established earlier that ` time ' in measurement based quantum computation is a transitive binary relation .
if a given mbqc is to be deterministically runnable , then we require a strict partial ordering among the measurement events . if , on the other side , in pursuing an analogy with general relativity we allow for closed time - like curves , then antisymmetry and irreflexivity are not required . given this setting , in the previous sections of this paper
we have analyzed the constraints on mbqc temporal relations imposed by a group of gauge transformations . in this section ,
we provide another angle at the question of what time in mbqc is .
namely , we introduce a second group of symmetry transformations on mbqcs which is related to local complementation @xcite - @xcite .
these transformations in general change the mbqc on which they act , but leave temporal relations invariant .
the possible temporal relations in mbqc thus label representations of this group . according to eq .
( [ obs ] ) , for any qubit @xmath15 in a given resource state , the local observable measured to drive the computation is @xmath474= \cos \varphi_a\ , \sigma_\phi^{(a ) } + ( -1)^{q_a}\sin \varphi_a\ , \sigma_{s\phi}^{(a)}.\ ] ] therein , @xmath141 is a linear function of the measurement outcomes @xmath475 , c.f .
( [ to7a ] ) .
let s see what happens if we use a different rule for the adjustment of measurement bases , namely @xmath476= ( -1)^{q_a } \cos \varphi_a\ , \sigma_\phi^{(a ) } + \sin \varphi_a\ , \sigma_{s\phi}^{(a)},\ ] ] that is , if @xmath73 , to obtain @xmath477 $ ] we are flipping the observable @xmath478 $ ] about the @xmath479-axis rather than the @xmath30-axis . comparing eqs .
( [ obs ] ) and ( [ lobspr ] ) , we find that @xmath480 \equiv ( -1)^{q_a}o_a[q_a],\ ] ] independent of the measurement angle @xmath23 .
the measurements of @xmath481 $ ] and @xmath482 $ ] are always in the same basis for the same @xmath141 , and the measured eigenvalues differ by a factor of @xmath483 .
we call the transformation @xmath484 : o_i[q_i ] \longrightarrow o'_i[q_i]$ ] _ flipping of the measurement plane at qubit @xmath232_. on the elementary degrees of freedom , namely the resource state @xmath18 , the pauli observables @xmath485 , @xmath486 ( action on @xmath485 is implied ) , and the measurement angle @xmath487 , the flipping @xmath484 $ ] acts as @xmath488 : \begin{array}{lcl } \sigma_\phi^{(i ) } & \longleftrightarrow & \sigma_{s\phi}^{(i)},\\ \varphi_i & \longrightarrow & ( -1)^{q_i}\frac{\pi}{2 } - \varphi_i,\\ of @xmath484 $ ] on the pauli operators @xmath489 and the measurement angles @xmath490 , for @xmath491 is trivial .
all our considerations are independent of the values of the measurement angles . in particular , the second part of the transformation eq .
( [ tau_s ] ) does not affect temporal order .
we now discuss the effect of the flipping @xmath492 $ ] , @xmath33 on a given mbqc .
denote the measurement outcome of a measurement of @xmath493 by @xmath494 . by eq .
( [ obsid ] ) , the following two measurement procedures are always equivalent .
( i ) measuring @xmath482 $ ] and outputting @xmath140 , and ( ii ) measuring @xmath481 $ ] and outputting @xmath495 .
we may call the device that performs procedure i a @xmath496-box , and the device that performs procedure ii a @xmath497-box . then , . the prime in the @xmath497-box accounts for the fact that not the measurement outcome @xmath494 itself is outputted , but rather the locally post - processed value @xmath498 .
now , instead of outputting @xmath499 , the device at @xmath15 may only output @xmath494 ( that is an @xmath500-box ) , and the classical post - processing relations for the adaption of measurement bases are modified accordingly , i.e. @xmath501 .
can the resulting relations again be written in a form @xmath502 ?
we now attempt transforming a @xmath496-box into an @xmath500-box at qubit @xmath15 .
the vectors of measurement outcomes @xmath77 and @xmath503 are related via @xmath504 . inserting this relation into eq .
( [ to7a ] ) , we obtain @xmath505 case i : @xmath506 .
physically , this means that the measurement basis at the flipped qubit @xmath15 does not depend on the measurement outcome at @xmath15 , before the transformation .
multiplying eq .
( [ star ] ) with @xmath507 from the left yields @xmath508 . inserting back into eq .
( [ star ] ) , we obtain @xmath509 likewise , @xmath510 eqs .
( [ tlc ] ) and ( [ tlccons ] ) completely describe the effect of the flipping @xmath492 $ ] of the measurement plane at qubit @xmath15 on the classical processing relations eq .
( [ to7a ] ) , ( [ to7b ] ) . _
remark : _ if the matrices @xmath305 are given their normal form eq .
( [ hrtzshape ] ) wrt the pair @xmath307 then the flipping of the measurement plane at any vertex @xmath15 with @xmath506 leaves this normal form intact .
we can therefore state a transformation rule equivalent to eqs .
( [ tlc ] ) , ( [ tlccons ] ) for the matrices @xmath511 .
this rule is , in fact , simpler .
we again consider the composite matrix @xmath302 , c.f .
( [ tppl ] ) , @xmath512 the effect of flipping of the measurement plane at @xmath15 then is @xmath513 thus , the rule is just the same as eq .
( [ tlc ] ) for the original influence matrix @xmath80 .
case 2 : @xmath514 . in this case
, the measurement basis at @xmath15 does depend on the measurement outcome at @xmath15 .
this is an example for a closed time - like curve ( only involving the measurement device at @xmath15 ) , and an obstacle to deterministic runnability .
now , the matrix @xmath515 on the left side in eq .
( [ star ] ) is not invertible , @xmath516 .
hence , the relation ( [ star ] ) can not be solved for @xmath74 in this case .
there is no relation eq .
( [ to7a ] ) with the same sets @xmath149 , @xmath148 before and after flipping .
we now discuss the consequences of flipping measurement planes for the above two cases .
we now return to the above case 1 , namely when flipping of a measurement plane yields a computation with a new relation eq .
( [ to7a ] ) .
note that the computation before and after the flip generate the same output distribution . flipping a @xmath496-box into an @xmath497-box
is an equivalence transformation , only based on the operator identity eq .
( [ obsid ] ) .
changing an @xmath497-box into an @xmath500-box is again an equivalence transformation , provided it can be carried out .
the influence matrices @xmath80 and @xmath517 before and after the flipping , respectively , are in general not the same , c.f .
( [ tlc ] ) .
however , @xmath80 and @xmath517 still generate the same temporal order , as we now show .
[ topres ] be @xmath80 an influence matrix with @xmath518
. then , @xmath80 and @xmath519 generate the same temporal relation under transitivity .
_ proof of lemma [ topres ] .
_ let s introduce a shorthand @xmath520 for @xmath521 ( meaning that the measurement outcome at @xmath15 influences the measurement basis at @xmath522 ) .
now , we have to show that @xmath523 , for any @xmath517 generated from @xmath80 by the transformation eq .
( [ tlc ] ) .
\(i ) `` @xmath524 '' : assume that @xmath525
. then there exists a sequence of measurement events @xmath526 .
can we break the arrow @xmath520 , say ? to investigate this , let us rewrite the transformation rule eq .
( [ tlc ] ) for the flip @xmath484 $ ] as @xmath527 : \begin{array}{lclr } \textbf{fc}(a ) & \longrightarrow & \textbf{fc}(a ) \oplus
\textbf{fc}(i ) , & \mbox{if } i \in fc(a),\\ \textbf{fc}(a ) & \longrightarrow & \textbf{fc}(a ) , & \mbox{if } i \not\in fc(a ) , \end{array}\ ] ] case 1 : @xmath528 before the transformation @xmath484 $ ] .
then , @xmath529 .
since @xmath518 by assumption , @xmath528 after the transformation @xmath484 $ ] .
sub - case 1a : @xmath530 before the transformation @xmath484 $ ] . since @xmath518 ( @xmath531 ) , @xmath530 after the transformation @xmath484 $ ] .
thus @xmath532 after the transformation , and hence @xmath533 .
sub - case 1b : @xmath534 before @xmath484 $ ] .
then , @xmath520 remains after the transformation .
case 2 : @xmath535 before the transformation @xmath484 $ ] .
then @xmath520 after @xmath484 $ ] .
thus , in all cases @xmath533 , and therefore @xmath536 .
\(ii ) `` @xmath537 '' : from eq .
( [ tpr3 ] ) , @xmath484 ^ 2 = i$ ] .
@xmath200 apply a series of transformations eq .
( [ tlc ] ) on an initial influence matrix @xmath80 with vanishing diagonal part may produce an influence matrix with a non - vanishing diagonal part .
thus , the application of the transformation eq .
( [ tlc ] ) is restricted . to circumvent this problem
, we introduce a modified transformation @xmath538 : t \longrightarrow t ' = t + t\textbf{e}_i \textbf{e}_i^tt + { \cal{d}}(t\textbf{e}_i \textbf{e}_i^tt ) \mod 2.\ ] ] clearly , this transformation takes influence matrices with vanishing diagonal part to influence matrices with vanishing diagonal part , and thereby avoids the problem of restricted applicability of transformation eq .
( [ tlc ] ) .
note that the transformation eq .
( [ tlc2 ] ) has the form of local complementation , albeit the influence matrix @xmath80 that it acts on will in general not be symmetric . but
what is the physical significance of transformation eq .
( [ tlc2 ] ) ?
the only additional effect of the transformation @xmath539 $ ] over @xmath484 $ ] is the cancelling of the diagonal part of the influence matrix after the transformation , c.f .
the last term in eq .
( [ tlc2 ] ) . this can be achieved by a local unitary that exchanges @xmath540 on a respective qubit .
the action of @xmath539 $ ] on the elementary degrees of freedom therefore is @xmath541 : \begin{array}{lcll } \sigma_\phi^{(i ) } & \longleftrightarrow & \sigma_{s\phi}^{(i)},\\ \varphi_i & \longrightarrow & ( -1)^{q_i}\frac{\pi}{2 } - \varphi_i , \\
\sigma_s^{(j ) } & \longleftrightarrow & \sigma_{s\phi}^{(j ) } , & \forall j \in fc(i ) \cap bc(i),\\ \end{array}\ ] ] we find that the local measured operators for all qubits @xmath542 change in a way that can not be accommodated by a change of the respective measurement angle . for those qubits ,
the new measured observables lie in a different equatorial plane of the bloch sphere .
therefore , the transformation eq .
( [ tlc3 ] ) , unlike the transformation eq .
( [ tlc ] ) , does _ not _ necessarily map a given computation onto itself .
what it does , however , is mapping a given computation to a computation with the same temporal relation .
[ ti ] be @xmath80 an influence matrix with @xmath518
. then , @xmath80 and @xmath543 generate the same temporal relation under transitivity .
proof of lemma [ ti ] . assuming the initial influence matrix has vanishing diagonal part , we split the transformation @xmath544 into two steps , namely @xmath545 and @xmath546 . by lemma [ topres ] , @xmath80 and
@xmath517 generate the same temporal order .
now assume that @xmath547 for some @xmath548 , @xmath549 .
this requires that @xmath550 .
then , we also have @xmath551 and @xmath552 .
thus , @xmath553 and @xmath554 .
the closed time - like curve involving @xmath555 is not changed by setting @xmath556 .
all other relations trivially remain unaffected by the transformation @xmath557 .
@xmath200 we may now consider the orbit @xmath558 of a particular influence matrix @xmath80 under local complementation eq .
( [ tlc3 ] ) . every local complementation @xmath539 $ ] permutes the elements of @xmath558 .
therefore , we have a homomorphism @xmath539 \longrightarrow p[i]$ ] where @xmath559 $ ] is a permutation matrix of size @xmath560 .
the matrices @xmath561|\ , i \in \omega\}$ ] generate a representation of the local complementation group .
the representation acts on a given orbit @xmath558 of influence matrices , and all @xmath562 give rise to the same temporal relation under transitivity .
therefore , this temporal relation is a label of the given representation .
all possible temporal relations in mbqc arise as labels of representations of the local complementation group .
in this section we return to the case of @xmath563 before flipping the measurement plane at qubit @xmath232
. such a qubit @xmath232 can not be in @xmath149 , since @xmath272 by definition . if @xmath563 then @xmath564 .
the backward cones of all qubits in @xmath232 are empty by definition of @xmath2 , however .
likewise , @xmath565 . if @xmath563 then @xmath566 .
however , @xmath273 for all @xmath163 .
thus , there is only one case to consider , namely @xmath567 . in this case
, there exists a correction operator @xmath568 for qubit @xmath232 before the flipping , @xmath569 . after flipping at @xmath232 ,
this operator turns into @xmath570(k(i ) ) = \sigma^{(i)}_\phi \otimes k(i)|_{\omega \backslash i}=:\overline{k}'(i).\ ] ] that is , the operator @xmath484(k(i))$ ] resulting from flipping at @xmath232 is a gauge type operator , c.f .
( [ corrgauge ] ) .
thus , the flipping transformation @xmath484 $ ] ( when @xmath563 ) enlarges @xmath149 by one qubit , @xmath570:\ ; i_{\text{gauge } } \longrightarrow i_{\text{gauge } } \cup \{i\},\;\ ; \mbox{if } t_{ii}=1.\ ] ] furthermore , after the flipping at @xmath232 there no longer is a correction operation for qubit @xmath232 , hence @xmath570:\ ; o_{\text{comp } } \longrightarrow o_{\text{comp } } \cup \{i\},\;\ ; \mbox{if } t_{ii}=1.\ ] ] this has two consequences .
first , the forward cone of @xmath232 becomes empty .
in particular @xmath518 after the flipping .
thus , the closed time - like curve consisting of qubit @xmath232 has been removed .
second , an additional bit of optimal classical output is being created by the flipping at @xmath232 .
what does that output bit signify ?
recall that before the flipping at @xmath232 , the rule for adjusting the measurement basis at @xmath232 is @xmath571 for some set @xmath572 . here , we have dropped a constant offset @xmath573 on the r.h.s .
the symbol `` ! '' above the equality means that equality is a requirement for the correctness of the computation , but it can not be deterministically implemented . as follows from eq .
( [ obsid ] ) , the measurement outcomes before and after the flip , @xmath230 and @xmath574 are related via @xmath575 . for all the other qubits , @xmath576 . substituting this into the above relation
, we obtain @xmath577 thus , the additional output bit @xmath578 is a flag . if @xmath579 then the computation succeeded , and if @xmath580 then it did not .
now suppose that the problem solved by the given mbqc is in np .
then , this flag bit is not necessary .
the remaining output may be efficiently checked for correctness anyway .
thus , one may safely discard the extra bit @xmath231 of output .
not post - selecting on @xmath579 can , if anything , only increase the success probability of the computation .
we thus arrive at [ ctcrem ] be @xmath581 an mbqc with a classical output @xmath101 and influence matrix @xmath80 such that @xmath563 , i.e. , @xmath581 has a closed time - like curve involving a single qubit @xmath582 .
be @xmath583 the mbqc with the same classical output @xmath101 , obtained from @xmath581 by flipping the measurement plane at @xmath232 .
then , the closed time - like curve of @xmath232 in @xmath581 is removed in @xmath583 . furthermore
, if @xmath581 solves a problem in the complexity class np with probability @xmath584 , then @xmath583 solves the same problem with probability @xmath585 . _ remark : _
lemma [ ctcrem ] does not guard against the inefficiencies of post - selection , in particular if multiple ctcs of length 1 are being removed . while the success probability after removing the ctcs is guaranteed not to be smaller than for the original computation with the ctcs ( which can only be executed using post - selection ) , neither it is provably significantly larger .
_ event horizons .
_ let us consider the flow of information between qubit @xmath232 whose measurement plane has been flipped and the other qubits . before the flip ( mbqc @xmath581 of lemma [ ctcrem ] ) , @xmath586 . after the the flip ( mbqc @xmath583 of lemma [ ctcrem ] ) , @xmath587 . in @xmath583 , since @xmath169 , no information for the adaption of measurement basis is flowing into site @xmath232 from the other sites .
likewise , since @xmath229 , no information for the adaption of measurement bases is flowing out of site @xmath232 .
finally , because of the normal form eq .
( [ hrtzshape ] ) , the measurement outcome @xmath230 appears in only one readout bit .
this readout bit is @xmath231 as given in l.h.s .
( [ s_tau ] ) , which is precisely the bit of classical output that can be discarded if the problem solved by the quantum computation is in np .
if @xmath231 is discarded , then no information is flowing out of the site @xmath232 at all .
thus , in summary , from the viewpoint of classical processing , qubit @xmath232 in @xmath583 becomes entirely disconnected from the computation .
it vanishes behind the mbqc counterpart of an event horizon . breaking a closed time - like curve of length 1 .
the looped qubit becomes an element of @xmath588 after flipping the measurement plane .
as such both its forward cone ( @xmath274 ) , and backward cone ( @xmath272 ) must be empty.,width=566 ] a closed time - like curve @xmath590 of length @xmath591 is a set of qubits such that the relations @xmath592 , @xmath593 , .. , @xmath594 ,
.. , @xmath595 , @xmath596 hold .
we call a closed time - like curve @xmath13 minimal if no proper subset of @xmath13 is a closed time - like curve .
minimal closed time - like curves are of the form @xmath597 that is , for a minimal closed time - like curve , no arrows other than the defining ones are present .
proof : assume @xmath598 such that @xmath599 .
then , @xmath600 is also a closed time - like curve , and a proper subset of @xmath13 .
hence , @xmath13 is not minimal .
@xmath200 regarding the embedding of @xmath13 in the overall temporal structure , we have @xmath601 proof : @xmath602 , @xmath603 , hence @xmath604 .
hence , @xmath605 .
furthermore , @xmath602 , @xmath606 .
hence , @xmath607 .
hence , @xmath608 .
@xmath200 now , the ctc @xmath13 can be broken up between qubits 1 and l. this time , no transformation @xmath609 is necessary .
the starting point 1 " for the labelling is of course arbitrary . to break up the ctc , we enlarge the sets of gauge input and computational output , @xmath610 and modify the correction and gauge operators @xmath611 the first line in eq .
( [ ktrans ] ) says that @xmath612 is re - interpreted as @xmath613 .
hence , qubit @xmath614 no longer has a correction operation and thus becomes a member of @xmath148 , as required in eq .
( [ iotrans ] ) . the second line in eq .
( [ ktrans ] ) ensures that @xmath615 for all @xmath154 such that @xmath616 .
thus , @xmath617 , as required in eq .
( [ iotrans ] ) . the third line in eq .
( [ ktrans ] ) makes the gauge operators compatible with the new normal form for @xmath618 ( based on @xmath619 and @xmath620 ) , and has no effect on temporal order . since , after the reshuffling of the gauge and correction operators according to eq .
( [ ktrans ] ) , @xmath621 and @xmath622 and we have successfully broken up the ctc @xmath13 between 1 and @xmath614 . since the computational output set has been enlarged by one qubit , @xmath614 , we have one bit @xmath623 of additional classical output .
what does it signify ? - when retracing the transformations eq .
( [ ktrans ] ) in the normal form eq .
( [ gnf ] ) of the stabilizer generator matrix @xmath240 , we find that the matrix @xmath222 is updated according to @xmath624 where @xmath625 is the characteristic vector of @xmath626 , restricted to @xmath627 , and @xmath628 is the matrix obtained from @xmath222 by deleting column @xmath614 .
thus we find that the new output bit @xmath629 is @xmath630 we compare this to the relation @xmath631 before the transformation eq .
( [ ktrans ] ) , and find the following interpretation of @xmath629 : the temporal relations before the transformation eq .
( [ ktrans ] ) contain a closed time - like curve and can therefore only be implemented probabilistically .
so one may as well assume @xmath632 , perform the measurement and later check whether the relation between @xmath633 and @xmath77 was obeyed . with eq .
( [ addlout ] ) , this check amounts to @xmath634 , and @xmath629 is thus a flag for the correctness of the computation . if the computational problem solved is in np then we can afford to discard this extra bit @xmath629 of output .
the solution can be efficiently checked anyway . in
no longer post - selecting on @xmath634 , we retain all the ` good ' cases while admitting additional ones .
the success probability of the algorithm does not decrease .
we thus arrive at [ ctcl2 ] be @xmath581 an mbqc with a classical output @xmath101 and influence matrix @xmath80 giving rise to a minimal closed time - like curve @xmath590 of length @xmath589 .
be @xmath583 the mbqc with the same classical output @xmath101 , obtained from @xmath581 by enlarging @xmath635 , @xmath636 .
then , the closed time - like curve @xmath13 is not present in @xmath583 .
furthermore , if @xmath581 solves a problem in the complexity class np with probability @xmath584 , then @xmath583 solves the same problem with probability @xmath585 .
we still need to check that by the transformation eq .
( [ ktrans ] ) of the correction and gauge operators we do not create additional closed time - like curves in exchange for removing the minimal ctc @xmath13 . in this regard , note that the middle line of eq .
( [ ktrans ] ) amounts to @xmath637 therefore , for any @xmath638 , @xmath639 only if @xmath92 or @xmath640 .
thus , @xmath639 only if @xmath91 before the enlargement of @xmath149 , @xmath148 .
therefore , no new closed time - like curves are being created . combining the above two cases of ctcs of length 1 and @xmath589 , respectively
, we can remove ctcs repeatedly until none remains .
the process must stop , since at some point every qubit is in the gauge input or / and the computational output set .
we thus arrive at the following [ ctcrem5 ] be @xmath581 an mbqc solving a problem in the complexity class np with probability @xmath584 .
then , all closed time - like curves can be removed from the influence matrix @xmath80 of @xmath581 , resulting in a deterministically runnable mbqc @xmath583 which solves the same problem with probability @xmath585 . _
theorem [ ctcrem5 ] does not guard against the inefficiencies of post - selection . while the success probability after removing the ctcs is guaranteed not to be smaller than for the original computation with the ctcs ( which can only be executed using post - selection )
, we can not prove that it is significantly larger .
in this paper we have suggested mbqc as a toy model for spacetime emerging from quantum mechanics .
we have analyzed the constraints on mbqc temporal order that arise from the quantum mechanical randomness of measurement , in combination with the computational requirement of preventing this randomness from affecting the quantum information processing .
we have identified two groups of symmetry transformations , a gauge transformation which leaves every mbqc including the temporal relations
invariant , and a second transformation , related to local complementation , which only preserves the temporal relations .
we have shown that for any given resource stabilizer state @xmath18 and set @xmath115 of measurement planes , the compatible temporal relations all arise from the bases of a matroid encompassing @xmath18,@xmath115 . finally , we have identified event horizons as a further piece of the phenomenology of general relativity which has a counterpart in mbqc . 1 .
we introduced a group of gauge transformations eq .
( [ g1],[g2 ] ) , and a group symmetry transformations eq .
( [ tlc2 ] ) , generated by flipping measurement planes .
both transformations preserve mbqc temporal orders .
can the two groups be unified ?
2 . some of the temporal relations admitted by the matroid @xmath240 contain closed time - like curves .
given a stabilizer state @xmath18 and set of measurement planes @xmath115 , can we find a similar algebraic ( or other ) structure which comprises only the partial orders ? can the partial order of measurements with the smallest set @xmath148 be efficiently computed ?
we have established that all temporal relations free of self - loops appearing in mbqc arise as representations of the local complementation group . however
, these representations are in general reducible . if we decompose them into irreps , where does temporal order go ?
[ qspace]in proposing mbqc as a toy model for spacetime , here we have focussed on the temporal part . what is a suitable notion of _ space _ that can be associated with a resource state and a set of measurement planes in mbqc ? 5
[ qdet]in mbqc , the link between the randomness in quantum mechanical measurement and temporal order is the principle that the randomness of measurement outcomes should not affect the logical processing . in a context more general than quantum computation , what could this principle be replaced by ?
_ remark regarding question [ qspace ] : _ with theorem [ onetoonetwo ] we arrive at a situation closely resembling malament s theorem @xcite .
the latter states that for a continuous spacetime manifold , the temporal order of spacetime events determines the spacetime metric up to a conformal factor , i.e. , determines the topology of spacetime . in mbqc , the classical processing relations which are essentially temporal information encoded in the influence matrix @xmath80 , plus the extra matrices @xmath251 , @xmath252 , @xmath1 related to the boundary sets @xmath149 and @xmath148determine the entire computation up to the measurement angles . to further pursue an analogy with malament s theorem
it is therefore desirable to have a characterization of ` space ' in mbqc that admits a discussion in terms of topology .
cellular complexes thus seem a suitable choice .
indeed , certain stabilizer states considered as resource states admit the characterization in terms of a cellular complex , such as the planar code state @xcite ( two - dimensional complex ) as well as computationally universal two - dimensional cluster states ( three - dimensional @xcite and four - dimensional complexes ) . _ remark regarding question [ qdet ] : _ a possible criticism of mbqc as a toy model for a quantum spacetime is that the laws which govern it , namely the classical processing relations eq . ( [ to7a ] ) , ( [ to7b ] ) are not laws of nature , but only rules imposed by the requirement of shielding the processed quantum information from the randomness of measurement . as such , the processing relations may be obeyed or violated at will by an operator running the mbqc . to this
we respond that if the operating unit for a measurement - based quantum computer was a dedicated device , it would not have the freedom to violate the classical processing rules . for it
, they _ would _ be the laws of nature .
they would follow straight from newton s axioms if the device was mechanical , and from maxwell s equations if it was electrical .
the processing rules could only be violated by a conscious being , such as a human or trained animal .
such beings , bound to the pull of gravity and the consequences of the no - cloning theorem by their belonging to this ` real ' universe , but entitled by their free will to disobey man - made regulations , can violate the processing rules of mbqc precisely because they have an existence outside it .
therefore , a theory of mbqc should not be required to describe them .
neither quantum mechanics nor the theory of gravity make statements about hypothetical objects that jump in and out of spacetime . to conclude
, we would like to recall the main idea underlying this work . in attempts to unify the theory of general relativity with quantum mechanics
, one may take the viewpoint that spacetime is not an independent construct , but rather a consequence of the laws of quantum mechanics .
once this assertion is spelled out , the natural next step is to identify the key quantum property which yields a mechanism for generating temporal order , and to illustrate this mechanism in a toy model .
measurement based quantum computation provides such a toy model .
therein , the key property which leads to temporal order is the inherent randomness of quantum measurement . | we propose measurement - based quantum computation ( mbqc ) as a quantum mechanical toy model for spacetime . within this framework
, we discuss the constraints on possible temporal orders enforced by certain symmetries present in every mbqc .
we provide a classification for all mbqc temporal relations compatible with a given initial quantum state and measurement setting , in terms of a matroid .
further , we find a symmetry transformation related to local complementation that leaves the temporal relations invariant . after light cones and closed time - like curves have previously been found to have mbqc counterparts , we identify event horizons as a third piece of the phenomenology of general relativity that has an analogue in mbqc . |
the dynamics of a two - state quantum system resulting from the action of a resonant external pulse of precise temporal area is a classical example in quantum mechanics .
this simple two - state quantum system has an interesting theoretical extension , from a nondegenerate two - level quantum system ( where only two two quantum states /ground and excited/ are available ) , to one with a degenerate ground level and a degenerate excited level .
these physical systems can be found at isolated atoms and molecules , where the states of well defined angular momentum j , due to the rotational symmetry produces a degeneracy of 2j+1 magnetic sublevels . for correct description of laser - induced transitions between quantum states with defined angular momentum , one must also incorporate several magnetic sublevels ( labeled by m ) , that may serve as a possible initial states .
hence each of these magnetic sublevels have a laser - driven excitation dynamic into to set of excited magnetic sublevels .
other source of multilevel systems can be the systems , whose mathematical description uses the usual rotating - wave approximation ( rwa ) . within the rwa the degeneracies can also be found for more general multistate quantum systems . to handle such multilevel systems , in 1983 morris and shore have introduced a coordinate transformation with the general property that could factorize the dynamic of any two degenerate sets of quantum states , to an equivalent description involving only independent uncoupled pairs of equations . generally , the morris - shore ( ms ) transformation is able to reduces the coherent quantum dynamics of a coupled degenerate two - level system to a set of independent nondegenerate two - state systems and a set of uncoupled states .
the mathematical description of ms transformation only requires deriving of eigenvalues and the eigenstates of a hermitean matrix , which is a product of interaction matrices .
the eigenstates of this hermitean matrix are the ms states , that represent the independent two - state systems and the dark states .
the eigenvalues represent the ms interactions in the independent ms two - state systems .
the ms transformation is quite general , but requires that all initial interactions are constant , or share the same time dependence , and also that all interactions are resonant , or equally detuned from the upper states .
the condition of equally detuned set of states implies that the lower set of states is degenerate in rwa sense , and so is the upper set of states .
having explain the constrains of ms transformation , it is natural to ask whether they can be relaxed .
it turns out that within the framework of weak deviation of equally detuned set of statrs , perturbative solution can be derived .
this paper is organized as follows . in sec .
ii we review the two - level ms transformation and in details define the problem for its extension . in sec .
iii we describe the perturbative solution to the problem of generalized ms transformation .
section iv presents an examples for the generalized ms transformation .
finally , sec .
v presents a summary of the results .
we consider a two - level quantum system in the general case when both levels are degenerate .
the ground @xmath0 level consists of @xmath1 degenerate states , the excited @xmath2 level of @xmath3 degenerate states and we assume , without any loss of generality , that @xmath4 every ground state @xmath5 @xmath6 is coupled via all excited states @xmath7 @xmath8 with pulsed interactions @xmath9 , each pair of which are on two - photon resonance , as depicted in [ fig1 ]
. the excited level may be off single photon resonance by some detuning @xmath10 which should be the same for all couplings .
we assume that the couplings share same time dependence @xmath11 @xmath12but we allow for different amplitudes @xmath13 and thereby pulse areas .
the dynamics of the system is governed by the time - dependent schrdinger equation which in the usual rotating - wave approximation ( rwa ) has the form@xmath14where the elements of the @xmath15-dimensional state vector @xmath16 are the probability amplitudes of the states and the hamiltonian in a block - matrix form is given by@xmath17here @xmath18 is the @xmath1-dimensional square zero matrix , in which the zero off - diagonal elements indicate the absence of couplings between the @xmath0 states , while the zero diagonal elements show that the @xmath19 states have the same energy , which is taken as the zero of the energy scale .
the matrix @xmath20 is an @xmath3-dimensional square diagonal matrix , with @xmath21 on the diagonal , @xmath20 @xmath22 @xmath23 in eq.([h ] ) @xmath24 is an @xmath25-dimensional matrix , which elements are the couplings between the ground and the excited states , @xmath26 is its hermitian conjugate @xmath27 , \label{v - vectors}\end{aligned}\ ] ] where @xmath28 @xmath29 are @xmath3-dimensional vectors , composed of the interaction energies of the @xmath30th state of the @xmath2 set with all states of the @xmath0 set,@xmath31 if the hamiltonian describes the atom - laser interaction , in the presence of weak magnetic field the degeneracy of the states is lifted . due to the external field
we have an extra terms , which can be assumed as a perturbation to the exact ms solution .
in this case we have @xmath32where @xmath33 is diagonal matrix with dimension @xmath34 and generally @xmath33 @xmath35 @xmath36 .
the aim of this paper is to analyze the ms transformation in the case of weak deviation of the @xmath20 @xmath37 @xmath23 condition , which in necessary for factorization towards set of two - state systems and set of decouple states .
some elements concerning the general solution of the ms transformation will be also discussed .
we will start our discussion by a brief description of the exact morris - shore ( ms ) transformation .
morris and shore @xcite have shown that any degenerate two - level system , in which all couplings share the same time dependence and the same detuning , can be reduced with a constant unitary transformation @xmath38 to an equivalent system comprising only independent two - state systems and uncoupled ( dark ) states , as shown in fig .
[ fig - system ] .
this time - independent transformation is given by @xmath39where the tildas denote the ms basis hereafter .
the constant transformation matrix @xmath38 can be represented in the block - matrix form @xmath40 , \label{w}\]]where @xmath41 is a unitary @xmath1-dimensional square matrix and @xmath42 is a unitary @xmath3-dimensional square matrix , @xmath43 and @xmath44 .
the constant matrices @xmath41 and @xmath45 mix only sublevels of a given level : @xmath41 mixes the @xmath46 sublevels and @xmath45 mixes the @xmath47 sublevels .
the transformed ms hamiltonian has the form@xmath48 , \label{h - ms}\]]where @xmath49the @xmath50 matrix @xmath51 may have @xmath52 null rows ( if @xmath53 ) , which correspond to decoupled states .
the decomposition of @xmath54 into a set of independent two - state systems requires that , after removing the null rows , @xmath55 reduces to a @xmath3-dimensional diagonal matrix ; let us denote its diagonal elements by @xmath56 ( @xmath57 ) .
it follows from eq .
( [ m ] ) that [ mm ] @xmath58hence @xmath41 and @xmath45 are defined by the condition that they diagonalize @xmath59 and @xmath60 , respectively .
it is important to note that the square matrices @xmath61 and @xmath60 have different dimensions , @xmath1 and @xmath3 , respectively . because all elements of @xmath62 are constant , @xmath41 and @xmath45 are also constant ; the elements @xmath56 are constant too .
it is straightforward to show that the @xmath3 eigenvalues of @xmath60 are all non - negative ; according to eqs .
( [ m ] ) and ( [ mm ] ) they are @xmath63 ( @xmath57 ) .
the matrix @xmath59 has the same eigenvalues and additional @xmath64 zero eigenvalues .
the ms hamiltonian ( [ h - ms ] ) has the explicit form @xmath65 .
\label{h - ms explicit}\ ] ] the structure of the ms hamiltonian ( [ h - ms explicit ] ) shows that in the ms basis the dynamics is decomposed into sets of @xmath66 decoupled single states , and @xmath3 independent two - state systems @xmath67 ( @xmath68 ) , each composed of an @xmath46 state @xmath69 and a @xmath47 state @xmath70 , and driven by the rwa hamiltonians , @xmath71 \quad ( n=1,2,\ldots , n_{b } ) .
\label{hn}\]]each of these two - state hamiltonians has the same detuning @xmath10 , but they differ in the couplings @xmath56 .
each of the new @xmath46 states @xmath72 is the eigenstate of @xmath61 corresponding to the eigenvalue @xmath73 , whereas each of the new @xmath47 states @xmath70 is the eigenstate of @xmath60 , corresponding to the same eigenvalue @xmath73 . the square root of this common eigenvalue , @xmath74 , represents the coupling in the respective independent ms two - state system @xmath75 ( @xmath57 ) .
the @xmath66 zero eigenvalues of @xmath59 correspond to decoupled ( dark ) states in the @xmath46 set ( we assume thoughout that @xmath76 ; therefore , dark states , if any , are in the @xmath46 set ) .
the dark states are decoupled from the dynamical evolution because they are driven by one - dimensional null hamiltonians .
the ms decomposition described above allows us to reduce the degenerate two - level problem to a set of @xmath66 nondegenerate two - state problems , wherein the detuning is unchanged while the couplings @xmath56 are combinations of the initial couplings between the @xmath46 and @xmath47 states and defined as the square roots of the eigenvalues of @xmath77 . from the vector form ( [ v - vectors ] ) of @xmath78
we obtain @xmath79 .
\label{v+v}\end{aligned}\ ] ] note that @xmath60 is the gramm matrix for the set of vectors @xmath80 .
thus if all these vectors are linearly independent then @xmath81 and all eigenvalues of @xmath82 are nonzero @xcite ; however , this assumption is unnecessary .
we assume that we can find the eigenvalues @xmath73 @xmath83 of the matrices ( [ vv+ ] ) and ( [ v+v ] ) , and the corresponding orthonormalized eigenvectors : the @xmath3 coupled eigenstates @xmath84 of @xmath59 and @xmath85 of @xmath60 , and the @xmath66 decoupled eigenstates @xmath86 of @xmath61 .
we use these eigenstates to construct the transformation matrices as @xmath87 , \quad \mathbf{b}=\left [ \begin{array}{c } \langle \widetilde{\psi } _ { 1}^{b}| \\ \vdots \\
\langle \widetilde{\psi } _ { n_{b}}^{b}|% \end{array}% \right ] .
\label{a b}\ ] ] the propagator the independent ms two - state systems @xmath67 ( @xmath68 ) , defined by @xmath88 = \widetilde{\mathbf{u}}_{n}(t_{f},t_{i})\left [ \begin{array}{c } \widetilde{c}_{n}^{a}(t_{i } ) \\
\widetilde{c}_{n}^{b}(t_{i})% \end{array}% \right ] , \ ] ] is unitary and can be parameterized in terms of the cayley - klein parameters@xmath89here , @xmath90 and @xmath91 are the cayley - klein parameters and they obey the relation@xmath92 by taking into account the ms propagators ( [ un ] ) for the two - state ms systems , the ordering of the states , and the ms hamiltonian ( [ h - ms explicit ] ) , the propagator of the entire system in the ms basis is written as@xmath93 .
\label{u - ms}\]]by taking into account the completeness relations [ completeness ] @xmath94it is straightforward to show that the propagator in the original basis @xmath95 reads @xmath96 .
\label{u}\ ] ] note that the propagator does not depend on the decoupled states @xmath97 @xmath98 , which are excluded by using eq .
( [ completeness a ] ) .
this has to be expected because , owing to their degeneracy , the choice of the decoupled states is not unique , since any superposition of them is also a zero - eigenvalue eigenstate of @xmath59 . because the dynamics in the original basis must not depend on such arbitrariness , the propagator @xmath99 must not depend on the decoupled states at all .
we @xmath100 where @xmath101 is diagonal matrix , given by @xmath102 @xmath103 , \label{d } \\
\mathbf{d}_{2}(t ) & = & \text{diag}\left [ d_{1}^{2}(t),d_{2}^{2}(t), .. ,d_{n_{b}}^{2}(t)\right ] , \end{aligned}\ ] ] the main difficulty for finding exact solution is that in the ms basis we have@xmath104 \\ & = & \mathbf{s}(\mathbf{h}_{0}(t)+\mathbf{d}(t))\mathbf{s}^{\dagger } = \mathbf{h% } _ { 0}^{ms}(t)+\mathbf{sd}(t)\mathbf{s}^{\dagger } \end{aligned}\ ] ] the matrices @xmath105 @xmath106 and @xmath107 @xmath108 are not diagonal this simply means in the ms basis the dynamics can not be decomposed into sets of @xmath66 decoupled single states , and @xmath3 independent two - state systems . hereafter for convenience we will assume @xmath109 and the notation @xmath110 will be used . instead of exact factorization
we look at solution that can be expressed in perturbative form .
the hamiltonian in a block - matrix form is given by @xmath111 we will look for perturbative solution for the ms problem and for this reason we will have the following definition for the transformation matrix @xmath112 @xmath113 where the additional condition @xmath114 is imposed .
this simply means that all terms from the series expansion in eq.([s - series ] ) are constant matrices .
also in reason to simplify the calculations and without loss of generality the @xmath115 matrices have the block - diagonal form , similar to ( [ w])@xmath116 .
\label{s - matrix - form}\ ] ] according to the eq.([transformation ] ) and using the series expansion for the transformation matrix @xmath38 given in eq.([s - series ] ) we will obtain @xmath117 from the expression above is clear the meaning of the @xmath118 state vectors . using the series expansion for the transformation matrix @xmath38
we will obtain the following expression for the hamiltonian in the new basis@xmath119 \left ( \sum_{i=0}\varepsilon ^{i}% \mathbf{s}_{i}^{\dagger } \right ) \\ & = & \mathbf{s}_{0}\mathbf{h}_{0}\mathbf{s}_{0}^{\dagger } + \varepsilon \left [ \mathbf{s}_{0}\mathbf{ds}_{0}^{\dagger } + \mathbf{s}_{1}\mathbf{h}_{0}\mathbf{% s}_{0}^{\dagger } + \mathbf{s}_{0}\mathbf{h}_{0}\mathbf{s}_{1}^{\dagger } % \right ] \notag \\ & & + \varepsilon ^{2}\left [ \mathbf{s}_{0}\mathbf{ds}_{1}^{\dagger } + \mathbf{s}% _ { 1}\mathbf{ds}_{0}^{\dagger } + \right .
\notag \\ & & \left .
+ \mathbf{s}_{1}% \mathbf{h}_{0}\mathbf{s}_{1}^{\dagger } + \mathbf{s}_{2}\mathbf{h}_{0}\mathbf{s% } _ { 2}^{\dagger } \right ] + o(\varepsilon ^{3 } ) \notag\end{aligned}\ ] ] in the above expression for the hamiltonian only the first two terms in the power expansion of @xmath120 are given , but the reader can easily take more terms for this perturbative expansion . let the matrix @xmath121 be chosen such that @xmath122and @xmath123 has the form @xmath124 , \ ] ] where @xmath51 is given by @xmath125 .\ ] ] the square matrix @xmath126 is diagonal , @xmath127\ ] ] this choice for the matrix @xmath121 corresponds to neglecting the perturbative term @xmath128 having in mind that the matrix @xmath121 is computed , using eq.(h - msbasis - eps ) we can continue and take the corrections up to first order of @xmath120 by choosing matrix @xmath129 from the equation @xmath130 using the block - matrix forms for @xmath131 @xmath132 @xmath133 and @xmath129 matrices and writing the third term to the rhs of the equation the above expression reads @xmath134 + \left [ \begin{array}{cc } \mathbf{0 } & \mathbf{a}_{0}\mathbf{\mathbf{v}b}_{1}^{\dagger } \\
\mathbf{b}_{0}\mathbf{\mathbf{v}}^{\dagger } \mathbf{a}_{1}^{\dagger } & \mathbf{b}_{0}\mathbf{\delta b}_{1}^{\dagger } % \end{array}% \right ] = -\left [ \begin{array}{cc } \mathbf{0 } & \mathbf{0 } \\
\mathbf{0 } & \mathbf{b}_{0}\mathbf{db}_{1}^{\dagger } % \end{array}% \right ] \label{s1-matrix - form}\ ] ] the above block - matrix equation is equivalent to the system of matrix equations@xmath135 the first and the second equation from the above system are equivalent under the hermitian conjugate operation . we want to note that from eq.([s0-def ] ) the @xmath136 and @xmath137 matrices are known .
hence the system of equation given by eq.([s1-b ] ) and eq.([s1-c ] ) , can be solved for @xmath138 and @xmath139 matrices , which defines the transformation matrix @xmath140 before we continue with @xmath141 matrix computation we would like to simplify the matrix equation eq.([s1-c ] ) .
remember that the matrix @xmath142 @xmath22 @xmath23 , one can use the commutation relations @xmath143 and @xmath144 , and the unitary properties @xmath145 and @xmath139 to transform eq.([s1-c ] ) .
@xmath146 using the unitary properties for the matrices @xmath137 and @xmath147 for eq.([s1-c ] ) the system of matrix equation , which defines the transformation matrix @xmath129 can be written in the form @xmath148 the algorithm for deriving expression for the @xmath141 matrix is similar to this for @xmath129 matrix . having in mind that the matrices @xmath121 and @xmath129 are computed , we can continue and take the corrections up to @xmath149 by choosing matrix @xmath150 from the equation @xmath151 by similar methods to the presented for the matrix @xmath141 , one can derive the higher order correction i.e. , find matrices @xmath152 @xmath153 etc .
above , we described the method for deriving solution to the generalized ms problem , where the evolution of a two - level quantum system is subjected to pulsed external field and in the general case the ground and excited levels have arbitrary level of degeneracy , @xmath1 and @xmath3 respectively . in this section
, we will illustrate these results with a specific example : when the @xmath2 set consists of two degenerate states , i.e. @xmath155 this case is interesting , because of the possible realizations of our model in different real physical systems .
moreover , this special case has an exact analytical solution , which is very rare for multidimensional systems .
we label the ground states @xmath5 @xmath156 where @xmath157 is arbitrary and we denote the two excited states @xmath158 and @xmath159 . in this case
the interaction operator @xmath78 ( [ v ] ) has @xmath157 columns and two rows , and its explicit form is given by where @xmath161 and @xmath162 are @xmath157-dimensional vectors comprising the couplings between the ground states and the corresponding excited state@xmath163 , ~~~\left\vert \omega ^{\prime \prime } \right\rangle = \left [ \begin{array}{c } \omega _ { 1}^{\prime \prime } \\ \omega _ { 2}^{\prime \prime } \\ \vdots \\ \omega _ { n}^{\prime \prime } % \end{array}% \right ] .\]]we write the product of @xmath164@xmath165 , \]]with eigenvalues , @xmath166 , \label{eigenvalues}\]]where @xmath167 denotes \begin{array}{c } \left\langle \omega ^{\prime } |\omega ^{\prime \prime } \right\rangle \\ \lambda _ { -}-\left\vert \omega ^{\prime } \right\vert ^{2}% \end{array}% \right ] = \left [ \begin{array}{c } e^{i\sigma } \cos \theta \\
-\sin \theta% \end{array}% \right ] , \label{b-}\end{aligned}\]]with @xmath175 and @xmath176 normalization factors .
the next step is to find the householder vectors @xmath177 and @xmath178 which are the ground states in the ms two - state systems and define the propagator @xmath179 they are the eigenstates of the @xmath157-dimensional matrix , corresponding to the same non - zero eigenvalues ( [ eigenvalues ] ) .
we construct them as superpositions of the interaction vectors with coefficients @xmath181 @xmath182 and @xmath183 @xmath184we determine these coefficients from the eigenvalue equations , @xmath185and from the normalization conditions , @xmath186as a result we obtain@xmath187 the results given in equations eq.([b+ ] ) , eq.([b- ] ) , eq.([a+ ] ) and eq.([a- ] ) define the block - structure of the @xmath121 matrix via eq.([s - matrix - form ] ) . using eq.([b+ ] ) , eq.([b- ] ) the @xmath145 matrix reads c.p .
williams and s.h .
clearwater , _ explorations in quantum computing _ , ( springer - verlag , berlin , 1997 ) ; a. steane , rep .
. phys . * 61 * , 117 ( 1998 ) ; m.a .
nielsen and i.l .
chuang , _ quantum computation and quantum information _
( cambridge university press , cambridge , 2000 ) .
k. bergmann , h. theuer , and b.w .
shore , rev .
phys . * 70 * , 1003 ( 1998 ) ; n.v .
vitanov , t. halfmann , b.w .
shore , and k. bergmann , ann .
phys . chem . *
52 * , 763 ( 2001 ) ; n.v .
vitanov , m. fleischhauer , b.w . shore and k. bergmann , adv . at .
. phys . * 46 * , 55 ( 2001 ) .
s. stenholm , in _ frontiers of laser spectroscopy _
, les houches summer school session xxvii , edited by r. bailian , s. haroche and s. liberman ( amsterdam , north holland , 1975 ) , p. 399 ; r. lefebvre and j. savolainen , j. chem . phys . * 60 * , 2509 ( 1974 ) ; m. bixon and j. jortner , j. chem . phys . * 48 * , 715 ( 1968 ) . | the aim of this paper is to analyze the morris - shore ( ms ) transformation in the case of weak deviation of the condition for equal detunings . for this generalization of the ms transformation a perturbative solution
is derived .
some elements concerning the general solution of the ms transformation are also discussed . |
quantum entanglement is a particularly intriguing phenomenon @xcite . since its conception in the famous epr paper @xcite
, it has received a great deal of interest . over the years
both theoretical and experimental research was dedicated to its investigation @xcite . two distinct strategies are usually applied to quantum entanglement detection and quantification .
the first approach to entanglement characterization is based on quantum state tomography and density matrix estimation @xcite . knowing the state s density matrix , it is possible to apply various entanglement criteria and entanglement measures ( e.g. , the peres - horodecki citerion @xcite ) . performing full state tomography is however a lengthy procedure since the number of required measurements grows exponentially with the dimension of hilbert space .
the second strategy is based on so - called entanglement witnesses ( ew ) . measuring a simple linear ew
involves performing suitable local measurements .
their correlation across involved parties then reveals the entanglement [ e.g. clauser - horne - shimony - holt ( chsh ) experiment @xcite ] .
this strategy , however , requires some a priory information about the investigated state to properly choose the performed measurements .
a revised version of this approach allows to estimate the amount of the available entanglement from the maximal amount of the chsh inequality violation @xcite .
one can describe entanglement using collective ( nonlinear ) entanglement witnesses @xcite that depend on joint measurements on several copies of the entangled state . as demonstrated experimentally , e.g. , by bovio _
et al . _
@xcite , it is possible to apply a specific two - copy witness to detect quantum entanglement . in this way one can also measure entanglement in terms of concurrence @xcite ( related to the entanglement of formation ) or other two - copy witnesses @xcite
there is a number of other nonlinear entanglement witnesses and quantifiers that do not rely on quantum state tomography @xcite .
in particular , adaptive approach to two - qubit entanglement detection was proposed in ref .
@xcite . conceptual scheme for measuring the collectibility of polarization encoded two - qubit states with linear optics .
two copies of a two - qubit state @xmath0 and @xmath1 are prepared .
polarization of one the photons from each pair is measured locally . at the same time ,
two different measurements on the other two photons are performed interchangeably : ( i ) simple unconditional detection ( balanced beam splitter bs is removed ) and ( ii ) two - photon bunching measurement corresponding to singlet state projection ( bs inserted ) .
the resulting four - fold coincidences are registered and the collectibility is then calculated from these coincidences using the procedure described in the text . ]
all the above mentioned approaches have their merits , but here we will focus on the proposal of rudnicki _ et al . _
it makes use of suitable measurements performed simultaneously on two copies of the investigated quantum state ( see the conceptual scheme in fig .
[ fig : concept ] ) .
this nonlinear entanglement witness is referred to as the _ collectibility _ , because values of the projections of the analyzed state onto the set of separable states are accumulated collectively . in a subsequent work ,
the authors of ref .
@xcite have extended the notion of the collectibility to apply also to mixed states @xcite and intorduced a collective nonlinear entanglement witness @xmath2 .
this technique does not share the disadvantages of the previously mentioned methods , _
i.e. _ , it allows to detect two - qubit entanglement for an arbitrary state and for pure states it allows to measure the amount of entanglement in terms of the negativity @xcite , an entanglement measure corresponding to the entanglement cost under the positive partial transpose operations @xcite .
moreover , the notion of the collectibility is based on the same principles as the entropic uncertainty relations , thus , its value can have a physical interpretation .
however , a completely universal negativity measurement requires in general four copies of the two - qubit state @xcite . in comparison to other previously proposed methods ,
the collectibility requires only one two - photon interaction to be implemented . to measure the colletibility of two - qubit systems
, it is required to perform collective measurements on two copies of the state . on the platform of linear optics with qubits encoded as polarization states of individual photons ,
these measurements include various local polarization projections combined with two - photon interference ( hong - ou - mandel effect @xcite ) .
the set of performed measurements is state - independent and , thus , does not require any a priory knowledge .
the use of local projections is an additional experimental benefit , because these can be simultaneously treated as heralds when using imperfect single - photon sources ( e.g. those based on spontaneous frequency down - conversion ) . here
, we demonstrate that the entire collectibility measurement procedure is experimentally feasible with current level of quantum optics technology .
we test the rudnicki _
@xcite method on three distinct two - photon quantum states ( one pure entangled , one pure separable and one maximally mixed ) .
consequently , we compare the observed values of @xmath3 and their theoretical values .
we demonstrate that all the measurements can be corrected for typical ( and unavoidable ) experimental shortcomings such as imperfect two - photon indistinguishability .
the possibility to calibrate the measurement apparatus is an important prerequisite allowing this procedure to be used in realistic experimental conditions .
we focus on two - qubit states @xmath4 .
let us assume that qubit @xmath5 is projected onto @xmath6 or @xmath7 , where @xmath8 and @xmath9 stand for a horizontally - polarized and vertically - polarized photon , respectively .
these projections are set by the @xmath10 rotations marked in fig .
[ fig : concept ] .
the probability associated with these projections reads @xmath11 and the reduced state is described by the following single - qubit density matrix @xmath12 , where @xmath13 $ ] .
the collectibility for two - qubit states @xcite can be calculated as @xmath14 where @xmath15 , @xmath16 are gramm matrix elements @xcite that depend on probabilities @xmath17 and two - photon overlaps @xmath18 ( @xmath19 ) .
these overlaps are measured by using the following identity @xmath20,$ ] where @xmath21 and @xmath22 is a projector on polarization - encoded singlet state .
this projection is implemented by overlapping photons @xmath23 and @xmath24 on a balanced beamsplitter and measuring the rate of anticoalescence of the photons transmitted through the beamsplitter .
this technique is very useful in measuring quantum properties of microsopic systems ( see , e.g. , ref .
direct calculations reveal that the maximum of collectibility does not depend on the parameter @xmath25 , thus , we set @xmath26 @xcite . measuring the maximum value of @xmath27
allows to both detect and quantify the entanglement of pure two - qubit states @xcite .
however , for mixed states it is more convenient to apply a closely related collective witness @xmath3 @xcite , i.e. , @xmath28 where the gramm matrix elements are calculated for @xmath29 .
this witness depends on conditional purity @xmath30 of a qubit @xmath31 , obtained after projecting its counterpart from subsystem @xmath5 onto @xmath32 the parameter @xmath33}$ ] depends on @xmath34|_{\theta=0}$ ] and @xmath35|_{\theta=\pi/2}$ ] . measuring negative value of @xmath3
indicates the detection of entanglement .
this witness is also extendible to higher - dimensional states @xcite .
experimental scheme for colectibility measurement on two copies of two - photon states .
laser source used was the coherent mira laser and the respective components are labeled as follows : shg second harmonics generation , hwp half - wave plate , pdl polarization dispersion line ( two beam displacers with one hwp in the middle ) , bbo pair of bbo crystals , qwp quarter - wave plate , wps set of half and quarter - wave plates , pol
polarizer , fc fiber coupler , fbs fiber beam spitter .
photon - mode numbers , as used in the text , are denoted by encircled numbers . ]
we have constructed a four - photon source as depicted in fig .
[ fig : setup ] .
this source is powered by femtosecond laser pulses emitted by coherent mira laser with of repetition rate .
second harmonics generation ( shg ) is then used to convert the wavelength of these pulses to .
typical mean power of the pumping beam at the output of shg is about . in the next step ,
the pumping beam is subjected to a polarization dispersion line ( pdl ) to correct for intrinsic polarization dispersion of the bbo crystals ( @xmath36-bab@xmath37o@xmath38 ) used to generate photon pairs .
this dispersion line consists of two beam displacers and a half - wave plate placed in between .
suitable tilt of these beam displacers allows to elongate optical path of one polarization with respect to the other .
the laser beam then impinges on a bbo crystal cascade ( known as the kwiat source @xcite ) .
type i spontaneous parametric down - conversion occurring in the first and second bbo crystal coherently generates pairs of horizontally and vertically polarized photons respectively . by adjusting the pumping beam polarization
, one can tune the generation rate of horizontally and vertically polarized pairs as well as their mutual phase shift .
the pumping beam then propagates through a quarter - wave plate , gets reflected on a mirror , goes again through that quarter - wave plate and impinges on the bbo crystal cascade on its way back towards the source . at this point a second pair of photons
is generated .
the quarter - wave plate is used to rotate the pumping beam polarization so that again it compensates for the polarization dispersion in bbo crystals ( now in reverse order ) . using mirrors , pairs of photons generated by both the forward and backward propagating pumping beam
are directed towards fiber couplers that lead them to single - photon detectors via single - mode optical fibers .
these couplers are equipped with interference filters in case of photons 1 and 3 and by interference filters in case of photons 2 and 4 . before being collected by the fiber couplers , one photon from each pair ( photons 1 and 3 as depicted in fig .
[ fig : setup ] ) is subjected to polarization projection using a quarter - wave plate , a half - wave plate and a polarizing cube .
the other two photons ( 2 and 4 ) are collected directly to single - mode fibers and then overlapped on a balanced fiber beam splitter ( fbs ) before led towards detectors .
temporal two - photon overlap on this beam splitter is adjusted by suitable choice of a fiber delay line and a motorized translation stage positioning the pumping beam back - reflecting mirror .
typical four - photon detection events occur about once per 5 minutes depending on the adjusted quantum state and polarization projection .
_ measurement and results _ as the first step , two copies of the investigated quantum state have to be prepared .
we have performed the collectibility measurement on three different quantum states @xmath39 first copy of a selected state was encoded into photons 1 and 2 , the second copy into photons 3 and 4 . for more detailed description of the state preparation procedure
see the appendix .
.[tab : results_s ] summarized results obtained for states defined in eq .
( [ eq : states ] ) .
@xmath40 denotes experimentally measured state purity , @xmath41 stands for its theoretical value , @xmath42 is the observed value of entanglement witness and @xmath43 is its theoretical value . [ cols="<,^,^,^,^",options="header " , ]
kl and kb acknowledge financial support by the czech science foundation under the project no .
16 - 10042y and the financial support of the polish national science centre under grant dec-2013/11/d / st2/02638 .
a acknowledges financial support by the czech science foundation under the project no .
the authors also acknowledge the project no .
lo1305 of the ministry of education , youth and sports of the czech republic financing the infrastructure of their workplace . | we present a proof - of - principle experiment demonstrating measurement of the collectibility , a nonlinear entanglement witness proposed by rudnicki _
et al . _ [ phys .
rev .
lett . * 107 * , 150502 ( 2011 ) ] .
this entanglement witness works for both mixed and pure two - qubit states . in the later case
it can be used to measure entanglement in terms of the negativity .
we measured the collectibility for three distinct classes of photonic polarization - encoded two - qubit states , i.e. , maximally entangled , separable and maximally mixed states .
we demonstrate that the measurement procedure is feasible and robust against typical experimental shortcomings such as imperfect two - photon indistinguishability . |
it is a long standing problem how to move particles on the martian surface .
the most prominent mechanism is wind in analogy to transport of sand on earth .
numerous work has been carried out on this in the past especially in wind tunnel experiments @xcite .
recent images of the hirise camera onboard the mars reconnaissance orbiter show that martian sand transport is still active .
they find dunes which travel several meters in a few years @xcite .
however , a problem encountered in the explanation of particle lift is that wind alone requires rather large speed to initiate saltation .
the pressure on mars on average is only 6 mbar in contrast to 1000 mbar on earth .
this reduces the dynamic pressure of a gas flow strongly . a speed of 30 m / s is supposed to be necessary to pick up the most susceptible particles of 100 @xmath1 in size @xcite .
high tangential wind speeds in vortices might also mobilize particles .
obviously , dust devils bear witness of dust lifting .
visible dust devils come in a variety of sizes @xcite .
the largest ones might easily lift dust .
however , @xcite and @xcite find wind speeds ( tangential and transversal ) which are not always large enough . also , @xcite observed dust devil activity on arsia mons .
this relates to an atmospheric pressure of only 2 mbar which requires still larger wind speeds .
+ there have been suggestions to support or ease particle lift one way or the other .
the choice of particles to be picked up has been varied in wind tunnel experiments . as an example
, the rolling of volcanic glass particles might reduce threshold speeds @xcite .
the pressure within dust devils is reduced compared to ambient conditions .
it has been proposed that the traverse of such a pressure minimum might be sufficient to lift dust @xcite .
last not least and connecting to the work presented here , @xcite found that illumination of a dust bed at low pressure provides a lift .
this was applied to mars by @xcite .
especially this latter effect is strongly depending on ambient pressure in a maybe non - intuitive way .
wind or gas drag and dynamic pressure decrease with decreasing pressure .
the induced lifting force of an insolated surface can increase to lower pressure by orders of magnitude in strength .
the force peaks around knudsen numbers of @xmath2 , where @xmath3 is the ratio between the mean free path of the gas molecules and the size of a particle or pore within the dust bed .
hence , for micrometer dust particles insolation supported lift is not important on earth but maximized on mars .
+ the model discussed for this lifting force so far included photophoretic forces , solid state greenhouse effects and gas compression by thermal creep @xcite .
these are important on long timescales ( hours ) as current research is supporting ( koester et al . personal communication ) .
however , here we consider gas flow through the dust bed and related pressure differences which was not included in the earlier models .
this provides lift for a sample where illumination changes on short time scales of seconds or even fractions of seconds .
the importance of this became obvious in microgravity experiments where @xcite observed an efficient gas flow through an illuminated dust bed directed upwards .
the effect is tied to the temperature profile within the illuminated dust bed .
the basic pumping is provided by thermal creep . for
a constant temperature gradient pure thermal creep pumping does not require any pressure differences within the soil . however , the temperature profile along the top layer of an illuminated soil is rather flat as radiation is absorbed and thermally emitted .
this layer can not pump by thermal creep but has to keep up the thermal creep gas flow from below nevertheless .
a pressure gradient is established close to the surface to do so . in accordance with darcy s law
the pressure increase below the surface transports the gas flow handed over by the knudsen pump ( thermal creep ) below .
once set , this overpressure not only moves the gas but also acts on the dust particles .
particles within this top layer are under constant tension and can be ejected if cohesion and gravity can be overcome by any means .
we call this an _ insolation activated layer_. + we quantify the thickness of this layer here based on laboratory experiments .
we illuminate a dust sample with sufficient light flux to compensate gravity .
we then remove cohesion by short vibrations .
this leads to a removal of the tension activated layer down to the pressure maximum and allows its thickness to be determined .
we detail the light induced sub - surface pumping in this section . the principle is shown in fig . [
fig : principle ] .
the light enters the dust bed and is absorbed .
the heat is conducted further down into the dust bed .
in addition , at the surface the dust bed can cool by thermal radiation . in consequence
a temperature gradient is established starting a few particle layers within the dust bed ( at depth @xmath4 in fig .
[ fig : principle ] ) and is directed to deeper layers .
temperature gradients always lead to a thermal creep gas flow in a porous medium , where gas is transported along the particles surface from cold to warm .
this was first introduced by @xcite as thermal transpiration where two gas reservoirs with different temperatures are connected by a small channel in a low pressure environment .
gas molecules creep along the channel wall from cold to warm . if the diameter of the channel is comparable to the mean free path of the gas molecules the pressure driven back flow can be smaller than the thermal creep flow . in an illuminated dust bed the gas molecules creep from cool layers deep within the dust bed ( @xmath5 in fig .
[ fig : principle ] ) upwards until the temperature levels off close to the surface at @xmath4 .
+ if the temperature increase in fig .
[ fig : principle ] would be linear from @xmath5 to the surface there would be no pressure differences .
every sub - layer would just transport the same amount of gas by thermal creep .
however , if there is a top layer of constant temperature the gas molecules do nt creep along these particles all the way to the surface , but only to @xmath4 .
the thermal creep gas flow leads to a concentration of molecules and the pressure is locally increased .
+ the increase of pressure below the surface at @xmath4 leads to a pressure driven gas flow through the top layer .
the pressure adjusts itself to a value where gas flow through the top layer matches the incoming thermal creep gas flow from below . both aspects , the temperature gradient driven thermal creep gas flow ( knudsen compression @xcite ) and the pressure driven gas flow ( darcy flow @xcite ) are usually occuring in the same capillary .
the mass flow of the gas through capillaries was described by @xcite and @xcite as @xmath6 where @xmath7 and @xmath8 are the average pressure and temperature within the dust bed , @xmath9 is a factor giving the amount of capillaries within the surface area @xmath10 , @xmath11 is the boltzmann constant , @xmath0 the molecular mass of the gas , @xmath12 and @xmath13 are the radius and length of the capillaries and @xmath14 and @xmath15 are the temperature and pressure differences within the dust bed , respectively . the coefficients @xmath16 and @xmath17 depend on the knudsen number and describe the pressure driven ( back ) flow and the flow by thermal creep , respectively .
it has to be noted that the length of the capillaries @xmath13 is different in the thermal driven ( @xmath18 , @xmath19 ) and the pressure driven ( @xmath20 , @xmath21 ) part .
+ if the dust bed is heated two mass flows develop . the first one @xmath22 is dominated by thermal creep ( knudsen pump ) due to the temperature gradient @xmath23 and the second one is driven by the pressure building up according to @xmath24 as the top layer temperature is flat or @xmath25 .
it might be noted that the latter is also equivalent to a description by darcy s law . in total , the flow velocity is set by the knudsen pump within the dust bed and the overpressure which maintains that flow also in the top layer activating it by putting tension on the dust particles .
if the force caused by this pressure gradient overcomes gravity and cohesion , particles can be lifted from the dust bed s surface .
+ to quantify the temperature profile and the resulting flow velocities and pressure differences we modeled the insolation of a dust bed as described in earlier work by @xcite .
figure [ fig : tempmax ] shows the numerical simulation of a temperature profile in a dust bed consisting of 25 @xmath0 m ( radius ) spheres .
the calculations were chosen to closely match the experimental setting : a composition of 25 @xmath0 m sized ( radius ) spheres in a dust bed with a thermal conductivity of 0.01 w k@xmath26m@xmath26 ( at @xmath27 mbar @xcite ) and illumination with 4 kw m@xmath26 ( see @xcite for details ) .
the temperature starts to decrease beyond 100 @xmath1 .
therefore , we expect that the activated layer would show a similar spatial scale .
simulation of temperature profile within a dust bed consisting of 25 @xmath0 m spheres .
the dust bed has a thermal conductivity of 0.01 w k@xmath26m@xmath26 .
the light flux is 4 kw m@xmath28 and illumination times are on the order of seconds . ]
the setup of the experiment is shown in fig.[fig : setup ] .
basaltic dust with grain sizes ranging from 0 to 125 @xmath0 m ( average size of 67 @xmath0 m ( diameter ) in a volume distribution , see fig .
[ fig : distribution ] ) is placed between two glass plates .
it forms a thin dust bed with a width of 2 mm and 25 mm height .
+ volume distribution of the used sample . ]
about 0.5 mm of dust emerges above the plates , forming a smooth surface which can be observed by a camera and microscope optics .
+ in a distance of 3 cm to the glass slides a weight of 30 g is locked 8 mm above the experimental table by an electromagnet .
if the magnet is turned off , the weight drops .
this results in a short vibration of the dust sample , removing cohesion between the dust particles and if present releasing the tension due to the overpressure .
this installation is placed within a vacuum chamber and experiments have been carried out at an ambient pressure of 0.1 mbar , 1 mbar and 10 mbar . a red laser ( 655 nm )
illuminates a spot of 8 mm width of the dust sample for @xmath29 20 seconds , providing a flux density of 4 kw m@xmath28 .
this flux is chosen to be close to the limit before a continuous particle ejection occurs .
it provides the most tension to ensure that the whole activated layer is ejected on tension release .
+ images of the dust surface before and after the impact are taken .
after each measurement the laser spot is moved further over the surface .
after 5 measurements new dust is added to the sample .
the images within the laser beam show a light surface and a black background ( fig .
[ fig : shape ] , 1a and 1b ) .
in addition images outside the laser beam are taken before as well as after the impact to measure the particle loss without radiation .
these are taken as bright field images with transmitted light to have a black surface on a bright background ( fig .
[ fig : shape ] , 2a and 2b ) .
+ example of the dust bed within ( 1a ) and outside ( 2a ) the laser beam at 1 mbar ambient pressure .
the images 1b and 2b show the results after tension release .
the dust bed within the laser beam ( 1a , 1b ) was illuminated by a red laser ( 680 nm ) with a light intensity of 4 kw m@xmath28 . ]
the surface lines before and after an ejection are traced .
the particle loss is calculated by substracting the lines of the surface before and after the impact . for each image
an average of the thickness of the removed layer with and without illumination is determined .
a total of 40 to 50 averaged measurements is taken for each pressure .
the mass flow of gas through capillaries can be calculated by eq .
( [ eq : massflow ] ) .
we assume that two gas reservoirs are connected by a small capillary with a constant temperature gradient .
for @xmath2 the gas is driven by thermal creep and flows with a certain rate through the capillary until the rising pressure difference @xmath15 balances the thermal creep flow as a pressure driven flow through the top layer .
there is a maximum @xmath15 at which no net gas flow is active ( @xmath30 , top layer as lid ) . from eq .
( [ eq : massflow1 ] ) we get @xcite @xmath31 we model the dust bed as follows : the mean radius of the dust particles is @xmath32 = 25 @xmath0 m as used for the temperature profile in fig.([fig : tempmax ] ) .
we consider this a suitable approximation for the dust sample of 33,5 @xmath0 m radius ( fig.([fig : distribution ] ) ) .
we further approximate the radius of the capillaries by a pore size @xmath33 @xcite .
the ambient pressure is @xmath27 mbar ( in one of the experimental sets ) .
the illuminated area of the dust bed is @xmath34 2 mm @xmath35 8 mm , the boltzmann constant @xmath36 j k@xmath26 .
we assume a temperature difference @xmath37 k over a length @xmath38 mm ( see fig .
[ fig : tempmax ] ) and an average temperature of @xmath39 k @xmath40 k within the heated part of the dust bed . according to @xcite ,
the coefficients are @xmath17 = 0.22 and @xmath16 = 1.6 .
this results in @xmath41 mbar .
+ in more detail the overpressure @xmath42 equals @xmath43 in eq .
( [ eq : massflow1 ] ) , and eventually , the gas velocity is the same in the temperature and the pressure driven part , as well as the mass flow . solving the equation @xmath44 for the overpressure @xmath15 , we get @xmath45 calculating this overpressure we take @xmath38 mm for the length of the capillaries with a temperature gradient and @xmath46 mm for the length of constant temperature ( see fig . [
fig : tempmax ] ) we get an overpressure of @xmath47 mbar .
this is consistently lower that the estimate of the maximum pressure difference of @xmath48 mbar . for an overpressure of about 0.05 mbar ,
the mass flow trough the pores is on the order of 10@xmath49 kg s@xmath26 .
the velocity of the gas flow can be calculated by dividing the mass flow by the illuminated area and the density of the gas at the given pressure @xmath50 with the molar mass of air @xmath51 amu and the molar gas constant @xmath52 = 8.3 j ( k mol)@xmath26 . at given parameters and @xmath27 mbar ,
the gas flow is 11 cm s@xmath26 .
this is consistent with measurements of tracer particles in drop tower experiments by @xcite .
+ further on , the force induced by the overpressure can be calculated for a particle column as @xmath53 with a particle cross section @xmath54 ( @xmath55 @xmath0 m ) we get @xmath56 n. + the gravitational force which has to be overcome can be approximated by @xmath57 with @xmath58 . with the density of basalt @xmath59 kg
m@xmath60 we get @xmath61 n , which is lower than the force caused by the overpressure by a factor of about 2 .
this is consistent with the fact that we adjusted the light flux to a value where no constant particle ejections occured , i.e. where gravity is compensated but not yet cohesion . as the overpressure is sufficient to compensate gravity of the top layer , only cohesion need to be overcome . as cohesion is removed by the impact the tension release leads to the ejection of the whole active particle layer .
probability to find a certain depth of particle layer ejected at 10 mbar ; blue bins are within the laser beam ; green bins show the depth without radiation for comparison ] probability to find a certain depth of particle layer ejected at 1 mbar ; blue bins are within the laser beam ; green bins show the depth without radiation for comparison ] the measurements show a significant difference between the depth of particle loss within and outside the laser beam for each pressure value as seen in fig .
[ fig : results10 ] , fig .
[ fig : results1 ] , fig .
[ fig : results2 ] and in tab .
[ tab : meandepth ] .
this proves that there is an activated dust layer .
the green bins in the figures show the probability of the depth to which particle loss occurs without light .
the blue bins show the probability to find a certain layer depth within the laser beam .
the negative depths refer to the fact that the surface can not be prepared perfectly smooth and therefore some aggregates can move during the tension loss , falling onto the measured spot and increasing the surface instead of decreasing it . without light
the measured values are strongly peaking around a depth within the dust bed of 50 @xmath0 m . within the illuminated area of the dust bed s surface the measured depth ranges from 50 up to 500 @xmath0 m for 1 and 10 mbar , where the variation of the values seems somewhat higher for 1 mbar . for 10 mbar
the layer size shows less variation and is more centered around 100 @xmath0 m .
the values for 0.1 mbar show no significant loss attributed to illumination .
this is due to the decrease in absolute pressure and pressure difference . while the temperature profile is similar as in the other two cases it is not imprinting itself enough on an overpressure to dominate over earth gravity .
+ probability to find a certain depth of particle layer ejected at 0.1 mbar ; blue bins are within the laser beam ; green bins show the depth without radiation for comparison ] the temperature profile for the relevant time scales in our experiment shown in fig .
[ fig : tempmax ] is consistent with the experimental results .
the flat temperature region has an extension of around 100 @xmath0 m which is the same as the measured depths for 1 and 10 mbar .
average values are given in tab .
[ tab : meandepth ] .
.the mean thickness of the active dust layers for the different pressures .
each data point is based on about 40 measurements and the mean thickness was calculated after subtracting the mean offset . [ cols="^,^,^,<,^,<",options="header " , ]
same as fig . [ fig : tempmax ] but for a light flux of 700 w m@xmath28 , average temperature of 200 k. the illumination times are 0.5 h , 1 h , 1.5 h@xmath62 12.5 h from bottom to top . ] the laboratory setting was chosen to quantify the active layer depth in a convenient way .
on mars the light flux is and the average temperature are lower and the durations of insolation are longer .
we therefore adjusted our simulation to scale the experimental results to martian conditions .
we choose 700 w m@xmath28 for the solar insolation and illuminate the dust bed in our simulation for up to 12.5 hours at an average temperature of t = 200 k. the temperature profiles are shown in fig .
[ fig : marstemp ]
. the maximum temperatures are somewhat larger than temperatures found on mars .
this is due to the simplifications of the model but we consider the results to capture the essential trend .
additionally we consider co@xmath63 instead of air with a molecular mass of @xmath64 44 @xmath65 , a geometric radius of @xmath66 m and a resulting cross section @xmath67 m@xmath68 . at p
= 6 mbar with a particle density @xmath69 the mean free path of co@xmath70 is @xmath71 19 @xmath0 m . hence , the ratio between q@xmath72 and q@xmath73 is 1.68 ( @xcite ) .
according to fig .
[ fig : marstemp ] we expect typical temperature differences of @xmath74 k for a length @xmath75 mm ( capillaries with a temperature gradient ) and @xmath76 mm for the activated layer of constant temperature .
+ equation [ eq : pressure ] gives a pressure difference of @xmath77 mbar .
this corresponds to a lifting force for a particle pile in the active layer of @xmath78 n. gravity on the same particles is @xmath79 n. that means that the activation accounts for 13% of gravity in a 1 mm thick layer of the soil .
variations in strength are expected for changing insolation and soil parameters .
our laboratory experiments showed that the illumination of a dust bed at low ambient pressure leads to an overpressure below the surface .
this pressure provides a lifting force on particles in insolated dust beds .
the pressure and thickness of this activated layer can well be described by a thermal creep model . in the experiments we used 4 kw m@xmath28 and illuminated dust beds for about 20 s.
this results in an activated layer of about 100 micrometer thickness . in the laboratory
the overpressure at 1 mbar and 10 mbar is large enough to compensate ( earth ) gravity .
+ at martian conditions the activated layer is supposedly deeper up to 1 mm .
the supporting force can be on the 10% level compared to gravity .
we expect that this mechanism supports other lifting mechanisms significantly , e.g. by decreasing the threshold wind speed for saltation by 10% .
this project is supported by dlr space management with funds provided by the federal ministry of economics and technology ( bmwi ) under grant number dlr 50 wm 1242 and by the german research foundation ( dfg ) under grand number ke 1897/1 - 1 .
we thank ralph lorenz and the anonymous referee for their thoughtful reviews . | the illuminated dusty surface of mars acts like a gas pump .
it is driven by thermal creep at low pressure within the soil . in the top soil layer
this gas flow has to be sustained by a pressure gradient .
this is equivalent to a lifting force on the dust grains .
the top layer is therefore under tension which reduces the threshold wind speed for saltation .
we carried out laboratory experiments to quantify the thickness of this activated layer .
we use basalt with an average particle size of 67 @xmath0 m . we find a depth of the active layer of 100 to 200 @xmath1 .
scaled to mars the activation will reduce threshold wind speeds for saltation by about 10% . |
when low- and intermediate mass stars enter the agb phase at the end of their lives , the mass - loss rate exceeds the nuclear burning rate and so dominates the subsequent evolution . in this process , a circumstellar envelope ( cse ) of gas and dust is formed which expands at a rate of about @xmath5km@xmath6 ( see for example @xcite ) .
the agb mass loss undergoes variations on different time scales .
the mass loss gradually increases over a time scale of several hundred thousands of years to reach a maximum on the tip of the thermally pulsing - agb @xcite .
the pulsations of the central star cause variations over several hundreds of days .
strong variations in the mass - loss rate , probably related to thermal pulses occurring every @xmath7-@xmath5 @xmath8yr @xcite , can lead to the formation of circumstellar detached shells .
@xcite have recently studied 7 carbon stars with known detached molecular shells .
they find that the shells are caused by periods of very intense mass loss ( @xmath9 ) and that these shells are running into a previous low mass - loss - rate agb wind .
after the intense mass - losing period , the mass loss decreases to a few @xmath10 , on a time scale of a few thousand years .
the question then arises if these types of mass - loss variations can be deduced from the sed . to answer this question
, we modelled the sed of a carefully selected sample of nearby agb stars , post - agb stars and supergiants . for modelling purposes it proved desirable to obtain millimetre observations of the stars in our sample , as there are not many observations available in the literature in this wavelength range .
still , a considerable part of the dust envelope is so cold that it emits its radiation at millimetre wavelengths , making these fluxes necessary to constrain the models for the seds .
+ the outline of this article is as follows .
first , the obtained fluxes at @xmath0 mm are presented .
sect.[observations ] contains information on the sample selection , the observations with simba and the data reduction with mopsi .
sect.[aperture ] is devoted to the aperture photometry . in sect.[models ] the constructed models for the seds are presented and the quality of these models and the indications for variable mass loss are discussed . in sect.[variations ] , the variability at @xmath11 mm is analysed on the basis of our observations .
the conclusions are summarised in sect.[conclusions ] .
the agb stars in the sample include m- and c - stars .
we have also observed some post - agb stars and supergiants that are known to have circumstellar shells . to improve the sensitivity to very faint circumstellar emission , the selection was restricted to objects with low mm - background emission based on the _ iras _ 60 and @xmath12 m data ( typically the cirr3 flag , which is the total @xmath13 m sky surface brightness , @xmath14 ) and the sample was further restricted to stars within 1kpc ( see table [ results ] ) .
there are a few exceptions , such as the unique f - type supergiant irc@xmath15 .
the data were taken with the 37-channel hexagonal bolometer array simba installed at the 15-m sest telescope , during @xmath16 observation runs : 2002 september 510 , 2003 may 913 and 2003 july 1315 .
the channels have a half power beam width ( hpbw ) of 24arcsec and two adjacent channels are separated by 44arcsec .
the filter bandpass is centered on 1.2 mm or 250ghz and has a full width half max ( fwhm ) of 90ghz .
simba works without a wobbling secondary mirror ; the maps were made with the fast - scanning technique .
the scans were performed in azimuthal direction by moving the telescope at a speed of @xmath17 with a separation in azimuth of 8arcsec .
the size of the scans varies between 500arcsec x 360arcsec and 600arcsec x 480arcsec .
4 to 8 consecutive scans were made of each source .
this procedure was repeated several times for each source during the observation runs .
although simba has a hexagonal array design , there are still gaps in the spatial coverage . because the telescope is set up in horizontal coordinates , the scan direction on the sky changes with the hour angle . by assembling scans taken at different hour angles ,
the gaps in the spatial coverage are reduced .
of course the assembling of the different scans also increases the signal - to - noise ratio .
the mopsi software was used to reduce the data .
basically the procedure described in the section on mopsi in the simba user - manual @xcite was followed , but the reduction was further divided into 3 different steps . in the first step some fundamental operations like despiking , opacity correction and sky - noise reduction
are performed on each of the individual scans .
also , the 4 to 8 consecutive scans of each source are put together to form one map . in a second step ,
the maps made during different nights are assembled .
if the source was detected in the individual maps , the maps were recentered before the assembling . out of this map information
is gained about the position of the source , which is used in the last step to define a proper baseline and to improve the sky - noise reduction .
for the absolute calibration , scans were made of uranus , typically at the beginning and/or end of each observing night .
the data reduction is basically the same as for the scientific sources , although a few steps were altered because uranus is a relatively strong and extended source .
only low order baseline fits were used ( with an appropriate base range definition ) and neither despiking nor sky - noise reduction were carried out , as recommended in the manual .
the flux of uranus at 250ghz was estimated with the subprogram ` planet ' in mopsi .
table [ tabelvarianties ] gives the resulting calibration factors for the 3 observation runs .
k , amc @xmath20 , @xmath21 , @xmath22k , @xmath23 between @xmath24 and @xmath25 , @xmath26 between @xmath25 and @xmath27 .
the upper figure shows the photometric data ( asterisk ) and the model ( full line ) .
error - bars are shown , but most of the time they fall within the symbols .
a reversed triangle represents an upper limit .
the lower figure shows the lrs spectrum ( plus signs ) and the model ( full line ) . as a comparison ,
the dotted line shows the corresponding hydrodynamical model . ] | it is generally acknowledged that the mass loss of asymptotic giant branch ( agb ) stars undergoes variations on different time scales .
we constructed models for the dust envelopes for a sample of agb stars to assess whether mass - loss variations influence the spectral energy distribution ( sed ) . to constrain the variability , extra observations at millimetre wavelengths ( @xmath0 mm ) were acquired . from the analysis of the dust models , two indications for the presence of mass - loss variations can be found , being ( @xmath1 ) a dust temperature at the inner boundary of the dust envelope that is far below the dust condensation temperature and ( @xmath2 ) an altered density distribution with respect to @xmath3 resulting from a constant mass - loss rate . for 5 out of the 18
studied sources a two - component model of the envelope is required , consisting of an inner region with a constant mass - loss rate and an outer region with a less steep density distribution . for one source an outer region with a steeper density distribution was found .
moreover , in a search for time variability in our data set at 1.2 mm , we found that wx psc shows a large relative time variation of @xmath4% which might partially be caused by variable molecular line emission .
[ firstpage ] stars : agb and post - agb stars : mass - loss stars : variables : other radio continuum : stars . |
the theory of games provides a general structure within which both cooperation and competition among independent entities may be modeled , and provides powerful tools for analyzing these models .
applications of this theory have fundamental importance in many areas of science .
this paper considers games in which the players may exchange and process quantum information .
we focus on competitive games , and within this context the types of games we consider are very general .
for instance , they allow multiple rounds of interaction among the players involved , and place no restrictions on players strategies beyond those imposed by the theory of quantum information . while classical games can be viewed as a special case of quantum games , it is important to stress that there are fundamental differences between general quantum games and classical games .
for example , the two most standard representations of classical games , namely the _ normal form _ and _ extensive form _ representations , are not directly applicable to general quantum games .
this is due to the nature of quantum information , which admits a continuum of pure ( meaning extremal ) strategies , imposes bounds on players knowledge due to the uncertainty principle , and precludes the representation of general computational processes as trees . in light of such issues ,
it is necessary to give special consideration to the incorporation of quantum information into the theory of games .
a general theory of quantum games has the potential to be useful in many situations that arise in quantum cryptography , computational complexity , communication complexity , and distributed computation .
this potential is the primary motivation for the work presented in this paper , which we view as a first step in the development of a general theory of quantum games .
the following facts are among those proved in this paper : * every multiple round quantum strategy can be faithfully represented by a single positive semidefinite operator acting only on the tensor product of the input and output spaces of the given player .
this representation is a generalization of the choi - jamiokowski representation of super - operators .
the set of all operators that arise in this way is precisely characterized by the set of positive semidefinite operators that satisfy a simple collection of linear constraints .
* if a multiple round quantum strategy calls for one or more measurements then its representation consists of one operator for each of the possible measurement outcomes .
the probability of any given pair of measurement outcomes for two interacting strategies is given by the inner product of their associated operators .
* the maximum probability with which a given strategy can be forced to output a particular result is the _ minimum _ value of @xmath1 for which the positive semidefinite operator corresponding to the given measurement result is bounded above ( with respect to the lwner partial order ) by the representation of a valid strategy multiplied by @xmath1 .
we give the following applications of these facts : * a new and conceptually simple proof of kitaev s bound for strong coin - flipping , which states that every quantum strong coin - flipping protocol allows a bias of at least @xmath2 . *
the exact characterization @xmath0 of the class of problems having quantum refereed games ( i.e. , quantum interactive proof systems with two competing provers ) .
this establishes that quantum and classical refereed games are equivalent in terms of expressive power : @xmath3 .
it is appropriate for us to comment on the relationship between the present paper and a fairly large collection of papers written on a topic that has been called _
quantum game theory_. meyer s _ pq penny flip _
game @xcite is a well - known example of a game in the category these papers consider .
the work of eisert , _
et al . _
@xcite is also commonly cited in this area .
some controversy exists over the interpretations drawn in some quantum these papers see , for instance , refs .
@xcite .
a key difference between our work and previous work on quantum game theory is that our focus is on multiple - round interactions .
understanding the actions available to players that have quantum memory is therefore critical to our work , and to our knowledge has not been previously considered in the context of quantum game theory .
a second major difference is that , in most of the previous quantum game theory papers we are aware of , the focus is on rather specific examples of classical games and on identifying differences that arise when so - called quantum variants of these games are considered . as a possible consequence , it may arguably be said that none of the results proved in these papers has had sufficient generality to be applicable to any other studies in quantum information .
in contrast , our interest is not on specific examples of games , but rather on the development of a general theory that holds for all games .
it remains to be seen to what extent our work will be applied , but the applications that we provide suggest that it may have interesting uses in other areas of quantum information and computation .
a different context in which games arise in quantum information theory is that of _ nonlocal games _
@xcite , which include _ pseudo - telepathy games _ @xcite as a special case .
these are cooperative games of incomplete information that model situations that arise in the study of multiple - prover interactive proof systems , and provide a framework for studying bell inequalities and the notion of nonlocality that arises in quantum physics .
while such games can be described within the general setting we consider , we have not yet found an application of the methods of the present paper to this type of game .
possibly there is some potential for further development of our work to shed light on some of the difficult questions in this area .
this section gives a brief overview of various quantum information - theoretic notions that will be needed for the remainder of the paper .
we assume the reader has familiarity with quantum information theory , and intend only that this overview will serve to establish our notation and highlight the main concepts that we will need .
readers not familiar with quantum information are referred to the books of nielsen and chuang @xcite and kitaev , shen and vyalyi @xcite .
when we speak of the vector space associated with a given quantum system , we are referring to some complex euclidean space ( by which we mean a finite - dimensional inner product space over the complex numbers ) .
such spaces will be denoted by capital script letters such as @xmath4 , @xmath5 , and @xmath6 .
we always assume that an orthonormal _ standard basis _ of any such space has been chosen , and with respect to this basis elements of these spaces are associated with column vectors , and linear mappings from one space to another are associated with matrices in the usual way .
we will often be concerned with finite sequences @xmath7 of complex euclidean spaces .
we then define @xmath8 for nonnegative integers @xmath9 , and define @xmath10 for @xmath11 .
it is convenient to define various sets of linear mappings between given complex euclidean spaces @xmath4 and @xmath5 as follows .
let @xmath12 denote the space of all linear mappings ( or _ operators _ ) from @xmath4 to @xmath5 , and write @xmath13 as shorthand for @xmath14 .
we write @xmath15 to denote the set of hermitian operators acting on @xmath4 , @xmath16 to denote the set of all positive semidefinite operators acting on @xmath4 , and @xmath17 to denote the set of all density operators on @xmath4 ( meaning positive semidefinite operators having trace equal to 1 . )
an operator @xmath18 is a _ linear isometry _ if @xmath19 .
the existence of a linear isometry in @xmath12 of course requires that @xmath20 , and if @xmath21 then any linear isometry @xmath18 is unitary .
we let @xmath22 denote the set of all linear isometries from @xmath4 to @xmath5 .
the operator @xmath23 denotes the identity operator on @xmath4 .
transposition of operators is always taken with respect to standard bases . the hilbert - schmidt inner product on @xmath13 is defined by @xmath24 for all @xmath25 .
for given operators @xmath26 , the notation @xmath27 means that @xmath28 .
this relation is sometimes called the _ lwner partial order _ on @xmath15 . when we refer to _ measurements _ , we mean povm - type measurements .
formally , a measurement on a complex euclidean space @xmath4 is described by a collection of positive semidefinite operators @xmath29 satisfying the constraint @xmath30 here @xmath31 is a finite , non - empty set of _ measurement outcomes_. if a state represented by the density operator @xmath32 is measured with respect to such a measurement , each outcome @xmath33 results with probability @xmath34 .
a _ super - operator _ is a linear mapping of the form @xmath35 where @xmath4 and @xmath5 are complex euclidean spaces .
a super - operator of this form is said to be _ positive _ if @xmath36 for every choice of @xmath37 , and is _ completely positive _ if @xmath38 is positive for every choice of a complex euclidean space @xmath6 .
the super - operator @xmath39 is said to be _
admissible _ if it is completely positive and preserves trace : @xmath40 for all @xmath41 .
admissible super - operators represent discrete - time changes in quantum systems that can , in an idealized sense , be physically realized . the choi - jamiokowski representation @xcite of super - operators is as follows .
suppose that @xmath42 be a given super - operator and let @xmath43 be the standard basis of @xmath4 .
then the _ choi - jamiokowski representation _ of @xmath39 is the operator @xmath44 it holds that @xmath39 is completely positive if and only if @xmath45 is positive semidefinite , and that @xmath39 is trace - preserving if and only if @xmath46 . for two complex euclidean spaces @xmath4 and @xmath5 , we define a linear mapping @xmath47 by extending by linearity the action @xmath48 on standard basis states . we make extensive use of this mapping in some of our proofs , as it is very convenient in a variety of situations .
let us now state some identities involving the @xmath49 mapping , each of which can be verified by a straightforward calculation .
[ prop : identities ] the following hold : 5 mm for any choice of @xmath50 , @xmath51 , and @xmath52 for which the product @xmath53 makes sense we have @xmath54 [ item : trace ] for any choice of @xmath55 we have @xmath56 for any choice of @xmath57 we have @xmath58 [ item : choi ] let @xmath59 and suppose @xmath60 is given by @xmath61 for all @xmath41 .
then @xmath62 for any non - empty set @xmath63 of hermitian operators , the _ polar _ of @xmath64 is defined as @xmath65 and the _ support _ and _ gauge _ functions of @xmath64 are defined as follows : @xmath66 these functions are partial functions in general , but it is typical to view them as total functions from @xmath15 to @xmath67 in the natural way .
for any set @xmath64 of positive semidefinite operators , we denote @xmath68 [ prop : polar ] let @xmath4 be a complex euclidean space and let @xmath64 and @xmath69 be non - empty subsets of @xmath15 .
then the following facts hold : 1 . [
item : polar - contain ] if @xmath70 then @xmath71 .
[ item : polar - pos ] if @xmath72 for each @xmath37 then @xmath73 .
[ item : polar - polar ] if @xmath64 is closed , convex , and contains the origin , then the same is true of @xmath74 . in this case
we have @xmath75 , @xmath76 the first two items in the above proposition are elementary , and a proof of the third may be found in rockafellar @xcite .
in this section we define the notions of a quantum _ strategy _ and the _ choi - jamiokowski representation _ of quantum strategy .
the remainder of the paper is concerned with the study of these objects and their interactions .
we begin with our definition for quantum strategies , which we will simply call _ strategies _ given that the focus of the paper is on the quantum setting .
[ def : strategy ] let @xmath77 and let @xmath78 and @xmath79 be complex euclidean spaces .
@xmath80-turn non - measuring strategy _ having input spaces @xmath78 and output spaces @xmath79 consists of : 5 mm complex euclidean spaces @xmath81 , which will be called _
memory spaces _ , and an @xmath80-tuple of admissible mappings @xmath82 having the form @xmath83 an _ @xmath80-turn measuring strategy _ consists of items 1 and 2 above , as well as : 5 mm a measurement @xmath84 on the last memory space @xmath85 .
we will use the term _ @xmath80-turn strategy _ to refer to either a measuring or non - measuring @xmath80-turn strategy .
figure [ fig : strategy ] illustrates an @xmath80-turn non - measuring strategy .
( 23734,2939)(-600,1700 ) ( 18612,4212)(19812,4212)(19812,3012)(18612,3012)(18612,4212 ) ( 1812,4212)(3012,4212)(3012,3012)(1812,3012)(1812,4212 ) ( 5412,4212)(6612,4212)(6612,3012)(5412,3012)(5412,4212 ) ( 9012,4212)(10212,4212)(10212,3012)(9012,3012)(9012,4212 ) ( 2412,3612)(0,0)@xmath86 ( 6012,3612)(0,0)@xmath87 ( 9612,3612)(0,0)@xmath88 ( 19212,3612)(0,0)@xmath89 ( 16512,3612)(18312,3612 ) ( 18072.000,3552.000)(18312.000,3612.000)(18072.000,3672.000 ) ( 18072.000,3552.000 ) ( 17712,1512)(18912,2712 ) ( 18784.721,2499.868)(18912.000,2712.000)(18699.868,2584.721 ) ( 18784.721,2499.868 ) ( 19512,2712)(20712,1512 ) ( 20499.868,1639.279)(20712.000,1512.000)(20584.721,1724.132 ) ( 20499.868,1639.279 ) ( 912,1512)(2112,2712 ) ( 1984.721,2499.868)(2112.000,2712.000)(1899.868,2584.721 ) ( 1984.721,2499.868 ) ( 2712,2712)(3912,1512 ) ( 3699.868,1639.279)(3912.000,1512.000)(3784.721,1724.132 ) ( 3699.868,1639.279 ) ( 4512,1512)(5712,2712 ) ( 5584.721,2499.868)(5712.000,2712.000)(5499.868,2584.721 ) ( 5584.721,2499.868 ) ( 6312,2712)(7512,1512 ) ( 7299.868,1639.279)(7512.000,1512.000)(7384.721,1724.132 ) ( 7299.868,1639.279 ) ( 8112,1512)(9312,2712 ) ( 9184.721,2499.868)(9312.000,2712.000)(9099.868,2584.721 ) ( 9184.721,2499.868 ) ( 9912,2712)(11112,1512 ) ( 10899.868,1639.279)(11112.000,1512.000)(10984.721,1724.132 ) ( 10899.868,1639.279 ) ( 3312,3612)(5112,3612 ) ( 4872.000,3552.000)(5112.000,3612.000)(4872.000,3672.000 ) ( 4872.000,3552.000 ) ( 6912,3612)(8712,3612 ) ( 8472.000,3552.000)(8712.000,3612.000)(8472.000,3672.000 ) ( 8472.000,3552.000 ) ( 10512,3612)(12312,3612 ) ( 12072.000,3552.000)(12312.000,3612.000)(12072.000,3672.000 ) ( 12072.000,3552.000 ) ( 20112,3612)(21912,3612 ) ( 21672.000,3552.000)(21912.000,3612.000)(21672.000,3672.000 ) ( 21672.000,3552.000 ) ( 15012,3612 ) ( 15012,3612 ) ( 13212,3612 ) ( 13212,3612 ) ( 14112,3612 ) ( 14112,3612 ) ( 15912,3612 ) ( 15912,3612 ) ( 1112,2112)(0,0)[b]@xmath90 ( 4712,2112)(0,0)[b]@xmath91 ( 8312,2112)(0,0)[b]@xmath92 ( 17912,2112)(0,0)[b]@xmath93 ( 3662,2112)(0,0)[b]@xmath94 ( 7262,2112)(0,0)[b]@xmath95 ( 10862,2112)(0,0)[b]@xmath96 ( 20612,2112)(0,0)[b]@xmath97 ( 4212,4012)(0,0)@xmath98 ( 7862,4012)(0,0)@xmath99 ( 11462,4012)(0,0)@xmath100 ( 17462,4012)(0,0)@xmath101 ( 21012,4012)(0,0)@xmath85 although there is no restriction on the dimension of the memory spaces in a quantum strategy , it is established in the proof of theorem [ theorem : characterization ] that every measuring strategy is equivalent to one in which @xmath102 for each @xmath103 .
we also note that our definition of strategies allows the possibility that any of the input or output spaces is equal to @xmath104 , which corresponds to an empty message .
one can therefore view simple actions such as the preparation of a quantum state or performing a measurement without producing a quantum output as special cases of strategies .
when we say that an @xmath80-turn strategy is described by _ linear isometries _
@xmath105 , it is meant that the admissible super - operators @xmath106 defining the strategy are given by @xmath107 for @xmath108 .
notice that , when it is convenient , there is no loss of generality in restricting ones attention to strategies described by linear isometries in this way .
this is because every admissible super - operator can be expressed as a mapping @xmath109 for some linear isometry @xmath50 , followed by the partial trace over some `` garbage '' space that represents a tensor factor of the space to which @xmath50 maps . by including the necessary
`` garbage '' spaces as tensor factors of the memory spaces , and therefore not tracing them out , there can be no change in the action of the strategy on the input and output spaces .
along similar lines , there is no loss of generality in assuming that a given measuring strategy s measurement is projective .
a given @xmath80-turn strategy expects to interact with something that provides the inputs corresponding to @xmath78 and accepts the strategy s outputs corresponding to @xmath79 .
let us define an @xmath80-turn _ co - strategy _ to be the sort of object that a strategy interfaces with in the most natural way .
let @xmath77 and let @xmath78 and @xmath79 be complex euclidean spaces .
the spaces @xmath78 are viewed as the input spaces of some @xmath80-turn strategy while @xmath79 are to be viewed as its output spaces .
@xmath80-turn non - measuring co - strategy _ to these spaces consists of : 5 mm complex euclidean _ memory _ spaces @xmath110 , a density operator @xmath111 , and an @xmath80-tuple of admissible mappings @xmath112 having the form @xmath113 an _ @xmath80-turn measuring co - strategy _ consists of items 1 , 2 and 3 above , as well as : 5 mm a measurement @xmath114 on the last memory space @xmath115 . as for strategies
, we use the term _ @xmath80-turn co - strategy _ to refer to either a measuring or non - measuring @xmath80-turn co - strategy .
figure [ fig : interaction ] represents the interaction between an @xmath80-turn strategy and co - strategy .
( 23734,4539)(-600,200 ) ( 20412,1212)(21612,1212)(21612,12)(20412,12)(20412,1212 ) ( 18612,4212)(19812,4212)(19812,3012)(18612,3012)(18612,4212 ) ( 1812,4212)(3012,4212)(3012,3012)(1812,3012)(1812,4212 ) ( 3612,1212)(4812,1212)(4812,12)(3612,12)(3612,1212 ) ( 5412,4212)(6612,4212)(6612,3012)(5412,3012)(5412,4212 ) ( 7212,1212)(8412,1212)(8412,12)(7212,12)(7212,1212 ) ( 9012,4212)(10212,4212)(10212,3012)(9012,3012)(9012,4212 ) ( 10812,1212)(12012,1212)(12012,12)(10812,12)(10812,1212 ) ( 12,1212)(1212,1212)(1212,12)(12,12)(12,1212 ) ( 612,612)(0,0)@xmath116 ( 4212,612)(0,0)@xmath117 ( 7812,612)(0,0)@xmath118 ( 11412,612)(0,0)@xmath119 ( 21012,612)(0,0)@xmath120 ( 2412,3612)(0,0)@xmath86 ( 6012,3612)(0,0)@xmath87 ( 9612,3612)(0,0)@xmath88 ( 19212,3612)(0,0)@xmath89 ( 16512,3612)(18312,3612 ) ( 18072.000,3552.000)(18312.000,3612.000)(18072.000,3672.000 ) ( 18072.000,3552.000 ) ( 18312,612)(20112,612 ) ( 19872.000,552.000)(20112.000,612.000)(19872.000,672.000 ) ( 19872.000,552.000 ) ( 17712,1512)(18912,2712 ) ( 18784.721,2499.868)(18912.000,2712.000)(18699.868,2584.721 ) ( 18784.721,2499.868 ) ( 19512,2712)(20712,1512 ) ( 20499.868,1639.279)(20712.000,1512.000)(20584.721,1724.132 ) ( 20499.868,1639.279 ) ( 912,1512)(2112,2712 ) ( 1984.721,2499.868)(2112.000,2712.000)(1899.868,2584.721 ) ( 1984.721,2499.868 ) ( 2712,2712)(3912,1512 ) ( 3699.868,1639.279)(3912.000,1512.000)(3784.721,1724.132 ) ( 3699.868,1639.279 ) ( 4512,1512)(5712,2712 ) ( 5584.721,2499.868)(5712.000,2712.000)(5499.868,2584.721 ) ( 5584.721,2499.868 ) ( 6312,2712)(7512,1512 ) ( 7299.868,1639.279)(7512.000,1512.000)(7384.721,1724.132 ) ( 7299.868,1639.279 ) ( 8112,1512)(9312,2712 ) ( 9184.721,2499.868)(9312.000,2712.000)(9099.868,2584.721 ) ( 9184.721,2499.868 ) ( 9912,2712)(11112,1512 ) ( 10899.868,1639.279)(11112.000,1512.000)(10984.721,1724.132 ) ( 10899.868,1639.279 ) ( 3312,3612)(5112,3612 ) ( 4872.000,3552.000)(5112.000,3612.000)(4872.000,3672.000 ) ( 4872.000,3552.000 ) ( 6912,3612)(8712,3612 ) ( 8472.000,3552.000)(8712.000,3612.000)(8472.000,3672.000 ) ( 8472.000,3552.000 ) ( 1512,612)(3312,612 ) ( 3072.000,552.000)(3312.000,612.000)(3072.000,672.000 ) ( 3072.000,552.000 ) ( 5112,612)(6912,612 ) ( 6672.000,552.000)(6912.000,612.000)(6672.000,672.000 ) ( 6672.000,552.000 ) ( 8712,612)(10512,612 ) ( 10272.000,552.000)(10512.000,612.000)(10272.000,672.000 ) ( 10272.000,552.000 ) ( 10512,3612)(12312,3612 ) ( 12072.000,3552.000)(12312.000,3612.000)(12072.000,3672.000 ) ( 12072.000,3552.000 ) ( 12312,612)(14112,612 ) ( 13872.000,552.000)(14112.000,612.000)(13872.000,672.000 ) ( 13872.000,552.000 ) ( 11712,1512)(12912,2712 ) ( 12784.721,2499.868)(12912.000,2712.000)(12699.868,2584.721 ) ( 12784.721,2499.868 ) ( 20112,3612)(21912,3612 ) ( 21672.000,3552.000)(21912.000,3612.000)(21672.000,3672.000 ) ( 21672.000,3552.000 ) ( 21912,612)(23712,612 ) ( 23472.000,552.000)(23712.000,612.000)(23472.000,672.000 ) ( 23472.000,552.000 ) ( 16812,612 ) ( 16812,612 ) ( 15012,3612 ) ( 15012,3612 ) ( 13212,3612 ) ( 13212,3612 ) ( 14112,3612 ) ( 14112,3612 ) ( 15912,3612 ) ( 15912,3612 ) ( 15012,612 ) ( 15012,612 ) ( 15912,612 ) ( 15912,612 ) ( 17712,612 ) ( 17712,612 ) ( 1112,2112)(0,0)[b]@xmath90 ( 4712,2112)(0,0)[b]@xmath91 ( 8312,2112)(0,0)[b]@xmath92 ( 11962,2112)(0,0)[b]@xmath121 ( 17912,2112)(0,0)[b]@xmath93 ( 3662,2112)(0,0)[b]@xmath94 ( 7262,2112)(0,0)[b]@xmath95 ( 10862,2112)(0,0)[b]@xmath96 ( 20612,2112)(0,0)[b]@xmath97 ( 4212,4012)(0,0)@xmath98 ( 7862,4012)(0,0)@xmath99 ( 11462,4012)(0,0)@xmath100 ( 17462,4012)(0,0)@xmath101 ( 2412,212)(0,0)@xmath122 ( 6012,212)(0,0)@xmath123 ( 9612,212)(0,0)@xmath124 ( 13212,212)(0,0)@xmath125 ( 19212,212)(0,0)@xmath126 ( 21012,4012)(0,0)@xmath85 ( 22662,212)(0,0)@xmath115 we have arbitrarily defined strategies and co - strategies in such a way that the co - strategy sends the first message .
this accounts for the inevitable asymmetry in the definitions . while it is possible to view any @xmath80-turn co - strategy as being an @xmath127-turn strategy having input spaces
@xmath128 and output spaces @xmath129 , it will be convenient for our purposes to view strategies and co - strategies as being distinct types of objects .
similar to strategies , there will be no loss of generality in assuming that the initial state @xmath116 of a co - strategy is pure , that each of the admissible super - operators @xmath130 takes the form @xmath131 for some linear isometry @xmath132 , and , in the case of measuring co - strategies , that the measurement @xmath133 is a projective measurement .
an @xmath80-turn strategy and co - strategy are _ compatible _ if they agree on the spaces @xmath78 and @xmath79 . by the _ output _
of a compatible strategy and co - strategy , assuming at least one of them is measuring , we mean the result of the measurement or measurements performed after the interaction between the strategies takes place . in particular , if both the strategy and co - strategy make measurements , then each output @xmath134 results with probability @xmath135 the definitions of strategies and co - strategies given above are natural from an operational point of view , in the sense that they clearly describe the actions of the players that they model . in some situations
, however , representing a strategy ( or co - strategy ) in terms of a sequence of admissible super - operators is inconvenient .
we now describe a different way to represent strategies that is based on the choi - jamiokowski representation of super - operators .
let us first extend this representation to @xmath80-turn non - measuring strategies . to do this ,
we associate with the strategy described by @xmath82 a single admissible super - operator @xmath136 this is the super - operator that takes a given input @xmath137 and feeds the portions of this state corresponding to the input spaces @xmath78 into the network pictured in figure [ fig : strategy ] , one piece at a time .
the memory space @xmath85 is then traced out , leaving some element @xmath138 .
such a map is depicted in figure [ fig : choijam ] for the case @xmath139 .
the choi - jamiokowski representation of the strategy described by @xmath82 is then simply defined as the choi - jamiokowski representation @xmath140 of the super - operator @xmath141 we have just defined .
an alternate expression for the choi - jamiokowski representation of a strategy exists in the case that it is described by linear isometries @xmath142 .
specifically , its representation is given by @xmath143 where @xmath144 is defined by the product @xmath145 ( 6999,2697)(0,-10 ) ( 1587,1137)(2262,1137)(2262,12)(1587,12)(1587,1137 ) ( 3162,1812)(3837,1812)(3837,687)(3162,687)(3162,1812 ) ( 4737,2487)(5412,2487)(5412,1362)(4737,1362)(4737,2487 ) ( 1925,575)(0,0)@xmath86 ( 3500,1250)(0,0)@xmath87 ( 5075,1925)(0,0)@xmath88 ( 6350,2262)(0,0)[l]traced out ( 687,1587)(0,0)[r ] @xmath146{0mm}{0mm}\right.$ ] ( 6312,912)(0,0)[l ] @xmath147{0mm}{0mm}\right\}\xi(\xi)$ ] ( 2362,912)(3062,912 ) ( 2942.000,882.000)(3062.000,912.000 ) ( 2942.000,942.000)(2942.000,882.000 ) ( 787,912)(1487,912 ) ( 1367.000,882.000)(1487.000,912.000 ) ( 1367.000,942.000)(1367.000,882.000 ) ( 3937,1587)(4637,1587 ) ( 4517.000,1557.000)(4637.000,1587.000 ) ( 4517.000,1617.000)(4517.000,1557.000 ) ( 5512,2262)(6212,2262 ) ( 6092.000,2232.000)(6212.000,2262.000 ) ( 6092.000,2292.000)(6092.000,2232.000 ) ( 5512,1587)(6212,1587 ) ( 6092.000,1557.000)(6212.000,1587.000 ) ( 6092.000,1617.000)(6092.000,1557.000 ) ( 2362,237)(6212,237 ) ( 6092.000,207.000)(6212.000,237.000 ) ( 6092.000,267.000)(6092.000,207.000 ) ( 787,1587)(3062,1587 ) ( 2942.000,1557.000)(3062.000,1587.000 ) ( 2942.000,1617.000)(2942.000,1557.000 ) ( 3937,912)(6212,912 ) ( 6092.000,882.000)(6212.000,912.000 ) ( 6092.000,942.000)(6092.000,882.000 ) ( 787,2262)(4637,2262 ) ( 4517.000,2232.000)(4637.000,2262.000 ) ( 4517.000,2292.000)(4517.000,2232.000 ) ( 5862,2487)(0,0)@xmath100 ( 5862,1812)(0,0)@xmath96 ( 5862,1137)(0,0)@xmath95 ( 5862,462)(0,0)@xmath94 ( 1137,2487)(0,0)@xmath92 ( 1137,1812)(0,0)@xmath91 ( 1137,1137)(0,0)@xmath90 ( 4287,1812)(0,0)@xmath99 ( 2712,1137)(0,0)@xmath98 next we consider measuring strategies .
assume that an @xmath80-turn measuring strategy is given , where the measurement is described by @xmath148 for some finite , non - empty set @xmath31 of measurement outcomes .
in this case we first associate with the strategy a collection of super - operators @xmath149 , each having the form @xmath150 the definition of each super - operator @xmath151 is precisely as in the non - measuring case , except that the partial trace over @xmath85 is replaced by the mapping @xmath152 notice that @xmath153 where @xmath141 is the mapping defined as in the non - measuring case .
each super - operators @xmath151 is completely positive but generally is not trace - preserving .
the choi - jamiokowski representation of the measuring strategy described by @xmath82 and @xmath154 is defined as @xmath155 . in the situation where a measuring strategy is described by linear isometries @xmath105 and a measurement @xmath154 ,
its choi - jamiokowski representation is given by @xmath156 for @xmath157 where @xmath158 for @xmath50 as defined above .
it is not difficult to prove that for given input spaces @xmath78 and output spaces @xmath79 , a collection @xmath156 of operators is the choi - jamiokowski representation of some @xmath80-turn measuring strategy if and only if @xmath159 is the representation of an @xmath80-turn non - measuring strategy over the same spaces .
finally , we define the choi - jamiokowski representation of measuring and non - measuring co - strategies in precisely the same way as for strategies , except that for technical reasons the resulting operators are transposed with respect to the standard basis .
( this essentially allows us to eliminate one transposition from almost every subsequent equation in this paper involving representations of co - strategies . ) specifically , a given @xmath80-turn co - strategy is viewed as an @xmath127-turn strategy by including an empty first and last message as discussed previously .
this strategy s choi - jamiokowski representation is defined as above .
the operators comprising this strategy representation are then transposed with respect to the standard basis to obtain the choi - jamiokowski representation of the co - strategy .
as we work almost exclusively with the choi - jamiokowski representation of strategies and co - strategies hereafter , we will typically use the term _ representation _ rather than _ choi - jamiokowski representation _ for brevity .
the applications of choi - jamiokowski representations of strategies given in this paper rely upon three key properties of these representations , stated below as theorems [ theorem : interaction - output - probability ] , [ theorem : characterization ] , and [ theorem : gauge ] .
this section is devoted to establishing these properties .
the first property provides a simple formula for the probability of a given output of an interaction between a strategy and a co - strategy .
[ theorem : interaction - output - probability ] let @xmath156 be the representation of a strategy and let @xmath160 be the representation of a compatible co - strategy . for each pair @xmath161 of measurement outcomes , we have that the output of an interaction between the given strategy and co - strategy is @xmath162 with probability @xmath163 .
let us fix a strategy and co - strategy whose representations are @xmath156 and @xmath164 , respectively . without loss of generality , the strategy is described by linear isometries @xmath105 and a projective measurement @xmath165 , while the co - strategy is described by a pure initial state @xmath166 , linear isometries @xmath167 and a projective measurement @xmath168 . for each output pair @xmath134 ,
define @xmath169 as follows : @xmath170 the probability of the outcome @xmath162 is @xmath171 for each pair @xmath162 .
now , making use of the @xmath49 mapping , we may express @xmath172 in a different way : @xmath173 where @xmath174 the probability of outcome @xmath162 is therefore @xmath175 \op{vec}(i_{\y_{1\ldots n}\otimes\x_{1\ldots n}})\\ & = \tr\left [ \left(\tr_{\z_n } x_a x_a^{\ast}\right ) \left(\tr_{\w_n } y_b y_b^{\ast}\right)^{\t } \right ] \\ & = \ip{q_a}{r_b } \end{aligned}\ ] ] as required .
let us denote by @xmath176 the set of all representations of @xmath80-turn strategies having input spaces @xmath78 and output spaces @xmath79 .
we may abbreviate this set as @xmath177 or @xmath178 whenever the spaces or number of turns is clear from the context .
similarly , we let @xmath179 denote the set of all representations of co - strategies for the same spaces
. it will be convenient to define @xmath180 .
the second property of strategy representations that we prove provides a characterization of @xmath181 in terms of linear constraints on @xmath182 .
[ theorem : characterization ] let @xmath77 , let @xmath78 and @xmath79 be complex euclidean spaces , and let @xmath183
. then @xmath184 if and only if @xmath185 for @xmath186 .
first assume that @xmath187 , which implies that there exist memory spaces @xmath81 and admissible super - operators @xmath188 that comprise a strategy whose representation is @xmath189 .
let @xmath190 be the super - operator associated with this strategy as described in section [ sec : strategies ] , and let @xmath191 be the super - operator associated with the @xmath192-turn strategy obtained by terminating the strategy described by @xmath82 after @xmath193 turns .
we have @xmath194 and so @xmath195 for @xmath196 as required .
now assume that @xmath183 satisfies @xmath197 for some choice of @xmath186 .
our goal is to prove that @xmath198 .
this will be proved by induction on @xmath80 .
in fact , it will be easier to prove a somewhat stronger statement , which is that the assumptions imply that there exists a strategy whose representation is @xmath189 that ( i ) is described by linear isometries @xmath105 , and ( ii ) satisfies @xmath199 .
if @xmath200 , there is nothing new to prove : it is well - known that if @xmath201 satisfies @xmath202 , then @xmath203 for some admissible super - operator @xmath204 . any such super - operator can be expressed as @xmath205 for @xmath206 and some choice of a linear isometry @xmath207 .
the 1-turn strategy we require is therefore the strategy described by @xmath208 .
now assume that @xmath209 . by the induction hypothesis
, there exist spaces @xmath210 and linear isometries @xmath211 with @xmath212 such that @xmath213 for @xmath214 defined as @xmath215 as required , we let @xmath85 be a complex euclidean space with dimension equal to the rank of @xmath189 , and let @xmath216 be any operator satisfying @xmath217 such a choice of @xmath51 must exist given that the dimension of @xmath85 is large enough to admit a purification of @xmath189 .
note that @xmath218 next , let @xmath219 be a complex euclidean space with dimension equal to that of @xmath93 , and let @xmath220 be an arbitrary unitary operator .
we have @xmath221 and therefore @xmath222 at this point we have identified two purifications of @xmath223 .
we will use the isometric equivalence of purifications to define an isometry @xmath224 that will complete the proof . specifically , because @xmath225 has the minimal dimension required to admit a purification of @xmath223 , it follows that there must exist a linear isometry @xmath226 such that @xmath227 this equation may equivalently be written @xmath228 we now define @xmath229 .
this is a linear isometry from @xmath230 to @xmath231 that satisfies @xmath232 this implies that the strategy described by @xmath105 has representation @xmath233 and therefore completes the proof .
theorem [ theorem : characterization ] is equivalent to the following statement : an operator @xmath183 is the representation of some @xmath80-turn strategy with input spaces @xmath78 and output spaces @xmath79 if and only if there exist operators @xmath234 , where @xmath235 , such that the following linear constraints are satisfied : @xmath236 each operator @xmath237 in of course uniquely determined by the representation @xmath189 , and represents the strategy obtained by terminating any strategy represented by @xmath189 after @xmath238 turns . theorem [ theorem : characterization ] also gives a characterization of @xmath80-turn co - strategies , as stated in the following corollary . [ cor : characterization ]
let @xmath77 , let @xmath78 and @xmath79 be complex euclidean spaces , and let @xmath183 .
then @xmath239 if and only if @xmath240 for @xmath241 satisfying @xmath242 the fact that @xmath181 and @xmath243 are bounded and characterized by the positive semidefinite constraint together with finite collections of linear constraints yields the following corollary .
let @xmath77 , let @xmath78 and @xmath79 be complex euclidean spaces .
then the sets @xmath181 and @xmath243 are compact and convex . just as @xmath181 consists of all representations of non - measuring strategies , the set @xmath244 consists of all elements of representations of measuring strategies . in other words , for any @xmath80-turn measuring strategy representation @xmath245 , it holds that @xmath246 for each @xmath33 .
moreover , for each operator @xmath189 there exists an @xmath80-turn measuring strategy @xmath156 of which @xmath189 is an element if and only if @xmath247 . the set @xmath248 has similar analogous properties . the final property of strategy representations that we will prove concerns the maximum probability with which some interacting co - strategy can force a given measuring strategy to output a given outcome .
[ theorem : gauge ] let @xmath77 , let @xmath78 and @xmath79 be complex euclidean spaces , and let @xmath249 represent an @xmath80-turn measuring strategy with input spaces @xmath78 and output spaces @xmath79 .
then for each @xmath33 , the maximum probability with which this strategy can be forced to output @xmath250 , maximized over all choices of compatible co - strategies , is given by @xmath251\,:\ , q_a \leq p r\;\ ; \text{for some}\;r\in{\mathcal{s}}_n(\x_{1\ldots n},\y_{1\ldots n})\}.\ ] ] an analogous result holds when @xmath156 is a measuring co - strategy .
the remainder of the present section is devoted to a proof of this theorem .
[ lemma : polar - technical ] let @xmath219 and @xmath252 be complex euclidean spaces and let @xmath253 be any closed , convex set that contains the origin
. then for @xmath254 we have @xmath255 the assumption @xmath256 implies that @xmath257 for every @xmath258 , and therefore @xmath259 .
consider any choice of @xmath260 , and note that @xmath261 for every choice of @xmath262 .
if it is the case that @xmath263 then we have @xmath264 for all @xmath265 , and therefore @xmath266 . on the other hand , if @xmath266 then @xmath264 for all @xmath265 .
it follows from the fact that @xmath189 is positive semidefinite that @xmath267 for all @xmath268 , and therefore @xmath269 .
[ lemma : switch - polar ] let @xmath219 be a complex euclidean space , let @xmath270 be non - empty closed and convex sets , and suppose @xmath271 then @xmath272 let @xmath273 as @xmath274 for every @xmath275 , it follows that @xmath276 .
clearly @xmath277 , and therefore @xmath278 .
thus , @xmath279 on the other hand , we have that every @xmath280 is contained in @xmath281 , implying that @xmath282 for all @xmath283 . as @xmath284
, this implies that @xmath267 for @xmath285 .
consequently , @xmath286 .
thus @xmath287 , and so @xmath288 as required .
[ lemma : polar ] let @xmath77 and let @xmath78 and @xmath79 be complex euclidean spaces .
then for all @xmath289 we have 5 mm @xmath290 if and only if @xmath291 for some choice of @xmath292 .
@xmath293 if and only if @xmath291 for some choice of @xmath187 .
the proof is by induction on @xmath80 . as @xmath294 $ ] and @xmath295 $ ] , the lemma holds for the case @xmath296 .
now suppose that @xmath77 .
the two items in the statement of the lemma are equivalent by lemma [ lemma : switch - polar ] , so it will suffice to prove the first .
define @xmath297 as @xmath298 by lemma [ lemma : polar - technical ] we have @xmath299 also define @xmath300 as @xmath301 again applying lemma [ lemma : polar - technical ] , we obtain @xmath302 now , by theorem [ theorem : characterization ] we have @xmath303 , and so @xmath304 . by the induction hypothesis we have @xmath305 and therefore @xmath306 by corollary [ cor : characterization ] we have that @xmath307 which completes the proof .
let @xmath308 $ ] denote the maximum probability with which @xmath156 can be forced to output @xmath250 in an interaction with some compatible co - strategy .
it follows from theorem [ theorem : interaction - output - probability ] that @xmath309 . using lemma [ lemma : polar ] , along with the fact that @xmath310 is positive semidefinite
, we have @xmath311 which completes the proof .
quantum coin - flipping protocols aim to solve the following problem : two parties , alice and bob , at separate locations , want to flip a coin but do not trust one another .
a quantum coin - flipping protocol with bias @xmath312 is an interaction between two ( honest ) strategies @xmath50 and @xmath51 , both having output sets @xmath313 , that satisfies two properties : 1 .
the interaction between the honest parties @xmath50 and @xmath51 produces the same outcome @xmath314 for both players , with probability @xmath315 for each outcome .
( neither player outputs `` abort '' when both are honest . ) 2 . if one of the parties does not follow the protocol but the other does , neither of the outcomes @xmath314 is output by the honest player with probability greater than @xmath316 .
protocols that satisfy these conditions are generally referred to as _
strong _ coin - flipping protocols , because they require that a cheater can not bias an honest player s outcome toward either result 0 or 1 .
( in contrast , _ weak _ protocols assume that alice desires outcome 0 and bob desires outcome 1 , and only require that cheaters can not bias the outcome toward their desired outcome . )
kitaev @xcite proved that no strong quantum coin - flipping protocol can have bias smaller than @xmath2 , meaning that one cheating party can always force a given outcome on an honest party with probability at least @xmath317 .
kitaev did not publish this proof , but it appears in refs .
@xcite . here
we give a different proof based on the results of the previous section .
suppose @xmath318 is the representation of honest - alice s strategy and @xmath319 is the representation of honest - bob s co - strategy in some coin - flipping protocol .
these strategies may involve any fixed number of rounds of interaction .
the first condition above implies @xmath320 now , fix @xmath314 , and let @xmath1 be the maximum probability that a cheating bob can force honest - alice to output @xmath321 .
obviously we have @xmath322 .
theorem [ theorem : gauge ] implies that there must exist a strategy @xmath189 for alice such that @xmath323 .
if a cheating alice plays this strategy @xmath189 , then honest - bob outputs @xmath321 with probability @xmath324 given that @xmath325 for all @xmath326 , we have that either honest - alice or honest - bob can be convinced to output @xmath321 with probability at least @xmath317 .
this proof makes clear the limitations of strong coin - flipping protocols : the inability of bob to force alice to output @xmath321 directly implies that alice can herself bias the outcome toward @xmath321 .
weak coin - flipping does not directly face this same limitation .
currently the best bound known on the bias of weak quantum coin - flipping protocols , due to ambainis @xcite , is that @xmath327 rounds of communication are necessary to achieve a bias of @xmath312 .
the best weak quantum coin - flipping protocol currently known , which is due to mochon @xcite , achieves bias approximately 0.192 ( which surpasses the barrier @xmath328 on strong quantum coin - flipping ) .
next , we define quantum refereed games .
it will be noted that von neumann s min - max theorem for zero - sum quantum refereed games follows from the facts we have proved about representations of strategies together with well - known generalizations of the classical min - max theorem .
although it is completely expected that the min - max theorem should hold for quantum games , it has not been previously noted in the general case with which we are concerned .
( lee and johnson @xcite proved this fact in the one - round case . )
this discussion will also be helpful for the application to interactive proof systems with competing provers that follows .
let us first define specifically what is meant by a zero - sum quantum refereed game .
such a game is played between two players , alice and bob , and is arbitrated by a referee .
the referee s output after interacting with alice and bob for some fixed number of rounds determines their pay - offs .
[ def : referee ] an _ @xmath80-turn referee _ is an @xmath80-turn measuring co - strategy @xmath329 whose input spaces @xmath78 and output spaces @xmath79 take the form @xmath330 for complex euclidean spaces @xmath331 , @xmath332 , @xmath333 and @xmath334 , for @xmath335 . an _ @xmath80-turn quantum refereed game _ consists of an @xmath80-turn referee along with functions @xmath336 defined on the referee s set of measurement outcomes , representing alice s payoff and bob s payoff for each output @xmath33 . such a game is a _
zero - sum _ quantum refereed game if @xmath337 for all @xmath33 .
the referee s actions in a quantum refereed game are completely determined by its representation @xmath329 . during each turn
, the referee simultaneously sends a message to alice and a message bob , and a response is expected from each player .
the spaces @xmath331 and @xmath332 correspond to the messages sent by the referee during turn number @xmath238 , while @xmath333 and @xmath334 correspond to their responses .
after @xmath80 turns , the referee produces an output @xmath338 .
a refereed quantum game does not in itself place any restrictions on the strategies available to alice and bob .
for example , alice and bob might utilize a strategy that allows quantum communication , they might share entanglement but be forbidden from communicating , or might even be forbidden to share entanglement . specific characteristics of a given game , such as its nash equilibria , obviously depend on such restrictions in general .
the focus of the remainder of the paper is on the comparatively simple setting of zero - sum quantum refereed games . in this case
, we assume that alice and bob do not communicate or share entanglement before the interaction takes place .
more precisely , we assume that alice and bob play independent strategies represented by @xmath339 and @xmath340 , respectively .
the combined actions of alice and bob are therefore together described by the operator @xmath341 .
it is a completely natural assumption that alice and bob play independent strategies in a zero - sum quantum refereed game , given that it can not simultaneously be to both players advantage to communicate directly with one another or to initially share an entangled state .
this should not be confused with the possibility that entanglement among the players and referee might exist at various points in the game , or that the referee might choose to pass information from one player to the other .
these possibilities are not disallowed when alice and bob s joint strategy is represented by @xmath342 .
now , assume that a zero - sum quantum refereed game is given , and that alice and bob play independent strategies @xmath50 and @xmath51 as just discussed .
let us write @xmath343 and define @xmath344 alice s expected pay - off is then given by @xmath345 while bob s expected pay - off is @xmath346 .
now , @xmath347 is a real - valued bilinear function in @xmath50 and @xmath51 .
because the operators @xmath50 and @xmath51 are drawn from compact , convex sets @xmath348 and @xmath349 respectively , we have that @xmath350 this is the min - max theorem for zero - sum quantum games . note that the above expression does not immediately follow from von neumann s original min - max theorem , but follows from an early generalization due to j. ville @xcite and several subsequent generalizations such as the well - known min - max theorem of ky fan @xcite . the real number represented by the two sides of this equation
is called the _ value _ of the given game .
classical interactive proof systems with competing provers have been studied by several authors , including feige , shamir , and tennenholts @xcite , feige and shamir @xcite , feigenbaum , koller , and shor @xcite , and feige and kilian @xcite .
a quantum variant of this model is defined by allowing the verifier to exchange quantum information with the provers @xcite . in both cases
these interactive proof systems are generalizations of single prover interactive proof systems @xcite .
interactive proof systems with two competing provers are naturally modeled by zero - sum refereed games . to highlight this connection
we will refer to the verifier as the referee and the two provers as alice and bob .
the referee is assumed to be computationally bounded while alice and bob are computationally unrestricted .
alice , bob , and the referee receive a common input string @xmath351 , and an interaction follows .
after the interaction takes place , the referee decides that either alice wins or bob wins .
a language or promise problem @xmath352 is said to have a classical refereed game if there exists a referee , described by a polynomial - time randomized computation , such that : ( i ) for every input @xmath353 , there is a strategy for alice that wins with probability at least @xmath354 against every strategy of bob , and ( ii ) for every input @xmath355 , there is a strategy for bob that wins with probability at least @xmath354 against every strategy of alice . the class of promise problems having classical refereed games is denoted @xmath356 .
it is known that @xmath356 is equal to @xmath357 .
the work of koller and megiddo @xcite implies @xmath358 , while feige and kilian @xcite proved the reverse containment .
let us note that zero - sum classical refereed games , and therefore the class @xmath356 , are unaffected by the assumption that alice and bob may play quantum strategies , assuming the referee remains classical .
this assumes of course that the classical referee is modeled properly within the setting of quantum information , which requires that any quantum information that it touches immediately loses coherence .
equivalently , the referee effectively measures all messages sent to it by alice and bob with respect to the standard basis before any further processing takes place . as there also can not be a mutual advantage to alice and bob to correlate their strategies using shared entanglement
, there is no advantage to alice or bob to play a quantum strategy against a classical referee .
this is not the case in the cooperative setting , because there alice and bob might use a shared entangled state to their advantage @xcite .
quantum interactive proof systems with competing provers are defined in a similar way to the classical case , except that the referee s actions are described by polynomial - time generated quantum circuits and the referee may exchange quantum information with alice and bob .
the complexity class of all promise problems having quantum refereed games is denoted @xmath359 .
the containment @xmath360 follows from @xmath361 .
it was previously known that @xmath362 @xcite , and we will improve this to @xmath363 .
this establishes the characterization @xmath0 , and implies that quantum and classical refereed games are equivalent with respect to their expressive power . in the refereed game associated with a competing prover quantum interactive proof system ,
the referee declares either alice or bob to be the winner .
specifically , the referee outputs one of two possible values @xmath364 , with @xmath250 meaning that alice wins and @xmath321 meaning that bob wins . by setting @xmath365 and @xmath366 , we obtain a quantum refereed game whose value is the maximum probability with which alice can win .
we will show that this optimal winning probability for alice can be efficiently approximated : it is the value of a semidefinite programming problem whose size is polynomial in the total dimension of the input and output spaces of the referee .
it follows from the polynomial - time solvability of semidefinite programming problems @xcite that @xmath363 .
we will use similar notation to the previous subsection : for a fixed input @xmath367 , the referee is represented by operators @xmath368 , and assuming @xmath80 is the number of turns for this referee we let the input and output spaces to alice be denoted by @xmath369 while @xmath370 denote the input and output spaces to bob .
the assumption that the referee is described by polynomial - time generated quantum circuits implies that the entries in the matrix representations of @xmath371 and @xmath372 with respect to the standard basis can be approximated to very high precision in exponential time .
now , given any strategy @xmath339 for alice , let us define @xmath373 the functions @xmath374 and @xmath375 are linear and extend to uniquely defined super - operators . under the assumption that alice plays the strategy represented by @xmath339 and bob plays the strategy represented by @xmath340 , we have that the referee outputs @xmath250 with probability @xmath376 and outputs @xmath321 with probability @xmath377
. one may think of @xmath378 as being the co - strategy that results from `` hard - wiring '' alice s strategy represented by @xmath50 into the referee .
now , alice s goal is to minimize the maximum probability with which bob can win . for a given strategy @xmath50 for alice , the maximum probability
with which bob can win is @xmath379 which , by theorem [ theorem : gauge ] , is given by @xmath380 the following optimization problem therefore determines the maximum probability @xmath1 for bob to win , minimized over all strategies for alice : @xmath381 this optimization problem can be expressed in terms of linear and semidefinite constraints as follows : @xmath382 & \tr_{\c_k}(a_k ) = a_{k-1 } \otimes
i_{\a_k } \quad & ( 2\leq k\leq n),\\ & \tr_{\c_1}(a_1 ) = i_{\a_1},\\[2 mm ] & q_k
= p_k \otimes i_{\d_k } \quad & ( 1\leq k\leq n),\\ & \tr_{\b_k}(p_k ) = q_{k-1 } \quad & ( 2\leq k\leq n),\\[2 mm ] & a_k \in \pos{\c_{1\ldots k}\otimes\a_{1\ldots k } } \quad & ( 1\leq k\leq n ) , \\ & q_k \in \pos{\d_{1\ldots k}\otimes\b_{1\ldots k } } \quad & ( 1\leq k\leq n ) , \\ & p_k \in \pos{\d_{1\ldots k-1}\otimes\b_{1\ldots k}}\quad & ( 1\leq k\leq n).\end{aligned}\ ] ] quantum interactive proofs with competing provers . in _ proceedings of the 22nd symposium on theoretical aspects of computer science _ ( 2005 ) , vol .
3404 of _ lecture notes in computer science _ ,
springer , pp . | we study properties of _ quantum strategies _ , which are complete specifications of a given party s actions in any multiple - round interaction involving the exchange of quantum information with one or more other parties .
in particular , we focus on a representation of quantum strategies that generalizes the choi - jamiokowski representation of quantum operations .
this new representation associates with each strategy a positive semidefinite operator acting only on the tensor product of its input and output spaces .
various facts about such representations are established , and two applications are discussed : the first is a new and conceptually simple proof of kitaev s lower bound for strong coin - flipping , and the second is a proof of the exact characterization @xmath0 of the class of problems having quantum refereed games . |
isolated electron spin systems are of great interest because of their potential application as coherent quantum memory in novel solid state devices@xcite .
several quantum computation proposals make use of electron spins , either as the qubits themselves or as intrinsic components of one- and two - qubit gates @xcite .
furthermore , experimental efforts are just now succeeding at measuring single electron spin relaxation times @xcite .
unfortunately , the exponential size of the hilbert space of these spin systems and the variety of channels for decoherence makes an exact analysis of spin dynamics virtually impossible ( without the use of a quantum computer)@xcite . as a result , it is often necessary to resort to approximations or perturbative treatments to address the timescale of electron spin decoherence , even using a simplified hamiltonian .
because approximations are inevitable , it is important to know which of these can be rigorously justified . in particular
, it is necessary to know which terms in a very complex system hamiltonian can be ignored because they do not contribute to decoherence on the experimentally relevant timescale . in this paper , we consider electron spin decoherence induced by the nuclear spins . in particular , we focus on the relevance of the @xmath6 term in the hamiltonian ( see ) , which comes from the contact hyperfine coupling of the electron and its surrounding nuclei . previously , the effect of the hyperfine interaction on decoherence has been studied in special cases or using various approximation methods .
recently , erlingsson et .
al used a semi - classical approximation to study electron spin correlation functions at low fields ( @xmath7 , where @xmath3 is on the order of the overhauser field , see below ) @xcite .
they found that the correlation functions were characterized by complex dynamics which did not decay , even at long times .
work by khaetskii et al .
has characterized the timescale of the logitudinal spin correlation for an electron coupled to surrounding nuclei via the hyperfine interaction @xcite . in refs .
@xcite , exact solutions for the case of completely polarized nuclei and for the case of homogeneous coupling , as well as perturbative solutions for the general , inhomogeneous situation were presented . from these solutions , the behavior of the general unpolarized ,
inhomogeneous case was inferred .
their results indicate that the logitudinal spin correlation time is on the order of @xmath8 for a single quantum dot and on the order of @xmath9 for an ensemble of dots .
gate imperfections due to the hyperfine coupling were also treated using perturbation theory in @xcite .
however , it was noted in @xcite that the perturbative solution would diverge to second - order and was therefore potentially unreliable .
more recently , coish et al .
have treated the behavior of this system using a generalized master equation , again confirming a power - law decay for short timescales on the order of @xmath8 @xcite . in general , though , all treatments based on a time - dependent perturbation expansion will be valid only at times smaller than some maximum timescale specified by the order of the expansion [ w.a .
coish , private communication ] ] .
although such treatments are useful for examining short - timescale and small - magnitude fluctuations in electron polarization , in this paper we are interested in the long - timescale dynamics of the electron spin which leads to @xmath10 decay of the longitudinal or in - plane polarization of the electron .
note that we are not claiming any particular functional form for the decay of coherence , exponential or otherwise .
rather , we are evaluating the timescale on which the longitudinal and in - plane magnetization decays to a constant fraction ( in this case @xmath10 ) of its initial value .
moreover , we focus on the high magnetic field regime which is relevant for the recent experimental developments in single spin read - out ( @xmath11 in ref .
@xcite ) . in ref .
@xcite it was assumed that at sufficiently high magnetic fields the electron - nuclei flip - flop term ( @xmath12 in ) which is due to the hyperfine interaction can be neglected as a direct source for electron spin decoherence . in this paper , we would like to 1 ) justify this assumption and 2 ) determine how strong a magnetic field must be applied to safely neglect this term . rather than resort to approximation to estimate the numerical value of @xmath0 and @xmath1 , this paper will instead place rigorous , non - perturbative lower bounds on the spin relaxation time of the electron in the high - field regime ( @xmath13 ) .
furthermore , we will explicitly take into account the application of the spin echo sequence which is often used experimentally to remove inhomogeneous broadening . to determine the importance of the electron - nuclear flip - flop term
, we use the hamiltonian presented in @xcite which contains only the contact hyperfine term . in doing so
, we can effectively isolate the decoherence contribution that derives exclusively from the hyperfine interaction .
when the timescale of this hyperfine - induced decoherence is significantly greater than the timescale of decoherence due to other mechanisms ( such as spectral diffusion induced by inter - nuclear dipolar coupling@xcite ) then it seems reasonable to ignore the direct effects of hyperfine coupling as a channel for electron spin decoherence .
it is important to define what we mean by `` electron spin decoherence '' .
typically , electron spins are assumed to undergo simple exponential decay governed by the bloch equations .
thus the coherence times @xmath0 and @xmath1 can be defined as the timescales of this exponential decay . however , when decay is non - exponential , as in this system ( see @xcite ) , there is some flexibility in the definition of the coherence times .
electron spin dynamics includes both short - timescale fluctuations ( which are generally small in magnitude @xcite ) and long - timescale ( i.e. , @xmath10 ) decay ; in this paper , we take `` electron spin decoherence '' to mean the latter . in other words ,
the coherence time of an electron is defined as the time it takes for the longitudinal or transverse polarization of an initially polarized electron spin to decay to a specific fraction , e.g. @xmath10 , of its initial value .
the decay of the in - plane magnetization is understood to occur in the context of a spin echo experiment .
spin echo experiments are designed to remove the effects of inhomogeneous broadening due to variations in the local field experienced by an ensemble of electron - nuclei systems ( i.e. multiple independent systems ) . when applied to a single electron spin
, a spin echo can be understood as removing the inhomogeneous broadening resulting from an ensemble of initial states of the single electron - nuclei system .
the remaining decay of the in - plane magnetization comes from fluctuating overhauser fields , which can not in general be removed by the spin echo ( but see @xcite for numerical evidence that a substantial part of hyperfine - induced fluctuations may , in fact , be reversible ) . because only _ bounds _ on relaxation behavior are desired
, our analysis of this decay can be completely rigorous , despite the complexity of the problem and the size of the hilbert space .
our results are applicable to localized electron spins in semiconductors such as quantum dots and donor impurities .
this paper is organized as follows : sec .
ii outlines the general structure of the problem and presents the results of our analysis .
the derivation of the theorems used in sec .
ii can be found in the appendix on epaps for the derivations of the theorems presented in sec .
ii . a direct link to this document
may be found in the online article s html reference section .
the document may also be reached via the epaps homepage ( http://www.aip.org/pubservs/epaps.html ) or from ftp.aip.org in the directory /epaps/. see the epaps homepage for more information . ] .
applications to electron spins in semiconductors of relevance to solid state quantum information processing are made in sec .
discussion and conclusions follow in sec .
let @xmath14 be the electron spin precessing in a magnetic field @xmath2 applied in the @xmath5 direction .
the electron spin is coupled to @xmath15 nuclear spins @xmath16 via the hyperfine interaction , characterized by the coupling constants @xmath17 . due to the hyperfine coupling ,
the nuclei induce an effective overhauser field on the electron with magnitude @xmath18 .
here we consider the hamiltonian @xmath19 where @xmath20 and @xmath21 are the gyromagnetic ratios of the electron and nuclei , respectively ( @xmath22 ) . can be written as @xmath23 this hamiltonian conserves the total spin angular momentum @xmath24 ; hence , the total number of `` down '' spins is conserved . for convenience
, we will label each of these blocks by the quantum number @xmath25 , where @xmath25 is the total number of `` down '' spins ( i.e. @xmath26 , @xmath27 ) .
then the hamiltonian can be block diagonalized based on the @xmath28 operator @xcite .
there will be @xmath29 blocks , one for each possible value of @xmath25 , and the @xmath30 block will be an @xmath31 matrix .
for the remainder of the paper , it will be assumed that all analysis applies to some subspace labeled by a particular value of @xmath25 .
the unperturbed hamiltonian , @xmath32 , commutes with the @xmath33 operator .
thus , the eigenstates of @xmath32 can be labeled by the @xmath5-polarization of the electron . on the other hand
, the full hamiltonian @xmath34 does not commute with @xmath33 because @xmath35 connects electron spin - up states to electron spin - down states , and vice - versa . at high fields greater than some critical field @xmath36
, the zeeman energy of the electron will dominate the energy of the eigenstates , and the electron spin - up and electron spin - down manifolds will be well separated in energy .
thus , if the perturbation is in some sense smaller in magnitude than this energy gap , the effects of the perturbation will be small and @xmath33 will still be an approximately good label for eigenstates of @xmath34 ( see figure 1 ) . our analysis in sec .
ii.a demonstrates that if the magnetic field @xmath2 is greater than the critical field @xmath3 , then the electron spin magnetization @xmath37 never changes appreciably . because the eigenstates of @xmath34 are nearly eigenstates of @xmath33 at high fields , the @xmath5-polarization of the electron does not undergo substantial ( i.e. @xmath10 ) decay ( this result matches the perturbative result found in @xcite ) .
furthermore , based on these arguments , a field - dependent bound is placed on the maximum fluctuation of @xmath33 from its initial value : the greater the field , the smaller the maximum possible fluctuation .
again , it should be emphasized that there may be small magnitude fluctuations in @xmath38 , which occur on a finite timescale .
however , because our definition of @xmath0 ( the timescale for @xmath10 decay of @xmath37 ) quantifies the large - scale coherence properties of the electron spin , the absence of large - scale decay renders @xmath0 effectively infinite .
although @xmath0 is infinite for large magnetic fields , dephasing processes can still occur provided that they conserve the @xmath5-polarization of the electron spin .
although this statement seems like a contradiction given that the perturbation , @xmath35 , flips the electron spin , virtual transitions involving @xmath39 can nevertheless flip pairs of nuclear spins without affecting the electron spin . for this reason , it is also important to place a lower bound on @xmath1 .
derivation of the lower bound on @xmath1 in sec .
ii.b follows a different methodology than the bound on @xmath0 . because @xmath1 is defined as the timescale of the decay of the spin echo envelope @xcite , a lower bound on the magnitude of @xmath40 after a pulse echo sequence implies a lower bound on @xmath1 . specifically , the standard definition of @xmath1 is the minimal time , @xmath41 such that , @xmath42 where a @xmath43 pulse is applied at time @xmath41 ( we have dropped a factor of @xmath44 for convenience ) .
it is assumed that for an infinite system , there will be no large - scale recurrence in @xmath45 , but only dissipative behavior .
small magnitude recurrence , due for instance to echo modulation @xcite , may still exist , but large - scale recurrence on the order of @xmath10 is assumed to be absent .
even if this large - scale recurrence does exist , the lower bound on @xmath1 is still valid . in particular ,
if it can be shown that for some value of @xmath46 , @xmath47 , it follows that @xmath48 .
this bound on @xmath49 can be proven using a variational argument .
every trial state @xmath50 induces a probability distribution over the eigenenergies of @xmath34 .
the mean of this energy distribution is given by the expectation value @xmath51 likewise , the variance of this energy distribution is @xmath52 the variance of the energy will be equal to zero if and only if @xmath50 is an eigenstate of @xmath34 ( or a superposition of degenerate eigenstates ) . for all other states which are superpositions of non - degenerate eigenstates
, the variance will be greater than zero .
the essential feature of the theorems given in sec .
ii.b is that the dynamics undergone by any state @xmath50 are limited in timescale by the variance in energy of that state .
using this idea , it can be shown that the dephasing processes of the system have a timescale governed by the energy @xmath53 conceptually , our bound on @xmath1 makes mathematically rigorous the idea that the coherence time of a state is inversely proportional to its linewidth . before proceeding to the proof of these results
, two important points must be made .
first , this method constructs only a rigorous _ lower bound _ on @xmath1 ; it does not predict the actual value of @xmath1 nor does it claim to be a _
tight _ lower bound on @xmath1 . in fact , numerical work suggests that the spin echo pulse may actually remove nearly all hyperfine - induced decoherence , leading to only small magnitude decay of @xmath54 under the spin echo sequence for all times , @xmath41 @xcite .
however , this evidence is numerical and has been confirmed only for small numbers of spins ( @xmath55 ) . in this paper
we are interested in what rigorous statements can be made regarding the @xmath1 time for a large system ( @xmath56 ) .
second , it is the special structure of this system at high @xmath2 fields that allows a _ useful _ bound to be constructed . though this method is general and can be used to construct valid bounds on @xmath1 for other systems , these bounds may prove to be useless in practice ( for instance , the rigorous bound that @xmath57 is valid , but is a practically useless bound ) .
fortunately , we find that our method provides practically useful lower bound values for real systems of relevance to quantum information processing ( see )
. the proof of the bound on @xmath0 will be given first and will be organized as follows .
first , we will show that the eigenstates of @xmath34 can be divided into two subspaces , the @xmath58 subspace and the @xmath59 subspace ( theorem 1 ) .
eigenstates contained in the @xmath58 subspace contain large contributions from electron spin - up states , and eigenstates in the @xmath59 subspace contain large contributions from electron spin - down states . in other words
, @xmath33 is an approximate quantum number for the true eigenstates of @xmath34 .
second , we will show that this fact places bounds on the eigenvalues of @xmath34 ( theorem 2 ) .
next , we will show that every eigenstate of @xmath32 is almost completely contained within either the @xmath58 or @xmath59 subspace of @xmath34 ( theorem 3 ) . to complete the bound on @xmath0
, we will show that these previous statements imply a lower bound on the decay of @xmath33 from its initial state ( theorem 4 ) . for future convenience
, we define @xmath60 & = & \sum_{j \in k \ , { \rm{largest}}}{a_j}\\ a_2[k ] & = & \sum_{j \in k \ , \rm{largest}}{a^2_j}\\ a_{max } & = & \sum_{j}{a_j } \\ a_{2,max } & = & \sum_{j}{a_j^2}.\end{aligned}\ ] ] in other words , @xmath61 $ ] is the sum of the @xmath62 largest hyperfine coupling constants .
@xmath63 is the sum of all @xmath15 hyperfine coupling constants and therefore represents the maximum magnitude of the overhauser field .
the matrix @xmath32 commutes with the @xmath33 and @xmath64 operators ; hence eigenstates of @xmath32 can be labeled by the @xmath5-polarization of the electron spin and the @xmath5-polarization of the nuclei .
let @xmath65 and @xmath66 represent the @xmath67 and @xmath68 polarized states of the electron spin , respectively , and let @xmath69 be an @xmath15-bit string of @xmath70s and @xmath71s representing the @xmath5-polarization of the @xmath15 nuclei .
the eigenstates of @xmath32 are of the form @xmath72 or @xmath73 ( it should be remembered that @xmath69 will contain @xmath25 down spins and @xmath74 will contain @xmath75 down spins , since the total number of down spins , including the electron spin , is conserved ) . the eigenvalue @xmath76 corresponding to the eigenstate @xmath72 is @xmath77 similarly , the eigenvalue @xmath78 corresponding to the eigenstate @xmath73 is , @xmath79 from it is clear that the eigenvalues of @xmath32 fall within certain ranges . in particular , we can write @xmath80 where we have defined @xmath81 - a_{max}\right ) \\ \label{equation::e0bound2 } e_{0,max}^\uparrow & = & + \frac{1}{2}\hbar\gamma_s b + \frac{1}{2}\left(n-2l\right)\hbar\gamma_i b + \frac{1}{4}\hbar^2\left(2a[n - l ] - a_{max}\right ) \\
\label{equation::e0bound3 } e_{0,min}^\downarrow & = & -\frac{1}{2}\hbar\gamma_s b + \frac{1}{2}\left(n-2l+2\right)\hbar\gamma_i b - \frac{1}{4}\hbar^2\left(2 a[n - l+1 ] - a_{max}\right ) \\
\label{equation::e0bound4 } e_{0,max}^\downarrow & = & -\frac{1}{2}\hbar\gamma_s b + \frac{1}{2}\left(n-2l+2\right)\hbar\gamma_i b + \frac{1}{4}\hbar^2\left(2a[l-1 ] - a_{max}\right).\end{aligned}\ ] ] it is critical to all proofs that follow that there is some energy gap @xmath82 between the electron spin - up and spin - down subspaces ( see figure 1 ) . in other words , the lowest energy electron spin - up
state must be greater in energy than the highest energy electron - spin down state . using
, it can be demonstrated that an energy gap exists if @xmath83+a[n - l+1]-a_{max}\right).\ ] ] an additional condition that emerges from theorem 2 sets the critical field slightly higher than would imply .
furthermore , we would like the definition of @xmath3 to be independent of @xmath25 .
therefore , for the remainder of the paper , we will assume that we are in the high - field regime defined by @xmath84 where @xmath85 typical values of this critical field for various systems are given in table 1 .
using these facts , we now show that the eigenstates of @xmath34 are near - eigenstates of @xmath33 .
we define the projection operators @xmath86 and @xmath87 such that @xmath88 where @xmath89 is the identity operator on the nuclear spins .
then the following theorem can be proven : [ theorem::1 ] the eigenstates of @xmath34 can be divided into two subspaces , which we will label @xmath58 and @xmath59 .
an eigenstate @xmath90 in the @xmath58 subspace obeys the conditions @xmath91 and an eigenstate @xmath92 in the @xmath59 subspace obeys the conditions @xmath93 where @xmath94 such that @xmath95 , @xmath96 , and @xmath97 a[n+1-l]}.\ ] ] it is useful to note that @xmath98 for all l. above the critical field given in , we find that @xmath99 and @xmath100 .
hence , we can write the eigenstates of @xmath34 as either @xmath101 or @xmath102 , where the @xmath58 eigenstates are those which contain contributions primarily from electron spin - up states ( @xmath65 ) , and the @xmath59 eigenstates contain contributions primarily from the electron spin down states ( @xmath66 ) , @xmath103 to summarize , we have shown that at high fields , states can still be approximately labeled according to their electron spin polarization .
we now proceed with the second step of the proof .
[ theorem::2 ] the eigenvalue @xmath104 corresponding to eigenstate @xmath101 and the eigenvalue @xmath105 corresponding to eigenstate @xmath102 obey the following conditions @xmath106 and @xmath107 where @xmath108 theorem 2 demonstrates that the eigenvalues of the @xmath58 and @xmath59 eigenstates of @xmath34 fall in approximately the same ranges as the eigenvalues of the @xmath109 and @xmath110 eigenstates of @xmath32 ( c.f . ) . the third step in the proof is to show that any eigenstate of @xmath33 , in other words , any state with a well - defined electron polarization , is contained almost completely in either the @xmath58 or @xmath59 subspace ( note that this statement _ does not _ immediately follow from theorem 1 , in which we proved the converse ; namely that @xmath58 or @xmath59 eigenstates of @xmath34 are almost completely contained within the @xmath109 or @xmath110 subspace . )
[ theorem::3 ] let @xmath111 be an arbitrary state in which the electron is polarized in the @xmath67 direction , i.e. @xmath112 is some arbitrary state of the nuclear spins .
then , @xmath113 where @xmath114 is the projector onto the @xmath59 eigenstates of @xmath34 and @xmath115 similarly , @xmath116 where @xmath117 is the projector onto the @xmath58 eigenstates of @xmath34 and @xmath118 to simplify future analysis , we will use @xmath119 in place of @xmath120 and @xmath121 .
then theorem 3 demonstrates that an arbitrary state with a @xmath122 polarized electron has at most an amplitude of @xmath123 on the @xmath124 subspace .
finally , let us consider the time evolution of the state @xmath125 .
we will show that at high fields , the @xmath5-polarization of a state does not vary substantially from its initial polarization .
in other words , at fields @xmath36 , @xmath0 is infinite because the @xmath5-polarization of the electron does not undergo substantial decay .
[ theorem::4 ] let @xmath126 be the time evolution operator given by , @xmath127 given an arbitrary initial state @xmath125 with the electron polarized in the @xmath67 direction , the following inequality holds for all times @xmath41 , @xmath128 similarly , given an arbitrary initial state @xmath129 with the electron polarized in the @xmath68 direction , @xmath130 given that the expressions for @xmath131 and @xmath132 are complicated , it is useful to observe their general form . if we assume that @xmath133 , we can expand @xmath123 and @xmath132 to lowest order in @xmath134 . to the lowest order in @xmath134 we find that @xmath135 and @xmath136 @xmath131 and @xmath132 are essentially the squares of the ratio of the overhauser energy @xmath137 to the electron zeeman energy @xmath138 .
if this ratio is small , then the @xmath5 spin of the electron undergoes only small fluctuations on the order of @xmath131 .
thus , for fields greater than @xmath3 , decay is suppressed and @xmath0 is consequently infinite .
let us now turn our attention to the second relaxation time @xmath1 .
@xmath1 describes the decay of the in - plane magnetization under a spin echo experiment , i.e. the decay of the expectation value @xmath139 , where a single spin echo @xmath43-pulse is applied at time @xmath41 .
the bound on @xmath1 is based on the variational principle mentioned above .
it will be shown that the @xmath58 and @xmath59 eigenstates obey very similar evolution equations .
thus , the dephasing which occurs between @xmath58 and @xmath59 eigenstates is limited by the `` magnitude '' of this difference in dynamics .
we first begin with some general observations .
let @xmath50 be a wavepacket formed from a linear combination of eigenstates of the hamiltonian @xmath34 .
then we can write , @xmath140 where @xmath141 is an eigenket of @xmath34 with eigenvalue @xmath142 and where we have defined @xmath143 . we can view @xmath144 as a probability distribution over the eigenstates of @xmath34 induced by the state @xmath50 .
then we can describe the root mean squared width @xmath82 of the distribution of energies as , @xmath145 theorems 5 and 6 will apply this idea to bounding the dynamics of @xmath50 .
[ theorem::5 ] let @xmath50 be a trial state with energy width @xmath82 , let @xmath146 be some energy width cutoff with @xmath147 , and let @xmath148 be the set of eigenstates of @xmath34 states such that @xmath149 .
then , @xmath150 [ theorem::6 ] given a state @xmath151 with energy width @xmath82 , then for times @xmath41 which satisfy , @xmath152 with @xmath147 , the free evolution of @xmath50 is given by , @xmath153 where @xmath154 is some arbitrary phase , @xmath155 is a residual vector orthogonal to @xmath50 and the parameter @xmath156 satisfies the inequality , @xmath157 the parameter @xmath158 is a function of the root mean squared energy width @xmath82 and the time @xmath41 such that , @xmath159 and @xmath160 is a parameter that can be chosen to optimize the resulting bound on @xmath1 .
using theorem 6 , it is possible to bound the dynamics of any initial state @xmath50 by evaluating its energy width , @xmath82 .
we will come back to the choice of @xmath160 later when we evaluate the @xmath1 bound for specific systems in sec .
we now turn to the problem of using the results of theorems 5 and 6 to place bounds on @xmath1 for our system of an electron spin coupled to a lattice of nuclear spins .
we assume that we start with a thermally equilibrated ensemble of nuclear spins .
a spin echo experiment is begins with an initial @xmath161 pulse denoted by the operator , @xmath162 where @xmath163 the system then undergoes free evolution for a time @xmath41 .
this free evolution is followed by a @xmath43 pulse , denoted by the operator @xmath164 which flips the electron spin .
finally , there is a second period of free evolution for a time @xmath41 .
this sequence is described by the unitary spin echo evolution operator , @xmath165 , @xmath166 the decay function @xmath167 is given by the magnitude of the in - plane magnetization after this sequence . at high fields ,
it is possible to cool the system well below the zeeman energy of the electron , thereby approximately polarizing the electron in the @xmath68 direction .
however , because the zeeman energy of a nucleus is @xmath168 smaller than that of the electron , the nuclei will , in general , be highly thermalized @xcite .
for the purposes of our analysis , we can then assume that in the initial state of the system , the electron is polarized in the @xmath68 direction ( i.e. aligned with the field ) and is uncoupled from the nuclei , which are in a completely mixed state .
the density matrix for such a system is given by @xmath169 where @xmath62 is some normalization constant which we will drop for convenience .
thus , we need to calculate the quantity @xmath170}}.\ ] ] using the cyclic property of the trace and , we can simplify this expression to @xmath171}}\\ & = & \frac{1}{2}\abs{\trace{\left[\tilde{u}^\dagger s_+ \tilde{u}(\id_s + r_\pi)\tensorm \id_i\right]}}\\ & = & \frac{1}{2}\abs{\trace{\left[\tilde{u}^\dagger s_+ \tilde{u}\right ] } + \frac{1}{2}\trace{\left[\tilde{u}^\dagger s_+ \tilde{u}r_\pi\right ] } } \\
\label{equation::v2 t } & = & \frac{1}{2}\abs{\trace{\left[\tilde{u}^\dagger s_+ \tilde{u}r_\pi\right]}}\end{aligned}\ ] ] where @xmath172 is the identity operator on the electron spin .
if we evaluate the trace in using the eigenbasis of @xmath34 , we can separate the contributions of the @xmath58 and @xmath59 eigenstates , @xmath173 } } \\ & = & \frac{1}{2}\abs { \sum_i{\bra{\psi_i^-}\tilde{u}^\dagger s_+ \tilde{u}r_\pi\ket{\psi_i^- } } + \sum_i{\bra{\psi_i^+}\tilde{u}^\dagger s_+ \tilde{u}r_\pi\ket{\psi_i^+ } } } \\ \label{equation::v2tsep } & \geq & \frac{1}{2}\abs { \sum_i{\bra{\psi_i^-}\tilde{u}^\dagger s_+ \tilde{u}r_\pi\ket{\psi_i^- } } } - \frac{1}{2}\abs { \sum_i{\bra{\psi_i^+}\tilde{u}^\dagger s_+ \tilde{u}r_\pi\ket{\psi_i^+ } } } \end{aligned}\ ] ] using theorem 1 and theorem 4 , we find , after some tedious algebra , that the contribution of the @xmath58 eigenstates is negligible , @xmath174 if we let @xmath175 be an eigenstate in the @xmath59 subspace , we can also simplify the contribution of the @xmath59 states , @xmath176 combining we obtain @xmath177 the problem of evaluating @xmath167 has been reduced to determining the time evolution of the states @xmath178 and @xmath179 .
the physical interpretation of the derivation of is that we have separated out terms which primarily contribute to dephasing processes ( i.e. , the terms in the sum over i ) from terms due to the longitudinal relaxation of the electron ( i.e. , the remaining terms ) .
these longitudinal terms are on the order of @xmath132 or @xmath131 since this is the magnitude of longitudinal relaxation demonstrated in theorem 4 .
we must next evaluate the sum in .
it is clear that under free evolution , @xmath178 simply acquires an overall phase ( minus some small residual ) , since it is the projection of an eigenstate of @xmath34 onto the @xmath110 subspace .
this statement is formalized in theorem 7 .
a similar idea can be used to characterize the evolution of @xmath179 . because @xmath179 is not necessarily close to an eigenstate of @xmath34 ,
its evolution can be complicated and may not in general be characterized simply by an overall phase .
if we could calculate the energy width of @xmath179 , then we could use theorem 6 to bound its dynamics .
however , because the explicit form of the nuclear state @xmath180 is not known , we must use a more subtle method of obtaining the energy width .
theorem 8 shows that a trial state @xmath50 can be constructed with energy width given by a quantity @xmath181 that is determined by @xmath182 and @xmath183 .
it is also shown that @xmath179 is the projection of @xmath50 onto the @xmath109 subspace .
using this fact , theorem 9 then bounds the evolution of @xmath184 .
we now present these three results .
[ theorem::7 ] let @xmath185 be an eigenstate of @xmath34 in the @xmath59 subspace .
then @xmath186 where @xmath187 again , because @xmath178 is close to an eigenstate of @xmath34 , its evolution will essentially be characterized by an overall phase plus a small residual . because @xmath179 is not necessarily close to an eigenstate of @xmath34
, we can not use the same reasoning to characterize its evolution .
instead , we show that @xmath179 is close to a trial state @xmath188 whose evolution we can characterize .
[ theorem::8 ] there exists a trial state @xmath188 such that @xmath189 and @xmath190 where @xmath191 according to theorem 8 , the trial state @xmath188 has an energy width of @xmath181 . using theorem 6 ,
we know that for @xmath192 the evolution of @xmath188 will then be well - described by phase evolution with some arbitrary phase denoted by @xmath193 ( corresponding to @xmath154 in theorem 4 ) plus some small residual . because @xmath188 is close to the state @xmath179 , theorem 9 now uses the same reasoning as in theorem 7 to characterize the evolution of @xmath179 .
[ theorem::9 ] let @xmath194 be an eigenstate of @xmath34 in the @xmath59 subspace .
then for @xmath195 , @xmath196 where the following inequalities are satisfied , @xmath197 @xmath198 @xmath199 using theorems 7 and 9 , we obtain the following results for the evolution of @xmath178 and @xmath179 under the spin echo experiment : @xmath200 @xmath201 one effect of the spin echo evolution is especially important to note .
if we compare the free evolution of @xmath202 and @xmath203 , we find that they acquire phases of @xmath204 and @xmath193 , respectively .
thus , there is a relative phase difference between the two evolutions , which would lead to a fast dephasing time @xmath205 when we sum over @xmath206 in to account for our initial , incoherent thermal distribution .
however , when the spin echo experiment is performed , show that both states evolve with the phase factor of @xmath207 , leading to no relative phase difference .
as expected , the inhomogeneous dephasing due to an incoherent distribution over initial states is removed by the spin echo experiment . finally , we can use to obtain , @xmath208 combining and , we obtain as a final result @xmath209 for @xmath210 .
though this bound is rigorous , it is somewhat complicated . in practice , numerical evaluation of the expressions for @xmath211 and @xmath181
can be greatly simplified if we assume that @xmath212 , as in . in this case
, we can also use the approximation that @xmath213 where @xmath214 it should be emphasized that this approximation ( @xmath212 ) is not intrinsic to our proof and is used only to make our results more compact and easier to present . if a strict bound were desired , we could drop this approximation and keep only the weaker condition that @xmath36 . to obtain a bound on @xmath1 , we set @xmath215 , @xmath216 let @xmath217 be the positive root of this equation .
then we solve for @xmath218 in terms of @xmath160 such that @xmath219 , @xmath220 this value of @xmath218 will be a lower bound on @xmath1 .
to demonstrate how our bounding method can be used in practice , we apply it here to three systems of relevance to quantum information processing@xcite : ( 1 ) a donor impurity in si , ( 2 ) a donor impurity in gaas , and ( 3 ) a gaas quantum dot .
table 1 shows the important physical parameters for these systems . to obtain a lower bound on @xmath1 for each system
, we must evaluate for the given system s parameters . because we are free to choose the parameter @xmath160 in theorem 6 with only the restriction that @xmath147 , we can select @xmath160 to maximize @xmath218 . in each of the following examples ,
this optimization procedure was carried out to obtain the largest possible @xmath218 .
consider first the case of a phosphorous impurity atom in a lattice of natural silicon ( which contains @xmath221 of the spin-@xmath222 @xmath223si isotope ) .
spin coherence is affected by hyperfine coupling between the electron and nuclei and dipolar coupling between nuclei .
the hyperfine coupling constants are given by the equation @xcite , @xmath224 where @xmath225 is the position of nuclear spin @xmath226 and @xmath227 is a hydrogenic orbital for the impurity electron . using our bounding technique
, we can determine when the @xmath228 terms in the hamiltonian can be safely neglected as the primary source for electron spin decoherence .
the coherence time due to dipole - dipole coupling of nuclei for a sample of natural si in the presence of an external field aligned with the @xmath229 $ ] direction has been estimated to be @xmath230 @xcite .
this value is essentially independent of the magnetic field strength .
in contrast , figure 2 shows that our lower bound is directly proportional to @xmath2 . using the data from figure 2
, we find that at approximately @xmath231 , decoherence due to the dipolar nuclear coupling dominates that due to the hyperfine interaction .
thus , for this system , at fields above @xmath232 it is safe to neglect the @xmath228 terms in the hamiltonian as sources for decoherence .
next , we investigate the case of a donor impurity in gaas and a gaas quantum dot ( @xmath233 nm radius , @xmath234 nm thickness ) .
ga and as nuclei are both spin @xmath235 .
we can easily extend our spin-@xmath222 formalism to include higher spin nuclei if we note that a spin-@xmath235 particle can be treated as a composite system of three spin-@xmath222 particles .
however , we also note that the ga and as nuclei will have different gyromagnetic ratios @xmath236 , @xmath237 , and @xmath238 . as a result
the zeeman energy of the electron can be exchanged with the nuclear zeeman energy , providing a new channel for @xmath0 relaxation . in order to apply our bound to the gaas systems
, we therefore must make the assumption that at high fields nuclear polarization is not transferred between nuclei of different types .
we then let @xmath239 , and apply our proof as before . again , the important parameters are shown in table 1 .
figure 3 shows the lower bound on @xmath1 as a function of @xmath2 for the donor impurity in a gaas lattice ( solid line ) and a gaas quantum dot ( dot - dashed line ) .
a lower bound for @xmath1 is provided for fields greater than @xmath240 .
for both systems , our method also shows that @xmath0 is infinite above a critical field of @xmath241 .
previous studies have shown that nuclear spectral diffusion causes decoherence on a timescale of @xmath242 for both these gaas systems @xcite .
hence , figure 3 shows that for magnetic fields greater than approximately @xmath243 , nuclear spectral diffusion will be the dominant decoherence mechanism and the @xmath228 terms in the hamiltonian can be neglected .
it is useful to assign a physical significance to the quantities that appear in our bounds .
for instance , the relevant quantity in the @xmath0 bound is given alternatively by @xmath131 or @xmath132 , both of which quantify the mixing of the spin - up and spin - down eigenstates of @xmath32 .
these quantities scale as the square of the ratio of the overhauser energy @xmath244 and the electron zeeman splitting , @xmath245 . in the language of perturbation theory
, the large energy denominator suppresses mixing of spin - up and spin - down states via the perturbation @xmath35 . on the other hand , the operator which sets the scale for @xmath1 is @xmath246 , which can be written as , @xmath247 this expression motivates us to identify the operator @xmath246 with difference in effective hamiltonians felt by electron spin - up and electron spin - down states .
for instance , the second term in this equation describes the hamiltonian experienced by an electron spin - up state scattered via two applications of @xmath35 into a second electron spin - up state . on the other hand
, the first term applies the same scattering process to an electron spin - down state .
taking the difference of these two terms allows cancellation to occur , reducing the magnitude of @xmath246 , and corresponding to the refocusing of spin states accomplished by the spin echo experiment .
if no spin echo were applied , then the first of these terms would be absent , and the magnitude of @xmath246 would be substantially larger , leading to a faster decay of coherence . in summary , we have used a completely non - perturbative approach to place a rigorous lower bounds on the @xmath0 and @xmath1 coherence times of an electron spin coupled via the contact hyperfine interaction to a lattice of nuclear spins . for @xmath0
, we have shown that for @xmath36 , the @xmath5-polarization of the electron is nearly conserved by time evolution ; hence @xmath0 is infinite . for @xmath1 , we have obtained an analytic expression for a lower bound @xmath218 , having explicitly taken into account the effects of a spin echo experiment , which would be required to remove any additional inhomogeneous broadening .
this analysis holds above some critical field satisfying the high field condition , @xmath248 .
we have also assumed that at the start of an experiment , the electron is polarized and uncoupled from the completely mixed nuclear initial state ; similar results can be derived for any highly mixed nuclear initial state . in general , this method can be used to determine when the contact hyperfine interaction can be safely neglected as a direct mechanism for electron spin decoherence .
we have demonstrated the utility of the bounds for a donor impurity in si , and -within the approximation that spontaneous nuclear polarization transfer does not occur- for a donor impurity in gaas , and a gaas quantum dot .
future directions for this investigation are two - fold .
first , numerical simulations indicate that the coherence time @xmath1 due to hyperfine coupling only might , in fact , be substantially higher than the lower bound given here , possibly even infinite @xcite .
the primary contribution to @xmath1 comes from the diagonal matrix elements of @xmath246 , which corresponds to the self - energy of the perturbation .
removing this term would greatly improve the lower bound on @xmath1 ; unfortunately , it is unclear how this term can be suppressed . also , we have not taken advantage of the substantial symmetries present in the system @xcite . taking advantage of these symmetries in our analysis
could dramatically increase our lower bound on @xmath1 .
second , this method might be extended to yield bounds on the fidelity of single qubit operations on the electron spin , which would be important for any solid state quantum computing applications .
finally , it is likely that our method could be extended to any quasi - two level system with a significant energy gap that is coupled to a bath via an off - diagonal perturbation
. it would be useful to generalize the method so that it could be applied to other systems of interest to quantum computation .
.this table lists the relevant parameters for our calculations of the lower bounds on @xmath0 and @xmath1 for the electron spin .
@xmath3 is given by and @xmath249 by .
the parameters for the si : donor and gaas : donor systems were calculated using simple hydrogenic orbitals as described in @xcite .
the gaas quantum dot considered had a radius of @xmath250 and a thickness of @xmath251 .
the parameters for the quantum dot were calculated as described in @xcite . here
@xmath252 provides an approximate bound for @xmath1 for any field strength @xmath2 .
[ cols="<,^,^,^",options="header " , ] and @xmath34 .
the eigenstates of the unperturbed hamiltonian can be grouped into two subspaces , @xmath109 and @xmath110 based on the @xmath5-polarization of the electron . in theorems 1 and 2 , we demonstrate that the eigenstates of the full hamiltonian @xmath34 can be grouped into two subspaces , @xmath58 and @xmath59 , based on the approximate @xmath5-polarization of the electron . theorem 3 further states that the @xmath58 and @xmath59 eigenstates fall within the same range as the @xmath109 and @xmath110 eigenstates , respectively .
, width=672,height=576 ] versus magnetic field for a donor impurity in a sample of natural si .
note : this plot shows the dependence of the _ lower bound _ on the magnetic field , not the dependence of @xmath1 itself . at high magnetic fields
@xmath212 the lower bound scales approximately as @xmath2 .
the dashed line shows the predicted timescale for decoherence due to nuclear spectral diffusion arising from internuclear dipolar coupling @xcite .
thus , at fields greater than @xmath232 , this nuclear spectral diffusion will dominate hyperfine coupling as a source for decoherence . ] versus magnetic field @xmath2 for a donor impurity in gaas and a gaas quantum dot with radius @xmath253 and thickness @xmath254 .
the dashed line shows the approximate decoherence time of @xmath255 for both systems due to dipolar - induced nuclear spectral diffusion @xcite . at fields above @xmath256
, spectral diffusion will dominate hyperfine coupling as a source for decoherence . ]
25 natexlab#1#1bibnamefont # 1#1bibfnamefont # 1#1citenamefont # 1#1url # 1`#1`urlprefix[2]#2 [ 2][]#2 , , , , , , , , * * , ( ) . , , , * * , ( ) . , * * , ( ) . , , , * * , ( ) . , * * , ( ) . , , , * * , ( ) .
, , , , * * , ( ) . , , , , ( ) . , , , , , , * * , ( ) . , , ,
* * , ( ) . , * * , ( ) . , , , , * * , ( ) . , * * , ( ) . , * * , ( ) . , , , * * , ( ) . , * * , ( ) . , * * , ( ) . , , , ( ) . ,
* * , ( ) . , , , * * , ( ) . , , , , , , , , * * , ( ) . , , , ,
* * , ( ) . , , , ( ) . ,
* * , ( ) . , * * , ( ) .
let @xmath112 be an eigenstate of @xmath34 such that , @xmath257 without loss of generality , we can write @xmath112 as , @xmath258 where @xmath259 recall that @xmath260 are the electron spin projection operators defined in . multiplying on the left by either @xmath261 or @xmath262 , we obtain the two equations , @xmath263 it is straightforward to set a bound on the quantities @xmath264 and @xmath265 since the eigenspectrum of @xmath32 is known ( see ) .
specifically , it can be shown that , @xmath266 and @xmath267 where @xmath268 here , we pause to introduce several relations involving matrix norms which we will make use of in what follows .
let @xmath269 be an arbitrary hermitian matrix .
then following the standard definitions , we let @xmath270 represent the maximum absolute column sum norm @xmath271 the maximum absolute row sum norm is given by @xmath272 finally , the spectral norm , can be defined as @xmath273 when applied to a vector @xmath274 is simply the @xmath275 norm of @xmath112 .
we will also use two useful relations regarding these norms@xcite @xmath276 we can use these relations applied to the matrix @xmath35 to bound the remaining quantities in .
using we obtain @xmath277 because the absolute column sum and absolute row sum norms of @xmath35 are easy to calculate , we can use them to bound the spectral norm of @xmath35 within a given @xmath25 subspace using @xmath278 a[n - l+1 ] } \\
& \equiv & v_{max}\end{aligned}\ ] ] similarly , @xmath279 then we obtain the two bounding inequalities , @xmath280 combining , we obtain the inequality @xmath281 since the normalization of @xmath112 implies that @xmath282 , some brief algebra then yields @xmath283 solving the equality for @xmath284 , we obtain the two solutions @xmath285 where @xmath286 and @xmath132 are given in . thus , using ,
we have shown that every eigenstate @xmath141 of @xmath34 satisfies one of the two inequalities , @xmath287 or @xmath288 qed we will show only one of these bounds as the rest can be demonstrated by identical procedures . for a @xmath58 eigenstate , we know from the proof of theorem 1 that @xmath289 then gives us , @xmath290 where we have recalled that our high field condition ensures that @xmath291 ( see ) .
the other bounds are derived in exactly the same way .
qed we will demonstrate this theorem for the spin - up case ; the spin - down case follows the same argument .
let @xmath125 be an arbitrary state with the electron polarized in the @xmath67 direction .
since @xmath125 has no spin - down component , its expectation value over @xmath35 is @xmath292 .
given the eigenspectrum of @xmath32 , it is straightforward to show that , @xmath293 given the matrix norm of the perturbation @xmath35 in , it is equally straightforward to show that , @xmath294 however , we can also write the left hand side of as , @xmath295 where @xmath296 then using , we can bound this quantity as , @xmath297 combining , we obtain , @xmath298 where @xmath299 qed we will demonstrate the inequality for the initial state @xmath125 ; the spin - down inequality can be proved in a similar manner . given the initial state @xmath125 , the state at time @xmath41 will be given by , @xmath300 the amplitude in the electron spin - down subspace after time t is given by , @xmath301 where the maximization is taken over all normalized nuclear states @xmath50 .
we have also used and the fact that @xmath126 and @xmath302 commute .
similarly , we can obtain , @xmath303 using , we can now bound the z - polarization of the electron starting from an initial state which is polarized .
first , we note that the operator @xmath33 can be written in terms of the projection operators @xmath86 and @xmath87 , @xmath304 using , we find that if the initial state of the system is given by @xmath125 , then the z - polarization at an arbitrary time @xmath41 is bounded by , @xmath305 qed from the statement of theorem 5 , we have defined @xmath306 .
we have also defined the set of eigenstates @xmath148 such that for every eigenstate @xmath62 in @xmath148 , we have @xmath307
. then we can write @xmath308 so , @xmath309 qed let @xmath315 be the eigenvalue corresponding to @xmath312 .
let us consider the normalized trial state , @xmath316 where @xmath317 we select this trial state because it has the desirable property that @xmath318 .
our selection of the spin - down component of @xmath50 is designed to minimize the overall energy width of @xmath50 .
we then obtain , @xmath319 where we have used the fact that @xmath320 .
if we define the operator @xmath321 then we can evaluate the expectation value , @xmath322 these expectation values can be bounded using the l2 norm of @xmath246 ( see for a definition of the l2 matrix norm ) . to obtain this norm , we write @xmath246 in the @xmath69 basis , @xmath323 here we define @xmath324 as the diagonal portion of @xmath246 in the @xmath325 basis , @xmath326 where @xmath226 sums over nuclei for which @xmath327 and @xmath62 sums over nuclei such that @xmath328 .
@xmath329 is the off - diagonal portion of @xmath246 , @xmath330 where @xmath69 , and @xmath74 are nuclear spin configurations which differ only at nuclear positions @xmath226 and @xmath62 such that that @xmath331 and @xmath332 .
we note that the two terms in are nearly equal , leading to a significant cancellation in .
this cancellation is a result of the spin echo experiment , and will be discussed in .
because @xmath333 are negligible relative to @xmath334 , we can neglect these terms in the denominator .
we also note that @xmath335 . then using the inequality in ,
we can bound @xmath181 as @xmath336 the second term in this equality is strictly less than the first term by a factor of @xmath123 ( see ) .
so we can write the strict inequality , @xmath337 hence it follows from and that @xmath338 qed since we know the energy width of the trial state @xmath188 , we can apply theorem 6 to characterize its dynamics @xmath339 in , @xmath340 is an overall phase corresponding to @xmath154 in theorem 6 .
@xmath341 is the magnitude of the residual evolution , and is bounded by . in order to characterize the evolution of @xmath179
, we then note that @xmath342 using similar reasoning as in the proof for theorem 7 , we obtain @xmath343 the final element in our proof is to demonstrate the three inequalities in
. follows directly from theorem 6
. follows from the fact that we can bound the magnitude of @xmath344 , @xmath345 finally , follows from the consideration that ( from theorem 6 ) , @xmath346 .
then , @xmath347 qed | we address the decoherence of a localized electron spin in an external magnetic field due to the hyperfine interaction with a lattice of nuclear spins . using a completely non - perturbative method ,
rigorous bounds on the @xmath0 and @xmath1 coherence times for the electron spin are provided .
it is shown that for magnetic fields @xmath2 greater than some critical field @xmath3 ( @xmath4 for the systems studied here ) , the @xmath5-polarization of the electron spin can not relax , and hence @xmath0 is infinite . however , even at high fields dephasing can still occur .
we provide a lower bound on the @xmath1 coherence time that explicitly takes into account the effects of a spin echo pulse sequence . |
* b72 hereafter ) show that one need an extra @xmath0 relative velocity boost after the plunge in order for the inbound debris to reach the ergosphere negative - energy orbits , to ultimately extract energy from the negative - energy orbits to be carried away by the outbound debris .
note that , while for a kerr black hole the apparent " radii of the event horizon , the photon orbit and and isco ( @xmath1 ) on a prograde orbit are the same in the boyer - lindquist coordinates , they are separate in proper coordinates and their energies are distinctly different . for example , for a kerr black hole , the orbital velocity at isco is @xmath2 ( not @xmath3 ) and the minimum energy of a plunge orbit that results from decay of a bound stable orbit is @xmath4 ( where @xmath5 is the mass of the orbiting particle ) .
thus , in order to get to the negative - energy ergosphere orbits , the orbiting particle at isco need to cross over this velocity gap of @xmath6 .
this condition was deemed to be astrophysically unlikely ( b72 ) .
we suggest here that , when two black holes merge , if one or both have orbital material at close to their respective isco , then some of the orbital material may be able to jump " over this velocity gap of @xmath6 to land in the ergosphere negative - energy orbits , while portion of the orbiting material that escapes may be able to extract energy from the black hole via the @xcite process at the expense of the rotation of the black hole @xcite .
our subsequent argument , in a large part , hinges on the assumption that at least some smbh are kerr black holes , because only their isco orbits are energetic enough to possess orbiting matter with a velocity dispersion equal to @xmath3 .
this is supported by at least two observational lines of evidence : the observational inference of high radiative efficiency of luminous quasars of @xmath7 ( e.g. , * ? ? ?
* ) and x - ray observations of iron k line profiles ( e.g. , * ? ? ?
theoretically , kerr ( or near kerr ) smbh are fairly easily reached by gas accretion ( e.g. , * ? ? ?
the existence of accretion gas around smbh is beyond any reasonable doubt , given the fact that we see quasars shine and there is no reasonable alternative to smbh accretion .
we shall discuss briefly the possible mix of the matter that may be orbiting smbh in the inner region ; in other words , if things other than gas , such as stars or dense stellar debris , also exist there that are sufficiently long - lived .
we shall only consider solar - type stars subsequently .
compact objects such as neutron stars and white dwarfs are always swallowed whole by the smbh of mass of interest here and we assume that they do not produce astrophysically tangible signals ( even if they were launched from the center with very high velocities ) .
the tidal disruption radius for a solar - type star of mass @xmath8 and solar radius @xmath9 is @xmath10 where @xmath11 and @xmath12 is the schwarzschild radius .
equation ( [ eq : rt ] ) merely states the well known fact that @xmath13 smbh swallow solar - type stars whole , but smaller smbh tidally disrupt solar - type stars and are predicted to produce a distinct class of optical - uv flashes that have now been observationally confirmed ( e.g. , * ? ? ?
it seems likely that , for smbh with @xmath14 sufficiently smaller than one , the stellar debris from tidal disruption that becomes bound to the smbh is gasified through a variety of processes ( distortion , compression , precession , possibly nuclear flash and ultimately shocks ) and accreted by the smbh in a relatively short period of time ( e.g. , * ? ? ?
thus , it may be that there is not a significant amount of stars or stellar debris orbiting @xmath15 smbh at @xmath16 .
we will now turn our attention to smbh with @xmath17 , in light of the strong observational evidence of their existence at the center of every giant elliptical galaxy ( e.g. , * ? ? ?
for smbh of @xmath13 three primary routes may be able to capture a star without having to severely damaging it .
a star on a low angular momentum parabolic orbit may lose its orbital energy through repeated pericenter passages to be eventually captured . in order to be captured to an orbit of radius @xmath18 , it needs to lose an amount of orbital energy equal to @xmath19 . under the assumption that the star is not destroyed ( e.g. ,
not puffed up to become a giant ) , the maximum rate of energy loss may not exceed its luminosity @xmath20 , which translates to a tidal capture timescale @xmath21 being @xmath22 where @xmath23 is the lifetime of the star and @xmath24 is assumed .
this seems too long to be interesting , given that the star also has to have a pericenter distance not too much larger than @xmath25 to experience significant tidal effect for @xmath13 smbh ( see equation [ eq : rt ] ) and our assumed dissipation rate is likely on the generous side already .
of course , if the star gets overheated thus expands , the envelope region that is puffed up would be tidally stripped off to become gas " .
interaction between the smbh with a stellar binary may allow one member of the binary to be captured to a bound orbit around the smbh .
the tidal capture radius through this three - body interaction is @xmath26 ( e.g. , * ? ? ?
* ) , which is larger than the tidal radius of a single star ( equation [ eq : rt ] ) by a factor of @xmath27 with @xmath28 being the semi - major axis .
this factor , possibly substantially greater than unity , allows @xmath13 smbh to be able to capture stars , although the exact rate depends on too many factors to be certain . during the merger of two massive galaxies each hosting a smbh
, it is possible that a significant number of stars ( some fraction of them are binaries ) may be driven into low energy orbits to reach the inner regions of each smbh , given the significant amount of torque exerted by one galaxy on the other . repeated interaction between stars and an accretion disc around the smbh may also bring the star to a bound orbit .
in fact , @xcite show that for @xmath13 smbh , star - disc drag may become the dominant capture mechanism . given these considerations
, it seems that smbh of @xmath13 with an accretion disc ( i.e. , luminous quasars ) could possibly have both gas and stars at @xmath29 , while @xmath15 smbh may carry a disc that is mostly composed of gas .
what might happen to stars embedded in an accretion disc is a subject beyond this study .
we shall only note two points .
first , for a star of mass @xmath5 on a circular orbit at radius @xmath18 around a smbh of mass @xmath30 , the orbital decay time scale due to its own gravitational radiation is @xmath31 @xcite , where @xmath32 is the gravitational constant , @xmath3 is speed of light and @xmath33 .
thus , if the supply rate of solar - type stars to regions of @xmath34 , from either decay of outer orbits and/or directly captured by some processes , is larger than @xmath35yr around an @xmath13 smbh , one should expect to see some accumulation of solar - type stars near the innermost stable circular orbit @xmath36 . other drag process experienced by stars or dense debris , such as by the accretion disc , is likely to be less strong than the gravitational radiation drag at @xmath16 .
for our purpose , it suffices that it is astrophysically plausible that smbhs of @xmath13 prior to their merger possibly carry stars or stellar debris as well as gas in their inner orbits .
second , for @xmath37 , the orbiting bound stars may be significantly deformed or disrupted approaching @xmath1 , thus producing stellar debris orbiting at @xmath16 .
if that happens , the velocity dispersion of the debris will be of order @xmath38 .
therefore , the debris would be confined to a relatively narrow range in radius following the disruption .
consider the inspiral of the two kerr smbh , @xmath39 and @xmath40 ( @xmath41 ) , via gravitational radiation , with only @xmath40 ( for simplicity ) carrying a disc of gas and/or stellar debris orbiting at @xmath42 .
we assume that all angular momentum vectors are parallel and all orbits are on the same plane .
when @xmath40 arrives at @xmath1 of @xmath39 , one finds that at any instant some material is moving at the same direction as the orbital velocity of @xmath40 around @xmath39 with a relative velocity of @xmath2 , while some other material is moving at the opposite direction to the orbital velocity of @xmath40 around @xmath39 with a relative velocity of @xmath2 .
still , there is some other material that is in - between .
assuming at least some debris are still on some elliptical - like orbits ( may or may not be closed ) , given that the relative velocity of the debris around @xmath40 spans the whole range from @xmath43 to @xmath2 relative to the orbital velocity of @xmath40 about @xmath39 , it seems plausible that at least some of this material is able to fill in the required velocity gap of @xmath6 relative to the orbital velocity at @xmath36 of @xmath40 to reach the ergosphere negative - energy orbits of @xmath39 . assuming that the part of the orbiting material that reaches the ergosphere negative - energy orbits is able to communicate with the remainder of the material that eventually escapes , the originally proposed @xcite process to extract energy from the smbh may become possible .
however , caution should be taken .
it may be that , when the two smbh get that close , the assumptions on which our simple argument is based break down .
the primary assumption is that the orbiting material at @xmath29 around @xmath40 , at least some of it , manages not to plunge into @xmath40 during the process in the presence of a nearby @xmath39 , but instead plunges to a minimum - energy orbit of @xmath39 .
the other assumption is that the orbiting material around @xmath40 has not lost the velocity dispersion of @xmath3 that it had at isco , when plunging into @xmath39 .
the second assumption at least seems feasible , because it is rather unlikely that during the short time of the plunge one is able to remove " or dissipate the velocity dispersion of order the speed of light for the orbiting material around @xmath40 . on the contrary
, it would seem plausible that , during the plunge , en route to the event horizon , the velocity dispersion of the orbiting material may be further increased .
detailed black hole merger simulations ( e.g. , * ? ? ?
* ) , carrying some extra amount of orbiting material , will be necessary to provide some insight .
if what is outlined above really happens , what the escaping material that is leaving with a total energy higher than its initial total energy ( but with reduced rest mass ) will look like is just a guess at this time , in the absence of detailed simulations .
but it is conceivable that the debris may form a relativistic jet or , strictly speaking , a mixed flow of gas and stellar debris moving at ( nearly ) the speed of light , in a fashion perhaps not unlike what was suggested earlier @xcite .
if the escaping amount of material is substantial and especially a substantial fraction of that is stars or dense stellar debris , the flow may be quite bulletstic " hence would remain cruising in a straight line , like a jet , for a while .
we have argued that it is astrophysically plausible that kerr smbh carry a significant amount of material ( gas and stars or stellar debris ) in the inner region close to the innermost stable circular orbit , after their host galaxies merge .
the orbiting matter at isco has a velocity dispersion equal to the speed of light about the orbital velocity of their host smbh . as a result
, we suggest that , when the two smbh merge , some of the debris may be able to jump over the velocity gap between the isco orbit and negative - energy ergosphere orbits .
if that indeed happens , we would have found an astrophysically plausible way to tap into the vast rotational energy reservoir of smbh via the penrose process . ,
d. , ajhar , e. a. , bender , r. , bower , g. , dressler , a. , faber , s. m. , filippenko , a. v. , gebhardt , k. , green , r. , ho , l. c. , kormendy , j. , lauer , t. r. , magorrian , j. , & tremaine , s. 1998 , , 395 , a14 + | if a supermassive black hole has some material orbiting around it at close to its innermost stable circular orbit ( isco ) , then , when it plunges into a second supermassive black hole , the orbiting material has a velocity dispersion of order of speed of light about the orbital velocity of its host black hole .
it becomes plausible that some of the orbiting material will be catapulted " to the negative - energy ergosphere orbits of the second black hole at the plunge .
this may provide an astrophysically plausible way to extract energy from the black hole , originally suggested by penrose .
= 1 |
the system of degenerate coupled multi - field kdv equations is given as @xcite-@xcite @xmath6 where @xmath7 and @xmath8 . in @xcite-@xcite , it was shown that this system is also a degenerate kdv system of rank one . in a previous work @xcite we focused on the equation ( [ eqn1 ] ) for @xmath9 .
we reduced this equation into an ode @xmath10 where @xmath11 is a polynomial function of degree four .
we analyzed all possible cases about the zeros of @xmath11 . due to this analysis we determined the cases when the solution is periodic or solitary . when the polynomial has one double @xmath12 and two simple zeros @xmath13 , @xmath14 with @xmath15 , or one triple and one simple zeros , the solution is solitary .
other cases give periodic or non - real solutions . by using the jacobi elliptic functions
@xcite , we obtained periodic solutions and all solitary wave solutions which rapidly decay to some constants , explicitly .
we have also shown that there are no real asymptotically vanishing traveling wave solutions for @xmath9 .
indeed we have the following theorem for the degenerate coupled @xmath0-kdv equation , @xmath16 .
the degenerate coupled @xmath0-kdv equation can be reduced to the equation @xmath17 by taking @xmath0 functions as @xmath18 , @xmath19 , , @xmath20 , where @xmath21 in ( [ eqn1 ] ) .
here @xmath22 is a polynomial of @xmath3 of degree @xmath23 .
if we apply the asymptotically vanishing boundary conditions to ( [ genpoly ] ) , we have @xmath24 [ thmrealtravel ] when @xmath25 odd , we have @xmath26 , and then the degenerate coupled @xmath0-kdv equation has real traveling wave solution with asymptotically vanishing boundary conditions , but when @xmath25 even , the constant @xmath27 .
hence the equation does not have a real traveling wave solution with asymptotically vanishing boundary conditions .
* proof . * consider the degenerate coupled @xmath0-kdv equation ( [ eqn1 ] ) .
let @xmath18 , @xmath19 ,
, @xmath20 where @xmath21 . by using the first @xmath28 equations we obtain all functions @xmath29 , @xmath30 as a polynomial of @xmath31 .
we get @xmath32 where @xmath33 are constants , @xmath34 and @xmath35 are polynomials of @xmath3 of degree @xmath36 and @xmath37 , @xmath38 , @xmath39 , respectively .
now use the @xmath0th equation .
we obtain @xmath40 integrating above equation once we get @xmath41 by using @xmath42 as an integrating factor , we integrate once more . finally , we obtain @xmath43 where @xmath44 is a polynomial of @xmath3 of degree @xmath36 , @xmath45 . when @xmath0 is odd , the coefficient of @xmath46 , that is @xmath47 , is positive , and when @xmath0 is even , @xmath47 is negative . applying asymptotically vanishing boundary conditions , we get @xmath48 , where @xmath49 . hence for @xmath25 odd , the degenerate coupled @xmath0-kdv equation has real traveling wave solution with asymptotically vanishing boundary conditions , but when @xmath25 even , it does not .
@xmath50 in this work , we study the equation ( [ eqn1 ] ) for @xmath51 which is @xmath52 in detail .
it is clear from theorem [ thmrealtravel ] that unlike the case @xmath9 , we have real traveling wave solution with asymptotically vanishing boundary conditions in @xmath51 case . here the system ( [ l=3a])-([l=3c ] ) reduces to a polynomial of degree five , @xmath53 where @xmath54 , and @xmath55 are constants . when the degree of the polynomial in the reduced equation is equal to five or greater it is almost impossible to solve them .
as far as we know there is no known method to solve these equations .
we shall introduce two methods to solve such equations . the first one is based on the chebyshev s theorem @xcite which is used recently to solve the einstein field equations for a cosmological model @xcite , @xcite . by using the chebyshev s theorem we give several solutions of the reduced equations for @xmath51 and also for arbitrary @xmath0 .
the second method is based on the factorizing the polynomial @xmath22 as product of lower degree polynomials . in this way
we make use of the reduced equations of lower degrees .
we have given all possible such solutions for @xmath51 .
the layout of our paper is as follows : in sec .
ii , we study the behavior of the solutions in the neighborhood of the zeros of @xmath56 and discuss the cases giving solitary wave solutions . in sec .
iii , we find exact solutions of the system ( [ l=3a])-([l=3c ] ) by a new method proposed for any @xmath57 which uses the chebyshev s theorem and analyze the cases in which we may have solitary wave and kink - type solutions . in sec .
iv , we present an alternative method .
particularly , we find the solutions of the system ( [ l=3a])-([l=3c ] ) . here
we obtain many solutions for @xmath51 including solitary wave , kink - type , periodic , and unbounded solutions .
we present some of them in the text but the rest of the solutions are given in the appendices a and b.
here we will analyze the zeros of @xmath56 in ( [ p_5(f ) ] ) . *
( i ) * if @xmath59 is a _ simple zero _ of @xmath56 we have @xmath60 .
taylor expansion of @xmath56 about @xmath13 gives @xmath61 from here we get @xmath62 and @xmath63 .
hence we can write the function @xmath31 as @xmath64 thus , in the neighborhood of @xmath65 , the function @xmath31 has local minimum or maximum as @xmath66 is positive or negative respectively since @xmath63 . + * ( ii ) * if @xmath59 is a _
double zero _ of @xmath56 we have @xmath67 .
taylor expansion of @xmath56 about @xmath13 gives @xmath68 to have real solution @xmath3 , we should have @xmath69 . from the equality ( [ doublezero ] ) we get @xmath70 which gives @xmath71 where @xmath72 is a constant . hence @xmath73 as @xmath74 .
the solution @xmath3 can have only one peak and the wave extends from @xmath75 to @xmath76 .
+ * ( iii ) * if @xmath59 is a _ triple zero _ of @xmath56 we have @xmath77 .
taylor expansion of @xmath56 about @xmath13 gives @xmath78 this is valid only if both signs of @xmath79 and @xmath80 are same .
hence , to obtain real solution @xmath3 we have the following two possibilities : @xmath81 @xmath82 and @xmath83 + @xmath84 @xmath85 and @xmath86 + if @xmath82 and @xmath87 then we have @xmath88 which gives @xmath89 where @xmath90 is a constant .
thus @xmath73 as @xmath91 .
let @xmath85 and @xmath92 .
then @xmath93 which yields @xmath94 where @xmath95 is a constant .
thus @xmath73 as @xmath91 . * ( iv ) * if @xmath59 is a _
quadruple zero _ of @xmath56 , then we have @xmath96 . in this case taylor expansion of @xmath56
about @xmath13 gives @xmath97 this is valid only if @xmath98 .
then we have @xmath99 which gives @xmath100 where @xmath101 is a constant .
thus @xmath73 as @xmath91 .
* ( v ) * if @xmath59 is a _ zero of multiplicity 5 _ of @xmath56 , then we have @xmath102 .
this is valid only if @xmath103 .
so we obtain the solution @xmath3 as @xmath104 where @xmath105 is a constant .
hence @xmath73 as @xmath106 .
we analyze all possible cases about the zeros of @xmath56 that may give solitary wave solutions . here in each cases
we will present the sketches of the graphs of @xmath56 .
real solutions of @xmath107 occur in the shaded regions . *
( 1 ) one double and three simple zeros . * having one double and three simple zeros ] in figure 1.*(b ) * , @xmath13 , @xmath14 , and @xmath108 are simple zeros and @xmath12 is a double zero .
the real solution occurs when @xmath3 stays between @xmath13 and @xmath12 or @xmath12 and @xmath14 . at @xmath13 , @xmath109
hence graph of the function @xmath3 is concave up at @xmath110 . at double zero @xmath12 , @xmath111 as @xmath106 .
hence we have a solitary wave solution with amplitude @xmath112 .
similarly at @xmath14 , @xmath113 , hence graph of the function @xmath3 is concave down at @xmath114 .
therefore , we also have a solitary wave solution with amplitude @xmath115 .
now consider the graph * ( d ) * in figure 1 .
for @xmath14 we have @xmath116 thus graph of the function is concave up at @xmath114 .
at double zero @xmath108 , @xmath117 as @xmath106 .
hence we have a solitary wave solution with amplitude @xmath118 .
in other cases we have periodic solutions . * ( 2 ) two double and one simple zeros . * having two double and one simple zeros ] in figure 2.*(f ) * , @xmath13 and @xmath14 are double zeros and @xmath12 is a simple zero .
the real solution occurs when @xmath3 stays between @xmath12 and @xmath14 .
for @xmath12 we have @xmath119 thus graph of the function is concave up at @xmath120 .
at double zero @xmath14 , @xmath121 as @xmath106 .
hence we have a solitary wave solution with amplitude @xmath122 . now consider the graph * ( g ) * in figure 2 . here
@xmath12 and @xmath14 are double zeros and @xmath13 is a simple zero .
the real solution occurs when @xmath3 stays between @xmath13 and @xmath12 or @xmath12 and @xmath14 .
for @xmath13 we have @xmath123 thus graph of the function is concave up at @xmath110 . at double zero @xmath12 ,
@xmath111 as @xmath106 . hence we have a solitary wave solution with amplitude @xmath112 .
the other cases give kink , anti - kink type or unbounded solutions . *
( 3 ) one triple and two simple zeros .
* having one triple and two simple zeros ] consider the graph * ( h ) * in figure 3 . here
@xmath12 and @xmath14 are simple zeros and @xmath13 is a triple zero .
the real solution occurs when @xmath3 stays between @xmath13 and @xmath12 .
for @xmath12 we have @xmath124 so graph of the function @xmath3 is concave down at @xmath120 . at triple zero @xmath13 ,
@xmath73 as @xmath106 . hence we may have a solitary wave solution with amplitude @xmath125 .
in the graph in figure 3.*(i ) * , @xmath13 and @xmath14 are simple zeros and @xmath12 is a triple zero .
the real solution occurs when @xmath3 stays between @xmath13 and @xmath12 .
for @xmath13 we have @xmath109 so graph of the function @xmath3 is concave up at @xmath110 . at triple zero @xmath12 ,
@xmath111 as @xmath106 .
hence we may have a solitary wave solution with amplitude @xmath126 .
the other case gives periodic solution . *
( 4 ) one quadruple and one simple zeros .
* having one quadruple and one simple zeros ] in the graph figure 4.*(l ) * , @xmath13 is a simple zero and @xmath12 is a quadruple zero .
the real solution occurs between @xmath13 and @xmath12 . at @xmath13
we have @xmath109 , so graph of the function @xmath3 is concave up at @xmath110 . at quadruple zero @xmath12 ,
@xmath111 as @xmath106 . hence we may have a solitary wave solution with amplitude @xmath112 .
the other case gives unbounded solution .
* ( 5 ) one double and one simple zeros . *
having one double and one simple zeros ] consider the graph figure 5.*(n)*. here @xmath13 is a simple zero and @xmath12 is a double zero . the real solution occurs between @xmath13 and @xmath12 .
at @xmath13 we have @xmath109 , so graph of the function @xmath3 is concave up at @xmath110 . at double zero @xmath12 ,
@xmath111 as @xmath106 .
hence we have a solitary wave solution with amplitude @xmath112 .
the other case gives unbounded solution . to sum up
, we can give the following proposition for @xmath51 case .
[ prozeroanalysis ] equation ( [ p_5(f ) ] ) may admit solitary wave solutions when the polynomial function @xmath56 admits ( i ) one double and three simple zeros ( ii ) two double and one simple zeros ( iii ) one triple and two simple zeros ( iv ) one quadruple and one simple zeros ( v ) one double and one simple zeros .
the chebyshev s theorem is given as following @xcite .
[ chebthm ] let @xmath127 , @xmath128 , @xmath129 , @xmath72 , @xmath130 be given real numbers and @xmath131 .
the antiderivative @xmath132 is expressible by means of the elementary functions only in the three cases : @xmath133 the term @xmath134 is called a differential binomial .
note that the differential binomial may be expressed in terms of the incomplete beta function and the hypergeometric function .
let us define @xmath135 .
then we have @xmath136 where @xmath137 is the incomplete beta function and @xmath138 is the hypergeometric function . our aim is to transform the system ( [ eqn1 ] ) to @xmath139 by taking @xmath140 .
we can apply chebyshev s theorem to this equation if we assume that @xmath141 reduces to the form @xmath142 , where @xmath143 . for @xmath51 ,
let @xmath144 in ( [ p_5(f ) ] ) , then the equation becomes @xmath145 where @xmath146 to apply the chebyshev s theorem [ chebthm ] we assume that @xmath147 reduces to the following form , @xmath148 where @xmath127 , @xmath128 , @xmath129 are the constants in theorem [ chebthm ] , @xmath149 , and @xmath150 is a constant . here
we present the cases mentioned in the proposition [ prozeroanalysis ] .
other cases are given in appendix a. * let @xmath151 .
this form corresponds to the case of one simple or one simple and two double zeros .
we have @xmath152 so @xmath153 here @xmath154 , @xmath155 , and @xmath156 . hence @xmath157 for @xmath158 , from ( [ casea1 ] ) , by letting @xmath159 we obtain @xmath160 here choose @xmath161 , @xmath162 , the integration constant @xmath163 , plus sign in ( [ casea1 ] ) , we have @xmath164 and the graph of this solution for @xmath165 is given in figure 6 + ] + this is a kink - type solution .
* let @xmath166 .
this form corresponds to the case of one simple and one quadruple zeros .
we have @xmath167 so @xmath168 here @xmath154 , @xmath169 , and @xmath170 .
hence + @xmath171 for @xmath158 , from ( [ casea2 ] ) , by letting @xmath159 we obtain @xmath172 here choose @xmath173 , @xmath162 , the integration constant @xmath163 , plus sign in ( [ casea2 ] ) we have @xmath174 and the graph of this solution for @xmath175 is given in figure 7 + ] * let @xmath176 .
this form corresponds to the case of one simple and one double zeros .
we have @xmath177 so @xmath178 here @xmath179 , @xmath180 , and @xmath181 .
hence @xmath182 for @xmath183 , from ( [ casea3 ] ) , we obtain @xmath184 here choose @xmath185 , @xmath186 , the integration constant @xmath163 , and plus sign in ( [ casea3 ] ) we obtain @xmath187 + and the graph of this solution is given in figure 8 + ] + this is clearly a solitary wave solution .
* let @xmath188 .
this form corresponds to the case of one double and one triple zeros .
we have @xmath189 so @xmath190 here @xmath179 , @xmath169 , and @xmath191 .
hence @xmath192 for @xmath183 , from ( [ casea4 ] ) , we obtain @xmath193 here choose @xmath173 , @xmath162 , @xmath163 , and plus sign in ( [ casea4 ] ) we have @xmath194 and the graph of this solution is given in figure 9 + ] * let @xmath195 .
this form corresponds to the case of one simple and one quadruple zeros .
we have @xmath196 so @xmath197 here @xmath198 , @xmath169 , and @xmath181
. hence @xmath199 for @xmath183 , from ( [ casea5 ] ) , by letting @xmath159 we obtain @xmath200 + take @xmath173 , @xmath162 , @xmath163 , and plus sign in ( [ casea5 ] ) we get @xmath201 and the graph of this solution for @xmath202 is given in figure 10 + ]
when we have @xmath1 for the reduced equation it becomes quite difficult to solve such equations for @xmath203 . for this purpose
, we shall introduce a new method which is based on the factorization of the polynomial @xmath204 , @xmath203 .
let the polynomial @xmath22 have @xmath23 real roots i.e. @xmath205 , @xmath47 is a constant . define a new function @xmath206 so that @xmath207 .
hence we have @xmath208 . by taking @xmath209 where @xmath210 , @xmath211
, we get a system of ordinary differential equations .
solving this system gives the solution of the degenerate coupled @xmath0-kdv equation . for illustration , we start with a differential equation where we know the solution .
consider @xmath212 let @xmath213 so @xmath214 take @xmath215 where @xmath216 .
we start with solving the equation ( [ exalternative1 ] ) .
we have @xmath217 where @xmath218 and @xmath219 .
let @xmath220 .
this gives us @xmath221 .
after taking the integral we obtain @xmath222 so that @xmath223 now we insert this solution into eq.([exalternative2 ] ) .
let @xmath224 .
hence we get @xmath225 where @xmath226 , @xmath227 , and @xmath228 is a constant .
solving ( [ exampleintegral ] ) and using ( [ solnex ] ) gives @xmath229 which was obtained in @xcite .
here @xmath230 is the jacobi elliptic function and @xmath231 is the elliptic modulus satisfying @xmath232 .
now we apply our method to the case when the polynomial @xmath2 is of degree @xmath233 .
we have several possible cases , but here we shall give the case when the polynomial ( [ p_5(f ) ] ) has five real roots .
other cases are presented in appendix b. * case 1 . *
if ( [ p_5(f ) ] ) has five real roots we can write it in the form @xmath234 where @xmath235 and @xmath236 are the zeros of the polynomial function @xmath56 . now define a new function @xmath206 so that @xmath207 .
hence ( [ p_5(f)second ] ) becomes @xmath237 * i. * take @xmath238 in @xcite we have found the solutions of eq.([caseanewmethod1 ] ) .
one of the solutions is @xmath239 where @xmath231 is the elliptic modulus satisfying @xmath240 .
we use this solution in the equation ( [ caseanewmethod2 ] ) and get @xmath241 where @xmath242 and @xmath243 . for particular values ;
@xmath244 , @xmath245 , @xmath246 , @xmath247 , @xmath173 , @xmath248 , and choosing plus sign in ( [ rhocasea ] ) we get the graph of the solution @xmath31 given in figure 11 ) with ( [ rhocasea ] ) for particular parameters ] this is a solitary wave solution . *
* ii.**take @xmath249 where @xmath250 .
eq.([casebnewmethod1 ] ) has a solution @xmath251 where @xmath231 satisfies @xmath252 and @xmath253 is a constant .
we use this solution in eq.([casebnewmethod2 ] ) and get @xmath254 where @xmath255 is a constant and @xmath256 and @xmath257 . for particular values ; @xmath258 , @xmath259 , @xmath245 , @xmath260 , @xmath261 , @xmath262 , @xmath248 , and choosing plus sign in ( [ rhocaseb ] ) we get the graph of the solution @xmath31 given in figure 12 ) with ( [ rhocaseb ] ) for particular parameters ] this solution is periodic .
+ * iii . * take @xmath263 where @xmath250 .
consider first the equation ( [ casecnewmethod1 ] ) .
+ we have @xmath264 this equality can be reduced to @xmath265 where @xmath266 , @xmath267 . by making the change of variables @xmath268 , we get @xmath269 where @xmath253 is a constant . thus we obtain @xmath270 which yields @xmath271 now we use this result in ( [ casecnewmethod2 ] ) , @xmath272 where @xmath255 is a constant and @xmath273 and @xmath274 . for particular values ; @xmath275 , @xmath276 , @xmath277 , @xmath278 , @xmath279 , @xmath280 , @xmath281 , and choosing plus sign in ( [ rhocasec ] ) , we get the graph of the solution @xmath31 given in figure 13 ) for particular parameters ] * iv . *
take @xmath282 where @xmath250 .
consider the eq.([casednewmethod1 ] ) .
we have @xmath283 integrating both sides gives @xmath284 we use this result in the equation ( [ casednewmethod2 ] ) @xmath285 where @xmath255 is a constant and @xmath286 for particular values , @xmath258 , @xmath287 , @xmath288 , @xmath289 , @xmath279 , @xmath280 , and @xmath281 , and choosing plus sign in ( [ rhocased ] ) we get the graph of the solution @xmath31 given in figure 14 ) for particular parameters ] this is a solitary wave solution .
we study the degenerate @xmath0-coupled kdv equations .
we reduce these equations into an ode of the form @xmath290 , where @xmath291 is a polynomial function of @xmath3 of degree @xmath23 , @xmath57 .
we give a general approach to solve the degenerate @xmath0-coupled equations by introducing two new methods that one of them uses chebyshev s theorem and the other one is an alternative method , based on factorization of @xmath22 , @xmath57 .
particularly , for the degenerate three - coupled kdv equations we obtain solitary - wave , kink - type , periodic , or unbounded solutions by using these methods .
this work is partially supported by the scientific and technological research council of turkey ( tbitak ) . + * appendix a. other cases using chebyshev s theorem for degenerate three - coupled kdv equation * * let @xmath292 .
this form corresponds to the case of one simple zero or three simple zeros .
we have @xmath293 so @xmath294 here @xmath154 , @xmath295 , and @xmath181 .
so we have @xmath296 hence we can not obtain a solution through the chebyshev s theorem .
* let @xmath297 .
this form corresponds to the case of one triple or two simple and one triple zeros .
we have @xmath298 so @xmath299 here @xmath300 , @xmath155 , and @xmath181 .
so we have @xmath301 hence we can not obtain a solution through the chebyshev s theorem .
* let @xmath302 .
this form corresponds to the case of one double and one triple zeros .
we have @xmath303 so @xmath304 here @xmath300 , @xmath169 , and @xmath156 .
hence @xmath305 for @xmath158 , from ( [ caseap3 ] ) , by letting @xmath159 we obtain @xmath306 * let @xmath307 .
this form corresponds to the case of a zero with multiplicity five .
we have @xmath308 so @xmath309 here @xmath310 , @xmath169 , and @xmath311 .
hence @xmath312 for @xmath183 , from ( [ caseap4 ] ) , by letting @xmath159 we obtain @xmath313 here we get @xmath314.\ ] ] * let @xmath315 .
this form corresponds to the case of one simple zero .
we have @xmath316 so @xmath317 here @xmath310 , @xmath318 , and @xmath181 .
we have @xmath319 hence we can not obtain a solution through the chebyshev s theorem .
* case 2 . *
if ( [ p_5(f ) ] ) has three real roots we can write it in the form @xmath320 where @xmath13 , @xmath12 , and @xmath14 are the zeros of the polynomial function @xmath56 of degree five and @xmath321 . now define a new function @xmath206 so that @xmath207 .
hence ( [ p_5(f)threerealroots ] ) becomes @xmath322 * i. * take @xmath323 where @xmath250 .
consider the first equation above .
we have @xmath324 we know from case 1.ii that eq.([case2inewmethod1 ] ) has a solution @xmath325 where @xmath253 is a constant .
we use this solution in ( [ case2inewmethod2 ] ) and we get @xmath326 where @xmath255 is a constant and @xmath327 and @xmath257 . *
ii . * take @xmath328 where @xmath250 .
consider first the equation ( [ case2iinewmethod1 ] ) .
it has a solution @xmath329 as it is found in case 1.iii .
we use this solution in ( [ case2iinewmethod2 ] ) and get @xmath330 where @xmath255 is a constant and @xmath331 and @xmath332 .
* iii . * take @xmath333 where @xmath250 . consider the equation ( [ case2iiinewmethod1 ] ) .
it has a solution @xmath334 as it is obtained in case 1.iv .
we use this result in eq.([case2iiinewmethod2 ] ) @xmath335 where @xmath255 is a constant and @xmath336 and @xmath337 * iv . *
take @xmath338 where @xmath250 .
consider the first equation above .
we have @xmath339 where @xmath340 .
let @xmath341 .
hence the above equality becomes @xmath342 which gives @xmath343 hence after some simplifications and taking the integration constant zero we get @xmath344 we use this result in ( [ case2ivnewmethod2 ] ) @xmath345 where @xmath255 is a constant and @xmath346 and @xmath347 * case 3 .
* if ( [ p_5(f ) ] ) has just one real root we can write it in the form @xmath348 where @xmath13 is the zero of the polynomial function @xmath56 of degree five and the constants @xmath349 , @xmath350 are so that @xmath351 for real @xmath3 . * i. * take @xmath353 where @xmath250 .
consider the first equation above .
it has a solution @xmath354 as it is obtained in case 1.iv .
we use this result in ( [ case3inewmethod2 ] ) and get @xmath355 where @xmath255 is a constant . here
we do not get a simpler expression than the original equation .
hence it is meaningless to use the method of factorization of the polynomial for this case .
take @xmath356 where @xmath250 . here for simplicity , we will use the form @xmath357 instead of the polynomial function of degree four above . since @xmath358
, we will take @xmath359 to get an irreducible polynomial . if @xmath360 , then take @xmath361 . from @xmath362 @xmath363
we have @xmath364 where @xmath365 .
then we use this function @xmath3 in ( [ case3iinewmethod2 ] ) and get @xmath366 where @xmath367 . now take @xmath360 with @xmath361 . in this case
we have @xmath368 which gives the solution @xmath369 now use the equation ( [ case3iinewmethod2 ] ) .
we have @xmath370 where @xmath255 is a constant . solving the above equation
gives @xmath371 -b\arctan\big(\frac{2a_1a_3}{a_3 ^ 2-a_1 ^ 2+a_2 ^ 2}\big ) \big\}=\pm \sqrt{\mu}\xi+c_2,\ ] ] where @xmath372 and @xmath373 | traveling wave solutions of degenerate coupled @xmath0-kdv equations are studied . due to symmetry reduction
these equations reduce to one ode , @xmath1 where @xmath2 is a polynomial function of @xmath3 of degree @xmath4 , where @xmath5 in this work . here
@xmath0 is the number of coupled fields .
there is no known method to solve such ordinary differential equations when @xmath5 . for this purpose ,
we introduce two different type of methods to solve the reduced equation and apply these methods to degenerate three - coupled kdv equation .
one of the methods uses the chebyshev s theorem . in this case
we find several solutions some of which may correspond to solitary waves .
the second method is a kind of factorizing the polynomial @xmath2 as a product of lower degree polynomials .
each part of this product is assumed to satisfy different odes .
+ * keywords : * traveling wave solution , degenerate kdv system , chebyshev s theorem , alternative method .
[ section ] [ section ] [ section ] [ section ] [ section ] |
99 heyman p and garcia - molina h 2006 collaborative creation of communal hierarchical taxonomies in social tagging systems _ technical report infolab 2006 - 10 .
department of computer science , stanford university , stanford , ca , usa _
schmitz c , grahl m , hotho a , stumme g , cattuto c , baldassarri a , loreto v and servedio vdp 2007 network properties of folksonomies _ proceedings of the sixteenth international world wide web conference ( www2007 ) _ | we analyze _ citeulike _ , an online collaborative tagging system where users bookmark and annotate scientific papers . such a system can be naturally represented as a tripartite graph whose nodes represent papers , users and tags connected by individual tag assignments .
the semantics of tags is studied here , in order to uncover the hidden relationships between tags .
we find that the clustering coefficient reflects the semantical patterns among tags , providing useful ideas for the designing of more efficient methods of data classification and spam detection .. the recent development of the world wide web is characterized by a growing number of online social communities . in many such cases ,
individuals provide bits of information - about either their tastes , opinions or interests - and software applications gather and organize them into a database , allowing the browsing of the whole information collected so .
a class of such collaborative systems focusses on collecting users online bookmarks with either a general approach or a more specialized one . in particular , some websites have been recently born to store user generated scientific bibliographies . in these systems ,
the elementary contribution , the `` post '' , is made of three ingredients : a user , an article and an annotation of it by a number of tags chosen by users . in exchange for this voluntary contribution
, a user can browse others bibliographies and annotations .
tags are an alternative classification method with respect to traditional taxonomies , where items belong to `` taxa '' represented as a tree like set of categories : here , each category contains in turn a number of more specialized sub categories , and so on until the desired resolution of classification is been reached . instead , in tagging systems items are tagged by users characterized by diverse tagging strategies depending on a number of individual variables .
the set of tag resource relations in such a community is called a `` folksonomy '' .
such communities are now extremely popular , storing hundreds of thousands posts and more .
the tagging system we analyze here , _ citeulike _
@xcite , has been built , at the time of our survey , by ca .
180000 references annotated by ca .
48000 tags supplied by ca .
6000 users .
our dataset includes about 550000 `` tag assignments '' : each assignment is a t - uple ( user , resource , tag ) .
the sequence of chronologically ordered tags , in particular , can be interpreted as a stream of words , to which one can applies the traditional statistical text analysis to uncover how human behavior affects it .
the statistical analysis of word occurrences in a written text has shown that word frequencies are power law distributed according to the zipf s law , according to which a large number of words appears in a text only a few times , while a few words occur orders of magnitude more often @xcite .
such feature has been modeled by many models based on the preferential attachment principle , that is , the assumption that authors employ already used words with a probability proportional to the current word frequency .
moreover , it has been observed that the rate of new words decrease with the text
length @xcite , that is , the number of distinct word @xmath0 in a text of length @xmath1 scale as @xmath2 with @xmath3 .
however , models in literature assume that new words are introduced at a constant growth rate , so that their total number , i.e. the vocabulary , is a linear constant of the total number of words ( both new or repeated ones ) used so far @xcite . yet , to discover the semantical properties of _ citeulike _ , one rather represents it by means of the network formalism , which proved fruitful in the analysis of many natural and social phenomena involving unsupervised interacting units : in a network perspective , elementary interacting agents or objects are represented by nodes , interactions by edges connecting them
. the widespread success of such approach has been triggered by the discovery that many networks instances one encounters in reality share common statistical properties with no external tuning .
for example , the degree @xmath4 , that is , the number of edges pointing to a node , follows in many cases a broad distribution @xmath5 with long tail decaying algebraically as @xmath6 with @xmath7 .
if edges have varying intensities , each of them is attached a weight @xmath8 representing its intensity ; accordingly , a node is characterized by its @xmath9 , equal to the sum of the weights of edges pointing to it .
the distribution of @xmath10 , too , is power
law distributed in many real weighted network instances .
furthermore many such networks exhibit a strong transitivity , i.e. with high probability , the neighbors of a node are themselves connected by an edge , with respect to purely random realization of a network with equal number of nodes and links .
networks sharing the above properties are currently named `` complex networks '' @xcite .
the network approach has also been recently adopted to analyze the semantical structure of tagging systems @xcite .
tags can be represented by networks in different ways , in order to study how the behavior of users maps into the dynamical or topological features .
a more recent stream of research deals with the organization of tags , which are implicitly linked by hierarchical and logical associations emerging despite the diversity of users as their number is large enough .
the underlying semantical organization of tags reveals the dominant trends within a tagging community and allows to improve its navigability .
recently , algorithms have been introduced in order to infer a taxonomy of tags from a folksonomy @xcite .
the statistical properties we observed in the _ citeulike _ data are consistent with the findings obtained in similar surveys , confirming that tags in collaborative systems form complex networks indeed .
moreover , we have investigated how the underlying semantics of tags reflects on the topology of the network . as a matter of facts ,
tags provided by users come with no explicit hierarchy beside the chronological ordering , leading authors to analyze the stream of tags as a text like sequence of words .
interestingly , the time ordered sequence of tags displays statistical properties already observed in written texts , such as the fat tails in the word frequency distributions or the sub - linear vocabulary growth .
our analysis confirms the sub - linear vocabulary growth observed in written texts .
the number of distinct tags @xmath11 introduced by users after @xmath12 assignments grows approximately as @xmath13 , as shown in figure [ sublinear ] although the pace is slightly smaller than in other collaborative tagging systems already surveyed @xcite .
the frequency of tags , too , reported in [ wordsfreq ] , reminds that of words observed in written texts , algebraically decaying according to the zipf s law @xcite .
however , the sole frequency of tags as a function of time does not convey much information about the semantics , although it reflects the different centrality of associated concepts in the underlying knowledge organization . to investigate tag pair relations
, one has to represent the unit elements of a tagging system as nodes of a network .
the dataset we focus on can be naturally represented as a tri partite network , where each node represents either a user @xmath14 , a resource @xmath15 or a tag @xmath12 ; if a tag assignment @xmath16 exists , an edge is drawn from @xmath14 to @xmath15 , and from @xmath15 to @xmath12 . since in a single user s
post a resource can be tagged more than once , one post can correspond to multiple tag assignments @xcite .
although efficient algorithm have been developed to analyzed such tri partite network @xcite , the heterogeneity of nodes discourages in general the application of traditional network methods , mainly conceived to deal with network connections representing peer - to - peer relationships .
thus , to study how tags are organized we chose to project the tri partite networks on the tag space . as a result
, the tag co occurrence network we study is composed by nodes representing tags only , between which an undirected edge of weight @xmath8 is drawn if @xmath8 distinct resources are labeled by both tags .
the resulting network displays some of the typical features of weighted scale free networks .
we have measured the distribution of the sum @xmath17 of the weights of edges pointing to a given node , or the _ strength _ of the node : such distribution @xmath18 , plotted in figure [ strength - dist ] , exhibits a clear power
law decay @xmath19 , with @xmath20 for large values of @xmath17 .
interestingly , the heterogeneity of the observed nodes weights does not necessarily reflects the centrality of corresponding concepts , that are supposed to be assigned together with a wider range of more specialized concepts , in the underlying hierarchy of tags .
as it has been already shown @xcite , reshuffling the tag assignment in order to destroy the logical association among words does not change dramatically the shape of @xmath18 , which proves that the @xmath21 heterogeneity is more a consequence of frequency distribution broadness than of the varying roles of concepts in the semantical organization of the whole vocabulary .
nevertheless , the tag co occurrence network unveils some semantical feature of the underlying ontology if , instead of focussing on the properties of single nodes , one turns to the inspection of quantities involving its environment .
an example of such is represented by the analysis of the neighbor average degree @xmath22 of nodes with degree @xmath4 , where the degree is the number of incoming edges of a node . in our study
we examined instead the clustering properties of the tag co occurrence network through the clustering coefficient .
such coefficient @xmath23 counts the average density of triangles involving nodes with degree @xmath4 or , in other words , the probability that the nearest neighbors of a node with degree @xmath4 are in turn connected one to each other .
this reads @xmath24 where @xmath25 is 1 if a link exists between @xmath26 and @xmath27 and 0 otherwise , and @xmath28 is the frequency of nodes with degree @xmath4 .
this quantity has been found to characterize most complex networks found in nature and society , where it takes substantially larger values with respect to a purely random networks @xcite .
the properties of the clustering coefficient are often associated to the hierarchical organization of nodes @xcite .
indeed , the clustering coefficient appears to encode a signature of semantical relations between words . as represented in figure [ clustering - real ]
, the clustering coefficient @xmath23 in _ citeulike _ decays algebraically for large values of the degree @xmath4 , according to @xmath29 .
however , the clustering value displays an apparent fluctuation at @xmath30 . by inspecting the nodes corresponding to such value
, one discovers that the sharp rise taking place at al @xmath30 corresponds to a non existing resource labelled by @xmath31 distinct uncorrelated randomly chosen tags , which mimics the a spam contribution to the collaborative systems .
one is led thus to conjecture that the overall semantical organization of concept represented by tags is encoded in a characteristic behavior of the clustering coefficient @xmath23 , so that tags assigned in a semantically inconsistent way fall far away from this behavior . to verify such conjecture ,
we have performed the same statistical analysis after removing from the data set the tag assignments related to the spam like page . as shown in figure
[ clustering - real ] , after the removal the clustering coefficient follows a more regular behavior , confirming that the strong fluctuation observed above was due indeed to the presence of a single meaningless set of assignments involving a single resource .
tags assigned only to the spam resource form a complete co occurrence network , so that their clustering coefficient is equal to @xmath32 .
thus , the behavior of clustering coefficient of the tag co occurrence networks can be used as a test for models representing the tag semantical organization or , equivalently , how users choose tags when annotating a resource . as noted in literature @xcite
, users typically use tags hierarchically , labelling a resource by tags related to the same topics but with different generality , adding more specialized tags as the number of collected resources grows . on a very basic level , we have tested how such hierarchical tagging , affects the topology of the tag co occurrence network by a simple toy model defined in the following .
let us assume that tags are organized on a taxonomy , that is , a tree like structure stemming from a seed node , where each node corresponds to a tag and is an offspring of another tag belonging to the same branch of knowledge with higher generality . at discrete time steps
, a new post is added to the system , with a new resource and 2 tags .
the first tag can be a new one , with probability @xmath33 : in such case , the new tag is an offspring of a tag randomly chosen among the already employed ones .
otherwise , the first tag is chosen at random among the already employed ones .
the second tag is either chosen at random from within the whole set of used tags or , with probability @xmath34 , it is chosen according to hierarchy : in such case , the second tag is drawn at random among the nodes that lie on the shortest path length from the first tag to the seed node on the tree like taxonomy .
the tag co occurrence network resulting from the above algorithm share some features of the _ citeulike _ one , if we assume a time dependent @xmath33 which reproduces the sub - linear vocabulary growth observed in reality , and by a suitable choice of the parameter @xmath34 , which mimics the relevance of hierarchy in tagging activity . to reproduce the growth rule ,
we have set @xmath35 and imposed that @xmath36 and @xmath37 where the number of resources @xmath38 , the number of tag assignments @xmath39 , the number of tags @xmath40 and @xmath41 are set to the same values they take in _
citeulike_. as a result , this yields @xmath42 and @xmath43
. as shown in figure [ strength - dist ] , the strength distribution @xmath18 of tags is a scale free one with a good agreement with reality in the decaying exponent for large values of @xmath17 if one sets @xmath44 . for such choice of the parameter , the clustering coefficient reproduces qualitatively the algebraic decay observed in _
citeulike _ , as shown in figure [ summarymodel ] , although the absolute value differs of orders of magnitude .
we have found , thus , a simple model that captures the complex features of a tag co occurrence network issued from the dataset describing an online collaborative tagging system , _
citeulike_. in particular , by assuming that users label resources by hierarchically associated tags , the probability distribution of nodes strength is reproduced for a suitable choice of the parameters .
moreover , the model reproduces qualitatively the decaying asymptotic behavior of the clustering coefficient @xmath23 .
such quantity encodes a signature of the semantical organization of concepts represented by tags , so that malicious or meaningless tag assignments can be detected by inspecting the perturbation to the clustering coefficient they generate . establishing a relationship between clustering and semantics may suggest tools and algorithms for technological tasks such as automatic categorization of resources , recommendation and spam detection techniques .
the authors acknowledge useful discussions with francesca colaiori , stefano leonardi , ciro cattuto , vito d.p .
servedio and andrea baldassarri .
the authors acknowledge the european project delis for support and r. cameron for providing the data . as a function of time @xmath12 , where time is measure in chronologically ordered tags assignments ( plus symbols ) .
solid line represents @xmath45 for a comparison.,scaledwidth=80.0% ] of tag frequencies @xmath46 ( plus symbols ) .
solid line represents @xmath47 for a comparison.,scaledwidth=80.0% ] of the node strengths @xmath17 in the tag co occurrence network .
the best fitting power law exponent , represented by the solid curve , yields @xmath48 .
, scaledwidth=80.0% ] of the tag co occurrence network as a function of the nodes degree @xmath4 before ( circles ) and after ( plus symbols ) the removal of a spam post from the dataset .
the solid line represents a decay @xmath49 and the dashed vertical ruler is set at @xmath30.,scaledwidth=80.0% ] ( plus symbols ) ; the solid line represents the decay @xmath50 for a comparison with real data .
inset : the clustering coefficient in the co occurrence network derived from the model with @xmath51 ( plus symbols ) .
the solid line represents the decay @xmath49 for a comparison with real data.,scaledwidth=80.0% ] |
let @xmath2 be the set of all square matrices of order @xmath3 with entries in @xmath4 and @xmath5 be the group of all invertible matrices of @xmath2 .
a map @xmath6 is called an affine map if there exist @xmath7 and @xmath8 such that @xmath9 , @xmath10 .
we denote @xmath11 , we call @xmath12 the _ linear part _ of @xmath13 . the map @xmath13 is invertible if @xmath14 .
denote by @xmath15 the vector space of all affine maps on @xmath1 and @xmath16 the group of all invertible affine maps of @xmath17 .
let @xmath0 be an abelian affine sub - semigroup of @xmath18 . for a vector @xmath19
, we consider the orbit of @xmath0 through @xmath20 : @xmath21 .
denote by @xmath22 the closure of a subset @xmath23 .
the semigroup @xmath0 is called _ hypercyclic _ if there exists a vector @xmath24 such that @xmath25 . for an account of results and bibliography on hypercyclicity , we refer to the book @xcite by bayart and matheron .
we refer the reader to the recent book @xcite and @xcite for a thorough account on hypercyclicity .
costakis and manoussos in @xcite localize the concept of hypercyclicity using j - sets . by analogy
, we generalize this notion to affine case as follow : suppose that @xmath0 is generated by @xmath26 affines maps @xmath27 @xmath28 then for @xmath29 , we define the extended limit set j@xmath30 to be the set of @xmath31 for which there exists a sequence of vectors @xmath32 with @xmath33 and sequences of non - negative integers @xmath34 for @xmath35 with @xmath36 such that @xmath37 note that condition ( 1.1 ) is equivalent to having at least one of the sequences @xmath38 for @xmath39 containing a strictly increasing subsequence tending to @xmath40 .
we say that @xmath0 is _ locally hypercyclic _ if there exists a vector @xmath41 such that j@xmath42 .
so , the question to investigate is the following : when an abelian sub - semigroup of @xmath17 can be hypercyclic ?
the main purpose of this paper is twofold : firstly , we give a general characterization of the above question for any abelian _ sub - semigroup _ of @xmath17 using j - sets .
secondly , we generalize the results proved in @xcite , by a.ayadi and h.marzougi , for linear semigroups , which answers negatively , the question raised in the paper of costakis and manoussos @xcite : is it true that a locally hypercyclic abelian semigroup @xmath43 generated by matrices @xmath44 is hypercyclic whenever j@xmath45 for a finite set of @xmath10 whose vector space is equal @xmath1 ?
similarly for @xmath46 .
denote by @xmath47 the canonical basis of @xmath1 .
let @xmath48 .
let introduce the following notations and definitions .
denote by : + let @xmath49 be fixed , denote by : + @xmath50 , @xmath51 and @xmath52 .
+ @xmath53 the canonical basis of @xmath54 and @xmath55 the identity matrix of @xmath56 . + for each @xmath57 , denote by : + @xmath58 the set of matrices over @xmath4 of the form @xmath59 + @xmath60 the group of matrices of the form ( [ eq1 ] ) with @xmath61 .
+ let @xmath62 and @xmath63 such that @xmath64 in particular , @xmath65 .
write + @xmath66 in particular if @xmath67 , then @xmath68 and @xmath69
. + @xmath70 .
+ @xmath71 where @xmath72 , for @xmath73 .
so @xmath74 .
+ @xmath75 the second projection defined by @xmath76 .
+ define the map @xmath77 + @xmath78 we have the following composition formula @xmath79 then @xmath80 is an injective homomorphism of groups .
write + @xmath81 , it is an abelian subgroup of @xmath82 . + let consider the normal form of @xmath83 : by proposition [ p:2 ] , there exists a @xmath84 and a partition @xmath85 of @xmath86 such that @xmath87 .
for such a choice of matrix @xmath88 , we let + @xmath89 .
so @xmath90 , since @xmath91 .
+ @xmath92 .
we have @xmath93 .
+ @xmath94 .
+ denote by @xmath95 , @xmath96 and @xmath97 , then @xmath98 for such choise of the matrix @xmath99 , we can write @xmath100,\ \ \mathrm{with}\ \ \
q\in gl(n , { \mathbb{c}}).\ ] ] we have @xmath101 , @xmath102 and @xmath103 .
we have @xmath104 is an abelian semigroup of @xmath105 .
our principal results are the following : [ t:1 ] let @xmath0 be a finitely generated abelian semigroup of affine maps on @xmath1 . if @xmath106 for some @xmath107 then @xmath108 .
[ c:1 ] under the hypothesis of theorem [ t:1 ] , the following are equivalent : * @xmath0 is hypercyclic . *
* @xmath110 .
[ c:2 ] under the hypothesis of theorem [ t:1 ] , set @xmath111 .
if @xmath0 is not hypercyclic then @xmath112 , ( @xmath113 ) where @xmath114 are @xmath0-invariant affine subspaces of @xmath1 with dimension @xmath115 .
[ p:2 ] let @xmath0 be an abelian sub - semigroup of @xmath116 and @xmath81 .
then there exists @xmath117 such that @xmath118 is a sub - semigroup of @xmath119 , for some @xmath65 and @xmath120 .
[ p:002]@xmath121@xcite , proposition 2.1@xmath122 let @xmath0 be an abelian subgroup of @xmath123 and @xmath81 .
then there exists @xmath117 such that @xmath118 is a subgroup of @xmath124 , for some @xmath65 and @xmath120 .
suppose first , @xmath125 .
let @xmath126 be the group generated by @xmath83 .
then @xmath126 is abelian and by proposition [ p:002 ] , there exists a @xmath117 such that @xmath127 is an abelian subgroup of @xmath128 , for some @xmath129 and @xmath130 .
in particular , @xmath131 .
suppose now , @xmath132 .
for every @xmath133 , there exists @xmath134 such that @xmath135 ( one can take @xmath136 non eigenvalue of @xmath12 ) . write @xmath137 be the group generated by @xmath138 .
then @xmath137 is an abelian subsemigroup of @xmath82 .
hence by above , there exists a @xmath139 such that @xmath140 , for some @xmath130 . as @xmath141 then @xmath142 .
this proves the proposition .
let @xmath143 be the semigroup generated by @xmath83 and @xmath144 .
then @xmath143 is an abelian sub - semigroup of @xmath145 . by proposition [ p:2 ]
, there exists @xmath91 such that @xmath118 is a sub - semigroup of @xmath146 for some @xmath65 and @xmath120 and this also implies that @xmath147 is a sub - semigroup of @xmath146 .
+ + @xmath152 is obvious since @xmath153 by construction .
+ @xmath154 let @xmath31 and @xmath155 be a sequence in @xmath143 such that @xmath156 .
one can write @xmath157 , with @xmath158 and @xmath159 , thus @xmath160 , so @xmath161 .
therefore , @xmath162 .
hence , @xmath163 .
+ @xmath164 since @xmath165 and for every @xmath166 , @xmath167 , we get @xmath168 hence @xmath169
. therefore , @xmath178 and sequences of non - negative integers @xmath34 for @xmath39 such that @xmath179 hence @xmath180 .
+ @xmath181 since @xmath182 , then there exists a sequence of vectors @xmath183 with @xmath184 and sequences of non - negative integers @xmath34 for @xmath185 with @xmath186 such that @xmath187 suppose that @xmath106 with @xmath107 . by proposition [ p:2 ] , we can assume that @xmath194 and @xmath195 .
by lemma [ l:1 ] , we have @xmath196 . then by theorem [ t:5 ] , @xmath197 .
it follows by lemma [ ll1l:9 ] , that @xmath198 . with @xmath211^{t},\ x_{k}\in\{0\}\times\mathbb{c}^{n_{k}-1},\ x_{i}\in\mathbb{c}^{n_{i } } , \
\mathrm{if}\ i\neq k\right\}$ ] .
it follows that @xmath212 with @xmath213 is an affine space with dimension @xmath115 . | we give a characterization of hypercyclic using ( locally hypercyclic ) of semigroup @xmath0 of affine maps of @xmath1 .
we prove the existence of a @xmath0-invariant open subset of @xmath1 in which any locally hypercyclic orbit is dense in @xmath1 . |
the bulge is one of the major components of the milky way .
its integrated properties resemble those of giant ellipticals , which in turn are one of the dominant components of the universe .
however , in contrast to these distant environments , the relative proximity of the galactic bulge allows us to study individual stars in detail , and ( at least in some cases ) derive accurate abundances for the local version of old metal rich populations .
this capability is very helpful in reconstructing the history of formation , still quite controversial , of these very important constituents of the universe .
unfortunately , owing to the rather large distance and to substantial interstellar absorption characteristic of the bulge , without a significant boost from microlensing , detailed abundance analyses of bulge stars have to be limited to the evolved population .
an exploratory study of field bulge stars was carried out by mcwilliam & rich ( 1994 ) ; coupled with other studies at lower resolution ( rich 1988 ) , this study indicated that the bulge has a wide abundance distribution , peaking at a metal abundance ( as represented by [ fe / h ] ) slightly lower than solar , and may have an excess of @xmath7elements .
the latter would suggest that most of the bulge formed within a rather short interval of time , due to the interplay between star formation and the lifetime of the progenitors responsible for the synthesis of different elements .
while this analysis is of great value , the interpretation of the results is complicated by the fact that we are not able to determine the ages of the stars observed .
thus , for example , from these data alone , we are unable to deduce any strong constraint on the epoch of formation of the bulge .
this concern may at least in principle be overcome by studying stars in clusters .
the bulge of our galaxy has a rich population of globular clusters ( ortolani 1999 ) .
the colour - magnitude diagrams ( cmds ) of these clusters indicate that several of them are metal - rich , ( quite similar to the bulk of the field bulge population ) , and they are likely old ( ortolani _ et al . _
1995 ) . however , accurate metal abundances are needed to determine ages with the precision required to understand if the bulge is as old as ( or even older than ) the halo , or if instead it has the somewhat younger age of the oldest stars in the thin disk .
thus , accurate abundance determinations for bulge globular clusters represent a basic step in our understanding the formation of the milky way .
such an analysis also provides additional important pieces of information , since it allows us to : ( i ) extend the calibration of abundance scales at high metal abundance , poorly known at present since metal - rich clusters tend to be concentrated toward the galactic center , where reddening and crowding often hamper accurate observations ; ( ii ) derive quite accurate reddening estimates by comparing the observed colours with those expected for stars having temperatures determined from spectroscopically derived parameters , such as line excitation , which are independent of reddening : and ( iii ) further constrain the evolution of the bulge by determining the ratios of the abundances of @xmath7elements to iron ( crucial to determining the rate of chemical evolution ) . in order to address these questions , in a previous study ( cohen _ et al .
_ 1999 , henceforth paper 1 ) we used the keck telescope to acquire high resolution spectra of individual stars in ngc 6553 .
the large aperture of the keck telescope and the efficiency of its high resolution spectrograph ( hires , vogt _
et al . _
1994 ) allowed us to observe stars on the red horizontal branch ( rhb ) .
as explained in paper i , this choice is clearly advantageous as compared to observations of first ascent red giants , since rhb stars are warmer , making an abundance analysis of their spectra much easier .
furthermore , contamination of the rhb region of the cmd diagram by field stars is much less important than in the case of red giants , resulting in a higher probability of membership ( as confirmed a posteriori by the measured radial velocities ) . in paper
i we found that ngc 6553 has a metallicity of [ fe / h]=@xmath8 , somewhat higher than determined by analysis of two cool giants by barbuy _
et al . _ ( 1999 , henceforth b99 ) , and very similar to the bulk of bulge field giants observed by mcwilliam & rich ( 1994 ) .
also , we found an excess of the @xmath7elements similar to that found for the field stars in the bulge . in the present paper
we perform for the first time a similar study on ngc 6528 , a globular cluster having values of the metallicity indicators very similar to those of ngc 6553 ( harris 1996 ) .
ngc 6528 is a very highly concentrated cluster , between us and the galactic bulge , at about 7.8 kpc from the sun , hence very close to the galactic center ( e.g. ortolani _ et al . _ 1995 ) .
located in baade s window with @xmath9 deg ) , it seems certain that ngc 6528 is a bulge cluster .
moreover , the cluster velocity indicates that it is not a disk cluster .
et al . _ ( 1995 ) found from careful comparison of both cmds and luminosity functions that its stellar population is very similar to that of ngc 6553 and of baade s window .
in addition , recently davidge ( 1999 ) , noting that stars in ngc 6528 populate the same region in the two - colour plane ( j - h , h - k ) of the field bulge giants , supports the classification of this cluster as a true bulge cluster , rather than a very metal - rich disk cluster .
the metallicity of ngc 6528 is uncertain .
zinn & west ( 1984 ) derived an abundance of + 0.2 dex , but armandroff & zinn ( 1988 ) , using an analysis of the ca ir triplet in the integrated cluster spectrum , decreased that value significantly , to @xmath10 dex .
other methods , as that of sarajedini ( 1994 ) , can not be applied due to the lack of calibrating clusters in the high metallicity regime .
ngc 6553 and ngc 6528 are the only very high metallicity globular clusters that can be studied rather easily at optical wavelengths ; the remaining bulge clusters are either more metal - poor or very obscured .
hence the choice of ngc 6528 for the present study was obvious .
while the analysis presented here is very similar to that already done on ngc 6553 , it is based on higher s / n spectra ; moreover , we adjusted the instrumental set up in order to include lines of na as well as o. this was deemed important because observations of o and na in very metal - rich clusters may help to better understand the pattern of abundances of these elements found in more metal - poor globular clusters ( ivans _ et al . _ 1999
; kraft _ et al . _
in fact , serious doubts were cast on the up - to - now most favoured mechanism ( deep mixing ) by gratton _
et al . _ ( 2001 ) , who found that the o - na anticorrelation ( previously seen only for stars on the red giant branch of globular cluster ) is clearly present among dwarf , unevolved stars in ngc 6752 .
on the other hand , the alternative explanation suggested ( pollution due to mass lost by previous agb stars in the cluster ) is expected to be sensitive to the overall metal abundance of the cluster ( ventura _ et al . _
determination of o abundances in ngc 6528 rhb stars is favoured by the large radial velocity of this cluster , which shifts stellar lines away from telluric lines .
program stars were selected using high resolution images and vijk photometry of ngc 6528 from hst ( vi ) and irac2 ( jk ) kindly provided by montegriffo ( 1999 , private communication ) .
table 1 gives the most relevant parameters of these stars , while figure 1 shows their location within the cluster and figure 2 the position of these objects in the @xmath11 color - magnitude diagram of ngc 6528 . in figure 1 , the stars studied here are marked on this subset from a 100 sec wfpc2 image from the hst archive .
coordinates for the program stars are given in table 1 .
the field of the cluster is very crowded .
we selected for observation a number of apparently uncrowded stars having colours and magnitudes appropriate for the rhb in the cmd , in order to maximize the probability of cluster membership . beside the 4 program stars ( namely , 5422 , 3014 , 3025 and 3046 ) two other stars fell in the edge of the slit ( stars 3032 and 5425 ) on at least some of the spectra . while not good enough for abundance analysis , spectra of these extra stars are adequate for precise radial velocities . _ a posteriori _
, membership of the program stars in ngc 6528 was confirmed by their radial velocity : this is a very useful criterion in this case , due to the large @xmath12 of this cluster ; according to harris ( 1996 ) , the heliocentric @xmath12 is @xmath13 ( internal error ) km s@xmath14 , and its @xmath12 relative to the solar local standard of rest is 195 km s@xmath14 .
heliocentric @xmath12 for the program stars as measured on our spectra are given in the column 6 of table 1 : the average heliocentric @xmath12 from our spectra is @xmath15 km s@xmath14 with @xmath16 km s@xmath14 from 6 stars , giving double weight to velocity from the june 2000 spectra that have a higher s / n ratio .
the radial velocities were measured by cross correlating the region 6130 - 6170 in echelle order 57 and the region 6240 - 6280 in echelle order 58 , using the june 2000 spectrum of star 3025 as a template .
these measurements were carried out independently and with different software packages by ec and by jc .
the zero point was determined by fitting gaussians to 16 lines in these orders in the spectrum of the template judged to be unblended based on their fwhm .
the laboratory wavelengths of these lines were taken from the nist atomic spectra database .
( nist standard reference database no .
the agreement between orders is excellent and confirms the dispersion solution from the th - a lamp .
the internal radial velocity errors were calculated following the method of tonry & davis ( 1979 ) using the relation @xmath17 $ ] , where the parameter @xmath18 is a measure of the ratio of the height of the peak of the cross correlation to the noise in the cross correlation function away from the peak .
the constant @xmath4 was set at 15 km s@xmath14 , which represents a value typical of those found in other recent hires programs using similar instrumental configurations by mateo _
et al . _ ( 1998 ) , cook _ et al . _ ( 1999 ) , and ct _ et al . _ ( 1999 ) .
the maximum error in @xmath12 incurred by a point source which is on the edge of the 1 arcsec slit compared to an object that uniformly fills the slit is 4 km s@xmath14 , and this factor will be smaller for real stars under real seeing conditions that are partially in the slit .
the systematic errors are thus generously set at 1.5 km s@xmath14 for the program stars observed in 2000 , and 2.5 km s@xmath14 for the stars observed in 1999 as well as for those that fortuitously appeared in the slits .
our velocity dispersion measured for ngc 6528 is by far the smallest published for this cluster , and is consistent with a normal mass - to - light ratio for ngc 6528 .
however , our mean @xmath12 is distinctly higher that the literature value in the compilation by harris .
a search of the references quoted by harris revealed that all low values of @xmath12 for ngc 6528 are from older studies , while the most recent works tend to give a value quite similar to our own .
for example , minniti ( 1995 ) found an average @xmath12 of @xmath19 km s@xmath14 from 7 stars , noting also that @xmath12 of this cluster was found to have some discrepancy in previous analyses ( e.g. in armandroff & zinn 1988 ) .
the recent extensive work by rutledge _
_ ( 1997a ) found @xmath12 = @xmath20 km s@xmath14 with @xmath21 km s@xmath14(external error ) based on 8 stars .
both these results are in very good agreement with our @xmath12 .
we note that for ngc 6553 , rutledge _ et al .
_ ( 1997a ) found @xmath12 = 8.4 km s@xmath14 ( @xmath22 km s@xmath14 ) , in quite good agreement with the value of paper i. moreover , rutledge et al .
( 1997a ) already noted that their high value for ngc 6528 was different at the 3.4@xmath23 level from the mean value in harris compilation and expressed concern about the potential impact on measurements of @xmath12 of possible non - members in clusters so heavily contaminated by field stars .
this discussion further supports our adopted strategy of selecting target objects among the rhb stars .
the confirmation of this high @xmath12 is important as it is among the largest in absolute value for a bulge cluster so close to the galactic center .
the radial velocity of ngc 6528 ( @xmath24 km / s ) might be a reflection of the velocity dispersion of the metal rich bulge globular cluster population ( ct 1999 ) although it appears as rather extreme for a bulge object : in fact this value is slightly less than 2 @xmath23 ( of the scatter of values for individual field stars ) from the mean value for bulge k - giants ( terndrup _ et al . _
1995 ) , m - giants ( sharples _ et al . _ 1990
) , rr lyrae s ( gratton _ et al . _
1987 ) , and miras ( feast _ et al . _ 1980
) . while this large radial velocity clearly rules out the possibility that ngc6528 is a disk cluster , there is the possibility that it belongs to the inner halo rather than to the bulge and its orbit is by chance just passing through the bulge at the present time .
the distinction between inner halo and bulge is not clearly defined at present . to obtain better insight into this issue
, we examined more closely the large sample of bulge k - giants located in baade s window , as is ngc 6528 , observed by terndrup _
et al._. we will consider only those stars fainter than @xmath25 ( contamination by disk interlopers being important for brighter objects ) . for these stars , terndrup _ et al .
_ found an average velocity of @xmath26 km / s , with a dispersion of @xmath27 for individual objects .
20 out 334 stars ( that is , 6% ) have radial velocities in excess of 200 km / s ( in absolute value ) , i.e. have kinematics similar to or more extreme than that of ngc 6528 .
the average metallicity for these stars ( sadler _ et al . _
1996 ) is [ fe / h]=@xmath28 , with a scatter for individual stars of 0.53 dex .
the average value is not significantly different from the average metal abundance they found for the whole sample of bulge k - giants ( [ fe / h]=@xmath29 ) .
these stars are clearly much more metal - rich than the traditional value for halo stars ( [ fe / h]@xmath30 ) , and have a metal abundance quite similar to that we found for ngc6528 .
note that these considerations are based on the high radial velocity of ngc 6528 , which makes its membership in the disk population unlikely . on the other hand ,
the low radial velocity of ngc 6553 is compatible with that of a disk object .
some support for this view is given by a recent study by beaulieu _
et al._(2001 ) .
their color - magnitude diagram for ngc 6553 does not show a good match to the field bulge population and favors a metallicity comparable to solar , in good agreement with the revised value that we suggest in the present work ( see below ) .
observations were carried out with the hires spectrograph at keck i. the hires detector is not large enough to yield full spectral coverage in a single exposure .
the rhb stars in ngc 6528 are faint , and hence the exposures are long , so a single compromise instrumental configuration is mandatory . in our first effort described in paper i
, we wished to avoid crowding of lines , and hence centered the spectra rather far toward the red .
it turned out that , ignoring fe i absorption lines , there were few useful features beyond 8000 and also that line crowding was tolerable even at the blue end of the hires spectra of the rhb stars in ngc 6553 .
thus for the observations of the ngc 6528 rhb stars , the instrumental configuration was set to shift the spectra blueward , so that the o triplet at 7770 appears in the reddest order included .
this configuration had the advantage of adding features of several important elements with no lines in the wavelength regime covered in paper i , such as na , as well as important additional fe ii lines to improve the analysis of ionization equilibrium .
however , the downside of this layout was that the unblended mg i line at 8717 was lost , while the mg i lines included in this setup are more blended and more saturated .
these fields are rather crowded .
given the freedom to rotate the slit to a desired position angle and the ability to track at a fixed position angle , in many cases it might be possible to get two or more candidate rhb stars within the length of the hires slit .
at the time of our initial observations for ngc 6553 , hires did not have an image rotator to compensate for the rotation of the field at the nasmyth focus , and hence only one star could be observed at a time .
however , by 1999 , the hires image rotator had been completed by david tytler and the lick observatory engineering group ( tytler 2000 , private communication ) . the maximum slit length that can be used with our instrumental configuration without overlapping echelle orders is 14 arcsec . leaving room at the ends of the slit for sky
, we therefore searched the list of rhb candidates for suitable pairs of rhb candidates located not more than 7 arcsec apart .
the observations were carried out in two runs 10 months apart : \a ) run of august 1999 . due to a hardware problem , the lower dome shutter of the keck i telescope was not movable , and at low elevations it vignetted the telescope .
this forced us to observe only one object in ngc 6528 per night , and that only during the hour centered on culmination of the cluster .
exposures were 1200 sec , and 3 such were obtained for star 3014 and star 3025 , while only two were obtained from 3046 and for 5422 .
the stars were dithered by one or two arcsec along the length of the slit between exposures .
\b ) run of june 2000 .
two exposures were obtained in ngc 6528 on the nights of june 2 and june 3 , 2000 ( i.e. 1 each night ) .
each was 7200 sec long , in 1200 sec segments , with small spatial dithers along the slit between each segment .
each exposure contained the rhb star 3025 and a second star , either 3014 or 5422 .
thus the total exposure for star 3025 was 14,400 sec .
the hires setup was identical for these two exposures .
the s / n of these spectra , based strictly on the count rate in the continuum near the center of the echelle orders , is 65/pixel ( 130/4 pixel resolution element ) for star 3025 and 95/4 pixel resolution element for the other two stars in ngc 6528 .
the spectra from both runs were reduced using the suite of routines for analyzing echelle spectra written by mccarthy ( 1988 ) within the figaro image processing package ( shortridge 1988 ) .
the stellar data are flat fielded with quartz lamp spectra , thereby removing most of the blaze profile , and the results are normalized to unity by fitting a 10th - order polynomial to line - free regions of the spectrum in each order .
the spectra of the three stars observed in june , 2000 ( stars 3014 , 3025 and 5422 ) are of high s / n , and equivalent widths were measured directly from them ; the august 1999 spectra were not used in the abundance analysis for these objects . however , star 3046 was only observed in 1999 .
the s / n of the 1999 spectra is not very large ( typically @xmath31 ) . for this star
only , we followed the same procedure successfully used in paper i to improve the reliability of measures of equivalent widths .
we filtered the spectra by convolving them with a gaussian having a fwhm of 0.3 ( this reduced the resolution of the spectra down to @xmath32 but enhanced the s / n per pixel to about 60 ) .
we also applied the above procedure to the spectra of the 3 other stars of the 1999 run in order to derive a linear relationship between the two sets of measured ews , and thus to correct the 1999 ews of star 3046 measured using convolved spectra to that of the higher precision june , 2000 spectra : @xmath33 with the correlation coefficient @xmath34 and @xmath23=19.7 from 361 lines .
since the spectra of the program stars are very line rich , the following procedure was adopted to measure reliable equivalent widths ( ews ) on these spectra .
first , we corrected the continuum slightly by using a spline interpolating function : appropriate `` continuum '' points were selected by comparing the spectra of different stars , so that the fiducial continuum level was set consistently for each star .
this was done for all program stars . however , for star 3046 , whose spectra are at somewhat lower resolution , greater care is needed .
we then selected a small number of clean lines , and used these lines to set a fiducial relation between the fwhm of lines and their ews .
ews for a larger number of lines were then measured by using a special fitting routine that measures ews using this relation between fwhm and ews , shown in figure 3 . with this scheme ,
the fwhm , which is very sensitive to the presence of blending lines , was constrained when measuring the ews of individual lines , resulting in much more stable measures .
note that for each line , the fitting routine used an interactively selected portion of the line profile , so that regions of the profile obviously disturbed by blending lines were not considered . in this way
, we are not using just the central pixel , but a portion of the line profile that is generally broader than the fhwm of the line ( by itself broader than a resolution element ) , and we are fitting the central part of the line adopting a model of the line profile which is a gaussian , whose fwhm is a ( linear ) function of the ew . as also noted by the referee , it is however possible that ews are somewhat overestimated . on the other side , since the continuum is likely to be somewhat underestimated , if the two effects are not mutually compensating , we should notice some trends of abundances as a function of the wavelength , and this is not the case .
apart from errors due to the continuum placement and to blends , random errors @xmath35ew in these ews are : @xmath35ew = @xmath36 , where n is the number of points used in the fitting , and s / n is per pixel . in our case n @xmath37 (
actually , somewhat larger , in general ) , where dx is the wavelength step ; that is , we may write : @xmath35ew = @xmath38 where now s / n * is the s / n per fwhm element ( generally larger than a resolution element ) .
table 2 lists the final values of the ews , along with the adopted @xmath39 values .
errors in the ews measured by this procedure may be estimated by comparing the ews of different stars , since all four rhb stars have very similar atmospheric parameters ( see below ) . through such comparisons ,
we obtain typical rms scatters about the linear relationship between two stars of 9 m .
if we assume that both sets of ews have equal errors , we can estimate that typical errors in ews are 6 m from the better quality spectra of june , 2000 .
these errors are mainly due to uncertainties in the positioning of the fiducial continuum .
errors in ews measured on previous 1999 spectra ( i.e. namely for star 3046 ) are somewhat larger , between 10 and 12 m .
the analysis of the cmd of ngc 6528 ( ortolani _ et al . _ 1992 ) revealed that the interstellar reddening toward this cluster is quite large , somewhat uncertain , and with strong variations even within small projected distances on the sky .
infrared photometry by cohen & sleeper ( 1995 ) confirmed the presence of substantial reddening variations within this cluster .
for this reason , it is not possible to derive accurate atmospheric parameters from the observed colours of stars in ngc 6528 . following the approach adopted in paper i ( where we had a similar problem ) , effective temperatures for the program stars were derived directly from the spectra , by forcing fe i lines of different excitation to provide the same abundance ( typically 90 - 100 fe i lines were measured for each star , 70 for star 3046 , with lower quality spectra ) .
the errors on the linear regression fits allow one to estimate the ( internal ) errors in these temperatures : they are @xmath40 k ( corresponding to 1@xmath23 rms uncertainty of 0.017 dex / ev in the slope ) . as an example , left panels of figure 4 shows the run of abundances from individual fe i lines with excitation potential for the rhb stars .
systematic errors mainly depend on the set of model atmospheres used in the analysis ( kurucz , 1992 : models with overshooting , for consistency with the analysis of carretta & gratton 1997 ) .
they are not easy to estimate , but we feel they are about @xmath41 k. observation at slightly shorter wavelengths than done in paper i allowed a larger number of fe ii lines to be measured ( typically at least 4 good fe ii lines for each star ) : we were then able to estimate surface gravities @xmath42 from the equilibrium of ionization of fe .
internal errors in these surface gravities are @xmath43 dex ( where we considered both the contributions due to errors in ews of individual lines , and those due to uncertainties in the adopted @xmath44 ) . again
systematic errors are mainly related to the adopted model atmospheres , and to the assumption of lte made throughout this paper .
( this seems a solid assumption for the stars we are currently considering : see gratton _ et al . _
the average gravity we derived ( @xmath45 ) is very close to that predicted by evolutionary models for rhb stars in such a metal - rich cluster ( about 2.3 dex , from the latest padova isochrones , salasnich _ et al . _ 2000 ) , supporting the temperature scale adopted in the present analysis .
microturbulent velocities @xmath46 were derived by eliminating any trend in the derived abundances from fe i lines with expected equivalent widths for the lines ( following the approach of magain 1984 ) . given the large number of fe i lines measured , internal error bars in these @xmath46 s are small ( @xmath47 km / s ) .
as an example , the central panels of figure 4 show the run of abundances from individual fe i lines with ews for the four program stars .
of course , it is well known that in abundance analyses the derived values of microturbulent velocity depend on the gf s one is using .
our gf s are obtained by combining mainly two sources ( see carretta & gratton 1997 , henceforth cg97 , for detailed references ) : ( i ) for all strong lines , and a few of the weak ones , we are using laboratory gf s from a compilation of data from the literature . only gf s with errors
@xmath480.05 dex were considered ; ( ii ) for the vast majority of weak lines ( most of the lines measurable in the spectra of ngc 6528 stars ) , we use solar gf s , obtained from an inverse analysis whose zero point is set by the lines having laboratory gf s ( and whose abundance generally agree well with meteoritic values ) . as noted also by the referee
, there is some correlation between line strength and excitation potential ( @xmath49 ) , in the sense that there is a generic tendency of low - excitation fe i lines to be stronger than those of high excitation .
this is shown by the curves - of - growth in the right - handed panels of figure 4 , where open squares are used for lines with @xmath50 ev , while filled circles are for lines with @xmath51 ev . to test the relevance of this problem in our analysis , we repeated it , considering only those fe i lines having @xmath52 ev ; we did not changed the temperature since the excitation range included is now too small to allow fitting of @xmath44 , but we re - optimized the microturbulent velocity .
when only lines with @xmath53 ev are considered , we are essentially using only solar gf s ; these lines are weak in the solar spectrum , so that these gf s are virtually independent of details in the solar analysis .
the @xmath46 values we derived from this subset of lines are smaller than the original ones by @xmath54 km / s , and the fe abundances are larger by @xmath55 dex .
both these values are barely significant and much lower than the other sources of errors .
the final adopted atmospheric parameters are shown in table 3 .
an analysis of the influence of errors on the derived abundances is given in table 4 .
this table was obtained in the conventional way , specifically by comparing the abundances derived for star 3025 with those derived by varying the atmospheric parameters , one at a time , by the amount given in table 4 . as expected , larger effects arise from uncertainties in @xmath44 ( in particular for neutral species ) and in gravity ( in particular for singly ionized species ) , whereas the overall metal abundance and the microturbulent velocity only play a minor role .
the last column of table 4 gives the quadratic sum of effects due to individual parameters listed ; this can be taken as an estimate of the total uncertainty due to errors in atmospheric parameters .
the uncertainty in o / fe is by far the largest entry in this column .
one of our goals is to extend the calibration of the metallicity scale for globular clusters of carretta & gratton ( 1997 ) to the metal - rich regime .
therefore for the sake of consistency we adopt the same atomic parameters ( listed in table 2 ) that they used , just as we did in paper 1 .
note that the only difference in the line lists used for ngc 6528 and ngc 6553 is in the slightly different wavelength region covered , due to the different instrumental configuration of hires used for the observations . however , both lists are simply subsets of that used in cg97 , securing the required homogeneity .
in addition , the same set of model atmospheres ( kurucz 1992 with convective overshooting ) , code for abundance analysis , _ etc . _
previously adopted are also used in the present work . combining our previous work on ngc 6553 ( cohen _ et al . _
1999 ) with the present study doubles the sample of clusters with metal abundances near the solar value , while retaining a highly homogeneous fine abundance analysis for all clusters studied to date . from the four rhb stars , we find that the mean [ fe / h ] for ngc 6528 is @xmath56 dex , with a star - to - star scatter of @xmath1 dex .
this is the first high dispersion analysis of a galactic globular cluster in which an abundance greater than solar has been obtained .
note however that the uncertainties are from statistics only .
a fair estimate of the total error bar should include also systematic errors , that are in general rather hard to quantify : including systematics , a conservative budget could be about 0.1 dex , mostly due to errors related to the temperature scale and to adopted model atmospheres .
the resulting abundances for each species in each star are listed in table 5 . as in paper
i , all element ratios are computed with respect to fe i , except for o i and sc ii where we used abundances from fe ii to minimize the uncertainties resulting from the choice of @xmath44 .
the rms dispersion in abundance among the measured set of lines for each ion is given in parentheses and the adopted solar abundances are shown in the last column of table 5 .
the results for star 3046 are from ews converted to the same system of the 3 other stars .
only abundances derived from at least 2 lines are shown for star 3046 . also , for comparison , we give in the last column of this table the abundances obtained by a similar analysis of the well known population i star @xmath57 cyg from paper i. for oxygen and sodium ( discussed in detail below ) , we give also the abundances ( from line analysis ) including corrections for departures from lte , following gratton _
et al . _ ( 1999 ) .
abundances for sc ii , v i and mn i include detailed corrections for the quite large hyperfine structure of their lines ( see gratton 1989 and gratton & sneden 1991 for references ) .
the ca abundances were derived applying to each line used the collisional damping parameter appropriate for that line ( smith & raggett 1981 ) .
an estimate of eu abundance was obtained by comparing the average spectrum of all the rhb stars in ngc 6528 observed in the region of the eu ii line at 6645.11 with synthetic spectra ( figure 5 ) . from this comparison , evidence for a mild ( [ eu / fe ] @xmath58 ) overabundance of eu is found . a potential problem affecting the reliability of our results
could be the contamination of fe i lines by blends .
mcwilliam and rich ( 1994 ) demonstrated that in bulge stars the effects of cn lines alone can be overwhelming at metallicities whose mean is similar to the values derived here .
we expect this concern to be much less important in our spectra since they have a much higher resolution than those used by mcwilliam and rich and the temperature of the stars is higher . however , in order to check the impact of this problem , we considered more in detail the fe i lines .
for each fe i line included , we synthesized a spectral region of 3.2 centered on the line , using line lists extracted from kurucz 1995 cd - roms ( cd - rom 23 for the atomic lines ; and cd - rom 15 for diatomic molecules ; note that these lists include lines due to cn as well as to other molecules ) and the same model atmospheres used for the program stars .
( in this exercise , we assumed [ c / fe]@xmath59 and [ n / fe]@xmath60 since , as reminded by the second referee , rhb stars have already experienced the phase along the red giant branch where c is usually depleted and n enhanced , and their sum is constant . ) .
the synthetic spectra were then convolved with a gaussian with a fwhm of 0.12 , in order to match the resolution of our spectra .
we examined the profiles , and flagged all lines whose profiles are in some way distorted by blends ; we then measured the ews of the lines on these synthetic spectra using a gaussian fitting routine ( all parameters left free ; ews obtained by this procedure are much more affected by the presence of blends than those obtained using the procedure we applied for the program stars ) .
these ews were then compared with those obtained from a spectral synthesis where the line list only consisted of the line under scrutiny ( using this time a simple integration over the profile ) .
next we flagged all lines where the two ews differ more than 2 m ( note that the weakest fe i lines have ew@xmath61 m ) .
finally we adopted as very clean lines those with no appreciable distortion in the profiles , and for which the ews are not changed by more than 2 m by blends .
we find a total of 53 very clean fe i lines ; such lines are marked with an asterisk in table 2 .
average abundances obtained from this subset of lines are within 0.020 dex of those derived from the original sample which in addition included lines of somewhat poorer quality .
furthermore the r.m.s .
scatter of individual values are not modified in a significant way .
this exercise suggests that ( i ) the procedure used to measure ews ( i.e. adopting an average relation between ews and fwhm to constrain the fwhm ) allows us to derive reliable ews even for moderately blended lines as in the spectrum of star 3046 ; and ( ii ) that the largest source of errors in the ews is the location of the continuum level rather than the presence of blends .
moreover , following the same procedure as in paper i , we checked our overall [ fe / h ] values by comparing our spectra with syntheses of a spectral region around the li doublet , which includes several weak or medium strength fe lines .
these comparisons are shown in figure 6 , only for the 3 stars having new , high quality spectra from the run of 2000 .
the remaining star , 3046 , was tied on the ew system defined from the 3 others as explained above .
the comparisons between the spectral syntheses and the observed spectra for the rhb stars in ngc 6528 shown in figure 6 support the abundances found by the analysis of the equivalent widths .
lines computed with [ fe / h][email protected] and [ fe / h]=+0.27 are clearly too shallow and too strong with respect to the observations . figure 6 also shows that no lithium is detectable in all our program stars .
the synthetic spectra in figure 6 are computed with log n(li ) = @xmath52 , which is the logarithmic ratio of the abundances of elements a and h in the star , minus the same quantity in the sun . ]
; however , since the li line is blended with a fe line , the upper limit determined from our spectra is log n(li)@xmath62 .
we also note that if we only use fe i lines with log(gf ) - @xmath63 ( where @xmath64 = 5040/t ) , which corresponds to the weakest lines that could be detected in the spectra taken in the june 2000 run , we obtain [ fe / h]=+0.08 dex using 14 lines for star 3025 , [ fe / h]=+0.09 for star 5422 using 11 lines , and [ fe / h]=+0.06 for 12 lines in star 3014 , which values are indistinguishable from those obtained with the full set of fe i lines .
abundances from such weak lines are almost independent of the choice of microturbulent velocity .
although expected by the manner in which @xmath46 is set , the close agreement in the derived fe abundance between the set of weak lines and the full set of fe i lines is reassuring .
the present results are summarized in table 6 , where we list also the mean abundances for ngc 6553 ( both from paper i and from b99 ) , and for 11 giants in baade s window studied by mcwilliam & rich ( 1994 ) , in order to provide a deeper insight into our findings . for ngc 6528 and our analysis of ngc 6553 ,
if there was only one line per star for a given ion , the value is given in parentheses .
the most meaningful and immediate comparison is with our results for ngc 6553 from paper i , since the analysis technique and data set are very consistent and homogeneous .
however , both the quality of the observational material and the approach to the ew measurements present some differences . therefore , the first test we performed was a direct comparison of spectra of stars in the two clusters .
rhb stars are objects in a well defined evolutionary phase .
hence , we can expect their stellar stucture and parameters to be very similar at similar metallicities .
this is evident from figure 7 , where the spectrum of star 71 in ngc 6553 ( from paper i ) is compared to that of star 3025 of ngc 6528 ( from the present work ) in the spectral region 6700 - 6740 . for a meaningful comparison ,
the spectra have been degraded to the same resolution .
this figure shows that spectra of the two stars are actually very similar , apart for some small differences in the positioning of the continuum .
we conclude that the two stars can hardly be considered different or even distinguishable on the basis of their spectra .
this idea is strongly supported by figure 8 , where average ews from stars of ngc 6553 ( paper i ) and ngc 6528 ( present study ) are compared . in order to make the comparison more robust and less dependent on details of the continuum tracing ,
we average only ews of lines measured in at least 3 stars out of 5 in ngc 6553 and in at least 2 stars out of the 3 with better quality spectra in ngc 6528 . again , this comparison is quite good and ews are quite similar for a hypothetical @xmath65 rhb star in the two clusters .
if the observed spectra are the same and the equivalent widths are very similar , then the difference of about 0.2 dex in the average overall metallicity of these two globular clusters ( [ fe / h]@xmath66 for ngc 6553 and [ fe / h]@xmath67 dex in ngc 6528 ) must arise from the different atmospheric parameters adopted in the two analysis .
in fact , both here and in paper i , the atmospheric parameters adopted were derived directly from spectra , i.e. from ews , that in turn were measured on spectra of different quality ( better for the ngc 6528 stars observed in the run of june 2000 ) and using different methods .
to verify this , we use table 4 to estimate the changes in [ fe / h ] as due to differences between the mean atmospheric parameters used for rhb stars in ngc 6553 ( namely 4727/2.3/@xmath50.13/1.82 for temperature , gravity , model abundance and microturbulent velocity , respectively ) and the mean set used for ngc 6528 ( namely 4620/2.21/+0.07/1.34 ) .
the resulting difference in [ fe / h ] is @xmath68 , in the sense that an @xmath65 star in ngc 6553 should be more metal - poor than the @xmath65 one in ngc 6528 .
the small difference in adopted microturbulence of only 0.5 km s@xmath14 gives rise to most of the abundance difference .
another experiment seems to confirm this finding .
we used the set of average ews for ngc 6528 and repeated the analysis .
we derive the values 4640/2.28/0.08/1.37 for @xmath44 , @xmath69 [ a / h ] and @xmath46 , and obtain [ fe / h]@xmath70 using the procedures discussed in section 3.1 of zeroing trends of abundance from single lines with @xmath49 and expected line strengths . with these values
we also obtained a very good ionization equilibrium .
if we now repeat the abundance analysis applying this set of atmospheric parameters to the average ews of ngc 6553 , we obtain a solution where the slopes of linear regressions of abundances @xmath71 @xmath49 and of expected line strengths are well within the 1@xmath23 rms uncertainty .
the resulting abundance for ngc 6553 is then [ fe / h]@xmath72 , i.e. we can conclude that an acceptable value for the overall metallicity of this cluster , as estimated from the fe i abundance , is only 0.04 dex lower than that of ngc 6528 . finally , figure 9 shows the parameter space @xmath46 , @xmath44 over the regime expected for rhb stars . using the average ews , we show the contours which give errors in slopes of linear regressions of abundances @xmath71 @xmath49 and of expected line strengths within the 1@xmath23 rms uncertainty . the loci for both ngc
6528 ( solid line ) and for ngc 6553 ( dotted line ) are elongated .
this is simply another , graphic representation of the coupling that exists between @xmath46 and @xmath44 for cool , metal - rich stars discussed above ( section 3.1 ) .
the higher quality spectra taken in the second run for ngc 6528 do alleviate this phenomenon , permitting high precision measurements of ews of weak lines also of high excitation .
this in turn allows us to better constrain the region of this parameter space within which solutions of the abundance analysis are acceptable .
these tests support the conclusion that within the error bars , the overall metal abundance of the two clusters expressed in terms of [ fe / h ] is virtually the same .
this is a firm result , since the number of measured fe lines is large .
taking into account the overall uncertainty , we can be quite confident that ngc 6528 is a close twin " of ngc 6553 , as far as the overall metal abundance is concerned .
both these bulge clusters have a metallicity slightly larger than solar .
extremely high abundances for ngc 6528 as well ngc 6553 can be safely ruled out , as confirmed also by their normal integrated colours ( see figure 8 in feltzig & gilmore 2000 ) . to be conservative
, we will adopt hereinafter a value of metallicity for ngc 6553 that is the mean of that obtained in paper i , and the one obtained with atmospheric parameters equal to those of ngc 6528 . as error bar , we will adopt one giving these two values as extremes of our confidence range .
hence , we will adopt for ngc 6553 a value of [ fe / h]@xmath73 dex .
the pattern described by the element ratios appears to be rather similar for the two clusters .
the average abundances are generally in good agreement , apart for the [ mg / fe ] ratio . however , as discussed in 2.2 , with the present instrumental set up , we had no access to the mg line at 8717.8 , the clean line used in the analysis of rhb stars in ngc 6553 .
instead we had to use two lines around 5600 that are located in a more crowded region and are saturated , hence not optimal for a good abundance derivation .
even if the abundance from the line at 7657 that we measured for stars observed in the 2000 run generally confirms results from the other lines , at present , and until further confirmation , we can not be very confident that the low mg abundance found for ngc 6528 is real .
the overabundance of si and of ca seems to be well established in ngc 6528 .
since , however , o abundances ( discussed later ) and ti abundances seems more likely solar , it is not clear that ngc 6528 presents the classical overabundance of the @xmath4-elements typical of more metal - poor clusters and field stars . on the other hand ,
mn is rather underabundant in ngc 6528 , and this result is a signature of a small contribution to nucleosynthesis by sn ia , suggesting that the @xmath4-elements overabundance might be the typical fossil record of a nucleosynthesis history heavily dominated by massive stars .
some support to this scenario could be given by the mild overabundance of eu found from the combined spectrum of the rhb stars in ngc 6528 , since eu is known to be a n - capture element predominantly produced by r - processes .
note however that the ratio [ ba / eu ] is almost solar , suggesting that maybe we are seeing also the contribution to nucleosynthesis from intermediate - mass stars .
a comparison of our results for ngc 6553 with those of b99 was given in section 7.1 of paper i. here , we note only that the choice of solar and stellar atmosphere model adopted in b99 ( hence the solar abundances ) results in fe abundances that are 0.1 to 0.15 dex lower than those of the present scale .
this by itself decreases the differences in the element ratios presented in table 6 . in other words , had the comparison for ngc 6553 been made , instead , in terms of [ element / h ] , the pattern of many elements would have been more similar when our analysis of paper i was compared to that of b99 .
the slightly different instrumental configuration of hires adopted for the observations of the rhb stars in ngc 6528 allowed the use of the weak na doublet at 6154 , 6160 as well as the infrared permitted o i triplet , and , for three of the four stars , the forbidden 6300 o i line as well . to take into account departures from the adopted assumption of lte ,
appropriate corrections were applied to o and na abundances following the prescriptions of gratton _ et al . _
non - lte abundances from the line analysis are also listed in table 5 .
note however , that the corrections are rather small , and do not exceed a few hundredths of a dex in all cases .
we tested the abundances we derived for na and o using synthetic spectra of the 6154 - 60 na i doublet , and of the [ o i ] line at 6300 .
the spectral synthesis were computed assuming lte .
comparisons with spectral synthesis for these two spectral regions are shown in figures 10 and 11 respectively . from the comparisons in figure 10 we found that na is overabundant by about + 0.35 to + 0.5 dex in stars 3025 , 5422 , and 3014 ( respectively , we estimated [ na / fe]= + 0.5 , + 0.35 and + 0.4 dex ) .
these results are consistent with that deduced from the ews .
the comparison between observed and synthesized spectra for the region including the [ oi ] line at 6300.3 is more difficult for three reasons .
first , there are several telluric lines in this region whose locations are marked with a t in figure 11 . due to the high @xmath12 of ngc 6528 , the wavelength of the [ oi ] line in the observed spectra of ngc 6528 is free of contamination by telluric lines ; they should not affect our results .
second , as mentioned in section 3.1 , o abundances are quite sensitive to uncertainties in the atmospheric parameters .
third , there is some coupling between c and o abundances , due to formation of co in the atmospheres of these cool , metal - rich k - giants .
we estimated that derived o abundances increase by 0.06 dex if the c abundances are raised by 0.2 dex .
unfortunately , we could not determine c abundances from our spectra , since ews of c i lines measured in the red are not reliable ; we can only fix the overall strength of cn lines .
we matched adequately cn lines by assuming [ c / fe][email protected] and [ n / fe]=0 when [ o / fe][email protected] ; [ c / fe][email protected] and [ n / fe]=0 when [ o / fe]=0.13 ; and [ c / fe][email protected] and [ n / fe]=0 when [ o / fe]=+0.33 . these are the values adopted in the syntheses shown in figure 11 .
figure 11 shows that with the adopted c and n abundances , a good fit of the [ oi ] line is obtained for [ o / fe]@xmath74 , in reasonable agreement with the results given by the analysis of the equivalent widths ( where also other o lines are taken into account ) .
an o excess of 0.3 dex seems excluded , unless the c abundance is not much larger than solar , which seems unlikely .
we conclude that the program stars in ngc 6528 have solar oxygen abundances , but a distinct excess of na ( [ na / fe]@xmath75 ) .
our need , after the completion of our analysis of ngc 6553 of paper i , to have an analysis for stars in at least one more metal - rich cluster is apparent if one examines figure 12 , where the metallicities obtained from high resolution spectra by cg97 , plus those for ngc 6553 and ngc 6528 ( this study ) are plotted against the parameter w(caii ) determined by rutledge _
et al . _ ( 1997 ) from the measured strength of the infrared ca triplet in individual globular cluster giants . for reasons explained in section 4.1 ,
we believe that the correct [ fe / h ] values to adopt for these two clusters are + 0.07 and @xmath50.06 dex for ngc 6528 and ngc 6553 , respectively . when ngc 6528 , which has the largest value of any galactic globular cluster observed to date on the rutledge _ et al .
_ ranking scale ( w(caii)=5.41 ) , is added , the conclusion proposed in paper i is still firmly supported .
specifically , the two scales seem to be linearly correlated over the range from [ fe / h]@xmath76 to [ fe / h]@xmath77 ( i.e. up to not so extremely metal rich clusters ) . at higher metallicities , however , where the caii index is known to lose its sensitivity to metal abundance ( see armandroff & da costa 1991 ) , the linear correlation seems to break down .
note that clusters like pal 12 or rup 106 , known to have anomalously low [ @xmath4/fe ] ratios , are not included in the sample . with our new results
we can now better assess the relationship between the two scales : in fact , ngc 6528 falls very near ngc 6553 , in the [ fe / h]w(caii ) plane , and strongly constrains the position of metal - rich globular clusters for the metallicity calibration in terms of this low resolution indicator .
this is not unexpected , since in section 4.1 we showed how rhb stars of these 2 clusters are similar .
we find that to bring the values from rutledge _
et al._onto a homogeneous metallicity scale which is based completely on high dispersion spectroscopy , one has to adopt a second order polynomial relation : @xmath78_{us } } = -2.08 -0.04 { \rm w(caii ) } + 0.078 { \rm w^2(caii)}\ ] ] with the correlation coefficient @xmath79 and @xmath23=0.12 dex for 22 clusters .
the second order term is significant at a confidence level exceeding 95% ; inclusion of this term yields @xmath80 for the sample of 22 globular clusters in common .
this is a reduction in @xmath81 of a factor of 1.6 compared to a linear fit .
notice that similar results are obtained even if we omit entirely ngc 6553 , so that the larger error bar of the revised [ fe / h ] value for this cluster does not affect much our calibration .
this relation , shown in figure 12 , allows us to derive directly from the ca index of a globular cluster its metallicity on the carretta & gratton scale , as extended here and in paper i. the range of application should be restricted to that of calibrating clusters , namely from w(caii ) @xmath82 ( ngc 7078,ngc 4590 ) to w(caii ) @xmath83 ( ngc 6528 ) .
moreover , using data from the compilation of carney ( 1996 ) , we checked that the differences between the w(caii ) values as observed by rutledge _ et al .
_ and the w(caii ) values predicted from our eq .
( 2 ) are constant as a function of [ ca / fe ] .
we can then conclude that , since the [ ca / fe ] ratio holds rather constant for all clusters in the sample , we are actually calibrating an index that is tied to the temperatures and luminosities of stars along the rgb , which in turn depend on the cluster metal abundance . adding a second well analyzed high metallicity cluster is also very useful in intercomparing the cg97 scale and the widely used scale of zinn & west ( 1984 ) ( based on integrated light indices ) .
we are now able to derive a relation to bring their values on the new scale entirely based on high dispersion analyses , without using any uncertain extrapolations . as in cg97
, we prefer to use directly final metallicities from zinn & west ( 1984 ) , with the update of armandroff & zinn ( 1988 ; collectively zw ) , due to the variety of indicators used in the analysis of zinn & west ( 1984 ) . when ngc 6528 and ngc 6553 are added to the other 24 calibrating clusters , the quadratic relation previously found by cg97 is no longer the most appropriate , as one can see in figure 13 , where mean metallicities from the cg97 scale as augmented here are compared with metallicities on the zw scale . in this case
a cubic polynomial is the best relation to transform the zw scale to the new high dispersion spectroscopic scale : @xmath78_{cg } } = + 0.61 + 3.04 { \rm [ fe / h]_{zw } } + 1.981 { \rm [ fe / h]_{zw}}^2 + 0.532 { \rm [ fe / h]_{zw}}^3\ ] ] with @xmath23=0.10 dex and a correlation coefficient @xmath84 for 26 clusters . when considering the adopted error bar of 0.15 dex for the revised value of [ fe / h ] of ngc 6553 , the second and third order terms are significant at the 99% confidence level .
the value of @xmath81 decreases by more than a factor of two in going from a linear fit to a third order fit , e.g. from 392 to 144 for the comparison of our abundances with those of zinn & west ( 1984 ) for the 26 clusters ( 25 without ngc 6553 ) in the sample with high dispersion abundance analyses .
the present work extends the range of application of this transformation by about 0.3 dex toward high metallicity , with respect to eq . 7 in cg97 . using the above relation
now allows one to transform the zw metallicities onto our high dispersion scale in the range @xmath85[fe / h]@xmath86 .
we present for the first time an abundance analysis for stars in the metal - rich bulge cluster ngc 6528 based on high resolution spectra of s / n high enough to allow a reliable fine abundance analysis .
we observed 4 red hb stars , from which we found a mean [ fe / h]@xmath87 dex , with a star - to - star scatter of @xmath1 ( not including systematic effects ) .
this metallicity is slightly larger than the mean abundance in the galactic bulge found by mcwilliam & rich ( 1994 ) , and that found in a previous paper for ngc 6553 .
however , we found that the spectra of stars in ngc 6553 and ngc 6528 , and the ews we measured on them , are very similar to each other .
the slightly different metal abundance is the result primarily of adoption of slightly lower microturbulent velocity in the analysis of ngc 6528 stars and secondarily of the adoption of slightly lower temperatures for the ngc 6528 stars .
note that the present analysis is based on higher quality spectra , so that the atmospheric parameters here used for ngc 6528 are more firmly established than those used in paper i for ngc 6553 .
if the same atmospheric parameters are adopted for both samples , nearly equal metal abundances are derived for the two clusters .
we then propose to revise upward the metal abundance of ngc 6553 to [ fe / h]@xmath73 , where the error bar encompasses both the original abundance for ngc 6553 we derived in paper i , and the value we obtained assuming in the analysis atmospheric parameters identical to those we used for stars in ngc 6528 .
these results imply that metal - rich globular clusters may reach or even exceed the mean abundance of the galactic bulge found by mcwilliam & rich , but are found to fall within the spread of their distribution .
the relative abundance for the best determined @xmath4-process elements ( ca ) indicates an excess of @xmath4 process elements of about a factor of two and a global pattern of abundances similar to that of ngc 6553 and the bulge field stars .
moreover , mn in ngc 6528 clearly shows an underabundance typical of a quite small contribution to nucleosynthesis by yields from sn ia .
when coupled with the overabundances of si and ca , our results strongly suggest that these bulge globular clusters seem to have experienced a history of chemical enrichment essentially identical to that typical of bulge field stars , probably under conditions of enrichment by type ii supernovae at early epochs .
we found a solar o abundance , while there is a clear excess of na ( [ na / fe]@xmath75 ) .
however the sample is not large enough to understand if this abundance pattern is characteristics of all stars in this cluster , or rather stems from peculiarities of the program stars or their evolutionary state . our new data for ngc 6553 ( paper i ) and
ngc 6528 allow us to re - calibrate the widely used metallicity scales of zinn & west ( 1984 ) and of rutledge _
et al._(1997 ) , both based on low dispersion or integrated indices , onto a scale fully based on high dispersion spectroscopy .
we give the functions required to transform w(caii ) and [ fe / h]@xmath88 values into our updated scale , superseding the previous calibration by carretta & gratton ( 1997 ) .
like other bulge clusters heavily reddened and/or affected by differential reddening , the value of e(b - v ) for ngc 6528 in the literature ranges over more than 0.2 mag , e.g. 0.46 mag ( richtler _ et al . _
1998 ) from @xmath89 photometry , 0.55 mag ( ortolani _ et al . _
1992 and zinn 1980 ) , and 0.73 mag from cluster integrated colours ( reed _ et al . _
1988 ) .
since an underestimate of 0.05 mag in the value of e(b - v ) translates into an underestimate of about 0.2 dex in metal abundance when temperatures are derived from colours ( cohen 1983 ) , this issue deserve special care .
even methods such as that of sarajedini ( 1994 ) , which uses fits to the shapes of unreddened giant branches to derive metallicity and reddening simultaneously , are unreliable for very metal - rich clusters , since the most metal - rich of his calibrating clusters is 47 tuc ( [ fe / h]@xmath90 ) . with the present analysis
, we have a spectroscopic determination of the temperatures of four rhb stars that are derived only from the excitation of many fe i lines .
these estimates are therefore reddening - free and provide an independent route to the value of e(b - v ) in ngc 6528 . as a starting point we assumed a mean value of @xmath91 4610 k as representative of the spectroscopic temperatures of our rhb stars ( see table 3 ) .
this temperature , with the mean metallicity and gravity from table 3 , corresponds to a value ( v - k)@xmath92 based on the colour@xmath44 calibration of gratton , carretta & castelli ( 1997 ) calibrations .
hence , the v - i and b - v colours do not add useful information to what we are deriving from v - k colours . ] .
on the other hand , from the v , k photometry of montegriffo ( 1999 ) the mean v - k colour for the 4 program stars is 3.88 ; so that , assuming a standard relation e(v - k)=2.7 e(b - v ) ( e.g. cardelli _ et al . _
1989 ) , we deduce a value of e(b - v ) = 0.50 mag as an independent estimate for the reddening of ngc 6528 . repeating the above exercise using the @xmath93 colours and parameters of each star
, we could not isolate any significant difference in the resulting values of e(b - v ) .
this value is somewhat lower than the one listed in the compilation by harris ( 1996 ) , as well most of the values quoted above , although only slightly smaller that that found in the field just n and w of the cluster by stanek ( 1996 ) .
( stanek s value in the field just s of the cluster is close to that of harris . )
we believe that this arises because of a selection bias . in a metal - rich globular cluster with differential reddening , the rhb shape in a cmd changes from the normal short horizontal one characteristic of a constant reddening to an extended distribution sloping towards fainter magnitudes for the redder stars .
such a rhb is clearly shown in the case of ngc 6553 in figure 2b of ortolani _
et al._(1990 ) and ngc 6528 ( in figure 2 ) .
our sample of rhb stars selected for high dispersion spectroscopy is biased towards the brighter stars , and hence the less reddened ones .
table 3 shows , however , that the program stars have a total range in @xmath44 of 70 k ( well within the adopted uncertainties ) , even though they cover a rather large range in the observed reddened cmd .
ngc 6528 is known to be affected by differential reddening .
cohen & sleeper ( 1995 ) establish that the range of e(b - v ) is about 0.25 mag .
consistent with the suggestion of differential reddening , it is interesting to note that three of the program stars are located in the same region of ngc 6528 . for differential reddening of the magnitude found in ngc 6528 , this selection effect far outweighs any tendency toward picking rhb stars that are slightly evolved off the zero age hb .
irrespective of the origin of this effect and of the mean value of e(b - v ) for ngc 6528 , we stress here that the determination of temperatures by line excitation is a purely spectroscopic , reddening - free method and that this is the method used for the four rhb stars we observed . as noted by the referee
, one may wonder how sensitive this method could be to errors in gf values . in previous section
, we performed a test aimed to understand if our values for the microturbulent velocity were biased due to the correlation existing between line strength and excitation potential , because lines of low excitation are all strong .
hence we compared the values for the @xmath46 derived from all lines with those we derive using only high excitation lines .
this second value is nearly independent of the assumed temperature , because all lines have nearly the same excitation potential .
we checked the zero - point of this calibration using field red clump stars ( carretta _ et al .
_ , cohen _ et al .
_ in preparation ) with atmospheric parameters similar to those of stars in ngc 6528 and with good parallaxes from hipparcos .
these stars are sufficiently close that it is reasonable to assume that the reddening is negligible .
we are grateful to paolo montegriffo for providing the hst and irac photometry of ngc 6528 prior to publication .
we warmly thank bernardo salasnich for constructing for us the isochrones from the latest padova models in advance of publication .
jgc and bbb are grateful for partial support from nsf grant ast9819614 .
the entire keck / hires user community owes a huge debt to jerry nelson , gerry smith , steve vogt , and many other people who have worked to make the keck telescope and hires a reality and to operate and maintain the keck observatory .
we are grateful to the w. m. keck foundation , and particularly its late president , howard keck , for the vision to fund the construction of the w. m. keck observatory .
this research has made use of the simbad data base , operated at cds , strasbourg , france .
lrrrrrrrr [ tab1 ] red hb stars 5422 & 16.95 & 15.34 & 14.11 & 13.14 & @xmath94 & 990817 & 2 x 1200 & 18 04 47.58 @xmath530 03 30 & & & & & @xmath95 & 000603 & 6 x 1200 & 3025 & 17.14 & 15.43 & 14.21 & 13.22 & @xmath96 & 990817 & 3 x 1200 & 18 04 47.62 @xmath530 03 36 & & & & & @xmath97 & 000603 & 12 x 1200 & 3014 & 17.09 & 15.38 & 14.21 & 13.21 & @xmath98 & 990815 & 3 x 1200 & 18 04 47.38 @xmath530 03 43 & & & & & @xmath99 & 000603 & 6 x 1200 & 3046 & 17.25 & 15.50 & 14.12 & 13.34 & @xmath100 & 990818 & 2 x 1200 & 18 04 51.52 @xmath530 03 20 5425 & 17.09 & 15.35 & 14.25 & 13.32 & @xmath101 & 000603 & @xmath102 & 18 04 47.62 @xmath530 03 25 3032 & 16.69 & 14.71 & 13.58 & 12.34 & @xmath103 & 000603 & @xmath102 & 18 04 47.34 @xmath530 03 33 lrrrrrrrrrrrrrrrrrr [ tab2 ] o i & 6300.31 & 0.00 & @xmath59.75 & & 40.0 & & 39.0 & & 37.0 & & o
i & 6363.79 & 0.02 & @xmath510.25 & & 9.0 & & 15.0 & & 13.0 & & o i & 7771.95 & 9.11 & 0.33 & & 40.0 & & 36.0 & & 35.0 & & 54.8 o i & 7775.40 & 9.11 & @xmath50.03 & & 30.0 & & 31.0 & & & & na i & 5682.65 & 2.10 & @xmath50.67 & & 182.9 & & 194.4 & & 200.7 & & na i & 5688.22 & 2.10 & @xmath50.37 & & 184.8 & & 187.1 & & 208.6 & & na i & 6154.23 & 2.10 & @xmath51.57 & & 114.5 & & 125.3 & & 116.5 & & 101.6 na i & 6160.75 & 2.10 & @xmath51.26 & & 140.8 & & 137.2 & & 142.9 & & 119.6 mg i & 5528.42 & 4.34 & @xmath50.62 & & 270.1 & & 264.5 & & 276.3 & & mg i & 5711.09 & 4.34 & @xmath51.83 & & 156.6 & & 157.5 & & 155.0 & & 132.0 mg
i & 7657.60 & 5.11 & @xmath51.28 & & 124.5 & & 115.7 & & 165.9 & & si i & 5645.62 & 4.93 & @xmath52.14 & & 84.4 & & & & 96.2 & & si i & 5665.56 & 4.92 & @xmath52.04 & & 93.5 & & 80.3 & & 99.5 & & 78.2 si i & 5684.49 & 4.95 & @xmath51.65 & & 75.8 & & 95.1 & & 112.5 & & si i & 5690.43 & 4.93 & @xmath51.87 & & 75.7 & & 95.3 & & 78.8 & & 90.2 si i & 5701.11 & 4.93 & @xmath52.05 & & 61.5 & & 73.1 & & & & si i & 5708.41 & 4.95 & @xmath51.47 & & 129.8 & & & & & & si i & 5772.15 & 5.08 & @xmath51.75 & & 95.1 & & 103.0 & & 83.9 & & 87.2 si i & 5793.08 & 4.93 & @xmath52.06 & & 91.8 & & 99.2 & & 94.8 & & si i & 5948.55 & 5.08 & @xmath51.23 & & 127.4 & & 117.2 & & 133.9 & & 108.7 si i & 6125.03 & 5.61 & @xmath51.57 & & 69.6 & & 59.9 & & 60.7 & & 54.5 si i & 6145.02 & 5.61 & @xmath51.44 & & 51.6 & & 56.2 & & 59.2 & & 51.9 si i & 6848.57 & 5.86 & @xmath51.75 & & 50.5 & & 47.2 & & 42.7 & & si i & 6976.50 & 5.95 & @xmath51.17 & & & & 83.8 & & & & si i & 7003.58 & 5.96 & @xmath50.87 & & & & 63.9 & & 70.2 & & si
i & 7034.90 & 5.87 & @xmath50.88 & & 93.7 & & 86.0 & & 110.2 & & si i & 7226.20 & 5.61 & @xmath51.51 & & & & & & & & 60.6 si i & 7405.79 & 5.61 & @xmath50.82 & & 111.4 & & 120.7 & & 126.6 & & ca i & 5590.13 & 2.51 & @xmath50.57 & & 128.1 & & & & 150.9 & & ca
i & 5594.47 & 2.51 & 0.10 & & 224.5 & & 215.0 & & 213.1 & & ca i & 5857.46 & 2.93 & 0.24 & & 182.9 & & 178.4 & & 208.0 & & 184.3 ca i & 5867.57 & 2.93 & @xmath51.49 & & 66.7 & & 75.2 & & 86.8 & & 62.8 ca i & 6161.30 & 2.52 & @xmath51.27 & & 122.1 & & 129.0 & & 114.0 & & ca i & 6166.44 & 2.52 & @xmath51.14 & & 107.4 & & 120.0 & & 122.5 & & 119.7 ca i & 6169.04 & 2.52 & @xmath50.80 & & 145.0 & & 142.9 & & 150.7 & & ca i & 6169.56 & 2.52 & @xmath50.48 & & 164.5 & & 164.3 & & 160.9 & & ca i & 6439.08 & 2.52 & 0.39 & & & & & & & & 234.9 ca i & 6462.57 & 2.52 & 0.26 & & 299.3 & & 282.5 & & 299.7 & & ca i & 6471.67 & 2.52 & @xmath50.69 & & 152.9 & & 153.1 & & 162.8 & & 128.5 ca i & 6493.79 & 2.52 & @xmath50.11 & & 174.9 & & 185.3 & & 197.4 & & 193.0 ca i & 6499.65 & 2.52 & @xmath50.82 & & 130.7 & & 144.6 & & 147.7 & & 145.7 ca i & 6572.80 & 0.00 & @xmath54.32 & & & & & & & & 139.7 ca i & 6717.69 & 2.71 & @xmath50.52 & & 197.9 & & & & & & sc ii & 5526.82 & 1.77 & 0.19 & & 140.0 & & 127.2 & & 128.6 & & sc ii & 5640.99 & 1.50 & @xmath50.86 & & 103.4 & & 101.5 & & & & sc ii & 5657.88 & 1.51 & @xmath50.29 & & 119.4 & & 121.5 & & 120.6 & & sc ii & 5667.15 & 1.50 & @xmath51.11 & & 80.8 & & 81.5 & & & & sc ii & 5669.04 & 1.50 & @xmath51.00 & & 86.8 & & 81.0 & & 92.3 & & 64.8 sc ii & 5684.20 & 1.51 & @xmath50.92 & & & & 82.6 & & 102.4 & & sc ii & 6245.62 & 1.51 & @xmath51.05 & & 88.1 & & 89.6 & & 98.1 & & 79.2 sc ii & 6604.60 & 1.36 & @xmath51.14 & & 88.7 & & 92.8 & & 100.9 & & 89.8 ti i & 5490.16 & 1.46 & @xmath50.93 & & 97.1 & & 101.3 & & 87.7 & & ti i & 5503.90 & 2.58 & @xmath50.19 & & 81.9 & & & & & & ti i & 5662.16 & 2.32 & @xmath50.11 & & & & 87.8 & & 102.1 & & ti i & 5689.48 & 2.30 & @xmath50.47 & & & & & & 57.4 & & 60.4 ti i & 5866.46 & 1.07 & @xmath50.84 & & 144.6 & & 139.3 & & 160.9 & & 123.3 ti i & 5922.12 & 1.05 & @xmath51.47 & & 102.5 & & 111.7 & & 106.3 & & 90.8 ti i & 5978.55 & 1.87 & @xmath50.58 & & 105.5 & & 96.3 & & 108.0 & & ti i & 6091.18 & 2.27 & @xmath50.42 & & 68.2 & & 69.9 & & & & ti i & 6126.22 & 1.07 & @xmath51.42 & & 93.1 & & 104.9 & & 107.9 & & 87.9 ti i & 6258.11 & 1.44 & @xmath50.36 & & 122.7 & & 131.7 & & 134.7 & & ti
i & 6261.11 & 1.43 & @xmath50.48 & & 166.8 & & & & 157.2 & & ti i & 6554.24 & 1.44 & @xmath51.22 & & & & & & & & 109.2 ti i & 7251.72 & 1.43 & @xmath50.84 & & 122.7 & & 126.3 & & 125.2 & & 148.7 v i & 5627.64 & 1.08 & @xmath50.37 & & & & & & & & 103.8 v i & 5670.86 & 1.08 & @xmath50.42 & & 116.1 & & 110.7 & & 105.0 & & 87.1 v i & 5703.59 & 1.05 & @xmath50.21 & & 111.3 & & 113.8 & & 113.9 & & v i & 5727.06 & 1.08 & @xmath50.01 & & & & & & & & 142.6 v i & 6081.45 & 1.05 & @xmath50.58 & & 98.9 & & 95.2 & & 116.8 & & v i & 6090.22 & 1.08 & @xmath50.06 & & 120.8 & & 115.0 & & 130.6 & & v i & 6119.53 & 1.06 & @xmath50.32 & & & & & & & & 107.4 v i & 6199.19 & 0.29 & @xmath51.28 & & 141.5 & & 134.0 & & 148.7 & & v i & 6216.36 & 0.28 & @xmath51.29 & & & & & & & & 138.5 v i & 6243.11 & 0.30 & @xmath50.98 & & & & & & 180.6 & & v i & 6251.83 & 0.29 & @xmath51.34 & & 116.2 & & 117.2 & & 116.9 & & 113.6 cr i & 5409.80 & 1.03 & @xmath50.71 & & & & 268.1 & & & & cr i & 5702.33 & 3.45 & @xmath50.68 & & & & 76.6 & & 76.1 & & cr i & 5781.19 & 3.32 & @xmath50.88 & & 79.4 & & 64.1 & & 55.6 & & cr i & 5781.76 & 3.32 & @xmath50.75 & & 81.1 & & 67.5 & & 69.1 & & cr i & 5783.07 & 3.32 & @xmath50.40 & & & & & & 59.6 & & 53.6 cr i & 5783.87 & 3.32 & @xmath50.30 & & & & 96.0 & & 89.5 & & 97.8 cr i & 5784.98 & 3.32 & @xmath50.38 & & 89.6 & & 80.5 & & 96.9 & & cr i & 5787.93 & 3.32 & @xmath50.08 & & 97.5 & & 80.9 & & 82.8 & & cr i & 6330.10 & 0.94 & @xmath52.87 & & & & & & & & 115.1 cr i & 6882.50 & 3.44 & @xmath50.38 & & & & 65.8 & & & & cr i & 6883.07 & 3.44 & @xmath50.42 & & 65.6 & & 64.2 & & 71.9 & & cr i & 6882.52 & 3.44 & @xmath50.38 & & & & 65.8 & & & & cr i & 6883.06 & 3.44 & @xmath50.42 & & 65.6 & & 64.2 & & 71.9 & & cr i & 6979.80 & 3.46 & @xmath50.22 & & 76.0 & & 86.8 & & 77.0 & & cr i & 6980.94 & 3.46 & @xmath51.09 & & 40.3 & & & & & & cr i & 7400.19 & 2.90 & @xmath50.11 & & 132.6 & & 151.6 & & 151.8 & & 162.5
cr ii & 5502.09 & 4.17 & @xmath51.96 & & 46.0 & & 46.1 & & & & cr ii & 5508.63 & 4.15 & @xmath52.07 & & 57.7 & & 38.7 & & 53.3 & & mn i & 5394.67 & 0.00 & @xmath53.50 & & & & 199.7 & & & & mn i & 5420.37 & 2.14 & @xmath51.46 & & & & 192.9 & & & & mn i & 5432.55 & 0.00 & @xmath53.80 & & 176.1 & & 188.4 & & & & mn i & 6013.50 & 3.07 & @xmath50.25 & & & & & & & & 146.1 mn i & 6016.65 & 3.07 & @xmath50.09 & & & & & & & & 146.2 mn
i & 6021.80 & 3.08 & 0.03 & & 183.7 & & 133.7 & & 165.4 & & 150.7 mn i & 7302.85 & 4.43 & @xmath50.37 & & 59.5 & & 58.8 & & 52.0 & & fe i & 5386.34 * & 4.15 & @xmath51.74 & & & & 70.7 & & & & fe i & 5389.49 & 4.41 & @xmath50.57 & & & & 105.5 & & & & fe i & 5395.22 & 4.44 & @xmath51.73 & & & & 49.0 & & & & fe i & 5398.29 * & 4.44 & @xmath50.72 & & & & 100.7 & & & & fe i & 5406.78 & 4.37 & @xmath51.40 & & & & 65.7 & & & & fe i & 5412.79 & 4.43 & @xmath51.80 & & & & 57.6 & & & & fe i & 5417.04 & 4.41 & @xmath51.42 & & & & 67.0 & & & & fe i & 5436.30 & 4.39 & @xmath51.36 & & 77.8 & & & & & & fe i & 5464.29 & 4.14 & @xmath51.62 & & 60.4 & & 75.0 & & & & fe i & 5470.09 * & 4.44 & @xmath51.60 & & 57.0 & & 55.7 & & & & fe i & 5491.84 & 4.19 & @xmath52.24 & & 48.1 & & 36.5 & & 39.0 & & fe i & 5494.47 & 4.07 & @xmath51.96 & & 61.1 & & 68.9 & & 63.6 & & fe i & 5522.45 * & 4.21 & @xmath51.47 & & 89.3 & & 86.9 & & 91.0 & & fe i & 5560.22 * & 4.43 & @xmath51.10 & & 71.9 & & 80.1 & & & & fe i & 5577.03 * & 5.03 & @xmath51.49 & & & & 33.1 & & & & fe i & 5587.58 & 4.14 & @xmath51.70 & & 72.6 & & & & 84.3 & & fe i & 5618.64 & 4.21 & @xmath51.34 & & & & 90.2 & & 100.0 & & 66.0 fe i & 5619.61 & 4.39 & @xmath51.49 & & 68.6 & & & & & & 59.5 fe i & 5650.00 & 5.10 & @xmath50.80 & & & & 69.0 & & 65.3 & & fe i & 5651.48 * & 4.47 & @xmath51.79 & & & & 39.2 & & & & fe i & 5661.35 & 4.28 & @xmath51.83 & & & & 64.9 & & 55.4 & & fe i & 5701.56 & 2.56 & @xmath52.16 & & 156.5 & & 151.0 & & 163.1 & & fe i & 5717.84 & 4.28 & @xmath50.98 & & & & & & & & 89.3 fe i & 5731.77 * & 4.26 & @xmath51.10 & & 101.6 & & 86.7 & & 103.1 & & fe i & 5738.24 & 4.22 & @xmath52.24 & & 50.8 & & 40.9 & & 50.9 & & fe i & 5741.86 * & 4.26 & @xmath51.69 & & & & 61.5 & & 73.9 & & 71.2 fe i & 5752.04 * & 4.55 & @xmath50.92 & & 73.2 & & 82.3 & & 86.0 & & 87.3 fe i & 5760.36 & 3.64 & @xmath52.46 & & 62.8 & & 67.3 & & 77.0 & & fe i & 5775.09 & 4.22 & @xmath51.11 & & 95.6 & & 105.9 & & 97.2 & & fe i & 5778.46 & 2.59 & @xmath53.44 & & 82.2 & & 76.0 & & 76.8 & & fe i & 5784.67 * & 3.40 & @xmath52.53 & & 77.0 & & 73.2 & & & & fe i & 5793.92 * & 4.22 & @xmath51.62 & & & & 78.5 & & 77.2 & & fe i & 5835.11 & 4.26 & @xmath52.18 & & & & & & 58.1 & & fe i & 5852.23 & 4.55 & @xmath51.36 & & & & & & & & 75.9 fe i & 5855.09 * & 4.61 & @xmath51.56 & & 54.1 & & 47.9 & & & & 33.6 fe i & 5856.10 * & 4.29 & @xmath51.57 & & & & 75.2 & & & & 61.9 fe i & 5858.78 & 4.22 & @xmath52.19 & & 36.8 & & 49.0 & & 41.9 & & 46.3 fe i & 5859.60 & 4.55 & @xmath50.63 & & 97.5 & & 105.5 & & 118.4 & & 89.5 fe i & 5862.37 * & 4.55 & @xmath50.42 & & 110.9 & & 108.5 & & 128.3 & & 102.6 fe i & 5881.28 & 4.61 & @xmath51.76 & & & & 41.3 & & & & fe i & 5905.68 & 4.65 & @xmath50.78 & & & & & & & & 71.2 fe i & 5927.80 & 4.65 & @xmath51.07 & & 66.5 & & 79.9 & & 69.9 & & 54.7 fe i & 5929.68 * & 4.55 & @xmath51.16 & & 83.2 & & 75.3 & & 71.9 & & fe i & 5930.19 * & 4.65 & @xmath50.34 & & 121.8 & & 123.3 & & 126.2 & & 110.1 fe i & 5934.66 & 3.93 & @xmath51.08 & & 121.0 & & 117.1 & & 123.0 & & 123.8 fe i & 5956.71 * & 0.86 & @xmath54.56 & & 129.1 & & 137.2 & & & & 128.4 fe i & 5976.79 * & 3.94 & @xmath51.30 & & 111.1 & & 109.5 & & 109.1 & & fe i & 6003.02 & 3.88 & @xmath51.02 & & & & & & & & 123.1 fe i & 6027.06 * & 4.07 & @xmath51.20 & & 99.5 & & 103.7 & & 119.5 & & 117.3 fe i & 6056.01 & 4.73 & @xmath50.46 & & 88.0 & & 104.5 & & 102.8 & & 110.9 fe i & 6065.49 * & 2.61 & @xmath51.49 & & 189.6 & & 192.7 & & 199.9 & & fe i & 6078.50 * & 4.79 & @xmath50.38 & & 100.9 & & 106.9 & & 111.3 & & fe i & 6079.02 * & 4.65 & @xmath50.97 & & 75.9 & & 80.9 & & 95.0 & & fe i & 6082.72 & 2.22 & @xmath53.53 & & 108.1 & & & & 104.9 & & fe i & 6089.57 & 5.02 & @xmath50.87 & & 69.7 & & 70.5 & & 78.0 & & fe i & 6093.65 * & 4.61 & @xmath51.32 & & 64.5 & & 73.5 & & & & fe i & 6094.38 & 4.65 & @xmath51.56 & & & & & & 53.4 & & fe i & 6096.67 * & 3.98 & @xmath51.76 & & 89.6 & & 78.1 & & 87.3 & & fe i & 6098.25 & 4.56 & @xmath51.81 & & & & 42.5 & & 50.0 & & fe i & 6137.00 & 2.20 & @xmath52.91 & & 147.6 & & & & 155.7 & & fe i & 6151.62 * & 2.18 & @xmath53.26 & & 109.3 & & 114.5 & & 115.5 & & 115.5 fe i & 6157.73 & 4.07 & @xmath51.26 & & & & & & & & 120.6 fe i & 6165.36 & 4.14 & @xmath51.48 & & 85.0 & & 86.9 & & 90.6 & & 63.4 fe i & 6173.34 * & 2.22 & @xmath52.84 & & 138.1 & & 148.6 & & 149.0 & & 139.7 fe i & 6187.99 & 3.94 & @xmath51.60 & & 99.1 & & 89.4 & & 93.8 & & fe i & 6200.32 * & 2.61 & @xmath52.39 & & 144.9 & & 146.6 & & 158.4 & & fe i & 6219.29 & 2.20 & @xmath52.39 & & & & & & & & 169.4 fe i & 6232.65 & 3.65 & @xmath51.21 & & 115.2 & & 140.2 & & 133.5 & & fe i & 6240.65 & 2.22 & @xmath53.23 & & 109.7 & & & & 116.1 & & fe i & 6246.33 * & 3.60 & @xmath50.73 & & 146.7 & & 158.2 & & 156.0 & & 145.9 fe i & 6252.56 * & 2.40 & @xmath51.64 & & & & & & 206.1 & & fe i & 6265.14 & 2.18 & @xmath52.51 & & & & & & & & 187.6 fe i & 6270.23 & 2.86 & @xmath52.55 & & 115.4 & & 102.8 & & 116.5 & & 102.2 fe i & 6297.80 * & 2.22 & @xmath52.70 & & 157.9 & & & & 156.1 & & fe i & 6301.51 * & 3.65 & @xmath50.72 & & 161.0 & & 162.2 & & 178.5 & & fe i & 6322.69 & 2.59 & @xmath52.38 & & & & & & & & 135.4 fe i & 6330.85 & 4.73 & @xmath51.22 & & & & & & & & 65.5 fe i & 6335.34 & 2.20 & @xmath52.28 & & & & & & & & 175.4 fe i & 6380.75 & 4.19 & @xmath51.34 & & 100.4 & & 98.5 & & 104.5 & & 110.6 fe i & 6392.54 * & 2.28 & @xmath53.97 & & 69.0 & & & & 92.4 & & fe i & 6400.32 & 3.60 & @xmath50.23 & & 211.9 & & 226.6 & & 241.2 & & fe i & 6411.66 & 3.65 & @xmath50.60 & & 155.9 & & 166.3 & & 171.8 & & fe i & 6421.36 & 2.28 & @xmath51.98 & & 194.2 & & 206.6 & & 229.0 & & fe i & 6481.88 & 2.28 & @xmath52.94 & & 143.9 & & 142.7 & & & & 123.1 fe i & 6498.95 & 0.96 & @xmath54.66 & & 133.7 & & 136.1 & & 151.2 & & 155.9 fe i & 6518.37 * & 2.83 & @xmath52.56 & & 111.7 & & 109.6 & & 112.3 & & fe i & 6533.94 & 4.56 & @xmath51.28 & & 80.8 & & & & 73.6 & & fe i & 6574.25 * & 0.99 & @xmath54.96 & & & & & & & & 104.3 fe i & 6581.22 & 1.48 & @xmath54.68 & & & & & & 119.3 & & 103.8 fe i & 6593.88 * & 2.43 & @xmath52.30 & & 164.9 & & 164.0 & & 171.0 & & 168.0 fe i & 6608.04 & 2.28 & @xmath53.96 & & 67.2 & & 76.2 & & 75.9 & & 66.3 fe i & 6625.04 & 1.01 & @xmath55.32 & & & & & & 117.1 & & 123.7 fe i & 6627.56 & 4.55 & @xmath51.50 & & & & 58.5 & & 62.9 & & fe i & 6633.76 * & 4.56 & @xmath50.81 & & 94.3 & & & & & & fe i & 6703.58 & 2.76 & @xmath53.00 & & 84.1 & & 98.1 & & & & 88.1 fe i & 6713.74 & 4.79 & @xmath51.41 & & 58.7 & & & & 49.2 & & fe i & 6725.36 * & 4.10 & @xmath52.21 & & 41.9 & & & & & & 46.6 fe i & 6726.67 * & 4.61 & @xmath51.05 & & 82.5 & & 77.6 & & 88.2 & & 74.5 fe i & 6733.15 & 4.64 & @xmath51.44 & & 55.5 & & 53.8 & & 50.6 & & 67.4 fe i & 6750.16 * & 2.42 & @xmath52.58 & & 144.7 & & 152.3 & & 162.0 & & 146.5 fe i & 6786.86 * & 4.19 & @xmath51.90 & & 64.6 & & 55.9 & & 58.6 & & fe i & 6806.86 & 2.73 & @xmath53.14 & & & & & & & & 109.0 fe i & 6810.27 & 4.61 & @xmath51.00 & & & & & & & & 100.6 fe i & 6820.37 & 4.64 & @xmath51.16 & & & & & & & & 81.9 fe i & 6837.01 & 4.59 & @xmath51.71 & & 42.9 & & 58.1 & & 46.2 & & fe i & 6839.84 & 2.56 & @xmath53.35 & & 95.6 & & 97.0 & & 98.3 & & 106.1 fe i & 6843.66 & 4.55 & @xmath50.86 & & 85.0 & & & & 100.9 & & 92.7 fe i & 6858.16 & 4.61 & @xmath50.95 & & & & 79.1 & & & & 97.7 fe i & 6861.95 & 2.42 & @xmath53.78 & & 67.0 & & 77.6 & & 84.4 & & fe i & 6898.29 & 4.22 & @xmath52.08 & & 44.4 & & 43.8 & & 64.2 & & fe i & 6916.69 & 4.15 & @xmath51.35 & & 82.3 & & 99.5 & & 113.5 & & fe i & 6945.20 & 2.42 & @xmath52.46 & & & & & & & & 175.0 fe i & 6951.25 & 4.56 & @xmath51.05 & & & & & & & & 99.8 fe i & 6971.94 * & 3.02 & @xmath53.34 & & 63.4 & & 60.7 & & 64.1 & & fe i & 6978.86 & 2.48 & @xmath52.49 & & & & 142.3 & & 143.0 & & fe i & 6988.53 * & 2.40 & @xmath53.42 & & 108.2 & & 103.4 & & 108.4 & & 109.3 fe i & 7007.97 & 4.18 & @xmath51.80 & & 65.0 & & 60.7 & & 64.7 & & 82.0 fe i & 7010.35 & 4.59 & @xmath51.86 & & & & 46.7 & & & & 52.1 fe i & 7022.96 * & 4.19 & @xmath51.11 & & 110.9 & & 100.2 & & 108.6 & & 106.2 fe i & 7024.07 & 4.07 & @xmath51.94 & & & & 78.3 & & & & fe i & 7090.39 & 4.23 & @xmath51.06 & & & & & & & & 133.6 fe i & 7114.56 * & 2.69 & @xmath53.93 & & 46.5 & & & & & & fe i & 7118.10 * & 5.01 & @xmath51.52 & & & & 35.4 & & & & fe i & 7132.99 & 4.07 & @xmath51.66 & & 81.9 & & 87.2 & & 89.3 & & fe i & 7142.52 & 4.95 & @xmath50.93 & & & & & & 81.9 & & 89.1 fe i & 7151.47 & 2.48 & @xmath53.58 & & 101.2 & & & & 107.0 & & 101.3 fe i & 7189.16 * & 3.07 & @xmath52.77 & & & & 90.7 & & & & fe i & 7284.84 & 4.14 & @xmath51.63 & & 89.7 & & & & 96.1 & & fe i & 7306.57 & 4.18 & @xmath51.55 & & 78.8 & & 82.5 & & 93.0 & & fe i & 7401.69 & 4.19 & @xmath51.60 & & 71.2 & & & & 93.5 & & 100.2 fe i & 7411.16 & 4.28 & @xmath50.48 & & 153.3 & & 153.6 & & & & fe i & 7418.67 & 4.14 & @xmath51.44 & & 78.5 & & 85.3 & & 91.1 & & 103.0 fe i & 7421.56 * & 4.64 & @xmath51.69 & & 37.3 & & & & & & 36.0 fe i & 7430.54 * & 2.59 & @xmath53.82 & & 80.5 & & 79.4 & & 91.1 & & fe i & 7447.40 & 4.95 & @xmath50.95 & & & & 60.7 & & & & 63.0 fe i & 7461.53 & 2.56 & @xmath53.45 & & & & 85.0 & & & & 100.2 fe i & 7491.66 & 4.30 & @xmath51.01 & & 99.6 & & & & 111.2 & & fe i & 7540.44 & 2.73 & @xmath53.75 & & & & & & & & 74.2 fe i & 7547.90 & 5.10 & @xmath51.11 & & & & & & & & 44.1 fe i & 7568.91 * & 4.28 & @xmath50.90 & & 121.2 & & & & 133.2 & & 125.7 fe i & 7582.12 * & 4.95 & @xmath51.60 & & & & 30.0 & & & & fe i & 7583.80 & 3.02 & @xmath51.93 & & 150.8 & & 149.4 & & 165.5 & & 171.2 fe i & 7719.05 * & 5.03 & @xmath50.96 & & & & 55.0 & & & & 64.9 fe i & 7723.21 & 2.28 & @xmath53.62 & & 100.3 & & & & 108.2 & & 111.9 fe i & 7745.52 & 5.08 & @xmath51.14 & & & & & & & & 65.0 fe i & 7751.11 * & 4.99 & @xmath50.74 & & 82.6 & & 85.6 & & 94.8 & & fe i & 7807.91 & 4.99 & @xmath50.51 & & 84.5 & & 93.1 & & & & fe ii & 5414.08 & 3.22 & @xmath53.61 & & & & 42.4 & & & & fe ii & 5425.26 & 3.20 & @xmath53.27 & & 55.4 & & 61.0 & & & & fe ii & 5991.38 & 3.15 & @xmath53.55 & & & & 62.9 & & 65.1 & & fe ii & 6084.10 & 3.20 & @xmath53.80 & & & & & & 52.9 & & fe ii & 6149.25 & 3.89 & @xmath52.72 & & 52.6 & & 50.0 & & 51.7 & & 52.0 fe ii & 6247.56 & 3.87 & @xmath52.32 & & 61.5 & & 65.0 & & 69.1 & & 74.3 fe ii & 6369.46 & 2.89 & @xmath54.21 & & 27.2 & & 30.2 & & 36.5 & & 40.2 fe ii & 6416.93 & 3.89 & @xmath52.70 & & 52.4 & & 60.1 & & & & fe ii & 6432.68 & 2.89 & @xmath53.58 & & & & & & & & 67.7 fe ii & 6516.08 & 2.89 & @xmath53.38 & & & & & & 98.9 & & 77.2 ni i & 5468.11 & 3.85 & @xmath51.63 & & 28.8 & & & & 31.0 & & ni i & 5578.73 & 1.68 & @xmath52.57 & & 122.5 & & 130.1 & & 137.1 & & ni i & 5587.87 & 1.93 & @xmath52.39 & & 116.4 & & 125.9 & & 124.5 & & ni i & 5593.75 & 3.90 & @xmath50.78 & & 66.0 & & 66.4 & & & & ni i & 5643.09 & 4.16 & @xmath51.25 & & & & 33.7 & & & & ni i & 5748.36 & 1.68 & @xmath53.25 & & 98.6 & & 101.4 & & 102.3 & & ni i & 5760.84 & 4.10 & @xmath50.81 & & 82.8 & & 77.6 & & 73.1 & & ni i & 5805.23 & 4.17 & @xmath50.60 & & 82.3 & & & & & & ni i & 5847.01 & 1.68 & @xmath53.44 & & 92.6 & & 94.1 & & 103.2 & & ni i & 5996.74 & 4.23 & @xmath51.06 & & 74.8 & & 47.8 & & & & 41.7 ni i & 6053.69 & 4.23 & @xmath51.07 & & 54.9 & & 75.6 & & 78.9 & & ni i & 6086.29 & 4.26 & @xmath50.47 & & 70.4 & & 73.4 & & 89.4 & & ni i & 6108.13 & 1.68 & @xmath52.47 & & & & & & & & 151.1 ni i & 6111.08 & 4.09 & @xmath50.83 & & & & & & & & 75.0 ni i & 6128.98 & 1.68 & @xmath53.39 & & 100.4 & & 97.1 & & 96.3 & & 74.2 ni i & 6130.14 & 4.26 & @xmath50.98 & & & & & & 51.3 & & 26.6 ni i & 6176.82 & 4.09 & @xmath50.24 & & 100.2 & & 100.7 & & 106.6 & & ni i & 6177.25 & 1.83 & @xmath53.60 & & 70.6 & & & & 69.5 & & ni i & 6186.72 & 4.10 & @xmath50.90 & & 67.4 & & 57.8 & & 65.7 & & ni i & 6204.61 & 4.09 & @xmath51.15 & & 60.6 & & 59.1 & & 63.6 & & ni i & 6327.60 & 1.68 & @xmath53.08 & & & & & & & & 122.7 ni i & 6378.26 & 4.15 & @xmath50.82 & & 62.7 & & 67.5 & & 64.7 & & 55.6 ni i & 6384.67 & 4.15 & @xmath51.00 & & 63.4 & & 72.1 & & 61.5 & & ni i & 6482.81 & 1.93 & @xmath52.78 & & 105.0 & & 108.7 & & 125.4 & & 112.6 ni i & 6532.88 & 1.93 & @xmath53.42 & & 82.2 & & 88.2 & & 84.5 & & ni i & 6586.32 & 1.95 & @xmath52.78 & & 111.7 & & 111.6 & & 110.0 & & 104.8 ni i & 6598.61 & 4.23 & @xmath50.93 & & 50.9 & & 45.9 & & 55.7 & & ni i & 6635.14 & 4.42 & @xmath50.75 & & 65.8 & & 60.0 & & 48.2 & & 77.2 ni i & 6767.78 & 1.83 & @xmath52.06 & & 140.7 & & 153.3 & & 161.9 & & ni i & 6772.32 & 3.66 & @xmath50.96 & & 84.0 & & 88.9 & & 107.9 & & ni i & 6842.04 & 3.66 & @xmath51.44 & & 66.5 & & 71.3 & & 67.7 & & ni i & 7001.55 & 1.93 & @xmath53.65 & & 64.3 & & 61.7 & & 67.8 & & 49.7 ni i & 7030.02 & 3.54 & @xmath51.70 & & 46.3 & & & & & & 50.7 ni i & 7034.38 & 3.54 & @xmath52.10 & & 61.5 & & 48.6 & & 59.7 & & ni i & 7110.91 & 1.93 & @xmath52.91 & & 132.5 & & 130.5 & & 122.3 & & 116.6 ni i & 7327.65 & 3.80 & @xmath51.75 & & 36.7 & & 38.2 & & & & ni i & 7381.94 & 5.36 & @xmath50.05 & & & & & & & & 30.3 ni i & 7401.12 & 5.36 & @xmath50.18 & & 51.7 & & 46.5 & & & & ni i & 7422.29 & 3.63 & @xmath50.29 & & 146.9 & & 160.6 & & 157.9 & & 162.9 ni i & 7525.12 & 3.63 & @xmath50.67 & & & & & & & & 119.0 ni i & 7555.61 & 3.85 & @xmath50.12 & & & & & & & & 134.8 ni i & 7574.05 & 3.83 & @xmath50.61 & & 108.1 & & 118.3 & & 114.9 & & 131.4 ni i & 7715.58 & 3.70 & @xmath50.98 & & & & & & & & 110.6 ni i & 7727.62 & 3.68 & @xmath50.30 & & 127.5 & & 124.3 & & 135.0 & & ni i & 7797.59 & 3.30 & @xmath50.82 & & 125.1 & & & & 123.8 & & ni
i & 7863.79 & 4.54 & @xmath50.94 & & & & & & & & 54.2 zn i & 6362.35 & 5.79 & 0.14 & & 67.3 & & 60.9 & & 77.1 & & y ii & 5509.91 & 0.99 & @xmath51.01 & & 110.7 & & 121.6 & & & & ba ii & 5853.69 & 0.60 & @xmath51.00 & & 127.2 & & 127.5 & & 129.7 & & 126.1 ba ii & 6141.75 & 0.70 & 0.00 & & 169.4 & & 184.3 & & 199.7 & & 155.4 ba ii & 6496.91 & 0.60 & @xmath50.38 & & 171.0 & & 185.3 & & 177.9 & & 199.7 lrrrrr [ tab2 ] [ o / fe ] & @xmath50.352 & @xmath50.131 & @xmath50.142 & @xmath50.073 & 0.41 & 0.000 & @xmath50.058 & @xmath50.030 & @xmath50.003 & 0.06 & 0.050 & @xmath50.078 & 0.037 & 0.047 & 0.11 & @xmath50.125 & @xmath50.078 & @xmath50.063 & @xmath50.013 & 0.16 & @xmath50.100 & 0.019 & 0.008 & 0.045 & 0.11 & 0.062 & @xmath50.040 & @xmath50.021 & @xmath50.035 & 0.08 ii & 0.123 & @xmath50.031 & @xmath50.009 & @xmath50.008 & 0.13 i & 0.134 & @xmath50.016 & @xmath50.018 & @xmath50.014 & 0.14 & 0.128 & @xmath50.013 & @xmath50.028 & @xmath50.039 & 0.14 & 0.083 & @xmath50.012 & @xmath50.013 & 0.029 & 0.09 & 0.037 & @xmath50.028 & @xmath50.005 & @xmath50.033 & 0.06 i & 0.050 & 0.018 & 0.013 & @xmath50.087 & 0.10 ii & @xmath50.123 & 0.111 & 0.042 & @xmath50.062 & 0.18 & @xmath50.027 & 0.034 & 0.008 & @xmath50.001 & 0.04 ii & @xmath50.050 & 0.042 & 0.020 & @xmath50.093 & 0.12 llrllrllrllrllrllc [ tab4 ] fe i & & 93 & + 0.05 ( 0.16 ) & & 100 & + 0.08 ( 0.11 ) & & 90 & + 0.09 ( 0.14 ) & & 70 & + 0.04 ( 0.24 ) & & 27 & + 0.05 ( 0.20 ) & & 7.50 fe ii & & 5 & + 0.12 ( 0.14 ) & & 7 & + 0.15 ( 0.16 ) & & 6 & + 0.15 ( 0.21 ) & & 5 & + 0.10 ( 0.04 ) & & 1 & @xmath50.06 & & 7.44 & o i & & 4 & + 0.18 ( 0.29 ) & & 4 & + 0.15 ( 0.12 ) & & 3 & @xmath50.08 ( 0.05 ) & & & & & & & & 8.62 o i ( nlte ) & & 4&+0.16 ( 0.27)&&4 & + 0.13 ( 0.11 ) & & 3 & @xmath50.09 ( 0.03 ) & & & & & 3 & + 0.38 ( 0.18 ) & & 8.62 na i & & 4 & + 0.44 ( 0.13 ) & & 4 & + 0.51 ( 0.16 ) & & 4 & + 0.54 ( 0.16 ) & & 2 & + 0.19 ( 0.01 ) & & & & & 6.33 na i ( nlte ) & & 4 & + 0.42 ( 0.12 ) & & 4 & + 0.47 ( 0.16 ) & & 4 & + 0.51 ( 0.15 ) & & 2 & + 0.20 ( 0.04 ) & & & & & 6.33 mg i & & 3 & + 0.11 ( 0.17 ) & & 3 & + 0.05 ( 0.20)&&3 & + 0.27 ( 0.25 ) & & & & & 1 & + 0.20 & & 7.54 si i & & 14 & + 0.44 ( 0.25 ) & & 14 & + 0.39 ( 0.20 ) & & 13 & + 0.37 ( 0.22 ) & & 7 & + 0.24 ( 0.12 ) & & 2 & + 0.11 ( 0.26 ) & & 7.54 ca i & & 13 & + 0.23 ( 0.26 ) & & 11 & + 0.27 ( 0.14 ) & & 12 & + 0.28 ( 0.18 ) & & 8 & + 0.14 ( 0.17 ) & & 3 & + 0.07 ( 0.04 ) & & 6.16 sc ii & & 5 & @xmath50.07 ( 0.06 ) & & 6 & @xmath50.09 ( 0.08 ) & & 5 & + 0.09 ( 0.06 ) & & 3 & @xmath50.13 ( 0.31 ) & & 1 & + 0.12 & & 2.96 ti i & & 10 & + 0.08 ( 0.29 ) & & 9 & + 0.08 ( 0.10 ) & & 10 & + 0.05 ( 0.25 ) & & 6 & @xmath50.08 ( 0.30 ) & & 1 & @xmath50.28 & & 4.98 v i & & 6 & @xmath50.20 ( 0.13 ) & & 6 & @xmath50.25 ( 0.14 ) & & 6 & @xmath50.08 ( 0.21 ) & & 6 & @xmath50.29 ( 0.25 ) & & & & & 4.00 cr i & & 9 & + 0.05 ( 0.25 ) & & 13 & + 0.01 ( 0.22 ) & & 11 & @xmath50.06 ( 0.23)&&4 & + 0.00 ( 0.39 ) & & 1 & + 0.02 & & 5.62 mn i & & 1 & @xmath50.39 & & 4 & @xmath50.42 ( 0.18 ) & & & & & 3 & @xmath50.29 ( 0.14 ) & & & & & 5.39 ni i & & 36 & + 0.15 ( 0.21 ) & & 32 & + 0.14 ( 0.20 ) & & 31 & + 0.05 ( 0.17 ) & & 20 & + 0.06 ( 0.29 ) & & 9 & @xmath50.08 ( 0.17 ) & & 6.25 ba ii & & 3 & + 0.18 ( 0.20 ) & & 3 & + 0.24 ( 0.14 ) & & 3 & + 0.08 ( 0.01)&&3 & + 0.07 ( 0.38 ) & & & & & 2.21 lclclclrr [ tab6 ] fe i & + 0.07 & 0.02 & @xmath50.16 & 0.08 & @xmath50.55 & 0.20 & @xmath50.33 & fe ii & + 0.13 & 0.02 & @xmath50.18 & 0.10 & & & & & & o i & + 0.07 & 0.11 & + 0.50 & 0.13 & & & + 0.03 & 0.18 na i & + 0.40 & 0.12 & & & + 0.65 & 0.05 & + 0.21 & 0.37 mg i & + 0.14 & 0.09 & ( + 0.41 ) & ( 0.10 ) & + 0.33 & 0.13 & + 0.35 & 0.14 si i & + 0.36 & 0.07 & + 0.14 & 0.18 & + 0.35 & 0.05 & + 0.18 & 0.24 ca i & + 0.23 & 0.06 & + 0.26 & 0.09 & + 0.32 & 0.12 & + 0.14 & 0.17 sc ii & @xmath50.05 & 0.10 & ( @xmath50.12 ) & ( 0.18 ) & & & + 0.29 & 0.20 ti i & + 0.03 & 0.07 & + 0.19 & 0.06 & + 0.51 & 0.09 & + 0.34 & 0.10 v i & @xmath50.20 & 0.09 & & & & & + 0.06 & 0.19 cr i & 0.00 & 0.04 & + 0.04 & 0.09 & & & @xmath50.04 & 0.19 mn i & @xmath50.37 & 0.07 & & & & & & ni i & + 0.10 & 0.05 & + 0.01 & 0.07 & & & @xmath50.04 & 0.08 ba ii & + 0.14 & 0.07 & & & @xmath50.10 & 0.30 & + 0.20 & 0.28 vogt , s.e . , allen , s. , bigelow , b. , bresee , l. , brown , b. , cantrall , t. , conrad , a. , couture , m. , delaney , c. , epps , h. , hilyard , d. , hilyard , d. , horn , e. , jern , n. , kanto , d. , keane , m. , kibrick , r. , lewis , j. , osborne , c. , osborne , j. , pardeilhan , g. , pfister , t. , ricketts , t. , robinson , l. , stover , r. , tucker , d. , ward , j. & wei , m. , 1994 , spie , 2198 , 362 | we present the results of the first analysis of high dispersion spectra of four red hb stars in the metal rich globular cluster ngc 6528 , located in baade s window .
we find that the mean [ fe / h ] for ngc 6528 is @xmath0 dex ( error of the mean ) , with a star - to - star scatter of @xmath1 dex ( 4 stars ) , although the total error is likely to be larger ( @xmath2 dex ) due to systematic errors related to the effective temperature scale and to model atmospheres .
this metallicity is somewhat larger than both the mean abundance in the galactic bulge found by mcwilliam & rich ( 1994 ) and that found in our previous paper for ngc 6553 .
however , we find that the spectra of clump stars in ngc 6528 and ngc 6553 are very similar each other , the slightly different metal abundances found being possibly due to the different atmospheric parameters adopted in the two analyses .
since the present analysis is based on higher quality material , we propose to revise our previous published metal abundance for ngc 6553 to [ fe / h]=@xmath3 . for ngc 6528
we find excesses for the @xmath4-process elements si and ca ( [ si / fe]=+0.4 and [ ca / fe]=+0.2 ) , whereas mn is found to be underabundant ( [ mn / fe][email protected] ) .
we find a solar abundance of o ; however this is somewhat uncertain due to the dependence of the o abundance on the adopted atmospheric parameters and to coupling between c and o abundances in these cool , metal - rich stars .
finally , we find large na excesses ( [ na / fe]@xmath6 ) in all stars examined . |
for decades , quantum entanglement has been the focus of much work in the foundations of quantum mechanics , being particularly associated with quantum nonseparability , the violation of bell s inequalities , and the so - called einstein - pololsky - rosen ( epr ) paradox . beyond this fundamental aspect ,
creating and manipulating of entangled states are essential for quantum information applications . among these applications are quantum computation @xcite , quantum teleportation @xcite , quantum dense coding @xcite , quantum cryptography @xcite , and quantum positioning and clock synchronization @xcite .
hence , quantum entanglement has been viewed as an essential resource for quantum information processing .
recently much attention has been paid to continuous variable quantum information processing in which continuous - variable - type entangled pure states play a key role .
for instance , two - state entangled coherent states are used to realize efficient quantum computation @xcite and quantum teleportation @xcite .
two - mode squeezed vacuum states have been applied to quantum dense coding @xcite .
especially , continuous variable teleportation @xcite has been experimentally demonstrated for coherent states of a light field @xcite by using entangled two - mode squeezed vacuum .
it is also has been shown that a two - state entangled squeezed vacuum state can be used to realize quantum teleportation of an arbitrary coherent superposition state of two equal - amplitude and opposite - phase squeezed vacuum states @xcite .
continuous - variable - type entangled states including squeezed states and coherent states have also been widely applied to and quantum cryptography @xcite .
therefore , it is an interesting topic to create continuous - variable - type entangled pure states .
on the other hand , it is well known that atoms and photons can be viewed as carriers of storing and transmitting quantum information .
atoms are suited for storing quantum information at a fixed location , and they are sources of local entanglement for quantum information processing while photons are natural sources for communications of quantum information since they can transverse over long distances .
one topic of particular interest in this context is the possibility to generate atom - photon entangled states where one member is readily used to couple a distant system while the other is stored by the local sender . in this aspect
, some progress has been made , and possible schemes have been proposed in recent years .
for instance , moore and meystre @xcite proposed a scheme to generate atom - photon pairs via off - resonant light scattering from an atomic bose - einstein condensate ( bec ) .
deb and agarwal @xcite showed that it is possible to entangle three different many - particle states by bragg spectroscopy with nonclassical light in an atomic bec .
gasenzer and co - workers @xcite investigated quantum entanglement characterized by the relative number squeezing between photons and atoms coupled out from an atomic bec . optical control over the bec quantum statistics and atom - photon quantum correlation in an atomic bec @xcite have been widely studied . in this paper , we describe a method to produce atom - photon - entangled states of continuous - variable - type pure states in an atomic bec by electromagnetically induced transparency ( eit ) @xcite .
eit is a kind of quantum interference effects @xcite .
it arises in three - level ( or multilevel ) systems and consists of the cancellation of the absorption on one transition induced by simultaneous coherent driving of another transition .
the phenomenon can be understood as a destructive interference of the two pathways to the excited level and has been used to demonstrate ultraslow light propagation @xcite and light storage @xcite in many systems including atomic becs @xcite .
the proposed system in present paper consists of an atomic bec with three internal states , one weak probe laser , and one relatively strong coupling laser with appropriate frequencies .
the probe laser field is quantized while the coupling laser field is treated classically .
they form a three - level lambda configuration .
when the bec is in eit with the coupling laser being much stronger than the probe laser , the upper and intermediate levels are unpopulated due to quantum interference .
adiabatic elimination of the two unpopulated levels enables an effective two - mode model involving only atoms in the lowest internal state and photons in the probe laser .
we show that it is possible to generate atom - photon continuous - variable - type entangled pure states in the atomic bec .
this paper is organized as follows . in sec .
ii , we present the physical system under our consideration , review the reduction of the problem from four to two modes , followed by an analytic solution of the model , and then discuss the theoretical mechanism to create quantum entanglement in the proposed scheme . in sec .
iii , we focus on a dynamical approach to creating a variety of atom - photon entangled pure states . two - state ,
three - state , and four - state atom - photon continuous - variable - typed entangled pure states are generated explicitly .
we shall conclude our paper with discussions and remarks in the last section .
we consider a cloud of bec atoms which have three internal states labeled by @xmath1 , @xmath2 , and @xmath3 with energies @xmath4 , @xmath5 , and @xmath6 , respectively .
the two lower states @xmath1 and @xmath3 are metastable states in each of which the atoms can live for a long time .
they are raman coupled to the upper state @xmath2 via , respectively , a quantized probe laser field and a classical coupling laser field of frequencies @xmath7 and @xmath8 in the lambda configuration .
the interaction scheme is shown in fig .
1 . the atoms in these internal states are subject to isotropic harmonic trapping potentials @xmath9 for @xmath10 , respectively .
furthermore , the atoms in the bec interact with each other via elastic two - body collisions with the @xmath11-function potentials @xmath12 , where @xmath13 with @xmath14 and @xmath15 , respectively , being the atomic mass and the @xmath16-wave scattering length between atoms in states @xmath17 and @xmath18 .
a good experimental candidate of this system is the sodium atom condensate for which there exist appropriate atomic internal levels and external laser fields to form the lambda configuration which is needed for reaching eit under our consideration .
the sodium atom condensate in eit has been used to demonstrate ultraslow light propagation @xcite and amplification of light and atoms @xcite in atomic becs .
-shaped atoms coupled to a quantized probe laser and a classical coupling laser with the detunings @xmath19 and @xmath20 , respectively.,width=264 ] 0.2 cm the second quantized hamiltonian to describe the system at zero temperature is given by @xmath21 where @xmath22 and @xmath23 gives the free evolution of the probe laser field and the atomic fields respectively , @xmath24 describes the dipole interactions between the atomic fields and the probe and coupling fields , and @xmath25 represents inter - atom two - body collisional interactions .
the free atomic hamiltonian is given by @xmath26 \hat{\psi}_i({\bf x}),\ ] ] where @xmath27 are internal energies for the three internal states , @xmath28 and @xmath29 are the boson annihilation and creation operators for the @xmath30-state atoms at position @xmath31 , respectively , they satisfy the standard boson commutation relation @xmath32=\delta_{ij}\delta({\bf x}-{\bf x}')$ ] and @xmath33 = 0=[\hat{\psi}^{\dagger}_i({\bf x } ) , \hat{\psi}^{\dagger}_j({\bf x}')]$ ] .
the free evolution of the probe laser field is governed by the hamiltonian @xmath34 where @xmath7 is the frequency of the probe laser , and @xmath35 and @xmath36 are the photon creation and annihilation operators for the probe laser field , satisfying the boson communication relation @xmath37=1 $ ] .
the atom - laser interactions in the dipole approximation can be described by the following hamiltonian @xmath38,\end{aligned}\ ] ] where the two dipole coupling constants are defined by @xmath39 and @xmath40 with @xmath41 denoting a transition dipole - matrix element between states @xmath30 and @xmath42 , @xmath43 being the electric field per photon for the quantized probe light of frequency @xmath7 in a mode volume @xmath44 , and @xmath5 being the amplitude of the electric field for the classical coupling light of frequency @xmath8 , @xmath45 and @xmath46 are wave vectors of correspondent laser fields .
the collisional hamiltonian is taken to be the following form @xmath47,\end{aligned}\ ] ] where @xmath48 is the @xmath16-wave scattering length of condensed atoms in the internal state @xmath30 and @xmath49 that between condensed atoms in the internal states @xmath30 and @xmath42 .
for a weakly interacting bec at zero temperature there are no thermally excited atoms and the quantum depletion is negligible , the motional state is frozen to be approximately the ground state .
one may neglect all modes except for the condensate mode and approximately factorize the atomic field operators as @xmath50 where @xmath51 is a normalized wavefunction for the atoms in the bec in the internal state @xmath30 , which is given by the ground state of the following schrdinger equation @xmath52 { \phi}_i({\bf x})=\hbar\nu_i{\phi}_i({\bf x}),\ ] ] where @xmath53 is the energy of the mode @xmath17 , and @xmath54 denotes the frequency to the free evolution of the condensate in the internal state @xmath30 . substituting the single - mode expansions of the atomic field operators into eqs .
( [ 2 ] ) , ( [ 4 ] ) , and ( [ 5 ] ) , we arrive at the following four - mode approximate hamiltonian @xmath55 \nonumber \\ & & + \hbar\sum^3_{i=1}\lambda_i\hat{b}^{\dagger 2}_i\hat{b}^2_i + \hbar\sum_{i\neq j}\lambda_{ij}\hat{b}^{\dagger}_i\hat{b}_i \hat{b}^{\dagger}_j\hat{b}_j,\end{aligned}\ ] ] where one mode corresponds to the probe laser field , the other three to atomic fields in the three internal state . here
@xmath56 and @xmath57 denote the laser - atom dipole interactions , respectively , defined by @xmath58 and @xmath59 and @xmath60 ( @xmath61)describe inter - atomic interactions given by @xmath62 the hamiltonian ( [ 6 ] ) can be decomposed to the sum of an interaction hamiltonian and a free evolution hamiltonian defined by @xmath63 going over to an interaction picture with respect to @xmath64 , we transfer the time - dependent hamiltonian ( [ 6 ] ) to the following time - independent hamiltonian @xmath65 \nonumber \\ & & + \hbar\sum^3_{i=1}\lambda_i\hat{b}^{\dagger 2}_i\hat{b}^2_i + \hbar\sum_{i\neq j}\lambda_{ij}\hat{b}^{\dagger}_i\hat{b}_i \hat{b}^{\dagger}_j\hat{b}_j,\end{aligned}\ ] ] where @xmath66 and @xmath67 are the detunings of the two laser beams , respectively .
we consider the situation of the ideal eit which is attained only when the system is at the two - photon resonance with the two - photon detuning @xmath68 .
initially the lasers are outside the bec medium in which all atoms are in their ground state , i.e. , the internal state @xmath1 .
the condensed atoms are generally in a superposition state of the state @xmath1 and the state @xmath3 when they are in eit . however , when the coupling laser is much stronger than the probe laser , atomic population at the intermediate level approaches zero while the upper level is unpopulated @xcite . hence , under the condition of @xmath69 , the terms which involve @xmath70 and @xmath71 in the hamiltonian ( [ 8 ] ) may be neglected , and from the hamiltonian ( [ 8 ] ) we can adiabatically eliminate the atomic field operators in the internal states @xmath2 and @xmath3 .
firstly , the atomic operators in the internal state @xmath2 @xmath72 and @xmath73 can be adiabatically eliminated with the replacements : @xmath74 and @xmath75 , which come from the adiabatic elimination process in heisenberg equations @xmath76=0 $ ] and @xmath77=0 $ ] , respectively . from eq.([8 ] ) we can obtain the following effective hamiltonian @xmath78 which contains only atomic field operators in internal states @xmath1 and @xmath3 and the probe field operators .
here we have set @xmath79 and introduced @xmath80 the first part @xmath81 in the third term of eq .
( [ 9 ] ) describes such a quantum transition process where an atom in the lowest internal state @xmath1 absorbs a photon in the probe laser and then it transits to the intermediate internal state @xmath3 .
the second part @xmath82 describes the inverse process of this process .
then from the hamiltonian ( [ 9 ] ) the atomic field operators in the third internal state can be adiabatically eliminated through the replacements : @xmath83 and @xmath84 , and the final effective hamiltonian is given by @xmath85 the effective hamiltonian ( [ 11 ] ) is diagonal in the fock space with the bases defined by @xmath86 where @xmath87 and @xmath14 take non - negative integers . from the effective hamiltonian ( [ 11 ] ) we can well understand the theoretical mechanism to create atom - photon quantum entanglement in present scheme . comparing ( [ 11 ] )
with ( [ 8 ] ) we can see that for the bec in the eit , through adiabatically eliminating the atomic field operators at the upper and intermediate levels the laser - atom dipole interactions are converted as an effective collisional interaction between photons in the probe laser and condensed atoms in the internal state @xmath1 which can be described as effective elastic collisions between photons in the probe laser and condensed atoms in the internal state @xmath1 . from eq .
( [ 10 ] ) we can know that the effective scattering length to describe the atom - photon collisions is proportional to the ratio of the square of the dipole interaction and the two - photon detuning .
one can manipulate signs of the effective scattering length by changing signs of the two - photon detuning .
therefore , atom - photon quantum entanglement is produced by atom - photon and atom - atom collisional interactions described by the second and third terms in the effective hamiltonian ( [ 11 ] ) , respectively , and one can control or manipulate atom - photon entanglement by varying the dipole interaction strength between the probe laser and the bec and the two - photon detuning to create desired atom - photon entangled states .
it should be mentioned that in most textbook introductions to eit @xcite , both the probe and coupling lasers are treated classically , the decay of the excited state is included in the dynamics of the internal state through a set of density - matrix equations of the atomic system , the decay promotes the trapping of the atom into a dark state .
contrary to the usual treatment of eit , the probe laser is quantized in this paper .
a limitation of our present treatment is that we have ignored the decay rates of various levels .
however , this ignorance of the decay rates is a good approximation for the adiabatic eit that we study in the present paper , since the adiabatic eit is insensitive to any possible decay of the top level @xcite .
in this section we investigate generation of various atom - photon entangled states when the condensate in internal states @xmath1 and the probe laser are initially in non - entangled coherent states and superposition states of coherent states .
we assume that the probe laser and the condensate are initially uncorrelated and in the product coherent state @xmath88 . with the laser fields turned on at @xmath89 , then at time
@xmath90 the state of the system becomes @xmath91\ } \nonumber \\ & & \times \exp\left[-\frac{1}{2}(|\alpha|^2+|\beta|^2)\right]\frac{\alpha^n\beta^m}{\sqrt{n!m ! }
} |n , m\rangle.\end{aligned}\ ] ] in what follows we shall see that starting with the state ( [ 14 ] ) different atom - photon entangled states can be obtained at different times through adjusting various interaction strengths and the two - photon detuning . in order to do this ,
we rewrite the state ( [ 14 ] ) as the following simple form @xmath92 where we have used a scaled time @xmath93 and a running frequency @xmath94 where we have introduced a real effective interaction parameter @xmath95 defined by @xmath96 which describes an effective interaction induced by three adjustable parameters : the dipole interaction strength @xmath97 , the inter - atomic interaction strength @xmath98 , and the two - photon detuning @xmath99 .
the effective interaction parameter @xmath95 can takes positive or negative values which depends on the signs of @xmath98 and @xmath99 .
for an atomic bec with a positive @xmath16-wave scattering length , @xmath95 is positive ( negative ) when the two - photon detuning @xmath99 is positive ( negative ) . from eqs .
( [ 15 ] ) and ( [ 16 ] ) we can see that the time evolution characteristic of the system under our consideration is completely determined by the effective interaction parameter .
we note that the wavefunction of the system ( [ 15 ] ) is a two - mode extension of generalized coherent states @xcite .
these generalized coherent states differ from a conventional glauber coherent state by an extra phase factor appearing in the decomposition of such states into a superposition of fock states .
they can always be represented as a continuous sum of coherent states . and
under appropriate periodic conditions , they can reduce to discrete superpositions of coherent states .
it should be kept in mind that we use to create continuous - variable - type entangled states what we expect .
for the two - mode case under our consideration , we can express the state ( [ 15 ] ) as the following integral form @xmath100 where the phase function is given by @xmath101.\ ] ] eq .
( [ 19 ] ) indicates that the state @xmath102 is a continuous superposition state of two - mode product coherent states .
note that entangled states what we expect are generally superposition states of two - mode product coherent states . from eqs .
( [ 16 ] ) to ( [ 19 ] ) we can see that the values of the real parameter @xmath95 may seriously affect the form of the wavefunction ( [ 18 ] ) .
of particular interesting is a situation where @xmath95 may take values of nonzero integers . in this case , making use of ( [ 16 ] ) from ( [ 15 ] ) we can see that @xmath103 , which implies that the time evolution of the wavefunction ( [ 15 ] ) is a periodic evolution with respect to scaled time @xmath104 with the period @xmath105 . on the other hand ,
suppose that the scaled time @xmath104 takes its values in the following manner @xmath106 where @xmath107 and @xmath108 are mutually prime integers , then we can find @xmath109 = \exp\left[i2\pi\frac{m}{n } \theta_{n , m}\right],\ ] ] which means that the exponential phase function in the state ( [ 19 ] ) @xmath110 $ ] is a periodic function with respect to both @xmath87 and @xmath14 with the same period @xmath108 .
if @xmath104 takes its values according to eq .
( [ 20 ] ) , as a fraction of the period , then the wavefunction ( [ 18 ] ) becomes a discrete superposition state of product coherent states which can be expressed as follows @xmath111 where the two running phases are defined by @xmath112 from eqs .
( [ 15 ] ) and ( [ 22 ] ) we can find the following equation to determine the coefficients @xmath113 , @xmath114\right\}=1.\ ] ] carrying out summations over @xmath87 and @xmath14 in both sides of the above equation from @xmath115 to @xmath108 , and taking into account the normalization condition @xmath116 , we find the coefficients @xmath117 to be given by @xmath118\right\}.\ ] ] it is straightforward to see that the discrete superposition state ( [ 22 ] ) is generally an entangled coherent state with @xmath119 independent product coherent states . as a specific example of creating continuous - variable - type entangled states , in what follows we discuss generation of entangled states for the case of @xmath120 .
_ two - state entangled states .
_ assume that the probe laser and the condensate are initially in coherent states @xmath121 and @xmath122 , respectively . when @xmath120 , @xmath123 , and @xmath124 , from eqs .
( [ 16 ] ) and ( [ 25 ] ) we find that nonzero @xmath125-coefficients are @xmath126 which results in the following unnormalized two - state entangled state @xmath127,\ ] ] where @xmath128 are two normalized atomic schrdinger cat states , i.e. , even and odd coherent states defined by @xmath129 hence , the state ( [ 27 ] ) is an entangled state between two photon - coherent states and two atomic schrdinger cat states where one member is photons in the probe laser the other is atoms in the condensate . the degree of quantum entanglement of the two - state entangled states ( [ 27 ] ) can be measured in terms of the concurrence @xcite which is generally defined for discrete - variable entangled states to be @xcite @xmath130 where @xmath131 is the complex conjugate of @xmath132 .
the concurrence equals one for a maximally entangled state . in order to calculate the concurrence of continuous - variables - type entangled states like ( [ 27 ] ) , we consider a general biparticle entangled state @xmath133 where @xmath134 and @xmath135 are _ normalized _ states of subsystem @xmath115 and @xmath136 and @xmath137 normalized states of subsystem @xmath138 with complex @xmath139 and @xmath140 .
after normalization , the biparticle state @xmath132 can be expressed as @xmath141,\ ] ] where the normalization constant is given by @xmath142 assume that @xmath134 and @xmath135 ( @xmath136 and @xmath137)are linearly independent and span a @xmath138-dim subspace of the hilbert space .
then one can choose a discrete orthogonal basis @xmath143 as in ref @xcite latexmath:[\ ] ] where @xmath172 with @xmath173 , @xmath162 , @xmath174 and @xmath175 are unnormalized schrdinger cat states .
hence , the state ( [ 39 ] ) is an entangled state between two pairs of photon schrdinger cat states and two pairs of atom schrdinger cat states .
it is worthwhile to mention that besides their importance on quantum entanglement for above two - state , three - state , and four - state entangled states , we note that all of them are superposition states of a macroscopic number of atoms with a macroscopic number of photons . in this sense , we have also given a scheme for generating atom - photon schrdinger cat states .
the problem of creating a macroscopic superposition of atoms and photons raises an interest itself , because rather than two states of a given object , the atom - photon system is a seemingly impossible macroscopic superposition of two different objects .
this is beyond the usual macroscopic superposition of two states of a given object , the atom - photon system actually leads to a more counterintuitive situation since atoms and photons are different objects .
we have presented a scheme for the generation of atom - photon entangled states of continuous - variable - type pure states in the atomic bec which exploits quantum interference , or eit , in three - level atoms .
we have shown how to generate multi - state entangled coherent states when the atom - photon system is initially in an uncorrelated product coherent state . as examples
, the generation of two - state , three - state , and four - state entangled states has been investigated explicitly .
we have created not only atom - photon entangled coherent states , but also entangled states between photon ( atom)-coherent states and atom ( photon)-cat states , and between photon - cat and atom - cat states .
all of theses entangled states are mqs states of the atom - photon system .
they may be regarded as an extension of the usual schrdinger cat states of one given object to two different objects ( atoms and photons ) . in the proposed scheme we use the atomic bec with three internal states which are coupled with a weak quantized probe laser and a strong classical coupling laser .
when the bec is in the eit , since there is no atomic population at the upper and intermediate levels , the atomic field operators at these levels may be adiabatically eliminated .
the system then becomes an effective two - mode system in which one member is the probe laser ( photons ) , the other is the condensate ( atoms ) in the internal state @xmath1 . in this process
the ( probe ) laser - atom dipole interaction is converted as an atom - photon effective collisional interaction .
the strength of the atom - photon collisional interaction is determined by the dipole interaction and the two - photon detuning .
atom - photon entanglement is produced by inter - atomic and _ atom - photon effective collisional interactions_. this mechanism to create quantum entanglement is different with that in our previous analysis of generating atom - atom continuous - variable - type entangled states in a three - level @xmath0-shaped bec system @xcite .
atom - atom entanglement in ref .
@xcite is produced by inter - atomic collisional interaction and _ an inter - atomic effective tunnelling interaction _ generating by adiabatically eliminating the atomic field operators at the upper level , and no eit is required .
in present scheme the technical requirements involve two lasers with a well - controlled frequency difference , a controllable dipole interaction between the probe laser and bec , and a manipulable scattering length for atoms in the internal state @xmath1 .
these are met by most existing eit and bec experiments .
we hope that the proposed method for generating atom - photon continuous - variable - type entangled states can find its applications in quantum information processing and studies of fundamental problems of quantum mechanics .
it is possible to apply the approach to create entangled states in the present paper to study quantum entanglement between two distant condensates and quantum teleportation of quantum states of bec s , however , it is outside the scope of the present paper and will be discussed elsewhere .
kuang would like to thank professor yong - shi wu and dr .
guang - hong chen for useful discussions .
this work is supported by the national fundamental research program grant no .
2001cb309310 , the national natural science foundation grant nos .
90203018 and 10075018 , the state education ministry of china , the educational committee of hunan province , and the innovation funds from chinese academy of sciences via the institute of theoretical physics , academia , sinica .
d. p. divincenzo , science * 270 * , 255 ( 1995 ) ; d. gottesman and i. l. chuang , nature * 402 * , 390 ( 1999 ) .
p. w. shor , in _ proceedings of the symposium on the foundations of computer science , 1994 , los alamitos , california _ ( ieee computer society press , new york,1994 ) , pp . 124 - 134 ; l. k. grover , phys .
lett . * 79 * , 325 ( 1996 ) . c. h. bennett , g. brassard , c. crepeau , r. jozsa , a. peres , and w. k. wootters , phys .
lett . * 70 * , 1895 ( 1993 ) .
d. bouwmeester , j .- w .
pan , k. mattle , m. eibl , h. weinfurter , and a. zeilinger , nature ( london ) * 390 * , 575 ( 1997 ) ; d. boschi , s. branca , f. de martini , l. hardy , and s. popescu , phys .
lett . * 80 * , 1121 ( 1998 ) . c. h. bennett and s. j. wiesner , phys .
lett . * 69 * , 2881 ( 1992 ) ; k. mattle , h. weinfurter , p. g. kwiat , and a. zeilinger , phys .
. lett . * 76 * , 4656 ( 1996 ) ; s. l. braunstein and h. j. kimble , phys .
a 61 , 042302 ( 2000 ) . c. h. bennett , g. brassard , and n. d. mermin , phys .
lett . * 68 * , 557 ( 1992 ) .
v. giovannetti , s. lloyd , and l. maccone , nature * 412 * , 417 ( 2001 ) ; v. giovannetti , s. lloyd , and l. maccone , phys .
a 65 , 022309 ( 2002 ) .
h. jeong and m. s. kim , phys . rev .
a * 65 * , 042305 ( 2002 ) .
w. j. munro , g. j. milburn , and b. c. sanders , phys .
a * 62 * , 052108 ( 2000 ) .
s. j. van enk and o. hirota , phys .
a * 64 * , 022313 ( 2001 ) .
m. ban , j. opt .
b * 1 * , l9 ( 1999 ) .
s. l. braunstein and h. j. kimble , phys .
lett . * 80 * , 869 ( 1998 ) .
a. furusawa , j. l. srensen , s. l. braunstein , c. a.fuchs , h. j. kimble , and e. s. polzik , science * 282 * 706 ( 1998 ) .
x. h. cai and l. m. kuang , phys .
lett . a * 300 * , 103 ( 2002 ) ; l. zhou and l. m. kuang , phys .
a * 302 * , 273 ( 2002 ) .
t. c. ralph , phys .
a * 61 * , 010303(r ) ( 2000 ) ; phys .
a * 62 * , 062306 ( 2000 ) .
m. hillery , phys . rev .
a * 61 * , 022309 ( 2000 ) .
m. d. reid , phys .
a * 62 * , 062308 ( 2000 ) .
f. grosshans and p. grangier , phys .
* 88 * , 057902 ( 2002 ) .
d. gottesman and j. preskill , phys .
rev a , * 63 * , 022309 ( 2001 ) .
n. j. cerf , m. levy , and g. van assche , phys .
rev a * 63 * , 052311 ( 2001 ) .
m. g. moore and p. meystre , phys .
lett . * 85 * , 5026 ( 2000 ) .
b. deb and g. s. agarwal , phys .
a * 65 * , 063618 ( 2002 ) .
t. gasenzer , d. c. roberts , and k. burnett , phys .
a * 65 * , 021605 ( 2002 ) .
g. a. prataviera , e. v. goldstein , and p. meystre , phys .
a * 60 * , 4846 ( 1999 ) .
h. zeng , f. lin , and w. zhang , phys .
a * 201 * , 397 ( 1995 ) .
m. g. moore and p. meystre , phys .
a * 59 * , r1754 ( 1999 ) ; m. g. moore , o. zobay , and p. meystre , phys .
a * 60 * , 1491 ( 1999 ) .
l. m. kuang , commun .
theor . phys . *
30 * , 161 ( 1998 ) .
harris , phys . today * 50 * , no . 7 , 36 ( 1997 ) ; m. d. lukin and a. imamolu , nature * 413 * , 273 ( 2001 ) . m. o. scully and s. y. zhu , science * 281 * , 1973 ( 1998 ) .
hau , s.e .
harris , z. dutton , and c.h .
behroozi , nature ( london ) * 397 * , 594 ( 1999 ) .
s. inouye , r.f .
lw , s. gupta , t. pfau , a. grlitz , t.l .
gustavson , d.e .
pritchard , and w. ketterle , phys .
* 85 * , 4225 ( 2000 ) .
kash , v.a .
sautenkov , a.s .
zibrov , l. hollberg , g.r .
welch , m.d .
lukin , y. rostovtsev , e.s .
fry , and m.o .
scully , phys .
lett . * 82 * , 5229 ( 1999 ) .
d. budker , d.f .
kimball , s.m .
rochester , and v.v .
yashchuk , phys .
lett . * 83 * , 1767 ( 1999 ) .
turukhin , v.s .
sudarshanam , m.s .
shahriar , j.a .
musser , b.s .
ham , and p.r .
hemmer , phys .
lett . * 88 * , 023602 ( 2001 ) . c. liu , z. dutton , c.h .
behroozi , and l.v .
hau , nature ( london ) * 409 * , 490 ( 2001 ) .
phillips , a. fleischhauer , a. mair , r.l .
walsworth , and m.d .
lukin , phys .
86 , 783 ( 2001 ) .
scully and m.s .
zubairy , _ quantum optics _ ( cambridge university press , england , 1997 ) p.220 - 244 .
e. arimondo , prog . opt . * 35 * , 2570 ( 1996 ) ; s. alam , _ laser without inversion and electromagnetically induced transparency _ ( spie - the international society for optical engineering , bellingham , washington , 1999 ) .
l. m. kuang , g. h. chen , and y. s. wu , j. opt .
b : quantum semiclass . opt . * 5 * , 341 ( 2003 ) .
u. m. titulaer and r. j. glauber , phys . rev . *
145 * , 1041 ( 1966 ) .
z. bialynicka - birula , phys . rev . * 173 * , 1207 ( 1968 ) .
d. stoler , phys .
d * 4 * , 2309 ( 1971 ) ; a. vourdas and r. f. bishop , phys .
a * 39 * , 214 ( 1989 ) .
x. wang , j. phys .
a : math . gen .
* 35 * , 165 ( 2002 ) ; x. wang , phys .
a * 64 * , 022302 ( 2001 )
. s. hill , w. k. wootters , phys .
lett . * 78 * , 5022 ( 1997 ) ; w. k. wootters , phys .
lett . * 80 * , 2245 ( 1998 ) .
a. mann , b. c. sanders , and w. j. munro , phys .
a * 50 * , 989 ( 1995 ) .
p. rungta , v. buzek , c. m. caves , m. hillery , and g. j. milburn phys .
a * 64 * , 042315 ( 2001 ) .
b. yurke and d. stoler , phys .
lett . * 57 * , 13 ( 1986 ) .
l. m. kuang and l. zhou , submitted to phys . | in this paper , we present a method to generate continuous - variable - type entangled states between photons and atoms in atomic bose - einstein condensate ( bec ) .
the proposed method involves an atomic bec with three internal states , a weak quantized probe laser and a strong classical coupling laser , which form a three - level @xmath0-shaped bec system .
we consider a situation where the bec is in electromagnetically induced transparency ( eit ) with the coupling laser being much stronger than the probe laser . in this case
, the upper and intermediate levels are unpopulated , so that their adiabatic elimination enables an effective two - mode model involving only the atomic field at the lowest internal level and the quantized probe laser field .
atom - photon quantum entanglement is created through laser - atom and inter - atomic interactions , and two - photon detuning .
we show how to generate atom - photon entangled coherent states and entangled states between photon ( atom ) coherent states and atom-(photon- ) macroscopic quantum superposition ( mqs ) states , and between photon - mqs and atom - mqs states .
pacs number(s ) : 03.65.ta , 03.65.ud , 03.75.fi |
the most fundamental element of physics is the reduction principle .
the large variety of macroscopic forms of matter can be traced back according to this principle , to a few microscopic constituents which interact by a small number of forces .
the reduction principle has provided a guide to unraveling of the structure of physics from the macroscopic world through atomic and nuclear physics to particle physics .
the laws of nature are summarized in the standard model of the particle physics .
standard model is one of the successful model of the 20th century proposed by the glashow , weinberg and salam .
the weak and electromagnetic interactions are unified in the electroweak standard model .
this model has provided the plenty of successful predictions with an impressive level of precision . in quantum electrodynamics ( qed )
the interactions are specified by the gauge principle .
the electroweak standard model is based on the gauge symmetry group @xmath9 .
@xmath10 is non - abelian electroweak - isospin group , to which three @xmath11 gauge fields are associated and @xmath12 is the abelian hypercharge group . the associated @xmath13 field and the neutral component of the @xmath11 triplet field mix to form the photon field @xmath14 and a new neutral electroweak field @xmath15 .
this gauge group is spontaneously broken by the higgs mechanism ( see for example 2.3.1 ) . in the standard model ,
all particles acquire their masses by interaction with another particle , the higgs boson . using the knowledge of basic symmetry group ,
the gauge invariant lagrangian has been written , giving not only the interactions of the various fields including fermions but also the mass relationships for the fermions , gauge bosons and higgs boson ( see for example 2.3.2 ) . despite lot of success of the standard model
, no body can say that it is the end of physics .
there are some limitations ( see for example 2.4 ) which necessiate the extension of the standard model .
when we advertize the standard model , we have said that it is a model whose foundation is symmetry . in this model
only those reactions are allowed which conserve the individual as well as the total lepton - flavour .
thus the reactions violating the lepton flavour can not be contained in the standard model and cause its extension .
this extension is commonly known as physics beyond the standard model .
the problem of the physics beyond the standard model has been studied for a considerable length of time . in the past few years
substantial progress has been made to understand this physics .
lepton flavor violating ( lfv ) interactions are among the most promising candidates to understand this new physics .
using the experimental bounds on the three boby lepton flavor violating decays we have found the bounds on the two body lfv decays .
these bounds are suppressed by the form factors and was named as the dynamical suppression of the lfv bounnds .
the detailed study of these lfv decays is presented in chapter 3 .
the bounds thus found on the two body lfv decays are used for further studies of lfv decays in some interesting reactions .
today there is evidence for a very important new property of the neutrinos i.e. they have mass and , as a result mix with each other to lead the phenomenon of neutrino osscillation .
evidences that the neutrinos are massive particles come from three anomalous effects , the lsnd excess , the atmospheric anomaly and solar neutrino deficit .
but the atmospheric and the solar results are the most convincing one.the lsnd has the small probability compared to the atmospheric and solar anomalies . in order to incorporate neutrino mass
we must violate the standard model and introduce the new concept of lfv . a seesaw model of neutrino masses @xcite which involve right handed singlet neutrinos of mass @xmath16 of the order tev , is also disscussed using the lfv physics at loop order .
the main purpose of this model is that there is no new physics beyond @xmath5 tev .
it is however , shown that even if one keeps @xmath16 at @xmath5 tev , the present bounds on @xmath0 requires that new higgs boson , necessary in this model , to have mass larger than @xmath7 tev .
this necessiates a new physics beyond the tev energy scale defeating the orgional motivation of the model .
this is discussed in chapter 4 , where it is also shown that the lfv anamolous muou decay @xmath17 can not significantly contribute to the lsnd dar result , at least when we use @xmath1 as constrained by @xmath0 in a model independent way .
the proposal of the symmetry group for the electroweak theory , @xmath18 , was made by glashow in 1961 .
his motivation was to unify weak and electromagetic interactions into a symmetry group that contained @xmath19 .
the prediction includes the existance of four physical vector boson eigenstates , @xmath20 @xmath21 and @xmath22 obtained from the rotations of the weak eigenstate .
in particular , the rotation by the weak angle @xmath23 which defines the @xmath15 weak boson was introduced already in this work .
furthermore the correct structure of weak neutral current mediated by @xmath15-boson was also obtained.the massive weak bosons @xmath24 and @xmath15 were considered as mediators of weak interactions .
this model has serious problem of giving masses to @xmath24 and @xmath15 , since the gauge symmetry would predict their masses to be zero.the vector boson masses @xmath25 and @xmath26 were parameters introduced by hand and the interaction lagrangian was that of the ivb theory .
the mass term for vector bosons in the lagrangian not only destroyed the gauge symmetry but also normalizability of the theory .
another key ingredient for the building of the electroweak theory is provided by the goldstone theorem which was introduced by nambu in 1960 and proved and studied with generality by goldstone in 1961 and by goldstone , salam and weinberg in 1962 .
this theorem states the existance of massless spinless particles as an implication of spontaneous symmetry breaking of global symmetries .
the spontaneous symmetry breaking of local ( gauge ) symmetries , needed for the breaking of the electroweak symmetry @xmath9 , was studied by p.higgs , f. englert and r. brout , guralnik , hagen and kibble in 1964 .
the procedure for this spontaneous breakdown of gauge symmetries is reffered to as the higgs mechanism .
the electroweak theory as it is now known was formulated by weinberg and independently by salam in 1968 who incorporated the gauge group @xmath18 introduced by the glashow earlier .
this theory , commonly called glashow - weinberg salam model or standard model ( sm ) , was built with the help of the gauge principle and incorporated all the good phenomenological properities of the pregauged theories of the weak interactions , and in particular those of the ivb theory .
it incorporated the idea of spontaneous breaking of the gauge symmetry by introducing higgs doublet . in this way the weak vector bosons acquaired their masses with out destroying the normalizability of the gauge
theory.the sm is indeed a gauge theory of the electroweak interactions based on the gauge symmetry @xmath9 and the intermediate vector bosons , @xmath27 and @xmath15 are the four associated gauge bosons .
the gauge boson masses , @xmath25 and @xmath26 , are generated by the higgs mechanism in the electroweak theory and , as a consequence , it respects unitarity at all energies and is renormalizable .
the important proof of renormlizability of gauge theories with and without spontaneous symmetry was provided by t hooft in 1971 . the first firm indication that the standard model was the correct theory of electroweak interaction was the discovery of weak neutral current in 1973 as predicted by the model .
this also provided the first measurement of @xmath28 . by using this experimental input for @xmath29 and the values of electromagnetic coupling and @xmath30 ,
the sm provided the first estimates for @xmath31 and @xmath32 which were discovered experimently in 1983 at the predicted masses .
another important ingredients of the sm are : fermion family replication , quark mixing and cp violation .
the success of the sm was clearly the discovery of the gauge bosons @xmath33 and @xmath15 at the sps collider at cern in 1983 . since then there have been plenty of tests of the sm even at quantum level @xcite . in order to give an argument for the choice of @xmath9 gauge group for electroweak unification , it is sufficient to consider the @xmath34 component of the charged weak current that we write now in the form , @xmath35 and we have introduced the lepton doublet notation and the @xmath36 ( @xmath37 ) are pauli matrices , @xmath38 the three generators @xmath39 of unitary , unimodular group @xmath10 satisfy the commutation relation @xmath40 = i\epsilon _ { ijk}i_{k}\ ] ] i.e. the same commutation relation as satisfied by @xmath36 viz @xmath41 = 2i\epsilon _ { ijk}\tau _ { k},\ ] ] since for the fundamental representation of @xmath10 @xmath42 note that in the charged currents there are just two generators @xmath43 and @xmath44 . a third generator @xmath45 is needed in order to close the @xmath10 algebera .
this implies the third current that is relevent for the electroweak interactions , @xmath46 obviously this can not be identified with the @xmath47 which is @xmath48 .
this clearly indicate that @xmath10 is not sufficient for electroweak unification and it must be extended and the simple extension is to consider the group @xmath49 the gell - mann nishijima relation , @xmath50 where , @xmath51 electric charge , @xmath52 weak isospin , @xmath53 weak hypercharge ; implies @xmath54 the corresponding relation amoung the current is @xmath55 thus @xmath56 therefore , if the following are used as inputs @xmath57 one can get @xmath58 as output , @xmath59 this clearly indicates that @xmath60 due to symmetry breaking the two neutral currents @xmath61 and @xmath62 will mix to give two physical currents of which one must be identified with electromagnetic current @xmath63 and the second current will be new currents .
these currents will be coupled to physical vector bosons @xmath64 and @xmath65 .
@xmath66 where , @xmath67 from here @xmath68 thus , eq . ( [ a1 ] ) becomes @xmath69 + \frac{1}{2}\acute{g}j_{\mu } ^{y}\left [ \cos \theta _ { w}a_{\mu } -\sin \theta _ { w}z_{\mu } \right ] & = & ej_{\mu } ^{em}a_{\mu } + g_{z}j_{\mu } ^{z}z_{\mu } .
\\ & & \\ & & \end{aligned}\ ] ] so , @xmath70 \\ & = & \frac{1}{2}g_{2}\sin \theta _ { w}\left [ -\bar{e}_{l}\gamma _ { \mu } e_{l}+\bar{\nu}_{l}\gamma _ { \mu } \nu _ { l}\right ] \\ & & + \frac{1}{2}\acute{g}\cos \theta
_ { w}\left [ ( -1)\bar{e}_{l}\gamma _ { \mu } e_{l}+(-2)\bar{e}_{r}\gamma _ { \mu } e_{r}+(-1)\bar{\nu}_{l}\gamma _ { \mu } \nu _ { l}\right ] .\end{aligned}\ ] ] this simply gives us @xmath71 hence @xmath72 similarly @xmath73 .\end{aligned}\ ] ] this will gives us the neutral current @xmath74and the corresponding coupling @xmath75 .
this is the main indication of the electroweak unification .
the electroweak standard model is the commonly accepted theory of the fundamental electroweak interactions @xcite .
when we want to advertise the virtue of the standard model , we say that it is a model whoes foundation is symmetry @xcite .
it is a gauge invariant quantum field theory based on the symmetry group @xmath9 , which is spontaneously broken by the higgs mechanism .
the electroweak standard model consists of three components .
1 ) : the basic constituents of matter are leptons and quarks which are realized in three families of identical structure : @xmath76 we will concentrate on the leptonic sector only .
2 ) : four different forces act between leptons and quarks : the electromagnetic and weak forces are unified in the standard model .
the fields associated with these forces are spin @xmath5 fields , describing the photon @xmath77 and the electroweak gauge bosons @xmath24 and @xmath15 .
3 ) : the third component of the standard model is the higgs mechanism . before going to the deep discussion of the standard model , we have to require some theoretical basis @xcite .
the fundamental forces of the electroweak standard model , the electromagnetic and the weak force , are mediated by gauge fields .
the concept could consistantly be extended to massive gauge field by introducing the higgs mechanism which generates masses with out destroying the underlying gauge symmetries of the theory @xcite .
[ [ gauge - sector ] ] * 1 ) : gauge sector * + + + + + + + + + + + + + + + + + + gauge invariant theories are invariant under gauge transformations of fermion fields : @xmath78 .
@xmath79 is either a phase factor for abelian transformations or unitary matrix for non - abelian transformation acting on the multiplets of the fermion field @xmath80 . now
if the theory guarantee the local transformation for which @xmath79 depends on the space time point @xmath81 , the usual space - time derivatives @xmath82 must be extended to covariant derivatives @xmath83 which includes a new vector field @xmath84 : @xmath85 @xmath86 defines the universal gauge coupling of the system . under local gauge transformations
the gauge field @xmath84 is transformed by a rotation plus a shift : @xmath87 u^{-1}.\ ] ] but in contrast to this , the curl @xmath88 of @xmath84 , @xmath89\ ] ] is just rotated under gauge transformation .
the lagrangian describing the systen of spin@xmath90 fermions and vectorial gauge bosons for massless particles can be written in the compact form as follows : @xmath91 = \bar{\psi}id\psi -\frac 12trf^2\ ] ] it incorporates the following interactions : fermion - gauge bosons @xmath92 three bosons couplings @xmath93 .\ ] ] four boson couplings @xmath94 ^2.\ ] ] [ [ higgs - mechanism ] ] * 2 ) : higgs mechanism * + + + + + + + + + + + + + + + + + + + + + what is called the higgs mechanism is the extension of the spontaneous symmetry breaking to creat massive vector bosons in a gauge invariant theory . as the sm is a gauge theory
, the @xmath9 gauge invariance requires masses of the gauge bosons to be zero , since the presence of an explicit mass term for the gauge bosons in the lagrangian violates gauge invariance .
the higgs mechanism circumvents this constraint by begining with a gauge invariat theory having massless gauge bosons.the @xmath24 and @xmath95 masses were generated by spontaneously breaking the local gauge symmetry from @xmath96 , which was achieved by the introduction of a self - interacting complex scalar field , @xmath97 , transforming as an @xmath10 doublet . the doublet field @xmath97 and
its complex conjugate togather comprise four independent fields .
spontaneous symmetry breaking was implemented by giving one of the neutral fields a nonzero vacuum expectation value , @xmath98 of the four fields in the lagrangian before spontaneous symmetry breaking , three fields become the longitudinal degrees of freedom of the vector bosons @xmath24and @xmath95 ; the photon coupled to the remaining symmetry group @xmath19-generators , remains massless .
one neutral scalar particle remains in the physical sector of the theory .
this is the so - called salam - weinberg higgs particle , which the @xmath18 model predicts to exist .
since the same higgs doublet is used to give masses to the bosons and fermions , which have yukawa couplings with the scalar fields , the @xmath9 model predicts the couplings of the higgs particles with all the known bosons and fermions but makes no prediction about its mass .
this could be traced back to the fact that in salam - weinberg theory , the higgs particle mass is function of the unknown quartic higgs - boson coupling constant .
[ [ the - matter - sector ] ] * the matter sector * + + + + + + + + + + + + + + + + + + + the fundamental fermions , as families with left handed isospin doublets and right handed isospin singlets appear in the fundemental representation of the the group @xmath9 .
it is realized that the symmetry pattern in the first , second and third generation of the fermions is same , @xmath99 _
l \begin{array}{c } \nu _ { er } \\ e_r^{- } \end{array } ; \left [ \begin{array}{c } \nu _ \mu \\
\mu ^{- } \end{array } \right ] _
l \begin{array}{c } \nu _ { \mu r } \\ \mu _ r^{- } \end{array } ; \left [ \begin{array}{c } \nu _ \tau \\ \tau ^{- } \end{array } \right ] _
l \begin{array}{c } \nu _ { \tau r } \\
\tau _ r^{- } \end{array}\ ] ] the symmetry structure can not be derived in the standard model .
it is an experimental fact that in weak interactions the parity is not conserved .
the different isospin assigned to the left handed and right handed field allows for maximal parity violation in the weak interactions .
so the experimental observation is incorporated in the natural way .
the relationship between the electric charge @xmath100 and basic quantum numbers is described by gell - mann - nishijima relation @xmath101 [ [ interactions ] ] * interactions * + + + + + + + + + + + + + + the interactions of the standard model are summarized by the three terms in the basic lagrangian : @xmath102 which are specified in the following way ; [ [ gauge - fields ] ] * gauge fields * + + + + + + + + + + + + + + @xmath9 is a non - abelian group which is generated by the isospin operators @xmath103 , @xmath104 , @xmath105 and the hypercharge @xmath106 .
each of these generlized charges is associated with a vector field : a triplet of vector fields @xmath107 with @xmath108 and a singlet field @xmath109 with @xmath106 .
the isotriplet @xmath110 , @xmath111 and isosinglet @xmath109 lead to the field strength tensors @xmath112 @xmath113 is defined as the coupling constant for non - abelian gauge group @xmath10 using the equation ( [ 202 ] ) the pure gauge field langragian can be written as follows @xmath114 it is invariant under the non - abelian @xmath9 transformation . [
[ fermion - fields - and - fermion - gauge - interactions ] ] * fermion fields and fermion - gauge interactions * + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + the left - handed fermion fields of each lepton family @xmath115 with family index @xmath116 are grouped into @xmath10 doublets with component index @xmath117 , and the right - handed fields into singlets @xmath118 each left and right - handed multiplet is an eigenstate of the weak hypercharge @xmath106 such that the relation ( [ 202 ] ) is fulfilled .
the covariant derivative @xmath119 induces the fermion - gauge field interaction via the minimal subsitution rule @xmath120 @xmath121 is the coupling constant for the abelian @xmath12 gauge group .
[ [ higgs - field - and - higgs - interaction ] ] * higgs field and higgs interaction * + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + for spontaneous breaking of the @xmath9 symmetry leaving the electromagnetic gauge group @xmath12 unbroken , a single complex scalar doublet field with hypercharge @xmath122@xmath123 is coupled to the gauge fields through @xmath124 with the covariant derivative @xmath125 the higgs field self interaction @xmath126 is constructed in such a way that @xmath127 has a non vanishing vacuum expectation value , i.e. @xmath128 with @xmath129 the field ( [ 206 ] ) can be written in the following way , @xmath130 where the field components @xmath131 , @xmath132 , @xmath133 have vacuum expectation values zero . exploiting the invariance of the langragian ,
the components @xmath131 , @xmath133 can be gauged away ; this means that they are unphysical ( higgs ghosts or would be goldstone bosons ) . in this particular gauge ,
the unitarity gauge , the higgs field has simple form @xmath134 the real field @xmath135 which describes small osscillations about the ground state defines the physical higgs field .
the higgs field components have triplet and quartic self couplings following from @xmath136 and couplings to the gauge fields via the kinetic term of eq.([207 ] ) .
in addition , yukawa couplings of the fermions are introduced in order to make the fermion massive .
the lagrangian for the yukawa term can be written as follows @xmath137 the fermion mass terms follow from the @xmath138- part of @xmath139 @xcite .
the lagrangian @xmath140 summarizes the laws of physics for the electromagnetic and the weak interactions between the leptons , and it predicts the form of self - interaction between the gauge fields .
morover , the specific form of the higgs interaction generates the mass of the fundamental particles , the leptons , the gauge bosons and the higgs boson itself , and it predicts the interactions of the higgs particle @xcite . [
[ masses - and - mass - eigenstates - of - particles ] ] * masses and mass eigenstates of particles * + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + in the unitary gauge the mass terms are extracted by substiuting @xmath141\ ] ] in the basic higgs lagrangian ( [ 207 ] ) .
the apparent @xmath10 seems to be lost thereby , but only superficially so and remain present in the hidden form ; the resulting lagrangian preserves an apparent local @xmath142 gauge symmetry which is identified with the electromagnetic gauge symmetry : @xmath143 @xcite . [
[ gauge - bosons ] ] * gauge bosons * + + + + + + + + + + + + + + the mass matrix of the gauge boson in the basis @xmath144 takes the form @xmath145 this gives the mass of the vector boson in non diagonal form .
the mass of the charged weak bosons is obvious @xmath146 as eigenstates related to the two masses @xmath147 the charged @xmath24 boson state may be defined as @xmath148 .\ ] ] for the neutral bosons , the mass term gives the matrix @xmath149 since @xmath150 , therefore one of the eigenvalue of @xmath151 is zero .
the above matrix is diagonalized by defining the fields @xmath64 , @xmath65 : @xmath152 in matrix form the above equations can be written as follows ; @xmath153 then we get @xmath154 where @xmath155 i.e. the electroweak mixing angle @xmath156 is defined by the ratio of the @xmath10 and @xmath12 couplings .
introduce a parameter @xmath157 now using the value of @xmath158 , we get @xmath159 this is the consquence of the fact that higgs field is a doublet under @xmath160 @xcite .
experimently the mixing angle turns out to be large , i.e. @xmath161 . the fact that the experimental value for @xmath162 is far away from the limits @xmath163 , or @xmath5 , indicate a large mixing effect .
this supports the interpretation that the electromagnetic and the weak interactions are indeed manifestations of a unified electroweak interaction even though the underlying symmetry group @xmath164 is not simple .
it may therefore be concluded that the electromagnetic and the weak interactions are truly unified in the glashow - salam - weinberg theory of the electroweak interactions .
the ground - state value of the higgs field is releated to the fermi coupling constant . from the low - energy relation @xmath165 in @xmath166 decay and
combine with the mass relation @xmath167 , the value of @xmath138 can be derived : @xmath168^{1/2 } \nonumber \\ & \simeq & 246gev\end{aligned}\ ] ] the typical range for electroweak phenomena , defined by the weak masses @xmath31 and @xmath32 , is of the order @xmath169 . [ [ fermions ] ] * fermions * + + + + + + + + + + the leptons are endowed with mass by means of yukawa interactions with the higgs ground state : @xmath170 though the masses of chiral fermion fields can be introducd in a consistant way via the higgs mechanism , the standard model does not provide predictions for the experimental values of the yukawa couplings @xmath171 and , as consquence , of the mases .
the theory of the masses is not available yet .
[ [ the - higgs - bosons ] ] * the higgs bosons * + + + + + + + + + + + + + + + + + + the real field @xmath135 which describes small osscillations about the ground state tells us the mass of the physical neutral scalar particles with mass @xmath172 it can not predicted in the standard model since the quartric coupling @xmath173 is an unknown parameter .
we conclude this session with the folowing remarks : 1 ) .
a definite prediction of electroweak unification is the existance of weak neutral currents with the same effective couplings as charged currents .
this current has been found experimentaly .
the existance of the vector bosons @xmath24 , @xmath15 with definite masses which have also been discovered . 3 ) .
the theory has one free parameter @xmath174 there is no confirmed experimental evidence from accelerators against the standard model , and several possible extensions have been ruled out .
nevertheless , there is no thinking physicst could imagine that the standard model is the end of physics .
even if one accepts the strange set of group representations and hypercharge that it requires , the standard model contains at least @xmath175 parameters .
moreover , many more parameters are required if one wishes to accommodate non - accelerator observations .
for example , the neutrino masses and mixing introduce at least @xmath176 parameters : @xmath177 masses , @xmath178 mixing angles and @xmath5 cp violating phase @xcite . in this chapter , the @xmath10 and @xmath12 gauge group have been reviewed , which provides a back ground knowledge for the understanding of the electroweak standard model .
the concept of how vector bosons @xmath24 , @xmath15 acquire masses have also been discussed in this chapter .
the neutrino masses and mixings which are the weaknesses of the standard model have been mentioned here and will be studied in detail in the last chapter .
9 sven bergmann and yuval grossman , hep - ph/9809524 .
m. herrero , hep - ph/9812242 .
the electroweak standard model , w. hollik . michael e. peskin , beyond the standard model , page 49 . h. spiesberger , m. spira and p. m. zerwas . john f. gunion , howard e. haber , gordon kane and sally dawson , the higgs hunter s guide .
fayyazddin and riazuddin , a modern introduction to particle physics .
john ellis , hep - ph/9812235 .
it is generally believed that the standard model of electroweak interactions is a low - energy approximation to a more fundamental theory .
yet there is no clear experimental evidence either to guide its extension to additional physical processes or to predict the model parameters .
the standard model incorporates the lepton family - number conservation , which has been empirically verified to high precision but is not a consequence of a known gauge theory .
indeed many theoretical extensions to the standard model allow lepton family - number violation within a range that can be tested by experiment @xcite .
the predications of the rate for a given family - number nonconserving process vary among these extensions , and thus provide a test of the model .
many possibilities have been explored , and the present experimental limits for a wide variety of the processes have been tabulated .
thus the lepton flavour violation ( lfv ) processes that are strongly suppressed in the standard model by powers of ( small ) neutrino masses may provide signals for new physics . at present
we have strigent bound for @xmath179 decays , e.g. @xmath180 and some what weaker @xmath181 bounds on lfv @xmath182 decays @xcite .
the possibility of large samples of decaying vector bosons [ @xmath183 and @xmath184 and the clear signature provided by @xmath185 final states suggest searching for lfv two - body decays @xmath186 here we show that rather simple considerations based on unitarity , provide rather strong constraints on two - body lfv processes .
hence , most three - body @xmath179 and @xmath182 lfv decays are likely to provide more sensitive tests of lepton flavour violation , rather than the two - body decays
. let us assume that a vector boson @xmath187 ( here @xmath187 could be either a fundamental state , such as the @xmath95 or the quark- antiquark bound state such as the @xmath188 or @xmath189 ) couples to @xmath190 .
if it couples also to @xmath191 , as all the states above do , then by unitarity @xmath136 exchange contributes also to @xmath0 .
let us define the effctive couplings between vector boson @xmath187 and @xmath185 as @xmath192 , and the corresponding effective lagrangian can be written as @xmath193 from the feynman diagram , the amplitude @xmath194 can be written as @xmath195 here @xmath196 is the effective coupling of the vector boson @xmath187 to @xmath191 , while @xmath197 ( there are , of course , also potential axial vector couplings of @xmath136 to @xmath191 , which contribute to this process .
these can be included in the above , but as we shall see they do not change qualitative discussion ) .
since we are dealing with the low energy process , as a first approximation , it is sensible to neglect @xmath198 in comparision with @xmath199 .
therefore the eq .
( [ 302 ] ) takes the form , @xmath200 in order to calculate the decay width we have to calculate @xmath201 which is given by @xmath202 taking the spin average for the initial state and summation over spins for final state particles , we get @xmath203 \right .
\nonumber \\ & & \left . \times tr\left [ \gamma ^\alpha ( \not{p}+m_{_\mu } ) \gamma ^\rho \left ( \not{k}_3+m_e\right ) \right ] \right ] \left ( \frac{\tilde{g}_{v_{\mu e}}g_{v_{ee}}}{m_v^2}\right ) ^2 .
\label{309}\end{aligned}\ ] ] assuming that the external particles ( electrons ) are massless .
therefore , eq . ( [ 309 ] ) takes the form @xmath204 .
\label{314}\ ] ] we next carry the phase space integrations , starting with the integrals over the electrons momenta , given by @xmath205 where @xmath206 it follows from the lorentz covariance of the integral ( [ 315 ] ) that the most general form is @xmath207 from this equation it follows that @xmath208 this can be obtained by using eq .
( [ 317 ] ) . since we have been taking the electron masses to be zero so that @xmath209 and , on account of the @xmath210-function in ( [ 315 ] ) , @xmath211 in order to find @xmath212 and @xmath213 , we calculate the expersion on the left hand sides of ( [ 318 ] ) . from eq .
( [ 315 ] ) we obtain @xmath214 using eq .
( [ 319 ] ) , the above equation takes the form @xmath215 we see from its defination that the integral @xmath216 is an invariant , so that it can be evaluated in any coordinate system . for our convienence
we shall choose the centre - of - mass system of two electrons .
in this system @xmath217 and for the massless electron @xmath218 also @xmath219 so from above two results , we can write @xmath220 hence by removing the integration on @xmath221 , we get @xmath222 and from eq .
( [ 321 ] ) @xmath223 similarly , calculating @xmath224 , one gets @xmath225 this is obtained by using eq .
( [ 315 ] ) . also using eq .
( [ 319 ] ) , we got @xmath226 substituting eqs .
( [ 324 ] ) and ( [ 327 ] ) in eq .
( [ 315 ] ) , we get @xmath227 solving ( [ 328 ] ) and ( [ 329 ] ) for the vaules of @xmath212 and @xmath213 , we have @xmath228 hence eq .
( [ 317 ] ) leads to @xmath229 then the partial muon decay rate can be written as @xmath230 .
\label{331}\end{aligned}\ ] ] using ( [ 330 ] ) , ( [ 331 ] ) becomes @xmath231 \nonumber \\ & = & \left ( \frac{\tilde{g}_{v_{\mu e}}g_{v_{ee}}}{m_{v}^{2}}\right ) ^{2}\frac{1}{\left ( 2\pi \right ) ^{5}}\frac{1}{2e_{\mu } } \frac{d^{3}{\bf k}_{{\bf 3}}}{e_{3}}\frac{\pi } { 6 } \nonumber \label{32 } \\ & & \times 4\left [ \left ( p\cdot k_{3}\right ) q^{2}+2\left ( k_{3}\cdot q\right ) \left ( p\cdot q\right ) \right ] .
\label{332}\end{aligned}\ ] ] this is obtained by using eq .
( [ 315 ] ) .
finally , we must integrate eq .
( [ 332 ] ) over all momenta @xmath232 of the emitted electron . for a muon at rest ,
i.e. in the rest frame of muon , we have @xmath233 also @xmath234 now , we will calculate the terms involving in eq .
( [ 332 ] ) , i.e. @xmath235 by substuting these values eq .
( [ 332 ] ) takes the form @xmath236
\nonumber \\ & = & \left ( \frac{\tilde{g}_{v_{\mu e}}g_{v_{ee}}}{m_{v}^{2}}\right ) ^{2}\frac{1}{\left ( 2\pi \right ) ^{5}}\frac{2\pi } { 6}e_{3}^{2}de_{3}d\omega \left [ 3m_{\mu } ^{2}-4m_{\mu } e_{3}\right ] .
\label{336}\end{aligned}\ ] ] integrating eq .
( [ 336 ] ) over all directions @xmath237 of the emitted electron and over its complete range of energies @xmath238 , we obtain the total decay rated @xmath239 de_{3 } \nonumber \\ & = & \left ( \frac{\tilde{g}_{v_{\mu e}}g_{v_{ee}}}{m_{v}^{2}}\right ) ^{2}\frac{4\pi } { \left ( 2\pi \right ) ^{5}}\frac{2\pi } { 6}\frac{m_{\mu } ^{5}}{16 } \nonumber \\
\gamma \left ( \mu \rightarrow 3e\right ) & = & \frac{1}{2}\left ( \frac{\tilde{g}_{v_{\mu e}}g_{v_{ee}}}{m_{v}^{2}}\right ) ^{2}\frac{m_{\mu } ^{5}}{192\pi ^{3 } } \label{337}\end{aligned}\ ] ] comparing the above contribution to the @xmath0 process to that of the ordionary muon decay , @xmath240 , which proceeds via @xmath11 exchange and ( almost ) identical kinematics , gives the relation @xmath241 _ { v - exch.}}{\gamma \left [ \mu \rightarrow e\nu \bar{\nu}\right ] } \approx \frac{\tilde{g}_{v_{\mu e}}^{2}g_{v_{ee}}^{2}}{m_{v}^{4}}/\frac{g_{w}^{4}}{m_{w}^{4}}. \label{338}\ ] ] this ratio is defined as the branching ratio .
therefore @xmath242 _ { v - exch.}\approx \frac{\tilde{g}_{v_{\mu e}}^{2}g_{v_{ee}}^{2}}{m_{v}^{4}}/\frac{g_{w}^{4}}{m_{w}^{4}}. \label{339}\ ] ] since @xmath243 and @xmath244 while @xmath245 we can rewrite the above expression as @xmath242 _ { v - exch.}\approx \frac{\gamma \left ( v\rightarrow e^{+}e^{-}\right ) \gamma \left ( v\rightarrow \mu ^{+}e^{-}\right ) } { \gamma ^{2}\left ( w\rightarrow e\nu \right ) } \left ( \frac{m_{w}}{m_{v}}\right ) ^{6}. \label{340}\ ] ] using @xmath246 _ { v - exch.}\leq 10^{-12}$ ] and other data pertaining to the @xmath191 widths of the various vector mesons @xmath187 , we find a set of bounds for the two body lfv branching ratios of these vector mesons , i.e. @xmath247 _ { v - exch.}\gamma ^{2}\left ( w\rightarrow e\nu \right ) } { \gamma \left ( v\rightarrow e^{+}e^{-}\right ) } \left ( \frac{m_{v}}{m_{w}}\right ) ^{6}. \label{341}\ ] ] these bounds are calculated by dividing the above decay width by the full width of the vector mesons .
using the numerical values these bounds become @xmath248 all the vector ( or psudoscalars ) used as intermediaries in deriving the bounds in eqs .
( [ 342][345 ] ) are not on - shell .
thus we must entertain the possibilty that their contribution to the three - body decays considerd are reduced .
this could at least weaken the various strong bounds obtained above .
now we will focus on the possible mechanism for the reduction of such bounds .
the source of suppression is connected to possible `` form factor '' effect due to the dynamics which would , for example , reduce the contribution of various @xmath187 states to @xmath0 compared to the naive expectations .
however the effect of these form factors should be minimal or controllable if the lfv is induced by physics at scales much higher then the ew scale of the @xmath15 mass .
these effects of dynamics are beutifully discussed by illuna , jack and rienmann in their recent papers @xcite .
following these two papers , first of all we will discuss the dynamical suppression in the two body decay of @xmath249 and then move to @xmath0 .
the rare process @xmath250 is the classic example of the lepton flavour violating process
. the previous limit on its branching ratio is @xcite @xmath251 this reaction has not been observed so far , and the best experimental upper limit of its branching fraction is @xcite @xmath252 i.e. it is improved by the factor of @xmath253 . at the @xmath15 factory lep , searches for similar processes , through the @xmath15 boson , became possible : @xmath254 the best experimental limit on its branching ratio is ( @xmath255 c.l . ) @xmath256 let s discuss this reaction in detail .
the most general matrix element for the interaction of an on - shell vector boson with a fermionic current , as shown in fig .
[ tree3 ] , may be described by four dimensionless form factors . at one loop level
it is convinent to parameterize the amplitude as @xmath257 with @xmath258 , @xmath259 being the boson polarization vector and @xmath260 where @xmath261 in eq .
( [ 3102 ] ) , the form factors @xmath262 and @xmath263 stands for vector and axial - vector couplings and @xmath264 and @xmath265 for magnetic and electric dipole moments / transitions of equal / unlike final fermions .
the form factors depend on the momentum transfer squared @xmath266 . in principle
all the four form factors must be non - zero . in order to find the decay width we have to find @xmath267 which is @xmath268
so @xmath269 .
\nonumber \\ & & \label{3108}\end{aligned}\ ] ] using the spin averages , above result becomes @xmath270 .
\nonumber \\ & & \label{3109}\end{aligned}\ ] ] the decay width can be calculated as @xmath271 .
\label{3110}\end{aligned}\ ] ] the branching ratio can be obtained by dividing the decay width obtained in eq .
( [ 3110 ] ) by the total width of @xmath15 boson . @xmath272 = \frac{\alpha _ { w}^{3}}{24\pi ^{2}}\frac{m_{z}}{\gamma _ { z}}\left [ \left| f_{v}\right| ^{2}+\left| f_{a}\right| ^{2}+\frac{1}{2c_{w}^{2}}\left ( \left| f_{e}\right| ^{2}+\left| f_{m}\right| ^{2}\right ) \right ] .
\label{3111}\ ] ] where @xmath273 these form factors are model dependent .
it is easily seen that the branching fraction can be approximated as @xmath274 this is in agreement with its experimental value .
now we shall discuss these suppression in @xmath0 decay .
case-1 : first discuss the vector - axial - vector form factor contribution at @xmath275 vertex to @xmath0 , i.e. instead of the @xmath276 we have @xmath277 . therefore the amplitude becomes @xmath278 using the same procedure as for the previous cases @xmath279 , we get @xmath280 \left [ \left ( k_1.p\right ) \left ( k_2.k_3\right ) + \left ( k_1.k_3\right ) \left ( k_2.p\right ) \right ] \left ( \frac{\tilde{g}_{v_{\mu e}}g_{v_{ee}}}{m_v^2}\right ) ^2 \label{3114}\ ] ] using the spin averages , the above expression becomes @xmath281 \left [ \left ( k_1.p\right ) \left ( k_2.k_3\right ) + \left ( k_1.k_3\right ) \left ( k_2.p\right ) \right ] \left ( \frac{\tilde{g}_{v_{\mu e}}g_{v_{ee}}}{m_v^2}\right ) ^2 .
\label{3115}\ ] ] solving the delta function using the technique described above , the decay width becomes @xmath282 \left ( \frac{\tilde{g}_{v_{\mu e}}g_{v_{ee}}}{m_v^2}\right ) ^2\frac{m_\mu ^5}{192\pi ^3}. \label{3116}\ ] ] the corresponding branching ratio becomes @xmath283 \frac{\tilde{g}_{v_{\mu e}}^2g_{v_{ee}}^2}{m_v^4}/\frac{g_w^4}{m_w^4}. \label{3117}\ ] ] case-2 : now consider the electric and magnetic form factor contribution to the @xmath8 at @xmath284 vertex .
the amplitude can be written as @xmath285 using the same procedure as before the decay width for this reaction becomes @xmath286 \frac{m_{\mu } ^{2}}{m_{w}^{2}}\frac{m_{\mu } ^{5}}{192\pi ^{3 } } \label{3122}\ ] ] the branching ratio for this case becomes @xmath287 \frac{m_{\mu } ^{2}}{m_{w}^{2}}\left ( \frac{\tilde{g}_{v_{\mu e}}^{2}g_{v_{ee}}^{2}}{m_{v}^{4}}/\frac{g_{w}^{4}}{m_{w}^{4}}\right ) .
\label{3123}\ ] ] hence , the branching ratios of the three body decays are suppressed by a factor @xmath288 phys .
* 83 * , ( 1999 ) 1521 .
s. nissinov , r. d. peccei and x. m. zhang , phys . rev . *
d63 * , ( 2000 ) 016003 - 1 .
j. i. illana , m. jack and t. riemann , hep - ph/0001273 .
j. i. illana and t. riemann , hep - ph/0010193 .
r. bolton et al.prd * 38 * , ( 1988 ) 2077 .
[ tree2 ] [ tree3 ]
the question of whether or not neutrinos have a nonzero mass has remained one of the most tantalizing issue in the present day physics . in the standard model of electroweak theory neutrinos
are considerd to be massless .
but there is no compelling theoratical reason behind this assumption @xcite .
hints that the neutrinos are massive particles comes from the observation of three anomalous efects , 1 . the lsnd ( liquid scentilator neutrino detector ) excess @xcite , 2 . the atmospheric anomaly @xcite and 3 . the solar neutrino deficit @xcite
. in particular , the atmospheric results are the most convincing ones .
all the three effects can be naturally explained in terms of neutrino flavour oscillations , which will occur when neutrino propagate through space , if there masses are non - degnerate and the weak and mass eigenstates are mixed .
however , in order to explain all three experimentally observed effects in terms of neutrino flavour oscillation , one is forced to invoke additional sterile neutrino states @xcite to accommodate the very different frequencies of oscillations , given by three different mass squared @xmath289 s , indicated by three different effects .
the existance of such neutrino is currently unresolved problem and clearly demonstrates that the neutrino sector is not fully understood .
so due to the unappealing theoretical feature of a light sterile neutrino , it is interesting to look for alternatives that could explain the lsnd excess with the known three light neutrinos . from a phenomenological point of view , we recall that the neutrino flavour oscillation hypothesis predicts a well defined dependence of phenomena as a function of neutrino energy , characterized by the so called @xmath290 behaviour , where @xmath140 is distance between source and detector and @xmath291 is neutrino energy .
so far , no experiment has conclusively demonstrated such a @xmath290 dependence of the anomalous effect , with may be expectation of superkamiokande data which favours the dependence @xmath292 @xmath293 where @xmath294 @xcite .
in such an unclear sitution , it is possible to envisage `` non flavour oscillation '' mechanism to explain part of neutrino data . aside from theoretical argument against sterile neutrino
we argue that , from a phenomenological point of view , the lsnd effect is peculiar : it has a small probability , measured to be @xmath295 @xcite , in contrast to the solar and atmospheric neutrino anomalies , which are large . hence , lsnd is a natural candidate for an interpretation involving a different physics than in atmospheric and solar neutrino flavour oscillation .
we recall that the lsnd effect was first reported as an excess of @xmath296 s in the @xmath297 flux from the @xmath298 decay at rest ( dar ) process .
the neutrino beam is obtained with @xmath299 kinetic energy protons hitting a series of targets , producing secondary pions .
most of @xmath300 come to rest and decay through the sequence @xmath301 ; followed by @xmath302 , supplying the experiment with the @xmath297 s with a maximum energy of @xmath303 @xmath304 .
the intrisnic contamination of @xmath305 s coming from the symmetrical decay chain starting with @xmath306 is estimated to be small since most negatively charged mesons are captured before they decay .
the excess of @xmath307 s , explained in terms of neutrino flavour transitions of the type @xmath308 , occur via the reactions : @xmath309 i.e. anti - neutrinos are produced by the @xmath310 and detected by @xmath311 .
there is another evidence in favour of the neutrino flavour oscillation which was reported in the decay in flight ( dif ) .
but due to its lower statistical significance , we concentrate on the hint from stopped muon , and ignore the dif result . the lsnd claim is contradicted by the latest karmen2 results @xcite , however the experimental sensitivity is marginal to conclusively exclude or confirm completely the lsnd excess . a new experiment , miniboone @xcite would confornt the flavour oscillation hypothesis with a very high statistical accuracy . a negative result from miniboone experiment would indicate that the neutrino flavour oscillation is not the correct hypothesis to explain the excess seen in lsnd .
it would however not contradict other possible non - flavour - oscillation interpretation of the effect .
in particular , lfv decays would play a role in the interpretation of lsnd excess .
a neutrino factory is an ideal mechaine to probe such anomalous decays of muon .
th aim of this work is to investigate another approch .
we assume that the three neutrinos interact through lepton flavour violation interactions , which are forbidden in the standard model .
this is an attractive possibility , because various extensions of the standard model which predict the neutrino masses also give rise to such new interactions .
these interactions can affect the lsnd excess .
the motivation for the lfv is that the branching probability for lsnd is too small ( @xmath312 ) compared to atmospheric and solar anomalies .
we analyze the consequence of small lepton flavour violating interactions to explain the lsnd excess , and check whether this scenario will be feasible .
let us assume that a vector boson @xmath187 ( here @xmath187 could be either a fundamental state , such as the @xmath95 or the quark- antiquark bound state such as the @xmath188 or @xmath189 ) couples to @xmath190 .
if it couples also to @xmath191 , as all the states above do , then by unitarity its exchange contributes also to @xmath0 .
let us define the effctive couplins between vector boson @xmath187 and @xmath313 as @xmath192 , and the corresponding effective lagrangian can be written as @xmath314 here , the reaction under consideration is @xmath315 therefore amoung these vector bosons only the @xmath95 will contribute .
@xmath316 is any of the three known leptons , i.e. @xmath317 , @xmath179 , or @xmath182 .
the coupling at the lfv vertex remains the same , but at the neutrino vertex it will be changed and is obtained by the standard model .
we also assume that the lfv decay proceds through a similar diagram as the standard model muon decay , however with interchanged neutrino flavour as shown in fig .
[ tree4 ] . using feynman rules ,
the corresponding amplitude can be written as @xmath318 \nonumber \\ & = & \frac{\tilde{g}_{z_{\mu e}}g_{z\nu \bar{\nu}}}{m_{z}^{2}-s}\left [ \bar{v}(p_{1})\gamma ^{\alpha } v(p_{2})\bar{u}(p_{4})\gamma _ { \alpha } \left ( 1-\gamma _ { 5}\right ) v(p_{3})\right ] \label{402}\end{aligned}\ ] ] here @xmath319 is the effective coupling of the @xmath15- boson to @xmath320 . also for low energy
, we have to neglect @xmath198 in comparison to the @xmath26 .
therefore eq .
( [ 402 ] ) takes the form @xmath321 \label{403}\ ] ] now as usual we have to calculate @xmath322 , which is @xmath323 solving these traces , the above eq .
becomes @xmath324 .
\label{407}\ ] ] then the partial decay rate for the muon can be written as @xmath325 where @xmath326 using the same technique as we have used in the previous chapter , we got the total decay width as follows @xmath327 comparing the above contribution to the @xmath2 process to that of the ordionary muon decay , @xmath240 , which proceeds via @xmath11 exchange and ( almost ) identical kinematics , gives the relation @xmath328 _ { v - exch.}}\right ] \times \left [ \frac{\gamma \left ( \mu \rightarrow 3e\right ) _ { v - exch.}}{\gamma ( \mu \rightarrow e\nu \bar{\nu})}\right ] \nonumber \\ { \cal b}\left ( \mu \rightarrow e\nu _ l\bar{\nu}_l\right ) _ { z - exch . } & = & \left [ \frac{\gamma \left ( \mu \rightarrow e\nu _ l\bar{\nu}_l\right ) _ { z - exch.}}{\gamma \left [ \mu \rightarrow 3e\right ] _ { v - exch.}}\right ] \times { \cal b}\left ( \mu \rightarrow 3e\right ) _ { v - exch . }
\label{410}\end{aligned}\ ] ] using the value of the @xmath329 _ { z - exch.}$ ] , eq .
( [ 410 ] ) becomes @xmath330 we know that @xmath242 _ { v - exch.}\leq 10^{-12}.\ ] ] since @xmath331 and @xmath332 so eq . (
[ 410 ] ) becomes @xmath333 as it is seen from the above equation the branching ratio for the this lepton flavour violating decay is the same as that of the @xmath334 .
this is because of the fact that the reaction kinematic is same , and also in both the cases the external particles are assumed to be the massless .
the factor of @xmath335 is due to the fact that the couplings between electron - positron is different then the two neutrinos .
it is already mentioned that , in order to explain the lsnd result , the effective new physics coupling should satisfy @xmath336 we define @xmath337 to be the effective coupling of the anomalous muon decay .
thus , at the @xmath338 c.l .
we need @xcite @xmath339 now let s calculate this for the reaction discussed here . as we know that @xmath340 thus following the eq .
( [ 414 ] ) , we can write the effective coupling as follows @xmath341 now the ratio of the two couplings become @xmath342 so @xmath343 becomes @xmath344 as and using the value of the @xmath345 and @xmath346 from the standard model , we get @xmath347 thus , comparing with eq .
( [ 414 ] ) which requires @xmath3 , we learn that the lfv anomalous muon decays @xmath2 can not significantly contribute to the lsnd dar result , at least when we use @xmath348 as constrained by @xmath0 in a model independent way . thus the excess found by the lsnd is not due to these lepton flavour violating decays .
so we can not say that whether this excess is due to neutrino flavour oscillations or due to lfv decays which might occur through some exotic mechanism leading to @xmath349 .
hence , we can say that it is still unresolved puzzle of the neutrino physics . in the minimal standard gauge model of quarks and leptons
, each of three known neutrinos @xmath350 appears only as a member of left - handed @xmath10 lepton doublet @xmath351 and the higgs sector contains only one scalar doublet @xmath352 as a result , neutrinos are massless in this model .
experimentally there is now a host of evidence for neutrino oscillations , and that is most naturally explained if neutrinos are massive and mix with each other . theoretically there is no compelling reason for massless neutrinos , and any extension beyond the minimal standard model often allows them to massive .
there exists already a vast literature on specific models of neutrino masses and mixing @xcite . here
we will make the following simple observations . in the minimal standard model of particle interactions , the massless neutrinos acquire naturally small majorana masses through the effective dimension - five operator @xmath353 where @xmath354 is a large effective mass scale , and @xmath355 is the usual higgs doublet with the non zero expectation value , @xmath356 .
all models of neutrino mass and mixing ( which have the same light particle content as at the minimal standard model ) can be explained by the operator @xmath357 different models are merely different reasons for this operator .
the intermediate heavy particle in case of the operator defined in eq . ( [ 4103 ] ) is clearly a fermion singlet .
let s call it @xmath358 and let its mass be @xmath16 and its coupling to @xmath359 be @xmath360 .
also we can identify @xmath361 or simply @xmath362 as a dirac mass @xmath363 linking @xmath359 to @xmath358 and the neutrino mass matrix is introduced so that @xmath364 so that @xmath365 in eq . (
[ 4103 ] ) .
this is , of course , just the well known canonical seesaw mechanism , with @xmath358 identified as the right - handed neutrino with a large majorana mass . given that @xmath366 is at most of order @xmath367 and @xmath368 should be too small , the usual thinking is that @xmath16 has to be very large , i.e. , @xmath369 . as such
, this famous mechanism must be accepted on faith , because there can not be any direct experimental test of its validity .
consider now the possibility that there is no new physics beyond the @xmath370 energy scale .
this is an intriguing idea proposed recently in theories of large extra dimensions @xcite .
instead of using an ingradient supplied by the large extra dimensions , a model has recently been proposed @xcite to show how eq .
( [ 4105 ] ) may be realized naturally with @xmath16 of order @xmath5 @xmath370 in a simple extension of the standard model .
this means that @xmath363 should be small i.e. , @xmath371 .
if it comes from zeorth component of @xmath97 i.e. , @xmath372 as in the standard model , that would not be natural ; but as shown below , it will come instead from another with naturally small vacuum expectation value .
this new realization of the seesaw mechanism will allow direct experimental test of its validity , as discussed below .
consider the minimal standard model with three lepton families : @xmath373 where their transformations under the standard @xmath374 gauge group are indicated above .
now add three neutral fermion singlets @xmath375 but instead of assigning them the lepton number @xmath376 , so that they can pair up with the lepton doublet through the interaction @xmath377 , @xmath378 is assigned to forbid this yukawa term . to complete the model , a new scalar doublet , called leptoquark @xcite @xmath379
is introduced with lepton number @xmath6 . hence the terms @xmath380 appear in the lagrangian .
the effective operator of eq .
( [ 4103 ] ) for neutrino mass is then replaced by one with @xmath381 instead of @xmath127 , and , if @xmath382 is naturally small , the corresponding scale @xmath354 will not have to be so large and @xmath383 of eq .
( [ 4109 ] ) may indeed be of order @xmath5 @xmath370 .
the higgs potential of this model is given by @xmath384 where the @xmath385 term breaks @xmath140 explicitly but softly .
note that , given the particle content of this model , the @xmath385 term is the only possible soft term which also breaks @xmath140 .
the equations of constraints for @xmath386 and @xmath382 are @xmath387 + \mu _ { 12}^2u & = & 0 \label{4111 } \\ u\left [ m_2 ^ 2+\lambda _ 2u^2+\left ( \lambda _ 3+\lambda _ 4\right ) v^2\right ] + \mu _ { 12}^2v & = & 0 .
\label{4112}\end{aligned}\ ] ] consider the case @xmath388 then @xmath389 hence it is very clear that @xmath390 may be very small compared to @xmath391 @xmath392 .
for example , if @xmath393 , @xmath394 , then @xmath395 . the relative smallness of @xmath396 may be attributed to the fact that it corrosponds to the explecit breaking of the lepton number in @xmath136 of eq .
( [ 4110 ] ) .
the usual argument here is that , if @xmath396 were zero , then as seen from eq .
( [ 4114 ] ) the model s symmetry is increased , i.e. , the lepton number would not be broken .
hence the assumption that it is small compared to @xmath397 or @xmath398 is `` natural . ''
one thing is very clear from here that if @xmath396 were much smallar , then neutrino masses would be too small to account for the present observation of the neutrino oscillations .
the @xmath399 mass matrix spaning @xmath400 $ ] is now given by @xmath401 . \label{4115}\ ] ] the mixing between @xmath402 and @xmath403 is thus of the order @xmath404 , which will allow the physical @xmath403 to decay through its small component of @xmath402 to @xmath405 .
the effective mass matrix spaning the light neutrino is then @xmath406 and is of order @xmath5 @xmath407 if @xmath368 is of order unity .
there are five physical higgs boson with masses given by @xmath408 the @xmath409 behaves very much like the ordionary higgs boson .
the new scalar particles of this model , i.e. @xmath410 , @xmath14 , and @xmath411 ( all with mass@xmath412 ) , as well as @xmath413 , are now accessible to direct experimental discovery in future accelators . in summary , a new seesaw model of neutrino mass
is proposed , where a second scalar doublet @xmath414 with lepton number @xmath6 is added to the minimal standard model togather with three neutral right - handed fermion singlets @xmath358 with lepton number @xmath378 .
thus @xmath358 is allowed to have a majorana mass @xmath16 as well as interaction @xmath415 .
hence @xmath366 is proportional to @xmath416 and , if @xmath417 @xmath16 may be of the order @xmath5 @xmath370 and be observable experimently .
this is accomplished with the higgs potential of eq .
( [ 4110 ] ) , where @xmath140 is broken explicitly and uniquely with the soft term @xmath418 . as the lepton flavor violation
is discussed in all models of neutrino mass .
it is argued that in this model , there is no lfv at tree level for charged leptons .
however , it does occur in one loop throug @xmath381 and @xmath403 exchange @xcite .
the aim of this work is to test this model for the lepton flavor violating decays , i.e. @xmath0 and @xmath419 in one loop process .
we have to calculate the branching ratio of @xmath0 and compare it to its experimenal bounds .
first of all we will discuss the reaction @xmath0 .
the corresponding box digrams can be shown in fig .
[ figure ] .
the couplings at each vertex can be taken from eq .
( [ 4115 ] ) . instead of @xmath77 materices
we have unity at each vertex , because each higgs boson is a scalar . using feynman rules ,
the amplitude can be written as follows : @xmath420 v(p_2 ) \nonumber \\ & & \times \bar{u}(p_3)\left [ \frac i{\not{k}-m_i}\right ] v(p_4)\left [ \frac{-i}{k^2-m_h^2}\right ] ^2 \nonumber \\ & = & 2\sum_i\left ( f_{\mu i}^{*}f_{ei}^{*}f_{ei}^{*}f_{ei}\right ) \int \frac{d^4k}{\left ( 2\pi \right ) ^4}\bar{v}(p_1)\left [ \not{k}+m_i\right ] v(p_2 ) \nonumber \label{21 } \\ & & \times \bar{u}(p_3)\left [ \not{k}+m_i\right ] v(p_4)\left [ \frac 1{k^2-m_h^2}\right ] ^2\left [ \frac 1{k^2-m_i^2}\right ] ^2 \nonumber \\ & = & \frac 2{\left ( 2\pi \right ) ^4}\sum_i\left ( f_{\mu i}^{*}f_{ei}^{*}f_{ei}^{*}f_{ei}\right ) \left [ \left\ { \bar{v}(p_1)\gamma ^\alpha v(p_2)\bar{u}(p_3)\gamma ^\beta v(p_4)i^{\alpha \beta } \right\ } \right . \nonumber \\ & & + m_i\left\ { \bar{v}(p_1)\gamma ^\alpha v(p_2)\bar{u}(p_3)v(p_4)i^\alpha + \bar{v}(p_1)v(p_2)\bar{u}(p_3)\gamma ^\beta v(p_4)i^\beta \right\ } \nonumber \\ & & \left .
+ m_i^2\left\ { \bar{v}(p_1)v(p_2)\bar{u}(p_3)v(p_4)i\right\ } \right ] , \label{4121}\end{aligned}\ ] ] where @xmath421 it is assumed that the loop momenta is very high , i.e. @xmath422 . so there is no change in momenta at the vertices , and also the dominating integral is the first one . before calculating the @xmath423
, we have to do the loop integration .
to do this we use the technique known as feynman parameterization @xcite .
@xmath424 ^4}. \label{4128}\ ] ] for our above mentioned case @xmath425 and @xmath426 are @xmath427 so @xmath428 & = & k^2-m_h^2+\left ( m_h^2-m_i^2\right ) z \nonumber \\ & = & k^2+m_h^2\left [ \left ( 1-x_i\right ) z-1\right ] \nonumber \\ & = & k^2+s , \label{4127}\end{aligned}\ ] ] where @xmath429 and @xmath430 $ ] . then , eq . ( [ 4122 ] ) becomes @xmath431 ^4 } \nonumber \\ & = & 6\int_0 ^ 1dz\times z\left ( 1-z\right ) \times \left [ \frac{i\pi ^2\gamma \left ( 1\right ) \times g^{\alpha \beta } } { 2\gamma \left ( 4\right ) \times s}\right ] \nonumber \\ & = & 6\int_0 ^ 1dz\times z\left ( 1-z\right ) \times \left [ \frac{i\pi ^2g^{\alpha \beta } } { 2\times 6m_h^2\left [ \left ( 1-x_i\right ) z-1\right ] } \right ] , \label{41128}\end{aligned}\ ] ] because @xmath432 ^n}=\frac{i\pi ^2\gamma \left ( n-3\right ) } { 2\gamma \left ( n\right ) } \times \frac{g^{\alpha \beta } } { s^{n-3}}\qquad n\geq 3.\ ] ] therefore , after solving the @xmath433 integration eq .
( [ 41128 ] ) takes the form @xmath434 . \label{41129}\ ] ] using the same technique for eq .
( [ 4123 ] ) becomes @xmath435 because @xmath436 ^n}=0\qquad n\geq 3.\ ] ] also define @xmath437 hence finally eq .
( [ 4121 ] ) becomes @xmath438 .
\label{4132}\end{aligned}\ ] ] now the square of the amplitude becomes @xmath439 .\end{aligned}\ ] ] define @xmath440 then the above result takes the form @xmath441 ^22\sum_{i , j}\xi _ i\xi _ j^{*}a\left ( x_i\right ) a\left ( x_j\right ) \times \left [ \left ( p_1p_4\right ) \left ( p_2p_3\right ) + \left ( p_1p_3\right ) \left ( p_2p_4\right ) \right ] .
\label{4133}\ ] ] the corresponding decay width becomes @xmath442 = \left [ \frac{\pi ^2}{\left ( 2\pi \right ) ^4m_h^2}\right ] ^2\sum_{i , j}\xi _ i\xi _ j^{*}a\left ( x_i\right ) a\left ( x_j\right ) \times \frac{m_\mu ^5}{192\times \pi ^3 } , \label{4134}\ ] ] and the branching ratio becomes @xmath443 & = & \frac{\gamma \left [ \mu \rightarrow 3e\right ] } { \gamma \left [ \mu \rightarrow e\nu \bar{\nu}\right ] } \nonumber \\ & = & \left [ \frac 1{\left ( 2\pi \right ) ^42m_h^4}\right ] \sum_{i , j}\xi _ i\xi _ j^{*}a\left ( x_i\right ) a\left ( x_j\right ) \frac{m_w^6}{\gamma ^2\left [ w\rightarrow e\nu \right ] \times 2 ^ 5}. \nonumber \\ & & \label{4135}\end{aligned}\ ] ] now let s calculate the other reaction , i.e. @xmath444 the corresponding box diagrams can be shown in fig .
[ fig31 ] . using feynman rules ,
the amplitude can be written as follows @xmath445 \nonumber \\ & & \times v(p_2)\bar{u}(p_3)\left [ \frac i{\not{k}-m_i}\right ] v(p_4)\left [ \frac{-i}{k^2-m_{h^{+}}^2}\right ] \left [ \frac{-i}{k^2-m_{h_2 ^ 0}^2}\right ] \nonumber \\ & = & 2\sum_i\left ( f_{\mu i}^{*}f_{li}^{*}f_{ei}^{*}f_{li}\right ) \int \frac{d^4k}{\left ( 2\pi \right ) ^4}\bar{v}(p_1)\left [ \not{k}+m_i\right ] v(p_2 ) \nonumber \\ & & \times \bar{u}(p_3)\left [ \not{k}+m_i\right ] v(p_4)\frac 1{k^2-m_{h^{+}}^2}\left ( \frac 1{k^2-m_i^2}\right ) ^2\frac{-i}{k^2-m_{h_2
^ 0}^2}. \nonumber \\ & & \label{4136}\end{aligned}\ ] ] using the same assumption here that the loop moment is very high , i.e. @xmath422 . the above integral can be solvd by using feynman parameterization @xmath446 ^4}\ ] ] therefore @xmath447 where @xmath448 and @xmath449 define @xmath450 using these results eq .
( [ 4136 ] ) takes the form @xmath451 \sum_i\xi _ ia(x_{1i},x_{2i } ) \nonumber \\ & & \times \left\ { \bar{v}(p_1)\gamma ^\alpha v(p_2)\bar{u}(p_3)\gamma
_ \alpha v(p_4)\right\ } .
\label{4140}\end{aligned}\ ] ] now , after calculating @xmath452 , the decay width becomes @xmath453 & = & \left [ \frac{\pi ^2}{2m_i^2\left ( 2\pi \right ) ^4}\right ] ^2\sum_{i , j}\xi _ i\xi _ ja(x_{1i},x_{2i})a(x_{1j},x_{2j})\frac{m_\mu ^5}{192\times \pi ^3 } \nonumber \\ & = & \left [ \frac 1{32m_i^2\pi ^2}\right ] ^2\sum_{i , j}\xi _ i\xi _ ja(x_{1i},x_{2i})a(x_{1j},x_{2j})\frac{m_\mu ^5}{192\times \pi ^3}. \nonumber \\ & & \label{4141}\end{aligned}\ ] ] then the branching ratio becomes @xmath454 = \left [ \frac 1{32m_i^2\pi ^2}\right ] ^2\sum_{i , j}\left ( \xi _ i\xi _ ja(x_{1i},x_{2i})a(x_{1j},x_{2j})\right ) \frac{m_w^6}{2\times \gamma ^2\left [ w\rightarrow e\nu \right ] } .
\label{4142}\ ] ] in order to calculate the numerical value of the branching ratio for @xmath8 , we will fix the value of @xmath455 tev and also assume that all the couplings are of the order unity . as @xmath456 , therefore if @xmath457 gev@xmath458 its effect on the radiative parameters @xmath459 is negligible small and will not change the excellent experimental fit of the minimal standard model . to have a feeling about the branching ratios
let us take @xmath460 gev , the branching ratio for @xmath0 becomes @xmath461 which is much higher then the experimental bound on the branching ratio of @xmath0 , i.e. @xmath462 .
for the mass smallar than @xmath463 gev the branching ratio grows higher and higher .
thus from here it is obvious that @xmath464 tev . by varing the higgs mass
the corresponding branching ratio becomes @xmath465 it can be eaisly seen that with the increase of higgs mass , the branching ratio decreases .
as the higgs mass reaches to @xmath7 tev , branching ratio becomes @xmath466 which is comparable to the experimental value .
thus we have concluded that the higgs mass is greater than @xmath7 tev to explain lfv decays in seesaw model of neutrino masses @xcite .
the aim of this work has been to understand the physics beyond the standard model by considering lepton flavour violating ( lfv ) decay processes .
using the experimental bounds on the three body lfv decays , we have calculated the bounds on the corresponding two body decays .
the dynamical suppression of three boby lfv decays due to momentum dependent couplings have also been discussed in detail . the experimental bounds on the three body lfv decay , @xmath0 constrained the coupling @xmath1 , which is helpful to calculate the anamolous muon decays . the anamolous muon decay @xmath467 could not significantly contribute to the lsnd dar result .
this has been concluded by comparing the effective coupling of anamolous muon decay with @xmath3 @xcite .
lfv decays at a loop order have also been discussed in seesaw model of neutrino masses which involve right handed singlet neutrinos of mass @xmath468 of order tev @xcite .
the main purpose of the model is that there is no new physics beyond @xmath5 tev .
it has been shown that even if one keeps @xmath468 at @xmath5 tev the present bounds on @xmath0 requires that new higgs bosons , necessary in this model should have their mass larger than @xmath7 tev .
thus it is necessary to have a new physics beyond the tev energy scale defeating the origional motivation of the model . | this dissertation reviews the standard model formalism as well as the lepton flavour violating ( lfv ) decay processes which cause its extension , known as the physics beyond the sm .
firstly , using the experimental bounds on three body lfv decays , the corrosponding bounds on two body lfv decays are reviewed .
the dynamical suppression of three body lfv decays due to momentum dependent couplings is also reviewed .
secondly , the role of the lfv decays to explain the lsnd excess is discussed in detail , for which the experimental bounds on three body lfv decays , i.e. @xmath0 are used to constraint the coupling @xmath1 , which is needed to calculate the anamolous muon decay @xmath2 . then comparing the effective coupling of anamolous muon decay to @xmath3@xcite , it is proved that lfv is not the correct hypothesis to explain the lsnd excess .
finally , lfv decays at loop order are studied in seesaw model of neutrino masses @xcite where the smallness of the seesaw neutrino mass may be naturally realized with @xmath4 ( mass of right - handed singlet neutrinos ) of order @xmath5 tev .
it is shown that the higgs mass of a new scalar doublet with lepton number @xmath6 needed in the model has to be larger than @xmath7 tev to get the branching ratio of @xmath0 to be consistant with the existing bound on @xmath8 .
this defeats the origional motivation of the model , namely that there is no physics beyond the tev energy scale .
tcilatex acknowledgements i am filled with the praise and glory to all mighty allah , the most merciful and benevolent , who created the uneverse , with ideas of beauty , symmetry and harmony , with regularity and without any chaos , and gave us the abilities to discover what he thought .
bless muhammad ( p.b.u.h ) the seal of the prophets and his pure and pious progeny .
first and the foremost , i want to express my immense gratitude to my worthy supervisor prof .
dr .
riazuddin for his professsional guidance and encouragement throughout this research work . without his kind and
invaluable help it would have almost been impossible for me to accomplish this job .
i wish to say thanks to all the teachers at the physics department because it is due to them that we have an excellent environment of study in the department .
i can never forget the guidance of dr .
fayyazuddin , dr .
pervez hoodbhoy and dr .
khurshaid husnain throughout my studies .
my heartiest thanks to all members of the particle physics group at qau , and especially to shahzad , irfan , ijaz and saba who helped me in one way or other .
i also wish to thank mr .
amjad hussain shah gilani for his unforgetable help , cooperation and useful discussions which enable me to hunt this task .
i extend my thanks to dr .
kashif sabeeh and dr .
hafeez hoorani for there encourging attitude .
my cordial thanks to my friends , farhan , moetasim , assad , hassan , anees , adnan , imran , ansar , irfan , anis , ali , tasawar , farooque , jawad , asia , azra , aisha , faiza , zubair , jamal , anwar and usman .
there warm companionship make the life at campus beautiful and unforgetable .
finally i wish to record my deepest obligation to my parents , brothers and sisters for their prayers , encouragment and finincial support during my studies .
i also wish to acknowledge the financial support provided by dr .
riazuddin during my research work . .
this work is submitted as a dissertation in partial fulfilment of the requirement for the degree of * master of philosophy * in * physics * department of physics quaid - e - azam university islamabad 45320 pakistan .
_ to my loving * maan * jee and * baba * jee _ certificate certified that the work contained in this dissertation was carried out by mr .
muhammad jamil aslam under my supervision .
( prof .
dr .
riazuddin ) supervisor national center for physics quaid - e - azam university islamabad 45320 submitted through : ( prof .
dr .
asghari masood ) chairperson department of physics quaid - e - azam university islamabad 45320 |
m emission in blue and the mips-_spitzer _ 24 @xmath2 m emission is shown in green .
the objects analysed in this work are indicated . the arc feature in n22 and almost the whole interior of g18.149 - 0.283
are saturated in 24 @xmath2m.,width=362 ] nowadays it is well known that massive stars in our galaxy are born predominantly within the dense cores of giant molecular clouds ( for a thorough discussion of the star - formation theories see @xcite ) .
they usually form and evolve in clusters , hence it is expectable to observe several hii regions in different evolutionary stages and probably also supernova remnants ( snrs ) in a same galactic neighborhood .
some examples of this spatial coincidence are the field toward w28 , with several snrs and hii regions ( e.g. @xcite ) , the snr w44 and the hii region g034.8 - 0.7 @xcite , the luminous blue variable ( lbv ) star candidate g24.73 + 0.69 , some hii regions and the snr g24.7 + 0.6 @xcite .
moreover , it is usually observed large amounts of molecular gas in the surroundings of hii regions and snrs .
the shock and ionization fronts from these objects , which compress and sweep up the molecular gas , can trigger the formation of a new generation of massive stars .
this has been extensively probed towards the borders of hii regions ( e.g. @xcite ) where processes such as `` collect and collapse '' @xcite and/or radiative driven implosion @xcite can be taking place . the goal of this work is to provide observational evidence supporting that massive stars usually born in clusters from material of the same molecular cloud , which then produce , along their evolution , neighboring hii regions , interstellar bubbles and snrs that can interact with the parental cloud . to perform this
we select a very rich region in which lie the snr g18.1 - 0.1 and several hii regions , two of them catalogued as infrared dust bubbles by @xcite . in section [ present ] , we describe the studied hii region / snr complex , section [ ism ] presents an analysis of the interstellar medium ( ism ) around these sources , in section [ obstars ] we present an optical spectroscopic search for ob - type stars ( the hii regions exciting star candidates ) and an analysis of two x - ray point sources , section [ larger ] presents a larger scale ism study around the complex , and finally section [ summ ] summarizes the obtained results .
in figure [ presentfig ] we present the 22 @xmath1 22 region where the hii region / snr complex lies in a colour composite image as seen in the _ spitzer_-irac 8 @xmath2 m emission ( blue ) , the radio continuum emission at 20 cm ( red ) , and the the _ spitzer_-mipsgal emission at 24 @xmath2 m ( green ) .
we used the mosaicked image from the galactic legacy infrared mid - plane survey extraordinaire ( glimpse ) in the _
spitzer_-irac band at 8 @xmath2 m which has an angular resolution of @xmath319 ( see @xcite ) .
mipsgal is a survey of the same region as glimpse , using mips instrument ( 24 and 70 @xmath2 m ) on _ spitzer _ @xcite .
the mipsgal resolution at 24 @xmath2 m is 6 .
the radio continuum data at 20 cm , with a fwhm synthesized beam of about 5 , was extracted from the new gps of the multi - array galactic plane imaging survey @xcite which was conducted using the very large array ( vla ) . in what follows
we describe each object based on the information that appears in the literature .
snr g18.1 - 0.1 or g18.16 - 0.16 @xcite is a poor studied supernova remnant that was firstly observed at 57.5 mhz by @xcite , who identified it as snr g18.1 - 0.2 .
the author also mentions the presence of the source sharpless 2 - 53 @xcite , which is related to the hii regions complex composed by n21 , n22 and g018.149 - 00.283 as seen in figure [ presentfig ] .
based on the distance estimated for the hii regions complex of 4.5 kpc @xcite , @xcite suggests this value as the minimum kinematic distance for the snr .
later , the snr g18.1 - 0.1 was associated with the x - ray source axj182435 - 1311 , which was observed with asca @xcite .
more recently , this shell snr was identified in the very large array survey of the galactic plane at 90 cm by @xcite , reporting fluxes of 7.6 , 3.9 , and 3.0 jy at 90 , 20 , and 11 cm , respectively .
they derived an spectral index of @xmath4 using the emission at 90 and 20 cm , and @xmath5 using the 90 and 11 cm data .
@xcite using glimpse data , catalogued almost 600 infrared dust bubbles : full or partial rings bordered by a photodissociation region ( pdr ) , seen mainly at 8 @xmath2 m , which usually encloses ionized gas and hot dust observed at 24 @xmath2 m .
most of these bubbles are hii regions as is the case of bubbles n21 and n22 displayed in figure [ presentfig ] .
the bubble n22 , g18.259 - 0.307 hii region @xcite , has a recombination line velocity of v@xmath6 @xmath7 , and is located at @xmath3 4.1 kpc .
the interstellar medium around n22 was very recently investigated by @xcite , finding some evidence of triggered star formation .
the bubble n21 is not catalogued in @xcite but it appears as the hii region g018.20 - 00.40 in @xcite with a systemic velocity of v@xmath6 @xmath8 .
the spectral resolutions in @xcite and @xcite are 2.5 and between 2 and 4 , respectively .
@xcite identified 21 yso candidates around n21 , and based on a sed fitting to numerical hot stellar photosphere models , they suggested some stars to be the possible exciting sources of this hii region . according to @xcite , who analysed a sample of 291 galactic hii regions with the aim of resolving the distance ambiguity ,
n21 is related to the source called c18.19 - 0.40 with a resolved distance [email protected] kpc , and n22 to c18.26 - 0.30 with [email protected] kpc .
hii region g018.149 - 00.283 is not catalogued as an infrared dust bubble , and according to @xcite , has a recombination line velocity of v@xmath6 @xmath10 and is located at the distance of @xmath3 4.1 kpc .
this is in agreement with the distance catalogued for u18.15 - 0.28 , the source related to this hii region in @xcite the source that is superimposed over the eastern border of the snr is the hii region g18.197 - 00.181 @xcite , which has a systemic velocity v@xmath6 @xmath11 .
this hii region is related to c18.20 - 0.18 in @xcite , whose distance ( [email protected] or 12.4 kpc ) is not resolved because it presents a discrepancy between the used methods to resolve the ambiguity . in figure [ presentfig ] , based on the morphology of the ir and radio continuum emission , we indicated another possible hii region ( it appears with an interrogation mark in the figure ) , which was not found in any catalogue .
this source has a pdr visible in the 8 @xmath2 m band , which encloses ionized gas observed at 20 cm and hot dust observed at 24 @xmath2 m , typical characteristics of hii regions ( see e.g. @xcite ) . in this case
it can be appreciated two peaks at 24 @xmath2 m , one toward the center and the other almost in a border of this possible hii region .
according to the coordinates , this object should be hii region g18.237 - 0.240 . in order to summarize , table [ sources ]
presents the sources , their galactic coordinates , the radius in the case of the snr , the average radius for the hii regions and the linear sizes .
the snr radius was extracted from @xcite , while the average radius of bubbles n21 and n22 from @xcite . in the case of g018.149 - 00.283 , g18.197 - 00.181 and the possible unknown hii region , the average radius was estimated as was done by @xcite for the infrared dust bubbles .
the linear size is the spatial diameter of the sources by assuming a distance of 4 kpc .
.analysed sources . [ cols="<,^,^,^,^",options="header " , ] in this section we analyse the hii region / snr complex surroudings in a larger spatial scale .
a detailed inspection of the whole data cube at larger scale reveals that the hii region / snr complex lies in a border of a big molecular shell that peaks in the velocity interval of 5155 . by inspecting the mid - ir emission in this large scale area we did not find any possible counterpart to this molecular shell .
figure [ large ] presents the j=10 emission of a large region of about 13 @xmath1 08 integrated between 51 and 55 .
it is very interesting to note that the hii region / snr complex is located southward a very likely molecular shell with an elliptical shape with semi - axes of 30@xmath112 , centered at @xmath12 18185 , and @xmath130104 . assuming a distance of 4 kpc , the molecular shell has an extension of about 70 pc @xmath1 28 pc .
following the standard nomenclature for galactic shells , we designate this new structure as gsh 18.2 - 0.1 + 53 . along the mentioned velocity range , the co shell is well defined as a large structure , although , as in many other shells , the typical behaviour of an expanding structure ( i.e. a ring that starts as a cap , reaches its greatest angular dimension at the systemic velocity and then shrinks back to a cap ) is not completely observed .
figure [ large ] also shows that the shell is composed of several molecular condensations among which are those associated with the hii region / snr complex .
following the same assumptions and equations as section [ ism ] we derive an average h@xmath14 column density for the whole molecular shell . assuming a constant t@xmath15 = 20 k over the shell and a @xmath16 , we obtain n(h@xmath17 ) @xmath3 5 @xmath18 @xmath19 .
using equation ( [ eq4 ] ) and performing the summation over all the observed positions within the 1.2 k contour level ( which define the shell boundary , see figure [ large ] ) we obtain a total molecular mass of about 2 @xmath20 .
following @xcite , we estimate the kinetic energy stored in the shell and its age using @xmath21 and @xmath22 , respetively , in which is considered a simple model that describes the expansion of a shell created by a continuous injection of mechanical energy , where @xmath23 is the effective radius .
table [ para ] summarizes the main physical parameters of gsh 18.2 - 0.1 + 53 .
the origin of most of these expanding structures would be related to the action of massive stars ( ob - type stars ) and their descendants ( wr stars ) , which strongly disturb the environment that surrounds them , first by the high rate of ionizing photons and strong winds , and subsequently at the end of their lives , when they explode as supernovae . in this context , we wonder for the origin of gsh 18.2 - 0.1 + 53 .
we have not found any ob star catalogued toward the interior of the shell .
but of course we can not discard that this shell has been originated by the action of massive stars .
our x - ray analysis suggests that sx1 , which lies in the interior of the molecular shell , can be an o - type star , probably one of the responsibles , together with other undetected ob - type stars and even the snr progenitor , in generate the shell . regarding to a snr origin we consider two possible scenarios : i ) the contribution of the snr g18.1 - 0.1 , and ii ) the action of an ancient snr not observed at the present day .
given that the effective action of snr g18.1 - 0.1 is limited to the borders defined by the radio continuum shell , and considering that typical ages for snrs are at most 10@xmath24 yrs , we discard that g18.1 - 0.1 has originated the big shell by itself .
however , as mentioned above , its massive progenitor could have contributed to generate it . by the other hand ,
considering the scenario where the origin of gsh 18.2 - 0.1 + 53 is attributed to an ancient snr that would have injected to the ism a mechanical energy of about @xmath25 erg , from which the shell still preserves about 1% of that energy , it could be a similar case as was proposed for the surroundings of the snr g349.7 + 0.2 @xcite . in that case
it was found that the snr , some iras sources and an uchii region lie in a border of an extended molecular shell , suggested to be the fossil remains of an ancient sn explosion . summarizing , the origin of gsh 18.2 - 0.1 + 53 still remains uncertain and it deserves a deeper study , e.g. new observations toward its interior in order to look for ob - type stars .
the snr g18.1 - 0.1 is located , along the plane of the sky , close to several hii regions ( infrared dust bubbles n21 and n22 , and the hii regions g018.149 - 00.283 and g18.197 - 00.181 ) .
taking into account the catalogued distances and/or systemic velocities of these estructures , and our analysis of the molecular gas related to them , we suggest that all of these objects belong to a same complex at a distance of about 4 kpc .
this could be a clear observational evidence supporting that massive stars born predominantly within dense cores of giant molecular clouds and evolve in clusters , which can generates several neighboring hii regions and snrs .
the main results of this work can be summarised as follows : + ( 1 ) analysing the @xmath0co j=10 emission toward the hii region / snr complex we found several molecular structures very likely related to the sources within the velocity range between 39 and 59 .
the molecular gas morphology suggests that it is very likely being swept and shaped by the expansion of the hii regions , mainly by the infrared dust bubbles n21 and n22 , and the hii region g018.149 - 00.283 .
we obtained a molecular mass for the whole structure related to the complex of about @xmath26 .
we estimated a density of about @xmath27 @xmath28 for the densest portion of this molecular structre .
\(2 ) using spectroscopic optical observations obtained with the 2.15 m telescope at casleo , argentina , we searched for ob - type stars in order to look for sources responsible of ionizing the gas in the hii regions .
we discovered three o - type stars verly likely exciting the bubbles n21 and n22 , and an uncatalogued hii region northward bubble n22 , respectively .
thus , we confirm the presence of an unknown hii region . additionally , we discovered four b0 - 5 stars , one toward the bubble n22 and the others within the hii region g018.149 - 00.283 .
\(3 ) taking into account that hot massive stars usually are x - ray emmiters , we looked for x - ray point sources in the chandra source catalog .
we found two sources , one of them coincides with the above mentioned star that is exciting the uncatalogued hii region . estimating an upper limit for its x - ray emission level we can only suggest that it may be an o - type star . on the other side
, the x - ray analysis of the other source , which lies over the snr shell , strongly suggests that it is an early o - type star ( o3-o5 ) .
\(4 ) by inspecting the ism surrounding the hii region / snr complex in a large scale , we discovered a big molecular shell peaking between 51 and 55 . assuming a distance of 4 kpc
, this shell has an extension of about 70 pc @xmath1 28 pc , and the analysed hii region / snr complex appears to be located in its southern border .
we wish to thank the anonymous referee whose comments and suggestions have helped to considerably improve this paper . s.p .
m.o . and j.f.a.c .
are members of the _ carrera del investigador cientfico _ of conicet , argentina .
is a postdoctoral fellow of conicet , argentina .
this work was partially supported by argentina grants awarded by uba ( ubacyt ) , conicet and anpcyt .
the ccd and data acquisition system at casleo has been financed by r. m. rich trough u.s .
nsf grant ast-90 - 15827 . | we perform a multiwavelength study toward the snr g18.1 - 0.1 and nearby several hii regions ( infrared dust bubbles n21 and n22 , and the hii regions g018.149 - 00.283 and g18.197 - 00.181 ) .
our goal is to provide observational evidence supporting that massive stars usually born in clusters from the same molecular cloud , which then produce , along their evolution , different neighboring objects such as hii regions , interstellar bubbles and supernova remnants .
we suggest that the objects analysed in this work belong to a same complex located at the distance of about 4 kpc . using molecular data we inspected the interstellar medium toward this complex and from optical and x - ray observations we looked for ob - type stars in the region . analysing public @xmath0co j=10 data we found several molecular structures very likely related to the hii region / snr complex .
we suggest that the molecular gas is very likely being swept and shaped by the expansion of the hii regions . from spectroscopic optical observations obtained with the 2.15 m telescope at casleo , argentina , we discovered three o - type stars very likely exciting the bubbles n21 and n22 , and an uncatalogued hii region northward bubble n22 , respectively . also we found four b0 - 5 stars , one toward the bubble n22 and the others within the hii region g18.149 - 0.283 . by inspecting the chandra source catalog we found two point x - ray sources and we suggest that one of them is an early o - type star .
finally we inspected the large scale interstellar medium around this region .
we discovered a big molecular shell of about 70 pc @xmath1 28 pc in which the analysed complex appears to be located in its southern border .
[ firstpage ] ism : supernova remnants - ( ism : ) hii regions - ism : clouds - stars : massive |
in nature there are few known processes that allow the production of black holes .
the best well - known is the gravitational collapse of a massive star or cluster of stars . due to fermionic degeneracy pressure these black holes
can not have a mass below the oppenheimer - volkoff limiting mass ( @xmath2 in recent calculations ) .
another one is the quantum schwinger - like process of black hole pair creation in an external field .
these black holes can have planck dimensions and thus their evolution is ruled by quantum effects .
moreover , gravitational pair creation involves topology changing processes , and allows a study of the statistical properties of black holes , namely : it favors the conjecture that the number of internal microstates of a black hole is given by the exponential of one - quarter of the area of the black hole horizon , and it gives useful clues to the black hole information paradox .
the evaluation of the black hole pair creation rate has been done at the semiclassical level using the instanton method .
an instanton is an euclidean solution that interpolates between the initial and final states of a classically forbidden transition , and is a saddle point for the euclidean path integral that describes the pair creation rate .
this instanton method has been first introduced in studies about decay of metastable termodynamical states , and it has been applied in the context of pair creation of particles and fields in the absence of gravity , by several authors ( see , e.g. , @xcite for a review ) . the instanton method has been also introduced as a framework for quantum gravity , with successful results in the analysis of gravitational thermodynamic issues and black hole pair production processes , among others ( see @xcite ) .
the regular instantons that describe the process we are interested in - the pair creation of black holes in an external field - can be obtained by analytically continuing ( i ) a solution found by ernst @xcite , ( ii ) the de sitter black hole solutions , ( iii ) a solution found by kinnersley and walker known as the c - metric @xcite , ( iv ) a combination of the above solutions , or ( v ) the domain wall solution @xcite .
to each one of these five families of instantons corresponds a different way by which energy can be furnished in order to materialize the pair of black holes and to accelerate them apart . in case ( i )
the energy is provided by the electromagnetic lorentz force , in case ( ii ) the strings tension furnishes this energy , in case ( iii ) the energy is provided by the rapid cosmological expansion associated to the positive cosmological constant @xmath3 , in case ( iv ) the energy is provided by a combination of the above fields , and finally in case ( v ) the energy is given by the repulsive gravitational field of the domain wall .
since these solutions play a fundamental role , we will now briefly discuss some of them that are less known .
the c - metric @xcite describes a pair of black holes ( neutral or charged ) uniformly accelerating in opposite directions .
the solution has conical singularities at its angular poles that , when conveniently treated , can be interpreted as two strings from each one of the black holes towards the infinity and whose tension provides the necessary force to pull apart the black holes . by appending a suitable external electromagnetic field ,
ernst @xcite has removed all the conical singularities of the charged flat c - metric .
the ernst solution then describes two oppositely charged black holes undergoing uniform acceleration provided by the lorentz force associated to the external field ( for the magnetic solution see @xcite , while the explicit electric solution can be found in brown @xcite ) .
asymptotically , the ernst solution reduces to the melvin universe @xcite .
the lorentz sector of the c - metric and ernst solution describe the evolution of the black holes after their creation .
the usual de sitter black hole solutions , when euclideanized , give also instantons for pair creation of black holes .
indeed , the de sitter black holes solutions can be interpreted as representing a black hole pair being accelerated by the cosmological constant .
it was believed that the only black hole pairs that could be nucleated were those whose euclidean sector was free of conical singularities ( instantons ) .
this regularity condition restricted the mass and charge of the black holes that could be produced , and physically it meant that the only black holes that could be pair produced were those that are in thermodynamic equilibrium .
however , wu @xcite , and bousso and hawking @xcite have shown that euclidean solutions with conical singularities ( sub - maximal instantons ) may also be used as saddle points for the pair creation process , as long as the spacelike boundary of the manifold is chosen in order to contain the conical singularity and the metric is specified there . in this way ,
pair creation of black holes whose horizons are not in thermodynamic equilibrium is also allowed .
we will describe the studies that have been done on pair creation of black holes in an external field .
\(i ) the suggestion that the pair creation of black holes could occur in quantized einstein - maxwell theory has been given by gibbons @xcite in 1986 , who has proposed that extremal black holes could be produced in a background magnetic field and that the appropriate instanton describing the process could be obtained by euclideanizing the extremal ernst solution .
this idea has been recovered by giddings and garfinkle @xcite that confirmed the expectation of @xcite and , in addition , they have constructed an ernst instanton that describes pair creation of nonextreme black holes .
the explicit calculation of the rate for this last process has lead garfinkle , giddings and strominger @xcite to conclude that the pair creation rate of nonextreme black holes is enhanced relative to the pair creation of monopoles and extreme black holes by a factor @xmath4 , where @xmath5 is the hawking - bekenstein entropy of the black hole and @xmath6 is the area of the black hole event horizon .
this issue of black hole pair creation in a background magnetic field and the above relation between the pair creation rate and the entropy has been further investigated by dowker , gauntlett , kastor and traschen @xcite , by dowker , gauntlett , giddings and horowitz @xcite , and by ross @xcite , but now in the context of the low energy limit of string theory and in the context of five - dimensional kaluza - klein theory . to achieve their aim
they have worked with an effective dilaton theory which , for particular values of the dilaton parameter , reduces to the above theories , and they have explicitly constructed the dilaton ernst instantons that describe the process .
the one - loop contribution to the magnetic black hole pair creation problem has been given by yi @xcite .
brown @xcite has analyzed the pair creation of charged black holes in an electric external field . hawking , horowitz and ross @xcite ( see also hawking and horowitz @xcite )
have related the rate of pair creation of extreme black holes with the area of the acceleration horizon . in the nonextreme case , the rate has an additional contribution from the area of the black hole horizon .
from these relations emerges an explanation for the fact , mentioned above , that the pair creation rate of nonextreme black holes is enhanced relative to the pair creation of extremal black holes by precisely the factor @xmath7 . for a detailed discussion concerning the reason why this factor involves only @xmath6 and not two times this value see also emparan @xcite .
it has to do with the fact that the internal microstates of two members of the black hole pair are correlated .
\(ii ) the study of pair creation of de sitter ( ds ) black holes has been also investigated .
notice that the ds black hole solution can be interpreted as a pair of ds black holes that are being accelerated apart by the positive cosmological constant .
the cosmological horizon can be seen as an acceleration horizon that impedes the causal contact between the two black holes , and this analogy is perfectly identified for example when we compare the carter - penrose diagrams of the c - metric and of the ds schwarzschild black hole , for example .
the study on pair creation of black holes in a ds background has begun in 1989 by mellor and moss @xcite , who have identified the gravitational instantons that describe the process ( see also romans @xcite for a detailed construction of these instantons ) .
the explicit evaluation of the pair creation rates of neutral and charged black holes accelerated by a cosmological constant has been done by mann and ross @xcite .
this process has also been discussed in the context of the inflationary era undergone by the universe by bousso and hawking @xcite .
garattini @xcite , and volkov and wipf @xcite have computed the one - loop factor for this pair creation process , something that in gravity quantum level is not an easy task .
booth and mann @xcite have analyzed the cosmological pair production of charged and rotating black holes .
pair creation of dilaton black holes in a ds background has also been discussed by bousso @xcite .
\(iii ) in 1995 , hawking and ross @xcite and eardley , horowitz , kastor and traschen @xcite have discussed a process in which a cosmic string breaks and a pair of black holes is produced at the ends of the string .
the string tension then pulls the black holes away , and the c - metric provides the appropriate instantons to describe their creation . in order to ensure that this process is physically consistent achcarro , gregory and kuijken @xcite , and gregory and hindmarsh @xcite have shown that a conical singularity can be replaced by a nielson - olesen vortix .
this vortix can then pierce a black hole @xcite , or end at it @xcite .
moreover , it has been suggested that even topologically stable strings can end at a black hole @xcite-@xcite .
\(iv ) we can also consider a pair creation process , analyzed by emparan @xcite , involving cosmic string breaking in a background magnetic field . in this case
the lorentz force is in excess or in deficit relative to the net force necessary to furnish the right acceleration to the black holes , and this discrepancy is eliminated by the string tension .
the instantons describing this process are a combination of the ernst and c - metric intantons .
\(v ) the gravitational repulsive energy of a domain wall provides another mechanism for black hole pair creation .
this process has been analyzed by caldwell , chamblin and gibbons @xcite , and by bousso and chamblin @xcite in a flat background , while in an anti - de sitter background the pair creation of topological black holes ( with hyperbolic topology ) has been analyzed by mann @xcite .
other studies concerning the process of pair creation in a generalized background is done in @xcite .
it has been noticed that oddly the pair creation of electric black holes was apparently enhanced relative to the pair creation of magnetic black holes .
this was a consequence of the fact that the maxwell action has opposite signs in the two cases .
now , this discrepancy between the two pair creation rates was not consistent with the idea that electric and magnetic black holes should have identical quantum properties .
this issue has been properly and definitively clarified by hawking and ross @xcite and by brown @xcite , who have shown that the magnetic and electric solutions differ not only in their actions , but also in the nature of the boundaries conditions that can be imposed on them .
more precisely , one can impose the magnetic charge as a boundary condition at infinity but , in the electric case , one instead imposes the chemical potential as a boundary condition . as a consequence they proposed that the electric action should contain an extra maxwell boundary term .
this term cancels the opposite signs of the maxwell action , and the pair creation rate of magnetic and electric black holes is equal .
the process of black hole pair creation gives also useful clues to the discussion of the black hole information loss problem @xcite .
due to the thermal hawking radiation the black holes evaporate .
this process implies that one of the following three scenarios occurs ( see @xcite for reviews ) : ( i ) the information previously swallowed to the black hole is destroyed , ( ii ) this information is recovered to the exterior through the hawking radiation , or ( iii ) the endpoint of the evaporation is a plank scale remnant which stores the information .
there are serious difficulties associated to each one of this scenarios .
scenario ( i ) implies non - unitarity and violation of energy conservation , scenario ( ii ) implies violation of locality and causality , and the main problem with scenario ( iii ) is that a huge energy is needed in order to store all the information that has been swallowed by the black hole , and a planck scale remnant has very little energy .
pair creation of black holes has been used to test these scenarios .
indeed , it has been argued @xcite that if one demands preservation of unitarity and of locality then a careful analysis of the one - loop contribution to the pair creation process indicates that the hawking process would leave behind a catastrophic infinite number of remnants .
so the remnant hypothesis seems to be discarded , although some escape solutions can be launched @xcite . on the other side , hawking , horowitz and ross @xcite have called attention to the fact that the same instantons that describe pair creation can , when reversed in time , describe their pair annihilation , as long as the black holes have appropriate initial conditions such that they come to rest at the right critical separation ( this annihilation process was also discussed by emparan @xcite ) .
one can then construct @xcite an argument that favors the information loss scenario : black holes previously produced as a particle - antiparticle pair can accrete information and annihilate , with their energy being given off as electromagnetic and gravitational radiation .
therefore , the information loss scenario seems to occur at least in this annihilation process . an important process that accompanies the production of the black hole pair and the subsequent acceleration that they suffer is the emission of electromagnetic and gravitational radiation . in an asymptotically flat background ,
an estimate for the amount of gravitational radiation radiated during the pair creation period has been given by cardoso , dias and lemos @xcite : @xmath9 , where @xmath10 is the mass of each one of the created black holes and @xmath11 is the lorentz factor .
this value can lead , under appropriate numbers of @xmath10 and @xmath12 to huge quantities , and is a very good candidate to emission of gravitational radiation .
for example , for black holes with 30 times the planck mass and with @xmath13 of the velocity of light , the gravitational energy released is @xmath14 , which is about 100 times the rest energy of the pair .
the gravitational radiative properties of the resulting accelerated black holes has been analyzed by bik , and pravda and pravdova @xcite . in a ds background ,
the gravitational radiation emitted by uniformly accelerated sources without horizons has been analyzed by bik and krtou @xcite , and the radiative properties of accelerated black holes have been studied by krtou and podolsk ' y @xcite . in this paper
we discuss the process in which a cosmic string nucleates in a de sitter ( ds ) background , and then breaks producing a pair of black holes at its ends . therefore , the energy to materialize and accelerate the pair comes from the positive cosmological constant and , in addition , from the string tension .
this process is a combination of the processes considered in ( ii ) @xcite-@xcite and in ( iii ) @xcite-@xcite .
the instantons for this process can be constructed by analytically continuing the ds c - metric found by plebaski and demiaski @xcite and analyzed by podolsk and griffiths @xcite , and in detail by dias and lemos @xcite .
the plan of this paper is as follows . in sec .
[ sec : instanton - method ] , we describe the semiclassical instanton method used to evaluate the pair creation rate . in section [ sec : ds c - inst ] we construct , from the ds c - metric , the instantons that describe the pair creation process . then , in section [ sec : calc - i ] , we explicitly evaluate the pair creation rate for each one of the cases discussed in sec .
[ sec : ds c - inst ] . in sec .
[ sec : entropy ] we verify that the usual relation between pair creation rate , entropy and total area holds also for the pair creation process discussed in this paper . finally , in sec .
[ sec : conc ] concluding remarks are presented . in the appendix
a heuristic derivation of the pair creation rates is given . throughout this paper
we use units in which @xmath15 .
the pair creation of black holes in a de sitter ( ds ) background is described , according to the no - boundary proposal of hartle and hawking @xcite , by the propagation from nothing to a 3-surface boundary @xmath16 .
the amplitude for this process is given by the wave function @xmath17d[a_{\mu}]e^{-i(g_{\mu\nu},a_{\mu})}\ : , \label{wave } \end{aligned}\ ] ] where @xmath18 and @xmath19 are the induced metric and electromagnetic potential on the boundary @xmath20 of a compact manifold @xmath21 , @xmath22 $ ] is a measure on the space of the metrics @xmath23 and @xmath24 $ ] is a measure on the space of the maxwell field @xmath25 , and @xmath26 is their euclidean action .
the path integral is over all compact metrics and potentials on manifolds @xmath21 with boundary @xmath16 , which agree with the boundary data on @xmath16 . for a detailed discussion of the no - boundary proposal applied to the study of black hole pair creation see bousso and chamblin @xcite . in the semiclassical instanton approximation ,
the dominant contribution to the path integral comes from metrics and maxwell fields which are near the solutions ( instantons ) that extremalize the euclidean action and satisfy the boundary conditions .
thus , considering small fluctuations around this solution , @xmath27 and @xmath28 , the action expands as @xmath29 where @xmath30 are quadratic in @xmath31 and @xmath32 , and dots denote higher order terms .
the wave function , that describes the creation of a black hole pair from nothing , is then given by @xmath33 , where @xmath34 is the classical action of the gravitational instanton that mediates the pair creation of black holes , and the prefactor @xmath35 is the one loop contribution from the quantum quadratic fluctuations in the fields , @xmath30 .
similarly , the wave function that describes the nucleation of a ds space with a string from nothing is @xmath36 , and the wave function describing the nucleation of a ds space from nothing is @xmath37 .
the nucleation probability of the ds space from nothing , of the ds space with a string from nothing , and of a space with a pair of black holes from nothing is then given by @xmath38 , @xmath39 and @xmath40 , respectively .
we may now ask four questions : what is the probability for ( i ) pair creation of black holes in a ds spacetime , ( ii ) the nucleation of a string in a ds background , ( iii ) the process in which a string in a ds background breaks and a pair of black holes is created , and ( iv ) the combined process ( ii)+(iii ) . in the process ( i ) the energy to materialize the pair comes only from the positive cosmological constant background , @xmath3 .
the system does not contain a string and the probability for this process has been found in @xcite .
the aim of the present paper is to compute explicitly the probability for processes ( ii)-(iv ) .
it is important to note that in the process ( iii ) , one assumes that the initial background contains a string , i.e. , the question that is being asked is : given that the string is already present in our initial system , what is the probability that it breaks and a pair of black holes is produced and accelerated apart by @xmath3 and by the string tension ?
on the other side , in ( iv ) one is asking : starting from a pure ds background , what is the probability that a string nucleates on it and then breaks forming a pair of black holes ? naturally , the probability for process ( iv ) is the product of the probability for process ( ii ) and the probability for process ( iii ) . according to the no - boundary proposal ,
the nucleation rate of a string in a ds background is proportional to @xmath41 , i.e. , @xmath42 the pair creation rate of black holes when a string breaks in a ds background is given by @xmath43 and the pair creation rate of black holes when process ( iv ) occurs is given by the product of ( [ pc - rate - string ] ) and ( [ pc - rate - breakstring ] ) , i.e. , @xmath44 in sec .
[ sec : calc - i ] we will find @xmath34 and @xmath45 . in the three relations above , @xmath46 , @xmath47 and @xmath48 are one - loop prefactors which will not be considered in this paper .
the evaluation of this one - loop prefactor has been done only in a small number of cases , namely for the vacuum background by gibbons , hawking and perry @xcite , for the schwarzschild instanton by gross , perry and yaffe @xcite , for other asymptotically flat instantons by young @xcite , for the ds background by gibbons and perry @xcite and christensen and duff @xcite , for the ds - schwarzschild instanton by ginsparg and perry @xcite , young @xcite , volkov and wipf @xcite and garattinni @xcite , and for the ernst instanton by yi @xcite . at this point
we must specify the euclidean action needed to compute the path integral ( [ wave ] ) .
this issue was analyzed and clarified in detail by hawking and ross @xcite and by brown @xcite .
now , due to its relevance for the present paper , we briefly discuss the main results of @xcite .
one wants to use an action for which it is natural to fix the boundary data on @xmath16 specified in ( [ wave ] ) .
that is , one wants to use an action whose variation gives the euclidean equations of motion when the variation fixes these boundary data on @xmath16 @xcite . in the magnetic case
this euclidean action is the einstein - maxwell action with a positive cosmological constant @xmath3 given by @xmath49 where @xmath50 is the determinant of the euclidean metric , @xmath51 is the determinant of the induced metric on the boundary @xmath16 , @xmath52 is the ricci scalar , @xmath53 is the trace of the extrinsic curvature @xmath54 of the boundary , and @xmath55 is the maxwell field strength of the gauge field @xmath56 .
variation of ( [ i ] ) yields @xmath57 , where @xmath58 represents terms giving the equations of motion plus gravitational boundary terms that are discussed in @xcite , and @xmath59 is the unit outward normal to @xmath16 .
thus , variation of ( [ i ] ) gives the equations of motion as long as it is at fixed gauge potential @xmath60 on the boundary .
now , for magnetic black hole solutions , fixing the potential fixes the charge on each of the black holes , since the magnetic charge is just given by the integral of @xmath61 over a 2-sphere lying in the boundary . however , in the electric case , fixing @xmath60 can be regarded as fixing a chemical potential @xmath62 which is conjugate to the charge @xcite .
holding the electric charge fixed is equivalent to fixing @xmath63 on @xmath16 , as the electric charge is given by the integral of the dual of @xmath64 over a 2-sphere lying in @xmath16 .
therefore in the electric case the appropriate euclidean action is @xcite @xmath65 where @xmath66 is defined in ( [ i ] ) .
variation of action ( [ i - electric ] ) yields @xmath67 , and thus it gives the equations of motion when @xmath68 , and so the electric charge , is held fixed . since @xmath69 has opposite signs for dual magnetic and electric solutions , if we took ( [ i ] ) to evaluate both the magnetic and electric actions we would conclude that the pair creation of electric black holes would be enhanced relative to the pair creation of magnetic black holes .
this physically unexpected result does not occur when one considers the appropriate boundary conditions and includes the extra maxwell boundary term in ( [ i - electric ] ) .
we have to be careful @xcite when computing the extra maxwell boundary term in the electric action ( [ i - electric ] ) .
indeed , we have to find a vector potential , @xmath56 , that is regular everywhere in the instanton , including at the horizons .
usually , as we shall see , this requirement leads to unusual choices for @xmath56 .
the need of this requirement is easily understood if we take the example of the electric reissner - nordstrm solution @xcite . in this case , normally , the gauge potential in schwarzschild coordinates is taken to be @xmath70 . however , this potential is not regular at the horizon @xmath71 , since @xmath72 diverges there .
an appropriate choice that yields a regular electromagnetic potential everywhere , including at the horizon is @xmath73 or , alternatively , @xmath74 .
to all these potentials corresponds the field strength @xmath75 .
the ds c - metric has been found by plebaski and demiaski @xcite . the physical properties and interpretation of this solution
have been analyzed by podolsk and griffiths @xcite , and in detail by dias and lemos @xcite .
the ds c - metric describes a pair of uniformly accelerated black holes in a ds background , with the acceleration being provided by the cosmological constant and , in addition , by a string that connects the two black holes along their south poles and pulls them away .
the presence of the string is associated to the conical singularity that exists in the south pole of the ds c - metric ( see , e.g. , @xcite ) . for a detailed discussion on the properties of the ds c - metric we ask the reader to see @xcite . here
we will only mention those which are really essential .
following sec .
[ sec : instanton - method ] , in order to evaluate the black hole pair creation rate we need to find the instantons of the theory , i.e. , we must look into the euclidean section of the ds c - metric and choose only those euclidean solutions which are regular in a way that will be explained soon . to obtain the euclidean section of the ds c - metric from the lorentzian ds c - metric we simply introduce an imaginary time coordinate @xmath76 .
then the gravitational field of the euclidean ds c - metric is given by ( see , e.g. , @xcite ) @xmath77^{-2 } ( { \cal f}d\tau^2 + { \cal f}^{-1}dy^2+{\cal g}^{-1}dx^2 + { \cal g}d\phi^2)\ : , \label{c - metric pcds}\ ] ] @xmath78 where @xmath79 is the cosmological constant , @xmath80 is the acceleration of the black holes , and @xmath10 and @xmath81 are the adm mass and electromagnetic charge of the non - accelerated black hole , respectively .
the maxwell field in the magnetic case is given by @xmath82 while in the electric case it is given by @xmath83 the solution has a curvature singularity at @xmath84 where the matter source is .
the point @xmath85 corresponds to a point that is infinitely far away from the curvature singularity , thus as @xmath86 increases we approach the curvature singularity and @xmath87 is the inverse of a radial coordinate . at most ,
@xmath88 can have four real zeros which we label in ascending order by @xmath89 .
the roots @xmath90 and @xmath91 are respectively the inner and outer charged black hole horizons , and @xmath92 is an acceleration horizon which coincides with the cosmological horizon and has a non - spherical shape .
the negative root @xmath93 satisfies @xmath94 and thus has no physical significance .
the angular coordinate @xmath95 belongs to the range @xmath96 $ ] for which @xmath97 ( when we set @xmath98 we have @xmath99 and @xmath100 ) . in order to avoid a conical singularity in the north pole ,
the period of @xmath101 must be given by @xmath102 and this leaves a conical singularity in the south pole with deficit angle @xmath103 that signals the presence of a string with mass density @xmath104 , and with pressure @xmath105 . when we set the acceleration parameter @xmath106 equal to zero , the ds c - metric reduces to the usual ds@xmath107reissner - nordstrm or ds - schwarzschild solutions without conical singularities .
so far , we have described the solution that represents a pair of black holes accelerated by the cosmological constant and by the string tension .
this solution describes the evolution of the black hole pair after its creation .
now , we want to find a solution that represents a string in a ds background . this solution will describe the initial system , before the breaking of the cosmic string that leads to the formation of the black hole pair . in order to achieve our aim
we note that at spatial infinity the gravitational field of the euclidean ds c - metric reduces to @xmath108 ^ 2}{\biggl [ } \left ( -\frac{\lambda}{3a_0^{\,2}}-1+y^2 \right ) dt^2
\nonumber\\ & & + \frac{dy^2}{-\frac{\lambda}{3a_0^{\,2}}-1+y^2 } + \frac{dx^2}{1-x^2}+ ( 1-x^2)d\phi_0^{\,2 } { \biggr ] } , \label{backg metric - ds}\end{aligned}\ ] ] and the maxwell field goes to zero .
@xmath109 is a constant that represents a freedom in the choice of coordinates , and @xmath110 .
we want that this metric also describes the solution before the creation of the black hole pair , i.e. , we demand that it describes a string with its conical deficit in a ds background .
now , if we want to maintain the intrinsic properties of the string during the process we must impose that its mass density and thus its conical deficit remains constant . after the pair creation we already know that the conical deficit is given by ( [ conic - sing - ds ] ) .
hence , the requirement that the background solution describes a ds spacetime with a conical deficit angle given exactly by ( [ conic - sing - ds ] ) leads us to impose that in ( [ backg metric - ds ] ) one has @xmath111 the arbitrary parameter @xmath109 can be fixed by imposing a matching between ( [ c - metric pcds ] ) and ( [ backg metric - ds ] ) at large spatial distances @xcite , yielding @xmath112 ^ 2/\left [ 2{\cal g}''(x_\mathrm{s } ) \right ] $ ] .
returning back to the euclidean ds c - metric ( [ c - metric pcds ] ) , in order to have a positive definite euclidean metric we must require that @xmath86 belongs to @xmath113 . in general , when @xmath114 , one has conical singularities at the horizons @xmath115 and @xmath116 . in order to obtain a regular solution we have to eliminate the conical singularities at both horizons .
this is achieved by imposing that the period of @xmath117 is the same for the two horizons , and is equivalent to requiring that the hawking temperature of the two horizons be equal . to eliminate the conical singularity at @xmath115 the period of @xmath117
must be @xmath118 ( where @xmath119 is the surface gravity of the acceleration horizon ) , @xmath120 this choice for the period of @xmath117 also eliminates simultaneously the conical singularity at the outer black hole horizon , @xmath91 , if and only if the parameters of the solution are such that the surface gravities of the black hole and acceleration horizons are equal ( @xmath121 ) , i.e. @xmath122 there are two ways to satisfy this condition .
one is a regular euclidean solution with @xmath123 , and will be called lukewarm c instanton .
this solution requires the presence of an electromagnetic charge .
the other way is to have @xmath124 , and will be called nariai c instanton .
this last solution exists with or without charge .
when we want to distinguish them , they will be labelled by charged nariai and neutral nariai c instantons , respectively .
we now turn our attention to the case @xmath125 and @xmath126 , which obviously requires the presence of charge .
when this happens the allowed range of @xmath86 in the euclidean sector is simply @xmath127 .
this occurs because when @xmath125 the proper distance along spatial directions between @xmath92 and @xmath91 goes to infinity .
the point @xmath91 disappears from the @xmath128 section which is no longer compact but becomes topologically @xmath129 .
thus , in this case we have a conical singularity only at @xmath92 , and so we obtain a regular euclidean solution by simply requiring that the period of @xmath117 be equal to ( [ period tau - ya pcds ] ) .
we will label this solution by cold c instanton .
finally , we have a special solution that satisfies @xmath130 and that is regular when condition ( [ period tau - ya pcds ] ) is satisfied .
this instanton will be called ultracold c instanton and can be viewed as a limiting case of both the charged nariai c instanton and cold c instanton .
below , we will describe in detail each one of these four c instantons , following the order : ( a ) lukewarm c instanton , ( b ) cold c instanton , ( c ) nariai c instanton , and ( d ) ultracold c instanton .
these instantons are the c - metric counterparts ( @xmath131 ) of the @xmath98 instantons that have been constructed from the euclidean section of the ds@xmath107reissner - nordstrm solution ( @xmath98 ) @xcite .
the original name of the @xmath98 instantons is associated to the relation between their temperatures : @xmath132 .
this relation is preserved by their c - metric counterparts discussed in this paper , and we preserve the @xmath98 nomenclature .
the ultracold instanton could also , very appropriately , be called nariai bertotti - robinson instanton ( see @xcite ) .
these four families of instantons will allow us to calculate the pair creation rate of accelerated ds@xmath107reissner - nordstrm black holes in sec .
[ sec : calc - i ] .
as is clear from the above discussion , when the charge vanishes the only regular euclidean solution that can be constructed is the neutral nariai c instanton .
the same feature is present in the @xmath98 case where only the neutral nariai instanton is available @xcite .
for the lukewarm c instanton the gravitational field is given by ( [ c - metric pcds ] ) with the requirement that @xmath88 satisfies @xmath133 and @xmath134 .
in this case we can then write ( onwards the subscript
@xmath135 " means lukewarm ) @xmath136 with @xmath137 the parameters @xmath106 , @xmath3 , @xmath10 and @xmath81 , written as a function of @xmath92 and @xmath91 , are @xmath138 thus , the mass and the charge of the lukewarm c instanton are necessarily equal , @xmath139 , as occurs with its @xmath98 counterpart , the lukewarm instanton @xcite . the demand that @xmath140 is real requires that @xmath141 so the lukewarm c instanton has a lower maximum mass and a lower maximum charge than the @xmath98 lukewarm instanton @xcite and , for a fixed @xmath3 , as the acceleration parameter @xmath106 grows this maximum value decreases monotonically .
for a fixed @xmath3 and for a fixed mass below @xmath142 , the maximum value of the acceleration is @xmath143 .
as we said , the allowed range of @xmath86 in the euclidean sector is @xmath113 .
then , the period of @xmath117 , ( [ period tau - ya pcds ] ) , that avoids the conical singularity at both horizons is @xmath144 and @xmath145 is the common temperature of the two horizons . using the fact that @xmath146 [ see ( [ fg ] ) ] we can write @xmath147 and the only real zeros of @xmath148 are the south and north pole @xmath149 when @xmath106 goes to zero we have @xmath150 and @xmath151 . the period of @xmath101 , ( [ period phi ] ) , that avoids the conical singularity at the north pole ( and leaves one at the south pole responsible for the presence of the string ) is @xmath152 when @xmath106 goes to zero we have @xmath153 and the conical singularity disappears .
the topology of the lukewarm c instanton is @xmath154 ( @xmath155 , @xmath113 , @xmath156 , and @xmath157 ) .
the lorentzian sector describes two ds black holes being accelerated by the cosmological background and by the string , so this instanton describes pair creation of nonextreme black holes with @xmath139 .
the gravitational field of the cold c instanton is given by ( [ c - metric pcds ] ) with the requirement that the size of the outer charged black hole horizon @xmath91 is equal to the size of the inner charged horizon @xmath90 .
let us label this degenerated horizon by @xmath158 : @xmath159 and @xmath160 . in this case
, the function @xmath88 can be written as ( onwards the subscript @xmath161 " means cold ) @xmath162 with @xmath163 and the roots @xmath158 , @xmath93 and @xmath92 are given by @xmath164 the mass and the charge parameters of the solution are written as a function of @xmath158 as @xmath165 and , for a fixed @xmath106 and @xmath3 , the ratio @xmath166 is higher than @xmath167 .
the conditions @xmath160 and @xmath168 require that , for the cold c instanton , the allowed range of @xmath158 is @xmath169 the value of @xmath92 decreases monotonically with @xmath158 and we have @xmath170 .
the mass and the charge of the cold c instanton are also monotonically decreasing functions of @xmath158 , and as we come from @xmath171 into @xmath172 we have @xmath173 so the cold c instanton has a lower maximum mass and a lower maximum charge than the @xmath98 cold instanton , and , for a fixed @xmath3 , as the acceleration parameter @xmath106 grows this maximum value decreases monotonically . for a fixed @xmath3 and for a fixed mass below @xmath174 ,
the maximum value of the acceleration is @xmath175 .
as we have already said , the allowed range of @xmath86 in the euclidean sector is @xmath127 and does not include @xmath116 .
then , the period of @xmath117 , ( [ period tau - ya pcds ] ) , that avoids the conical singularity at the only horizon of the cold c instanton is @xmath176 and @xmath177 is the temperature of the acceleration horizon . in what concerns the angular sector of the cold c instanton , @xmath178 is given by ( [ fg ] ) , and its only real zeros are the south and north pole , @xmath179 with @xmath180 where @xmath10 and @xmath81 are fixed by ( [ mq pcds ] ) , for a given @xmath106 , @xmath3 and @xmath158 .
when @xmath106 goes to zero we have @xmath150 and @xmath151 .
the period of @xmath101 , @xmath181 , that avoids the conical singularity at the north pole ( and leaves one at the south pole responsible for the presence of the string ) is given by ( [ period phi ] ) with @xmath182 defined in ( [ polos - cold ] ) .
the topology of the cold c instanton is @xmath183 , since @xmath184 is at an infinite proper distance ( @xmath185 , @xmath127 , @xmath156 , and @xmath186 ) .
the surface @xmath184 is then an internal infinity boundary that will have to be taken into account in the calculation of the action of the cold c instanton ( see sec .
[ sec : cold - rate ] ) .
the lorentzian sector of this cold case describes two extreme ( @xmath187 ) ds black holes being accelerated by the cosmological background and by the string , and the cold c instanton describes pair creation of these extreme black holes . in the case of the nariai c instanton , we require that the size of the acceleration horizon @xmath92 is equal to the size of the outer charged horizon @xmath91 .
let us label this degenerated horizon by @xmath158 : @xmath188 and @xmath189 . in this case
, the function @xmath88 can be written as ( onwards the subscript
@xmath190 " means nariai ) @xmath191 where @xmath12 is defined by ( [ gamma - pcds ] ) , the roots @xmath158 and @xmath93 are given by ( [ zerosy1-cold ] ) and ( [ zerosy2-cold ] ) , and @xmath90 is given by @xmath192 .
the mass and the charge of the solution are defined as a function of @xmath158 by ( [ mq pcds ] ) .
the conditions @xmath189 and @xmath193 require that for the nariai c instanton , the allowed range of @xmath158 is @xmath194 the value of @xmath90 decreases monotonically with @xmath158 and we have @xmath195 .
contrary to the cold c instanton , the mass and the charge of the nariai c instanton are monotonically increasing functions of @xmath158 , and as we go from @xmath196 to @xmath172 we have @xmath197 note that @xmath198 implies @xmath199 . for a fixed @xmath3 and for a mass fixed between @xmath200 ,
the acceleration varies as @xmath201 . at this point ,
one has apparently a problem that is analogous to the one that occurs with the @xmath98 neutral nariai instanton @xcite and with the @xmath98 charged nariai instanton @xcite .
indeed , as we said in the beginning of this section , the allowed range of @xmath86 in the euclidean sector is @xmath113 in order to obtain a positive definite metric .
but in the nariai case @xmath124 , and so it seems that we are left with no space to work with in the euclidean sector .
however , as in @xcite , the proper distance between @xmath92 and @xmath91 remains finite as @xmath202 , as is shown in detail in @xcite where the nariai c - metric is constructed and analyzed . in what follows we
briefly exhibit the construction .
we first set @xmath203 and @xmath204 , in order that @xmath205 measures the deviation from degeneracy , and the limit @xmath206 is obtained when @xmath207 .
now , we introduce a new time coordinate @xmath208 , @xmath209 , and a new radial coordinate @xmath210 , @xmath211 , where @xmath212 and @xmath213 correspond , respectively , to the horizons @xmath91 and @xmath92 , and @xmath214 condition ( [ range - gamma - cnariai ] ) implies @xmath215 with @xmath216 .
then , in the limit @xmath207 , from ( [ c - metric pcds ] ) and ( [ f - cnariai ] ) , we obtain the gravitational field of the nariai c instanton @xmath217 \ : .
\label{nariai - c - metric pcds}\end{aligned}\ ] ] where @xmath210 runs from @xmath218 to @xmath219 , and @xmath220^{-2}\ : .
\label{rfactor pcds}\end{aligned}\ ] ] in the new coordinates @xmath208 and @xmath210 , the maxwell field for the magnetic case is still given by ( [ f - mag ] ) , while in the electric case we have now @xmath221 the period of @xmath208 of the nariai c instanton is simply @xmath222 .
the function @xmath178 is given by ( [ fg ] ) , with @xmath10 and @xmath81 fixed by ( [ mq pcds ] ) for a given @xmath106 , @xmath3 and @xmath223 . under these conditions the south and north pole ( which are the only real roots ) are also given by ( [ polos - cold ] ) and ( [ acess - zeros - ang pcds ] ) .
the period of @xmath101 , @xmath224 , that avoids the conical singularity at the north pole ( and leaves one at the south pole responsible for the presence of the string ) is given by ( [ period phi ] ) with @xmath182 defined in ( [ polos - cold ] ) .
the topology of the nariai c instanton is @xmath154 ( @xmath225 , @xmath226 , @xmath156 , and @xmath227 ) .
the nariai c instanton transforms into the nariai instanton when we take the limit @xmath98 @xcite .
the lorentzian sector is conformal to the direct topological product of @xmath228 , i.e. , of a ( 1 + 1)-dimensional de sitter spacetime with a deformed 2-sphere of fixed size .
to each point in the sphere corresponds a @xmath229 spacetime , except for one point - the south pole - which corresponds a @xmath229 spacetime with a string @xcite .
when we set @xmath98 the @xmath230 is a round 2-sphere free of the conical singularity and so , without the string . in this case
it has been shown @xcite that the nariai solution decays through the quantum tunnelling process into a slightly non - extreme ds black hole pair ( for a complete review on this subject see , e.g. , @xcite ) .
we then naturally expect that an analogous quantum instability is present in the nariai c - metric .
therefore , the nariai c instanton describes the creation of a nariai c universe that then decays into a slightly non - extreme ( @xmath231 ) pair of black holes accelerated by the cosmological background and by the string .
we recall again that the neutral nariai c instanton with @xmath232 is the only regular euclidean solution that can be constructed from the ds c - metric when the charge vanishes .
the same feature is present in the @xmath98 case where only the neutral nariai instanton with @xmath233 is available @xcite . in the case of the ultracold c instanton
, we require that the size of the three horizons ( @xmath92 , @xmath91 and @xmath90 ) is equal , and let us label this degenerated horizon by @xmath158 : @xmath234 . in this case , the function @xmath88 can be written as ( onwards the subscript
@xmath235 " means ultracold ) @xmath236 where @xmath12 is defined by ( [ gamma - pcds ] ) and the roots @xmath158 and @xmath93 are given by ( [ zerosy1-cold ] ) and ( [ zerosy2-cold ] ) , respectively .
given the values of @xmath106 and @xmath3 , @xmath158 can take only the value @xmath237 the mass and the charge of the solution , defined by ( [ mq - cold ] ) , are then given by @xmath238 and these values are the maximum values of the mass and charge of both the cold and charged nariai instantons .
being a limiting case of both the cold c instanton and of the charged nariai c instanton , the ultracold c instanton presents similar features .
the appropriate analysis of this solution ( see @xcite for a detailed discussion ) requires that we first set @xmath239 and @xmath240 , with @xmath205 .
then , we introduce a new time coordinate @xmath208 , @xmath241 , and a new radial coordinate @xmath210 , @xmath242 , where @xmath243 .
finally , in the limit @xmath244 , from ( [ c - metric pcds ] ) and ( [ f - ultra ] ) , we obtain the gravitational field of the ultracold c instanton @xmath245 , \nonumber \\ & & { \rm with } \qquad { \cal r}^2(x)=\left ( ax+\sqrt{2(\lambda+3a^2)}\right ) ^{-2}\ : .
\label{ultracold}\end{aligned}\ ] ] notice that the spacetime factor @xmath246 is just euclidean space in rindler coordinates , and therefore , under the usual coordinate transformation , it can be putted in the form @xmath247 .
@xmath212 corresponds to the rindler horizon and @xmath248 corresponds an internal infinity boundary . in
the new coordinates @xmath208 and @xmath210 , the maxwell field for the magnetic case is still given by ( [ f - mag ] ) , while in the electric case we have now @xmath249 the period of @xmath208 of the ultracold c instanton is simply @xmath250 .
the function @xmath178 is given by ( [ fg ] ) , with @xmath10 and @xmath81 fixed by ( [ mq - ultra ] ) for a given @xmath106 and @xmath3 . under these conditions the south and north pole ( which are the only real roots ) are also given by ( [ polos - cold ] ) and ( [ acess - zeros - ang pcds ] ) .
the period of @xmath101 , @xmath251 , that avoids the conical singularity at the north pole is given by ( [ period phi ] ) with @xmath182 defined in ( [ polos - cold ] ) .
the topology of the ultracold c instanton is @xmath252 , since @xmath248 is at an infinite proper distance ( @xmath253 , @xmath254 , @xmath156 , and @xmath255 ) .
the surface @xmath248 is then an internal infinity boundary that will have to be taken into account in the calculation of the action of the ultracold c instanton ( see sec .
[ sec : ultracold - rate ] ) .
the ultracold c instanton transforms into the ultracold instanton when we take the limit @xmath98 @xcite .
the lorentzian sector is conformal to the direct topological product of @xmath256 , i.e. , of a ( 1 + 1)-dimensional minkowski spacetime with a deformed 2-sphere of fixed size .
to each point in the sphere corresponds a @xmath257 spacetime , except for one point - the south pole - which corresponds a @xmath257 spacetime with a string @xcite .
we can , appropriately , label this solution as nariai bertotti - robinson c universe ( see @xcite ) .
when we set @xmath98 the @xmath230 is a round 2-sphere free of the conical singularity and so , without the string . in an analogous way to the nariai c universe , the nariai bertotti - robinson c universe is unstable and decays into a slightly non - extreme ( @xmath258 ) pair of black holes accelerated by the cosmological background and by the string .
the ultracold c instanton mediates this decay .
the allowed range of @xmath10 and @xmath81 for each one of the four c instantons is sketched in fig .
[ mq - fig ] .
relation @xmath259 vs @xmath260 for a fixed value of @xmath106 and @xmath3 for the four ds c instantons .
@xmath261 represents the lukewarm c instanton ( @xmath139 ) , @xmath262 represents the cold c instanton , @xmath263 represents the nariai c instanton , and @xmath264 represents the ultracold c instanton .
point @xmath265 represents the neutral nariai c instanton , the only instanton for the uncharged case .
when we set @xmath266 we obtain the charge / mass relation of the flat c - metric and flat ernst instantons , and setting @xmath98 yields the charge / mass relation for the ds instantons @xcite .
this reveals the close link that exists between the instantons that describe the pair creation process in different backgrounds .
@xmath267 , @xmath268 and @xmath269 . , height=201 ]
in the last section we have found the regular ds c instantons that can be interpreted as describing a circular motion in the euclidean sector of the solution with a period @xmath270 .
each of these instantons is the euclidean continuation of a lorentzian solution that describes a pair of black holes that start at rest at the creation moment , and then run hyperbolically in opposite directions , due to a string that accelerates them .
now , in order to compute the pair creation rate of the corresponding black holes , we have to slice the instanton , the euclidean trajectory , in half along @xmath271 and @xmath272 , where the velocities vanish .
the resulting geometry is precisely that of the moment of closest approach of the black holes in the lorentzian sector of the ds c - metric , and in this way the euclidean and lorentzian solutions smoothly match together . in particular ,
the extrinsic curvature vanishes for both surfaces and therefore , they can be glued to each other . in this section
, we will compute the black hole pair creation rates given in ( [ pc - rate - breakstring ] ) or ( [ pc - rate ] ) for each one of the four cases considered in sec .
[ sec : ds c - inst ] .
moreover , we will also compute the pair creation rate of black holes whose nucleation process is described by sub - maximal instantons . for the reason explained in sec .
[ sec : instanton - method ] , the magnetic action will be evaluated using ( [ i ] ) , while the electric action will be computed using ( [ i - electric ] ) . in general , in ( [ i ] ) and ( [ i - electric ] ) we identify @xmath273 with the surfaces of zero extrinsic curvature discussed just above .
it then follows that the boundary term @xmath274 vanishes .
however , as we already mentioned in sec .
s [ sec : cold - inst ] and [ sec : ultracold - inst ] , in the cold and ultracold case we have in addition an internal infinity boundary , @xmath275 , for which the boundary action term does not necessarily vanish .
the domain of validity of our results is the particle limit , @xmath276 , for which the radius of the black hole , @xmath277 , is much smaller than the typical distance between the black holes at the creation moment , @xmath278 . in order to compute the black hole pair creation rate given by ( [ pc - rate - breakstring ] ) or ( [ pc - rate ] )
, we need to find @xmath279 and @xmath280 .
the evaluation of the action of the string instanton that mediates the nucleation of a string in the ds spacetime is done using ( [ backg metric - ds ] ) and ( [ delta phi0-ds ] ) , yielding @xmath281 where the integration over @xmath117 has been done in the interval @xmath282 $ ] with @xmath283 , the integration range of @xmath95 was @xmath284 $ ] , the integration interval of @xmath86 was @xmath285 $ ] , and @xmath286 .
the action of the @xmath287 gravitational instanton that mediates the nucleation of the ds spacetime is @xcite @xmath288 the nucleation rate of a string in a ds background is then given by ( [ pc - rate - string ] ) .
for the particle limit , @xmath276 , the mass density of the string , @xmath289 , is given by @xmath290 and @xmath291 thus , the nucleation probability of a string in the ds background decreases when its mass density increases .
we first consider the magnetic case , whose euclidean action is given by ( [ i ] ) , and then we consider the electric case , using ( [ i - electric ] ) , and we verify that these two quantities give the same numerical value .
the boundary @xmath292 that appears in ( [ i ] ) consists of an initial spatial surface at @xmath271 plus a final spatial surface at @xmath293 .
we label these two 3-surfaces by @xmath294 .
each one of these two spatial 3-surfaces is delimitated by a 2-surface at the acceleration horizon and by a 2-surface at the outer black hole horizon .
the two surfaces @xmath294 are connected by a timelike 3-surface that intersects @xmath294 at the frontier @xmath92 and by a timelike 3-surface that intersects @xmath294 at the frontier @xmath91 .
we label these two timelike 3-surfaces by @xmath295 .
thus @xmath296 , and the region @xmath297 within it is compact . with the analysis of section [ sec : lukewarm - inst ] , we can compute all the terms of action ( [ i ] ) .
we start with @xmath298 ^ 4 } \ : , \label{i1-luk } \end{aligned}\ ] ] where we have used @xmath286 , and @xmath92 and @xmath91 are given by ( [ ya - luk ] ) , @xmath299 and @xmath182 are defined by ( [ polos - luk ] ) , and @xmath300 and @xmath301 are respectively given by ( [ beta - luk ] ) and ( [ period phi - luk ] ) .
the maxwell term in the action yields @xmath302 where we have used @xmath303 [ see ( [ f - mag ] ) ] , and @xmath304 . adding all these terms yields for the magnetic action ( [ i ] ) @xmath305 and ,
given that the string is already present in the initial system , the pair creation rate of nonextreme lukewarm black holes when the cosmic string breaks is @xmath306 where ( [ i - string ] ) yields @xmath307 , and @xmath47 is the one - loop contribution not computed here .
@xmath308 is a monotonically increasing function of both @xmath10 and @xmath106 ( for a fixed @xmath3 ) .
when we take the limit @xmath98 of ( [ i - mag - luk ] ) we get @xmath309 recovering the action for the @xmath98 lukewarm instanton @xcite , that describes the pair creation of two non - extreme ds@xmath107reissner - nordstrm black holes accelerated only by the cosmological constant . in the electric case ,
the euclidean action is given by ( [ i - electric ] ) with @xmath310 [ see ( [ f - el ] ) ] .
thus , @xmath311 in order to compute the extra maxwell boundary term in ( [ i - electric ] ) we have to find a vector potential , @xmath56 , that is regular everywhere including at the horizons .
an appropriate choice in the lukewarm case is @xmath312 , which obviously satisfies ( [ f - el ] ) .
the integral over @xmath16 consists of an integration between @xmath92 and @xmath91 along the @xmath271 surface and back along @xmath313 , and of an integration between @xmath271 and @xmath313 along the @xmath116 surface and back along the @xmath115 surface .
the normal to @xmath294 is @xmath314,0,0,0)$ ] , and the normal to @xmath295 is @xmath315,0,0)$ ] .
thus @xmath316 on @xmath295 , and the non - vanishing contribution comes only from the integration along the @xmath313 surface .
the maxwell boundary term in ( [ i - electric ] ) , @xmath317 , is then @xmath318 adding ( [ i1-luk ] ) , ( [ i2-luk - elect ] ) and ( [ i - electric - luk ] ) yields for the electric action ( [ i - electric ] ) @xmath319 where @xmath308 is given by ( [ i - mag - luk ] ) . [ t ] plot of @xmath320 as a function of @xmath10 and @xmath106 for a fixed @xmath3 , where @xmath321 is the action of the lukewarm c instanton , given in ( [ i - mag - luk ] ) , and @xmath322 is the action of de sitter space . see text of sec . [ sec : lukewarm - rate ] .
@xmath323 , @xmath324 and @xmath325 .
, title="fig:",height=201 ] in fig .
[ lukewarm - fig ] we show a plot of @xmath320 as a function of @xmath10 and @xmath106 for a fixed @xmath3 , where @xmath322 is the action of de sitter space . given the pair creation rate , @xmath326 ,
we conclude that , for a fixed @xmath3 and @xmath106 , as the mass and charge of the lukewarm black holes increase , the probability they have to be pair created decreases monotonically .
moreover , for a fixed mass and charge , this probability increases monotonically as the acceleration provided by the string increases .
alternatively , we can discuss the behavior of @xmath327 . in this case , for a fixed mass and charge , the probability decreases monotonically as the acceleration of the black holes increases .
we first consider the magnetic case , whose euclidean action is given by ( [ i ] ) .
the boundary that appears in ( [ i ] ) is given by @xmath328 , where @xmath294 is a spatial surface at @xmath271 and @xmath329 , @xmath295 is a timelike 3-surface at @xmath115 , and the timelike 3-surface @xmath330 is an internal infinity boundary at @xmath184 . with the analysis of sec .
[ sec : cold - inst ] , we can compute all the terms of action ( [ i ] ) .
we start with @xmath331 ^ 4 } \ : , \label{i1-cold } \end{aligned}\ ] ] where we have used @xmath286 , and @xmath158 and @xmath92 are respectively given by ( [ zerosy1-cold ] ) and ( [ zerosy3-cold ] ) , @xmath299 and @xmath182 are defined by ( [ polos - cold ] ) , and @xmath332 and @xmath333 are , respectively , given by ( [ beta - cold ] ) and ( [ period phi ] ) .
the maxwell term in the action yields @xmath334 where we have used @xmath303 [ see ( [ f - mag ] ) ] , and @xmath304 . adding all these terms yields for the magnetic action ( [ i ] ) of the cold case @xmath335 given that the string is already present in the initial system , the pair creation rate of extreme cold black holes when the string breaks is @xmath336 , where @xmath279 is given by ( [ i - string ] ) , and @xmath47 is the one - loop contribution not computed here . in fig .
[ cold - fig ] we show a plot of @xmath337 as a function of @xmath10 and @xmath106 for a fixed @xmath3 . given the pair creation rate , @xmath338
, we conclude that for a fixed @xmath3 and @xmath106 as the mass and charge of the cold black holes increases , the probability they have to be pair created decreases monotonically .
moreover , for a fixed mass and charge , this probability increases monotonically as the acceleration of the black holes increases .
alternatively , we can discuss the behavior of @xmath339 .
in this case , for a fixed mass and charge , the probability decreases monotonically as the acceleration of the black holes increases .
when we take the limit @xmath98 we recover the action for the @xmath98 cold instanton @xcite , which lies in the range latexmath:[$-\frac{3\pi}{2\lambda } \leq i_{\rm mag}^{\rm c}{\bigl pair creation of extreme ds@xmath107reissner - nordstrm black holes accelerated only by the cosmological constant .
[ t ] plot of @xmath341 as a function of @xmath10 and @xmath106 for a fixed @xmath3 , where @xmath342 is the action of the cold c instanton , given in ( [ i - mag - cold ] ) , and @xmath322 is the action of de sitter space ( see text of sec .
[ sec : cold - rate ] ) .
the plot for the ultracold c action , given in ( [ i - mag - ultracold ] ) , is also represented by the curve @xmath343 ( see text of sec . [
sec : ultracold - rate ] ) .
@xmath323 , @xmath324 and @xmath325 .
, title="fig:",height=201 ] in the electric case , the euclidean action is given by ( [ i - electric ] ) with @xmath310 [ see ( [ f - el ] ) ] .
thus , @xmath344 in order to compute the extra maxwell boundary term in ( [ i - electric ] ) we have to find a vector potential , @xmath56 , that is regular everywhere including at the horizons .
an appropriate choice in the cold case is @xmath312 , which obviously satisfies ( [ f - el ] ) .
analogously to the lukewarm case , the non - vanishing contribution to the maxwell boundary term in ( [ i - electric ] ) comes only from the integration along the @xmath329 surface , and is given by @xmath345 adding ( [ i2-cold - elect ] ) and ( [ i - electric - cold ] ) yields ( [ i2-cold ] ) .
thus , the electric action ( [ i - electric ] ) of the cold instanton is equal to the magnetic action , @xmath346 , and therefore electric and magnetic cold black holes have the same probability of being pair created . the nariai c instanton is the only one that can have zero charge . we will first consider the charged nariai c instanton and then the neutral nariai c instanton .
we start with the magnetic case , whose euclidean action is given by ( [ i ] ) .
the boundary that appears in ( [ i ] ) is given by @xmath347 , where @xmath348 is a spatial surface at @xmath349 and @xmath350 , and @xmath295 is a timelike 3-surface at @xmath212 and @xmath213 . with the analysis of sec .
[ sec : nariai - inst ] , we can compute all the terms of action ( [ i ] ) . we start with @xmath351 ^ 4 { \cal{k } } } \ : , \label{i1-nariai } \end{aligned}\ ] ] where we have used @xmath286 , @xmath299 and @xmath182 are defined by ( [ polos - cold ] ) , and @xmath224 is given by ( [ period phi ] ) .
the maxwell term in the action yields @xmath352 where we have used @xmath353 [ see ( [ f - mag ] ) ] , and @xmath304 . adding these three terms yields the magnetic action ( [ i ] ) of the nariai case @xmath354 where @xmath10 and @xmath81 are subjected to ( [ mq - cnariai pcds ] ) .
given that the string is already present in the initial system , the pair creation rate of extreme nariai black holes when the string breaks is @xmath355 , where @xmath279 is given by ( [ i - string ] ) , and @xmath47 is the one - loop contribution not computed here . in fig .
[ nariai - fig - pcads ] we show a plot of @xmath356 as a function of @xmath10 and @xmath106 for a fixed @xmath3 .
given the pair creation rate , @xmath357 , we conclude that for a fixed @xmath3 and @xmath106 as the mass and charge of the nariai black holes increases , the probability they have to be pair created decreases monotonically .
moreover , for a fixed mass and charge , this probability increases monotonically as the acceleration of the black holes increases .
alternatively , we can discuss the behavior of @xmath358 . in this case , for a fixed mass and charge , the probability decreases monotonically as the acceleration of the black holes increases .
when we take the limit @xmath98 we recover the action for the @xmath98 nariai instanton @xcite , which lies in the range latexmath:[$-\frac{\pi}{\lambda } \leq i_{\rm mag}^{\rm n}{\bigl nucleation of a nariai universe that is unstable @xcite and decays through the pair creation of extreme ds@xmath107reissner - nordstrm black holes accelerated only by the cosmological constant .
plot of @xmath360 as a function of @xmath10 and @xmath106 for a fixed @xmath3 , where @xmath361 is the action of the cold c instanton , given in ( [ i - mag - nariai ] ) , and @xmath322 is the action of de sitter space .
the neutral nariai plot is sketched by the curve @xmath362 ( see text of sec . [
sec : nariai - rate ] ) .
the curve @xmath343 represents two times the ultracold c action , given in ( [ i - mag - ultracold ] ) .
@xmath323 , @xmath324 and @xmath325 .
, height=201 ] in the electric case , the euclidean action is given by ( [ i - electric ] ) with @xmath363 [ see ( [ f - el - nariai pcds ] ) ] .
thus , @xmath364 in order to compute the extra maxwell boundary term in ( [ i - electric ] ) , the appropriate vector potential , @xmath56 , that is regular everywhere including at the horizons is @xmath365 , which obviously satisfies ( [ f - el - nariai pcds ] ) .
the integral over @xmath16 consists of an integration between @xmath212 and @xmath213 along the @xmath349 surface and back along @xmath350 , and of an integration between @xmath349 and @xmath350 along the @xmath212 surface , and back along the @xmath213 surface .
the unit normal to @xmath348 is @xmath366 , and @xmath316 on @xmath295 .
therefore , the non - vanishing contribution to the maxwell boundary term in ( [ i - electric ] ) , @xmath367 , comes only from the integration along the @xmath350 surface and is given by @xmath368 adding ( [ i2-nariai - elect ] ) and ( [ i - electric - nariai ] ) yields ( [ i2-nariai ] ) .
so , the electric action ( [ i - electric ] ) of the charged nariai instanton is equal to the magnetic action , @xmath369 , and therefore electric and magnetic charged nariai black holes have the same probability of being pair created .
now , we discuss the neutral nariai c instanton .
this instanton is particulary important since it is the only regular euclidean solution available when we want to evaluate the pair creation of neutral black holes .
the same feature is present in the @xmath98 case where only the neutral nariai instanton is available @xcite . the action of the neutral nariai c instanton is simply given by ( [ i1-nariai ] ) and , for a fixed @xmath3 and @xmath106 , it is always smaller than the action of the charged nariai c instanton ( see line @xmath362 in fig .
[ nariai - fig - pcads ] ) : @xmath370
. thus the pair creation of charged nariai black holes is suppressed relative to the pair creation of neutral nariai black holes , and both are suppressed relative to the ds space .
we first consider the magnetic case , whose euclidean action is given by ( [ i ] ) .
the boundary that appears in ( [ i ] ) is given by @xmath371 , where @xmath348 is a spatial surface at @xmath349 and @xmath350 , @xmath295 is a timelike 3-surface at the rindler horizon @xmath212 , and the timelike 3-surface @xmath275 is an internal infinity boundary at @xmath372 . with the analysis of sec .
[ sec : ultracold - inst ] , we can compute all the terms of action ( [ i ] ) .
we start with @xmath373 which yields ( using @xmath286 ) @xmath374 where @xmath299 and @xmath182 are defined by ( [ polos - cold ] ) and ( [ mq - ultra ] ) , and @xmath375 is given by ( [ period phi ] ) . the maxwell term in the action yields @xmath376 where we have used @xmath353 [ see ( [ f - mag ] ) ] with @xmath377 .
due to the fact that @xmath378 it might seem that the contribution from ( [ i1-ultracold ] ) and ( [ i2-ultracold ] ) diverges .
fortunately this is not the case since these two terms cancel each other . trying to verify this analytically is cumbersome , but
for our purposes we can simply fix any numerical value for @xmath3 and @xmath106 , and using ( [ mq - ultra ] ) and ( [ polos - cold ] ) we indeed verify that ( [ i1-ultracold ] ) and ( [ i2-ultracold ] ) cancel each other .
now , contrary to the other instantons , the ultracold c instanton has a non - vanishing extrinsic curvature boundary term , @xmath379 , due to the internal infinity boundary ( @xmath275 at @xmath372 ) contribution .
the extrinsic curvature to @xmath275 is @xmath380 , where @xmath381 is the unit outward normal to @xmath275 , @xmath382 is the projection tensor onto @xmath275 , and @xmath383 represents the covariant derivative with respect to @xmath23 .
thus the trace of the extrinsic curvature to @xmath275 is @xmath384 , and @xmath385 the magnetic action ( [ i ] ) of the ultracold c instanton is then @xmath386^{-1 } \nonumber \\ & & \!\!\!\!\ !
\times \frac{x_\mathrm{n}-x_\mathrm{s } } { \left [ a x_\mathrm{n}+\sqrt{2(\lambda+3a^2)}\right ] \left [ a x_\mathrm{s}+\sqrt{2(\lambda+3a^2)}\right ] } \ , , \label{i - mag - ultracold } \end{aligned}\ ] ] where @xmath299 and @xmath182 are defined by ( [ polos - cold ] ) and ( [ mq - ultra ] ) .
when we take the limit @xmath98 we get @xmath99 and @xmath387 , and @xmath388 and therefore we recover the action for the @xmath98 ultracold instanton @xcite , that describes the pair creation of ultracold black holes accelerated only by the cosmological constant . in fig .
[ lukewarm - fig ] we show a plot of @xmath389 as a function of @xmath10 and @xmath106 for a fixed @xmath3 .
when we fix @xmath3 and @xmath106 we also fix the mass and charge of the ultracold black holes . for a fixed @xmath3 , when @xmath106 increases the probability of pair creation of ultracold black holes , @xmath390 , increases monotonically and they have a lower mass and charge .
alternatively , we can discuss the behavior of @xmath391 . in this case
, the probability decreases monotonically as the acceleration of the black holes increases . in the electric case
, the euclidean action is given by ( [ i - electric ] ) with @xmath363 [ see ( [ f - el - ultracold ] ) ] .
thus , @xmath392 in the ultracold case the vector potential @xmath56 , that is regular everywhere including at the horizon , needed to compute the extra maxwell boundary term in ( [ i - electric ] ) is @xmath393 , which obviously satisfies ( [ f - el - ultracold ] ) .
the integral over @xmath16 consists of an integration between @xmath212 and @xmath372 along the @xmath349 surface and back along @xmath350 , and of an integration between @xmath349 and @xmath350 along the @xmath212 surface , and back along the internal infinity surface @xmath372 .
the non - vanishing contribution to the maxwell boundary term in ( [ i - electric ] ) comes only from the integration along the internal infinity boundary @xmath330 , and is given by @xmath394 adding ( [ i2-ultracold - elect ] ) and ( [ i - electric - ultracold ] ) yields ( [ i2-ultracold ] ) .
thus , the electric action ( [ i - electric ] ) of the ultracold c instanton is equal to the magnetic action , @xmath395 , and therefore electric and magnetic ultracold black holes have the same probability of being pair created .
the lukewarm , cold , nariai and ultracold c - metric instantons are saddle point solutions free of conical singularities both in the @xmath91 and @xmath92 horizons .
the corresponding black holes may then nucleate in the ds background when a cosmic string breaks , and we have computed their pair creation rates in the last four subsections . however , these particular black holes are not the only ones that can be pair created .
indeed , it has been shown in @xcite that euclidean solutions with conical singularities may also be used as saddle points for the pair creation process . in this way ,
pair creation of nonextreme sub - maximal black holes is allowed ( by this nomenclature we mean all the nonextreme black holes other than the lukewarm ones that are in the region interior to the close line @xmath396 in fig .
[ mq - fig ] ) , and their pair creation rate may be computed . in order to calculate this rate ,
the action is given by ( [ i ] ) and ( [ i - electric ] ) ( in the magnetic and electric cases respectively ) and , in addition , it has now an extra contribution from the conical singularity ( c.s . ) that is present in one of the horizons ( @xmath91 , say ) given by @xcite @xmath397 where @xmath398 is the area of the 2-surface spanned by the conical singularity , and @xmath399 is the deficit angle associated to the conical singularity at the horizon @xmath91 , with @xmath400 and @xmath401 being the periods of @xmath117 that avoid a conical singularity in the horizons @xmath92 and @xmath91 , respectively .
the contribution from ( [ i ] ) and ( [ i - electric ] ) follows straightforwardly in a similar way as the one shown in subsection [ sec : lukewarm - rate ] with the period of @xmath117 , @xmath402 , chosen in order to avoid the conical singularity at the acceleration horizon , @xmath115 .
the full euclidean action for general nonextreme sub - maximal black holes is then @xmath403 where @xmath404 is given by ( [ period phi ] ) , and the pair creation rate of nonextreme sub - maximal black holes is given by ( [ pc - rate - breakstring ] ) or ( [ pc - rate ] ) with the use of ( [ i - string ] ) and ( [ i - ds ] ) . in order to compute ( [ i - sub - maximal ] ) ,
we need the relation between the parameters @xmath106 , @xmath3 , @xmath10 , @xmath81 , and the horizons @xmath92 , @xmath91 and @xmath90 . in general , for a nonextreme solution with horizons @xmath405 , one has @xmath406 with @xmath407 the parameters @xmath106 , @xmath3 , @xmath10 and @xmath81 can be expressed as a function of @xmath92 , @xmath91 and @xmath90 by @xmath408 the allowed values of parameters @xmath10 and @xmath81 are those contained in the interior region defined by the close line @xmath396 in fig .
[ mq - fig ] .
in previous works on black hole pair creation in general background fields it has been well established that the pair creation rate is proportional to the exponential of the gravitational entropy @xmath1 of the system , @xmath409 , with the entropy being given by one quarter of the the total area @xmath410 of all the horizons present in the instanton , @xmath411 . in what follows
we will verify that these relations also hold for the instantons of the ds c - metric . in the lukewarm case , the instanton has two horizons in its euclidean section , namely the acceleration horizon at @xmath115 and the black hole horizon at @xmath116 .
so , the total area of the lukewarm c instanton is @xmath412 where @xmath92 and @xmath91 are given by ( [ ya - luk ] ) , @xmath299 and @xmath182 are defined by ( [ polos - luk ] ) , and @xmath413 is given by ( [ period phi - luk ] ) .
it is straightforward to verify that @xmath414 , where @xmath415 is given by ( [ i - mag - luk ] ) , and thus @xmath416 , where @xmath417 . in the cold case
, the instanton has a single horizon , the acceleration horizon at @xmath115 , in its euclidean section , since @xmath116 is an internal infinity .
so , the total area of the cold c instanton is @xmath418 where @xmath92 is given by ( [ zerosy3-cold ] ) , @xmath299 and @xmath182 are defined by ( [ polos - cold ] ) , and @xmath181 is given by ( [ period phi ] ) .
thus , @xmath419 , where @xmath342 is given by ( [ i - mag - cold ] ) , and thus @xmath420 , where @xmath421 . in the nariai case , the instanton has two horizons in its euclidean section , namely the acceleration horizon @xmath92 and the black hole horizon @xmath91 , both at @xmath422 , and thus they have the same area .
so , the total area of the nariai c instanton is @xmath423 where @xmath424 , @xmath299 and @xmath182 are defined by ( [ polos - cold ] ) , and @xmath224 is given by ( [ period phi ] ) , with @xmath10 and @xmath81 subjected to ( [ mq - cnariai pcds ] ) .
thus , @xmath425 , where @xmath361 is given by ( [ i - mag - nariai ] ) , and thus @xmath426 , where @xmath427 . in the ultracold case , the instanton has a single horizon , the rindler horizon at @xmath212 , in its euclidean section , since @xmath372 is an internal infinity .
so , the total area of the ultracold c instanton is @xmath428 with @xmath377 [ see ( [ range - gamma - ultra ] ) ] , @xmath299 and @xmath182 are defined by ( [ polos - cold ] ) , and @xmath181 is given by ( [ period phi ] ) , with @xmath10 and @xmath81 subjected to ( [ mq - ultra ] ) .
it straightforward to verify that @xmath429 , where @xmath430 is given by ( [ i - mag - ultracold ] ) , and thus @xmath431 , where @xmath432 .
as we have already said , the ultracold c instanton is a limiting case of both the charged nariai c instanton and the cold c instanton ( see , e.g. , fig .
[ mq - fig ] ) .
then , as expected , the action of the cold c instanton gives , in this limit , the action of the ultracold c instanton ( see fig . [ cold - fig ] ) .
however , the ultracold frontier of the nariai c action is given by two times the ultracold c action ( see fig .
[ nariai - fig - pcads ] ) . from the results of this section
we clearly understand the reason for this behavior .
indeed , in the ultracold case and in the cold case , the respective instantons have a single horizon ( the other possible horizon turns out to be an internal infinity ) .
this horizon gives the only contribution to the total area , @xmath433 , and therefore to the pair creation rate . in the nariai case , the instanton has two horizons with the same area , and thus the ultracold limit of the nariai action is doubled with respect to the true ultracold action . in the lukewarm case , the instanton has two horizons in its euclidean section , namely the acceleration horizon at @xmath115 and the black hole horizon at @xmath116 .
so , the total area of the saddlepoint solution is @xmath434 and once again one has @xmath435 , where @xmath66 is given by ( [ i - sub - maximal ] ) , and thus @xmath436 , where @xmath437 .
we have studied in detail the quantum process in which a cosmic string breaks in a de sitter ( ds ) background and a pair of black holes is created at the ends of the string . the energy to materialize and accelerate the pair comes from the positive cosmological constant and from the string tension .
this process is a combination of the processes considered in @xcite-@xcite , where the creation of a black hole pair in a ds background has been analyzed , and in @xcite-@xcite , where the breaking of a cosmic string accompanied by the creation of a black hole pair in a flat background has been studied .
we remark that in principle our explicit values for the pair creation rates also apply to the process of pair creation in an external electromagnetic field , with the acceleration being provided in this case by the lorentz force instead of being furnished by the string tension .
indeed , there is no ds ernst solution , and thus we can not discuss analytically the process .
however , physically we could in principle consider an external electromagnetic field that supplies the same energy and acceleration as our strings and , from the results of the @xmath266 case ( where the pair creation rates in the string and electromagnetic cases agree ) , we expect that the pair creation rates found in this paper do not depend on whether the energy is being provided by an external electromagnetic field or by a string . we have constructed the saddle point solutions that mediate the pair creation process through the analytic continuation of the ds c - metric , and we have explicitly computed the nucleation rate of the process ( see also a heuristic derivation of the rate in the appendix ) .
globally our results state that the ds space is stable against the nucleation of a string , or against the nucleation of a string followed by its breaking and consequent creation of a black hole pair .
in particular , we have answered three questions .
first , we have concluded that the nucleation rate of a cosmic string in a ds background @xmath438 decreases when the mass density of the string increases .
second , given that the string is already present in our initial system , the probability @xmath439 that it breaks and a pair of black holes is produced and accelerated apart by @xmath3 and by the string tension increases when the mass density of the string increases .
in other words , a string with a higher mass density makes the process more probable , for a fixed black hole mass .
third , if we start with a pure ds background , the probability @xmath440 that a string nucleates on it and then breaks forming a pair of black holes decreases when the mass density of the string increases .
these processes have a clear analogy with a thermodynamical system , with the mass density of the string being the analogue of the temperature @xmath441 .
indeed , from the boltzmann factor , @xmath442 ( where @xmath443 is the boltzmann constant ) , one knows that a higher background temperature turns the nucleation of a particle with energy @xmath444 more probable .
however , in order to have a higher temperature we have first to furnish more energy to the background , and thus the global process ( increasing the temperature to the final value @xmath441 plus the nucleation of the particle ) becomes energetically less favorable as @xmath441 increases .
we have also verified that the relation between the rate , entropy and area , which is satisfied for all the black hole pair creation processes analyzed so far , also holds in the process studied in this paper .
indeed , the pair creation rate is proportional to @xmath0 , where @xmath1 is the gravitational entropy of the system , and is given by one quarter of the total area of all the horizons present in the saddle point solution that mediates the pair creation . to conclude
let us recall that the ds c - metric allows two distinct physical interpretations . in one of them one removes the conical singularity at the north pole and leaves one at the south pole .
in this way the ds c - metric describes a pair of black holes accelerated away by a string with positive mass density .
alternatively , we can avoid the conical singularity at the south pole and in this case the black holes are pushed away by a strut ( with negative mass density ) in between them , along their north poles . in this paper
we have adopted the first choice .
technically , the second choice only changes the period of the angular coordinate @xmath101 : it would be given by @xmath445 instead of ( [ period phi ] ) .
we have chosen the first choice essentially for two reasons .
first , the string has a positive mass density and , in this sense , it is a more physical solution than the strut .
second , in order to get the above string / pair configuration we only have to cut the string in a point .
the string tension does the rest of the work .
however , if we desire the strut / pair system described above we would have to cut the strut in two different points . then we would have to discard somehow the segment that joins the black holes along their south poles .
it is a pleasure to acknowledge conversations with vitor cardoso and with alfredo b. henriques .
this work was partially funded by fundao para a cincia e tecnologia ( fct ) through project cern / fis/43797/2001 and peso / pro/2000/4014 .
ojcd also acknowledges finantial support from the fct through praxis xxi programme .
jpsl thanks observatrio nacional do rio de janeiro for hospitality .
in order to clarify the physical interpretation of the results , in this appendix we heuristically derive the nucleation rates for the processes discussed in the main body of the paper . we know that an estimate for the nucleation probability is given by the boltzmann factor , @xmath446 , where @xmath444 is the energy of the system that nucleates and @xmath447 is the work done by the external force @xmath64 , that provides the energy for the nucleation , through the typical distance @xmath135 separating the created pair .
forget for a moment the string , and ask what is the probability that a black hole pair is created in a ds background .
this process has been discussed in @xcite where it was found that the pair creation rate is @xmath448 . in this case ,
@xmath449 , where @xmath10 is the rest energy of the black hole , and @xmath450 is the work provided by the cosmological background . to derive
@xmath451 one can argue as follows . in the ds case ,
the newtonian potential is @xmath452 and its derivative yields the force per unit mass or acceleration , @xmath453 , where @xmath454 is the characteristic ds radius , @xmath455 .
the force can then be written as @xmath456 , where the characteristic mass of the system is @xmath457 .
thus , the characteristic work is @xmath458 , where the characteristic distance that separates the pair at the creation moment is @xmath455 .
so , from the boltzmann factor we indeed expect that the creation rate of a black hole pair in a ds background is given by @xmath448 @xcite .
a question that has been answered in the present paper was : given that a string is already present in our initial system , what is the probability that it breaks and a pair of black holes is produced and accelerated apart by @xmath3 and by the string tension ?
the presence of the string leads in practice to a problem in which we have an effective cosmological constant that satisfies @xmath459 , that is , the acceleration @xmath106 provided by the string makes a positive contribution to the process .
heuristically , we may then apply the same arguments that have been used in the last paragraph , with the replacement @xmath460 . at the end ,
the boltzmann factor tells us that the creation rate for the process is @xmath461 .
so , for a given black hole mass , @xmath10 , and for a given cosmological constant , @xmath3 , the black hole pair creation process is enhanced when a string is present , as the explicit calculations done in the main body of the paper show . for @xmath266 this heuristic derivation yields @xmath462 which is the pair creation rate found in @xcite .
another question that we have dealt with in the present paper was : what is the probability for the nucleation of a string in a ds background ?
heuristically , the energy of the string that nucleates is @xmath463 , i.e. , its mass per unit length times the ds radius , while the work provided by the cosmological background is still given by @xmath450 .
the boltzmann factor yields for nucleation rate the value @xmath464 , in agreement with ( [ pc - rate - stringend ] ) .
o. j. c. dias , j. p. s. lemos , _ false vacuum decay : effective one - loop action for pair creation of domain walls _ , j. math .
* 42 * , 3292 ( 2001 ) ; o. j. c. dias , in _ proceedings of xth portuguese meeting on astronomy and astrophysics _ , edited by j. p.
s. lemos et al ( world scientific , singapore , 2001 ) , gr - qc/0106081 .
s. w. hawking , in _ black holes : an einstein centenary survey _ , edited by s. w. hawking , w. israel ( cambridge university press , 1979 ) ; _ euclidean quantum gravity _ , edited by g. w. gibbons , s. w. hawking ( cambridge university press , 1993 ) .
f. j. ernst , _ removal of the nodal singularity of the c - metric _ , j. math
* 17 * , 515 ( 1976 ) .
w. kinnersley , m. walker , _ uniformly accelerating charged mass in general relativity _ ,
d * 2 * , 1359 ( 1970 ) .
a. vilenkin , _ gravitational field of vacuum domain walls _ ,
b*133 * , 177 ( 1983 ) ; j. ipser , p. sikivie , _ gravitationally repulsive domain wall _ ,
d*30 * , 712 ( 1984 ) .
z. c. wu , _ quantum creation of a black hole _
d * 6 * , 199 ( 1997 ) ; _ real tunneling and black hole creation _
d * 7 * , 111 ( 1998 ) .
r. bousso , s. w. hawking , _
lorentzian condition in quantum gravity _
d * 59 * , 103501 ( 1999 ) ; * 60 * , 109903 ( 1999 ) ( e ) . g. w. gibbons , in _ fields and geometry _ , proceedings of the 22nd karpacz winter school of theoretical physics , edited by a. jadczyk ( world scientific , singapore , 1986 ) .
d. garfinkle , s. b. giddings , _ semiclassical wheeler wormhole production _ , phys .
b*256 * , 146 ( 1991 ) .
d. garfinkle , s. b. giddings , a. strominger , _ entropy in black hole pair production _ , phys .
d * 49 * , 958 ( 1994 ) h. f. dowker , j. p. gauntlett , d. a. kastor , j. traschen , _ pair creation of dilaton black holes _
d * 49 * , 2909 ( 1994 ) ; h. f. dowker , j. p. gauntlett , s. b. giddings , g. t. horowitz , _ pair creation of extremal black holes and kaluza - klein monopoles _ , phys .
d * 50 * , 2662 ( 1994 )
. s. ross , _ pair creation rate for @xmath465 black holes _ , phys .
d * 51 * , 2813 ( 1995 ) .
yi , _ toward one - loop tunneling rates of near - extremal magnetic black hole pair creation _ , phys .
d * 52 * , 7089 ( 1995 ) .
j. d. brown , _ duality invariance of black hole pair creation rates _ , phys . rev .
d * 51 * , 5725 ( 1995 ) .
s. w. hawking , g. t. horowitz , s. f. ross , _ entropy , area , and black hole pairs _ , phys .
d * 51 * , 4302 ( 1995 ) .
s. w. hawking , g. t. horowitz , _ the gravitational hamiltonian , action , entropy and surface terms _ , class .
* 13 * , 1487 ( 1996 ) .
r. emparan , _ correlations between black holes formed in cosmic string breaking _ , phys .
d * 52 * , 6976 ( 1995 ) .
f. mellor , i. moss , _ black holes and quantum wormholes _
b * 222 * , 361 ( 1989 ) ; _ black holes and gravitational instantons _ , class .
quant . grav .
* 6 * , 1379 ( 1989 ) .
l. j. romans , _ supersymmetric , cold and lukewarm black holes in cosmological einstein - maxwell theory _ , nucl .
b * 383 * , 395 ( 1992 ) .
r. b. mann , s. f. ross , _ cosmological production of charged black hole pairs _ , phys .
d * 52 * , 2254 ( 1995 ) .
r. bousso , s. w. hawking , _ the probability for primordial black holes _ , phys .
d * 52 * , 5659 ( 1995 ) ; _ pair production of black holes during inflation _
d * 54 * , 6312 ( 1996 ) ; s. w. hawking , _ virtual black holes _ , phys .
d * 53 * , 3099 ( 1996 ) .
r. garattini , _ energy computation in wormwhole background with the wheeler - dewitt operators _ , nucl .
b ( proc . suppl . ) * 57 * , 316 ( 1997 ) ; _ wormwholes and black hole pair creation _ , nuovo cimento b*113 * , 963 ( 1998 ) ; space - time foam , casimir energy and black hole pair creation , mod . phys . lett .
a * 13 * , 159 ( 1998 ) ; _ casimir energy and black hole pair creation in schwarzschild - de sitter spacetime _ , class .
* 18 * , 571 ( 2001 ) .
m. volkov , a. wipf , _ black hole pair creation in de sitter space : a complete one - loop analysis _ ,
b*582 * , 313 ( 2000 ) . i. s. booth , r. b. mann , _ complex instantons and charged rotating black hole pair creation _ , phys .
* 81 * , 5052 ( 1998 ) ; _ cosmological pair production of charged and rotating black holes _ ,
b * 539 * , 267 ( 1999 ) .
r. bousso , _ charged nariai black holes with a dilaton _ , phys . rev .
d * 55 * , 3614 ( 1997 ) . s. w. hawking , s. f. ross , _ pair production of black holes on cosmic strings _
lett . * 75 * , 3382 ( 1995 ) .
d. m. eardley g. t. horowitz , d. a. kastor , j. traschen , _ breaking cosmic strings without black holes _
75 * , 3390 ( 1995 ) .
a. achcarro , r. gregory , k. kuijken , _ abelian higgs hair for black holes _ , phys .
d * 52 * , 5729 ( 1995 ) .
r. gregory , m. hindmarsh , _ smooth metrics for snapping strings _
d * 52 * , 5598 ( 1995 ) .
j. preskill , a. vilenkin , _ decay of metastable topological defects _ ,
d * 47 * , 2324 ( 1993 ) .
r. emparan , _ pair production of black holes joined by cosmic strings _ ,
* 75 * , 3386 ( 1995 ) .
r. r. caldwell , a. chamblin , g. w. gibbons , _ pair creation of black holes by domain walls _
d * 53 * , 7103 ( 1996 ) .
r. bousso , a. chamblin , _ patching up the no - boundary proposal with virtual euclidean wormholes _ ,
d * 59 * , 084004 ( 1999 ) .
r. parentani , s. massar , _ the schwinger mechanism , the unruh effect and the production of accelerated black holes _ ,
d * 55 * , 3603 ( 1997 ) ; z. c. wu , _ pair creation of black holes in anti - de sitter space background ( i ) _ , gen .
. grav . * 31 * , 223 ( 1999 ) ; _ pair creation of black hole in anti - de sitter space background _ ,
b * 445 * , 274 ( 1999 ) ; _ quantum creation of topological black hole _ , mod .
a * 15 * , 1589 ( 2000 ) ; j. garriga , m. sasaki , _ brane - world creation and black holes _ , phys
d * 62 * , 043523 ( 2000 ) ; r. emparan , r gregory , c. santos , _ black holes on thick branes _ , phys .
d * 63 * , 104022 ( 2001 )
. s. w. hawking , _ breakdown of predictability in gravitational collapse _
rev . d*14 * , 2460 ( 1976 ) .
t. banks , _ lectures on black holes and information loss _
, hep - th/9412131 ; s. b. giddings , _ the black hole information paradox _ , hep - th/9508151 ;
_ why are nt black holes infinitely produced ? _ , phys .
d * 51 * , 6860 ( 1995 ) ; l. susskind , _
trouble for remnants _ , hep - th/9501106 .
j. f. plebaski , m. demiaski , _ rotating , charged and uniformly accelerating mass in general relativity _ , annals of phys .
( n.y . ) * 98 * , 98 ( 1976 ) .
j. podolsk , j.b .
griffiths , _ uniformly accelerating black holes in a de sitter universe _ ,
d * 63 * , 024006 ( 2001 ) .
o. j. c. dias , j. p. s. lemos , _ pair of accelerated black holes in a de sitter background : the ds c - metric _ , phys .
d * 67 * , 084018 ( 2003 ) . v. cardoso , o. j. c. dias , j. p. s. lemos , _ gravitational radiation in d - dimensional spacetimes _ , phys .
d * 67 * , 064026 ( 2003 ) .
j. bik , _ gravitational radiation from uniformly accelerated particles in general relativity _ , proc .
a * 302 * , 201 ( 1968 ) ; v. pravda , a. pravdova , _ boost - rotation symmetric spacetimes - review _ , czech . j. phys . * 50 * , 333 ( 2000 ) ; _ on the spinning c - metric _ , in _ gravitation : following the prague inspiration _ , edited by o. semerk , j. podolsk , m. zofka ( world scientific , singapore , 2002 ) , gr - qc/0201025 .
j. bik , p. krtou , _ accelerated sources in de sitter spacetime and the insuffiency of retarded fields _ , phys .
d * 64 * , 124020 ( 2001 ) ; _ the fields of uniformly accelerated charges in de sitter spacetime _ , phys .
lett . * 88 * , 211101 ( 2002 ) .
p. krtou , j. podolsk ' y , _
radiation from accelerated black holes in de sitter universe _ , gr - qc/0301110 .
g. w. gibbons , s. w. hawking , m. j. perry , _ path integral and the indifiniteness of the gravitational action _ , nucl .
b * 138 * , 141 ( 1978 ) .
d. j. gross , m. j. perry , l. g. yaffe , _ instability of flat space at finite temperature _ , phys .
d * 25 * , 330 ( 1982 ) .
g. w. gibbons , m. j. perry , _ quantizing gravitational instantons _ ,
b * 146 * , 90 ( 1978 ) .
s. m. christensen , m. j. duff , _ quantizing gravity with cosmological constant _ , nucl .
b * 146 * , 90 ( 1978 ) .
p. ginsparg , m. j. perry , _ semiclassical perdurance of de sitter space _ ,
b * 222 * , 245 ( 1983 ) .
r. e. young , _ semiclassical stability of asymptotically locally flat spaces _ ,
d * 28 * , 2420 ( 1983 ) .
r. e. young , _ semiclassical instability of gravity with positive cosmological constant _ , phys .
d * 28 * , 2436 ( 1983 ) .
j. d. brown , j. w. york , _ quasilocal energy and conserved charges derived from the gravitational action _
d * 47 * , 1407 ( 1993 ) ; , _ microcanonical functional integral for the gravitational field _ ,
d * 47 * , 1420 ( 1993 ) .
s. nojiri , s. d. odintsov , _ effective action for conformal scalars and anti - evaporation of black holes _ , int . j. mod .
a * 14 * , 1293 ( 1999 ) ; _ quantum evolution of schwarzschild - de sitter ( nariai ) black holes _ ,
d * 59 * , 044026 ( 1999 ) ; r. bousso , _ proliferation of de sitter space _
d * 58 * , 083511 ( 1998 ) ; _ adventures in de sitter space _ , hep - th/0205177 . | we analyze the quantum process in which a cosmic string breaks in a de sitter ( ds ) background , and a pair of neutral or charged black holes is produced at the ends of the string . the energy to materialize and accelerate the pair comes from the positive cosmological constant and , in addition , from the string tension .
the compact saddle point solutions without conical singularities ( instantons ) or with conical singularities ( sub - maximal instantons ) that describe this process are constructed through the analytical continuation of the ds c - metric .
then , we explicitly compute the pair creation rate of the process .
in particular , we find the nucleation rate of a cosmic string in a ds background , and the probability that it breaks and a pair of black holes is produced .
finally we verify that , as occurs with pair production processes in other background fields , the pair creation rate of black holes is proportional to @xmath0 , where the gravitational entropy of the black hole , @xmath1 , is given by one quarter of the area of the horizons present in the saddle point solution that mediates the process . |
photometric redshifts is a term that has received recent , wide - spread visibility and interest , though the use of multicolor photometry to estimate the redshifts of galaxies has a history that is relatively long . as of march 1999
, my search of the ads abstracts for the term `` photometric redshifts '' yielded only 9 papers ending in 1994 , with the first in 1982 , i.e. averaging less than one paper per year .
a sudden jump then occurs in 1995 with 6 papers , followed by 1996 with 7 , 1997 with 9 , and then another major jump to 35 papers in 1998 .
the broad participation and range of topics of this workshop validate this surge in activity . among these papers ,
the earliest reference to the term `` photometric redshifts '' is found in the abstract of puschell et .
al . ( 1982 ) .
their attempt to estimate redshifts of faint radio galaxies via broadband photometry was pioneering ( even by today s standards ) in three respects : 1 ) the use of near - infrared bands ( @xmath0 ) along with optical bands ( @xmath1 ) , 2 ) the adoption of @xmath2 fits to spectral energy distributions ( sed ) , and 3 ) the use of different sets of sed templates , ranging from non - evolving local ellipticals , theoretical seds from bruzual , to observed seds from known radio galaxies .
the first use of the term in the title was by loh and spillar ( 1986 ) : _ photometric redshifts of galaxies_. this work was pioneering in its use of ccds to reach quite faint limits of @xmath3 , along with the use of 6 medium - band filters and again @xmath2 template fitting , but only to three observed _ local _ sed templates to represent all galaxy types at all redshifts .
the use of multicolors to estimate redshifts actually predates the above papers and the true literature on the subject is significantly more extensive , but due to the ambiguity of the term `` photometric redshifts '' as detailed in the next section , many workers in this arena chose not to use this term .
probably the first use of multicolor photometry for redshift estimation was by baum ( 1962 ) .
he obtained photoelectric photometry through 9 medium - wide filters of galaxies in cluster 3c395 , assumed that the seds were that of ellipticals , and obtained an estimate of redshift @xmath4 , quite close to the eventual spectroscopic value of 0.46 .
nearly two decades passed before multicolor photometry was used to estimate redshifts . butchins ( 1981 , 1983 )
used uk schmidt plate @xmath5 photometry that reached @xmath6 .
because of the overlap in @xmath5 colors of low - redshift ( @xmath7 ) early - type galaxies with higher redshift ( @xmath8 ) later - type galaxies , butchins applied probablility constraints on luminosities that are qualitatively similar to what has more recently been termed `` bayesian '' .
i was able to reach fainter limits ( @xmath9 ) by using kpno 4-m plates and had both superior redshift accuracy @xmath10 and far less degeneracy by exploiting a set of filters ( roughly @xmath11 ) with much longer wavelength coverage ( koo 1981 , 1985 , 1986 ) . since baum is not here and butchins is no longer in astronomy , i qualify as the old - timer in the field at this workshop .
another important set of papers that does _ not _ use the term `` photometric redshifts '' , and yet constrains redshifts from broadband photometry , relies on the so - called `` lyman - break technique '' ( see contribution by steidel ) . the basic idea is to discern very high redshift galaxies via a significant drop in the bluest band for galaxies with otherwise very blue colors in two or more redder bands .
this situation occurs when the lyman break enters , e.g. , the @xmath12 band , at redshifts @xmath13 . since the apparent drop in the @xmath12 band flux ( or especially in redder bands at higher redshifts )
is also affected by the intergalactic lyman - alpha forest depression , a more accurate term should be `` lyman - drop technique '' .
early examples of its use include papers by partridge ( 1974 ) , koo(1986 ) , cowie ( 1988 ) , majewski ( 1988 , 1989 ) , guhathakurta et .
1990 ) , and steidel & hamilton ( 1993 ) .
i suggest the following : photometric redshifts are those derived from _ only _ images or photometry with spectral resolution @xmath14 . my choice of 20 is intended to exclude redshifts derived from slit and slitless spectra , narrow band images , ramped - filter images , fabry - perot images , fourier transform spectrometers , etc .
this definition still leaves a wide range of approaches to obtain redshifts , examples of which include : 1 . spatial correlations :
galaxies are assumed to have , _ statistically _ , the redshifts of their neighbors in apparent close pairs , groups , and especially the cores of rich clusters .
photometric redshifts is an unconventional term to apply in these cases and should thus probably be avoided .
magnitudes : although using just magnitudes alone to estimate redshifts might be expected to work only for standard candles , such as the brightest cluster galaxies in @xmath15 or strong - flux radio galaxies in @xmath16 , the convolution of accessible volumes with the shape of the luminosity function of galaxies results in a fairly tight correlation between magnitudes and redshift , even for more common field galaxies ( see the redshift - magnitude plot fig . 4 of koo & kron 1992 ) .
for any of the following techniques , the addition of magnitudes can thus be expected to provide additional constraints on the redshift probability distribution ( see , e.g. , connolly et .
al . 1995 ) .
just as in the previous approach , the use of the term `` photometric redshifts '' would be quite unconventional , but technically accurate .
3 . one color : in situations where the sed is known or unique , one color may be sufficient to estimate a redshift .
practical applications include redshift estimates for the reddest galaxies in clusters or field and also very high redshift searches , especially when at least one of the two bands is in the near - infrared .
a single - color is probably the minimal information to qualify the use of the term `` photometric redshift '' as understood by most astronomers today .
two or more colors : the use of three or more filters is necessary to break the degeneracy between instrinsic color and redshift .
this approach is probably the most commonly accepted one when referring to photometric redshifts that employ optical bands , and works surprisingly well even with only three bands ( straizys & sviderskiene 1983 ) .
this is because most galaxies ( at least locally ) occupy only a small fraction of the possible multicolor volume ( see fig .
2a ) ; galaxy spectra are often composites of old and young stellar populations which result in bowl - shaped or u - shaped sed ( see fig . 1 and note the more bowl - shaped value of the z = 0 locus compared to that of its consituent stars ) ;
the pivot point of this curvature lies near the 4000 break and is roughly independent of the galaxy s average color ; and bowl - shaped spectra , when redshifted , result in moving the iso - z loci in a direction perpendicular to these loci ( see fig .
1 ) . obviously one needs longer wavelength bands to sample the pivot point near 4000 as one wishes to discern higher redshifts ( see fig .
nature has been kind : if galaxy spectra had instead blackbody shapes of different temperatures or power - laws in shape , we would not be able to separate redshifts using multicolors .
surface brightness : if the intrinsic surface brightness is roughly constant or slowly varying with time , as might be expected for large spirals that undergo largely constant star formation rate histories , the @xmath17 surface brightness dimming can be exploited to yield redshifts .
one color with light profiles : in principle , the light profile might yield the type of galaxy ( e.g. , @xmath18 profiles might imply a luminous spheroidal or bulge to disk ratios might suggest a probable galaxy type ) , which in turn might be assumed to have a unique sed so that a single color is sufficient for estimating a redshift ( sarajedini et .
al . 1999 ) .
one color with image structure : this is a generalization of the last two approaches .
anderson et .
1996 ) has already been exploring the correlations of galaxies among color , surface brightness , and image concentration .
this combination appears to be useful to improve the efficiency of gathering photometric redshifts by not requiring deep @xmath12 photometry , which is very costly in telescope time .
etc . whether the term `` photometric redshifts '' should include such a diversity of approaches is debatable , but what should be obvious is that astronomers have yet to exploit the wealth of additional photometric / structural information in refining redshift estimates from image data alone
the most popular method for estimating redshifts from a set of photometric measurements is the @xmath2 ( or maximum likelihood ) template - fitting method adopted by puschell et .
al ( 1982 ) , largely because the method is simple , does not rely on having any spectroscopic redshifts , and needs only a few templates ( empirical or theoretical ) to yield results . with the availability of multicolor ccd photometry
, almost anyone can get into the photometric redshift business today .
the achilles heel of the technique is of course the template set .
empirical seds are relatively few , almost exclusively confined to local galaxies whose seds may not reflect those of more distant galaxies , and which are not necessarily representative in luminosity , dust , inclination , morphology , etc . to the targets of interest . while theoretical seds avoid some of these problems , they may not be correct and do not yield information on the probability distribution for the estimated redshifts , since this information relies on having the probability distribution of the seds . to show the level of uncertainty between empirical and theoretical seds , see fig .
1 . we note that the empirical spectra ( which have traditionally been based on only a handful of measured integrated spectra ) differ significantly ( 0.05 or more in z ) from the theoretical at all redshifts .
in contrast , the empirical fitting method ( e.g. , connolly et .
1995 , wang et .
al . 1998 ) uses a large enough pool of spectroscopic redshifts to calibrate the relationship of colors and magnitudes to redshifts . by construction
, this method should provide accurate redshifts , but more importantly , it should also yield realistic estimates of the redshift errors . to achieve this in practice ,
however , requires a sufficiently complete set of known redshifts to the depth of the desired photometry . except for the mean and rms error when compared to spectroscopic redshifts , photometric redshift errors remain poorly characterized .
in particular , the position , shape , and asymmetry of the redshift error distribution should be derived ; the sources of the errors and especially whether they are due to random measurement errors , intrinsic dispersions in the seds of galaxies , or unknown systematics should be tracked down ; and , ideally , the dependence of the errors on a variety of other parameters ( color , luminosity , morphology , structure , redshift , environment , etc . ) should also be measured .
more importantly , these redshift uncertainties should be incorporatedly explicitly into the analysis , not by using the derived maximum likelihood or minimum @xmath2 value of the photometric redshift for each galaxy , but rather the full probability distribution .
this idea has already been incorporated for the c - method of deriving the luminosity function from photometric redshifts ( subbarao et .
al . 1996 ) , but this approach should be adopted more universally by others in most statistical analyses .
of special concern in the area of redshift errors is the likelihood both that the sed s of distant galaxies are different on average from those today and that the intrinsic dispersion in the seds are greater .
if true , we would expect larger random and systematic errors for galaxies at higher redshifts .
figures 2 and 3 are telling and sobering . more specifically , figure 2b shows that morphologically peculiar galaxies exhibit a greater spread of values _
perpendicular to the line _ in the @xmath19 two - color plots than that of more normal galaxies shown in figure 2a ( larson and tinsley 1978 ) .
interactions and mergers appear to be a primary cause of these peculiarities , so if indeed such galaxies are more common in the past , we should expect greater photometric redshift errors at higher redshifts that result from the greater spread in intrinsic seds alone .
larson and tinsley ( 1978 ) explain the wider spread as the result of a greater diversity in the star formation histories of galaxies , namely brief starbursts interrupting an otherwise more quiet or more smoothly varying star formation history .
figure 3 shows the possible tracks of galaxies with varying amounts of such bursts in the @xmath19 two - color diagram ; note in particular the regions _ perpendicular _ to the wide , solid line on top of which most local , morphologically - normal galaxies lie .
such deviations result in systematic errors in photometric redshifts .
clearly what is desired for significant progress in our use and understanding of photometric redshifts is a large pool of spectroscopic redshifts that span the full range of redshifts , depth , photometric bands , etc . in 1985
, we had about 100 spectroscopic redshifts ( @xmath20 ) to @xmath21 from 4 - 5 m class telescopes to study photometric redshifts in four bands @xmath11 ( koo 1985 ) , while roughly the same number has been measured with the keck 10-m to study the hubble deep field to fainter limits @xmath22 and a wider range in redshifts ( @xmath23 ) .
our current lack of good calibration is , however , changing dramatically . at the lowest redshifts of @xmath24 ,
the sloan digital sky survey will yield nearly @xmath25 spectra for a photometric sample that includes 5 filters . at intermediate redshifts between @xmath26 to 0.7 , cnoc2 will have 6000 spectroscopic redshifts for calibrating their 5 band system ( lin et .
al . 1999 ) .
keck will be providing over 1000 redshifts for @xmath27 to over @xmath28 and @xmath29 to 4 for passbands that include some @xmath16 , while handfuls of redshifts are coming in for the desert at @xmath30 to over 2 and the highest redshifts @xmath31 . with these samples to calibrate the photometric redshift system and especially its errors , photometric redshifts will finally rest on a much more secure foundation .
the scientific potential of using photometric redshifts has been recognized for a long time , especially during an earlier era when redshifts for large samples of galaxies fainter than @xmath32 were difficult or impractical , while multicolor photometry easily sampled the distant universe .
the following list is meant to be illustrative , rather than representative or exhaustive , of pioneering projects that were based on older multicolor photographic photometry or small - format ccds .
the topics of other papers from this workshop provide a more current view of how photometric redshifts are being applied in a new era of hst images , large mosaic ccd cameras , and 8 - 10 m telescopes .
the use of photometric redshifts has not only undergone a recent revival , but it is also rapidly becoming a crucial tool of many programs in mainstream observational cosmology . * searches for primeval galaxies at redshifts @xmath33 via the @xmath12 band detecting the lyman break ( cowie 1988 , cowie & lilly 1989 ; guhathakurta et .
al . 1990 ; koo 1986 ; majewski 1988 , 1989 ; steidel & hamilton 1993 ) or even via data further to the red to probe higher redshifts ( partridge 1974 ) .
* searches for high redshift qso s ( see review by warren & hewett 1990 ) or distant radio galaxies ( puschell et .
1982 ; van der laan 1983 ) . * studies of the evolution of field galaxies ( butchins 1983 ; cowie et .
1988 , 1990 ; guiderdoni 1987 ; koo 1986 ; lilly et .
al . 1991 ) or their luminosity function ( subbarao et . al .
1996 ) * discrimination of cluster members and superclusters at moderate redshifts ( connolly et .
al . 1996 ; koo 1981 , 1986 , koo et . al . 1988 ) * estimate of the geometry of the universe via the volume test ( loh and spillar 1986 ) i would like to thank ray weymann for letting me be an old - timer , albeit only briefly , and his conference team for organizing such a fun workshop . with an old - timer brain , i have also probably missed a number of important old references , and to their authors , my apologies for any oversights .
this work has been partially supported by nsf ast-9529098 and by nasa through grant number ar-07532.01 - 96 from the space telescope science institute , which is operated by aura , inc .
, under nasa contract nas 5 - 26555 .
anderson , d. r. , et .
1996 , , 188 , 817 baum , w. a. 1962 , iau symp .
15 , 390 butchins , s. a. 1981 , , 97 , 407 butchins , s. a. 1983 , , 203 , 1239 connolly , a. j. , et .
1995 , , 110 , 2655 connolly , a. j. , et .
1996 , , 473 , l67 cowie , l. l. 1988 , _ the post - recombination universe _ , 1 cowie , l. l. , et .
1988 , , 332 , l29 cowie , l. l. , & lilly , s. j. 1989 , , 336 , l41 cowie , l. l. , et .
al . 1990 , , 360 , l1 guhathakurta , p. , tyson , j. a. , & majewski , s. r. 1990 , , 357 , l9 guiderdoni , b. 1987 , _
high redshift and primeval galaxies _ , 271 koo , d. c. 1981 , , 252 , l75 koo , d. c. 1985 , , 90 , 418 koo , d. c. 1986 , , 311 , 651 koo , d. c. , & kron , r. g. 1992 , , 30 , 613 koo , d. c. , et .
1988 , , 333 , 586 larson , r. b. , & tinsley , b. m. 1978 , , 219 , 46 lilly , s. j. , cowie , l. l. , & gardner j. p. 1991
, , 369 , 71 lin , h. , et .
al . 1999 , , 518 , 533 loh , e. d. , & spillar , e. j. 1986 , , 303 , 154 majewski , s. r. 1988 , _ towards understanding galaxies at high redshifts _
, 203 majewski , s. r. 1989 , _ the epoch of galaxy formation _ , 85 partridge , r. b. 1974 , , 192 , 241 puschell , j. j. , owen , f. n. , & laing , r. a. 1982 , , 275 , l57 sarajedini , v. l. , et .
1999 , , 121 , 417 steidel , c. c. , & hamilton , d. 1993 , , 105 , 2017 straizys , v. , & sviderskiene , z. 1983 , , 94 , 23 subbarao , m. et .
1996 , , 112 , 929 tinsley , b. m. 1980 , fund . cosmic phys . , 5 , 87 van der laan , h. , et .
1983 , iau symp .
104 , 73 wang , y. , bahcall , n. , & turner , e. l 1998 , , 116 , 2081 warren , s. j. , & hewett , p. c. 1990 , rep . | i review the early history of photometric redshifts ; specify a working definition that encompasses a broader range of approaches than commonly adopted ; discuss the pros and cons of template fitting versus empirically - based techniques ; and summarize some past applications . despite its relatively long history ,
the technique of photometric redshifts remains far from being a mature tool .
areas needing development include the use of spatial structure , the incorporation of large redshift samples with multicolor photometry for empirical calibrations of redshift errors , and improved analysis tools that directly include redshift probability distributions rather than singular values .
photometric redshifts has not only undergone a recent revival it is also rapidly becoming a crucial tool of mainstream observational cosmology . |
let us consider the inclusive process of the pion creation in the fragmentation region of polarized proton at high energy proton - proton collisions @xmath2 where @xmath3 is the transversal to beam ( implied by cms ) axes spin of initial proton @xmath4 where @xmath5 is the energy fraction of pion , @xmath6 are the invariant masses of the jets , which we will assume to be of the order of nucleon mass @xmath7 .
we study the two jet kinematics with jets @xmath8 , @xmath9 moving along the initial hadron directions . the jet created by the transversely polarized proton
is supposed to contain the detected pion .
moreover , we consider the case when its production is not related with the creation of nucleon resonances . in terms of pion transverse components
it corresponds to the condition @xmath10
> m_{res}^2-m^2,\\ { \nonumber}{\mathbf{p}}+{\mathbf{k_1}}+{\mathbf{p_1'}}=0,\end{gathered}\ ] ] where @xmath11 is the transfer momentum between protons . through this paper
we use sudakov parameterization of 4momenta of the problem @xmath12 production @xmath13 . ]
the azimuthal one - spin asymmetry arises from the interference of the amplitudes with one and two gluon exchanges between nucleons ( see fig .
1 ) @xmath14 where @xmath15 is the strong coupling constant , @xmath16 is the azimuthal angle between the 2vectors @xmath17 and @xmath18 , transverse to the beam axis .
the functions @xmath19 , @xmath20 as well as the color factors @xmath21 will be specified below . for convenience we put here the alternative form for the phase volume of pion @xmath22
the one gluon exchange matrix element has a form @xmath23 the currents @xmath24 , @xmath25 are associated with the jets created by particles 1 , 2 and index @xmath3 describes the color state of the jet . using gribov representation of the metric tensor @xmath26 and the gauge condition for the currents
@xmath27 we express @xmath28 in the form @xmath29 @xmath30 the quantities @xmath31 are the invariant masses of the jets .
we imply that the jet @xmath8 does not contain the detected pion . at this point
we need some model describing the jets .
we use the heavy fermion model , i.e. , we consider the jet as a result of heavy fermion decay . we assume the coupling constant of the interaction of the pion with a nucleon and with heavy fermion to be the same .
we do not specify it as well as it is cancelled in the asymmetry ( [ eq:5 ] ) .
so we have @xmath32 @xmath33 are the generators of color group @xmath34 and @xmath35,\quad d_2=-\frac{m^2}{\beta}[\rho^2+\beta^2 ] .
{ \nonumber}\end{aligned}\ ] ] we use here the spin density matrices of the jets ( i.e. , heavy fermions ) @xmath36 it is important to note that we impose the gauge invariant form of matrix element and after that we use the heavy fermion model .
this operations do not commute as well as the heavy fermion currents do not satisfy the current conservation condition .
it is a specific of the considered model . for the lowest order differential cross section we obtain @xmath37 with the explicit expressions for the impact factors @xmath38 , @xmath39 given in appendix a and @xmath40 .
the lowest order spin
dependent contribution to the cross section arises from the interference of imaginary part of 1loop radiative correction ( rc ) of feynman diagram ( fd ) fig 1b , c with the born amplitude fig 1a .
we do not consider the rc from fd fig .
1 , believing that such kind fd contribute to the nucleon resonances formation . besides it do not contribute in the leading logarithmical approximation ( lla ) @xmath41 we obtain in the lowest order @xmath42 with @xmath43 and the color factor @xmath44=\frac{(n^2 - 1)(n^2 - 2)}{8n}.\ ] ] the impact factors @xmath45 are given in appendix b. impact factors
@xmath45 contain the new mass parameters @xmath46 which are intermediate jet state masses .
we had shown that the lowest order unpolarized and polarized cross sections can be expressed in terms of impact factors of projectiles moving in opposite directions which where introduced first in the papers of h. cheng and t. t. wu @xcite .
the calculation of rc to them can be done following the method developed by j. balitski and l. n. lipatov @xcite .
it was shown by these authors that in the lla , the cross section has as well the form of conversion of impact factors of colliding particles with some universal kernel .
physically it corresponds to the replacement of exchanged gluons by the reggeized gluons .
the reggeization states are taking into account in two factors .
first the regge factor @xmath47 must be introduced , where @xmath48 is the regge trajectory of gluon with the momentum squared @xmath49 .
the second factor takes into account the contribution of inelastic processes of emission of real gluons .
these both contributions suffer from the infrared divergences , however the total sum is free of them .
, @xmath50 dependence of a partial contribution to the pomeron intercept . ] , @xmath50 dependence of a partial contribution to the odderon intercept . ] for the rc to the unpolarized cross section we have @xmath51 with color factor @xmath52 and @xmath53,\end{aligned}\ ] ] with @xmath54 and @xmath55 given in appendix c. this formula can be inferred from the result for the non forward high energy scattering amplitude obtained in the paper @xcite @xmath56 with definition @xmath57\frac{\phi^{bb'}({\mathbf{k'}},{\mathbf{q } } ) } { { \mathbf{k'}{\lower-.2em\hbox{}^{2}}}({\mathbf{q}}-{\mathbf{k'}})^2}\\ { \nonumber}&\quad - \frac{\phi^{bb'}({\mathbf{k}},{\mathbf{q}})}{({\mathbf{k}}-{\mathbf{k'}})^2 } \bigg[\frac{{\mathbf{k}}^2}{{\mathbf{k'}{\lower-.2em\hbox{}^{2}}}+({\mathbf{k}}-{\mathbf{k'}})^2}+ \frac{({\mathbf{q}}-{\mathbf{k}})^2}{({\mathbf{q}}-{\mathbf{k'}})^2+({\mathbf{k}}- { \mathbf{k'}})^2}\bigg]\bigg\}.\end{aligned}\ ] ] the formula ( [ eq:17 ] ) can be obtained from the last general one by putting @xmath58 .
note that in the formula obtained in @xcite the used color group was @xmath59 .
let now consider the lla rc to the polarized part of the differential cross section .
there are presented three types of contributions corresponding to three different choices of two gluons which are involved in the reggeization procedure in the lowest order rc @xmath60,\ ] ] with the color factor @xmath61 symmetry reasons lead to the conclusion that @xmath62 and @xmath63 contributions are equal to @xmath64 one .
for @xmath64 we have @xmath65\right .
{ \nonumber}\\ & \quad -\left .
\frac{\phi_2^{(12)}({\mathbf{k}},{\mathbf{k_1}})}{({\mathbf{k}}-{\mathbf{k'}})^2 } \left[\frac{{\mathbf{k}}^2}{{\mathbf{k'}{\lower-.2em\hbox{}^{2}}}+({\mathbf{k}}-{\mathbf{k'}})^2}+ \frac{({\mathbf{k_1}}-{\mathbf{k}})^2}{({\mathbf{k_1}}-{\mathbf{k'}})^2 + ( { \mathbf{k}}-{\mathbf{k'}})^2}\right]\right\}.\end{aligned}\ ] ] the details of impact factor calculations are given in the appendices a d .
in the appendix e we give some details used for performing the integration over the transversal component of the loop momenta .
the quantities @xmath66 may be interpreted as a partial contributions to the odderon and pomeron intercepts .
their dependence on @xmath67 , @xmath50 is illustrated in fig .
[ pom ] and fig . [ ode ] .
the jet s masses was supposed to be larger than 1gev .
the size of contributions to the polarized and unpolarized differential cross sections depends on the used jet model as well as on the choice of the vertices which describe the transition of the nucleon to the jet and on the choice of the vertex function which describes the conversion of one sort of jet to the jet of another sort . because of the here used choice @xmath68
, we are forced to put on the gauge conditions .
another possible choice , @xmath69/m$ ] , leads to the zero contribution to the asymmetry ( in the limit of infinite large @xmath70 ) .
we see that one - spin effect is rather large
. another mechanisms of one - spin asymmetry associated with the nucleon resonances in intermediate state and final state interaction in @xmath71 process was considered in papers @xcite where the effect was of the same order .
the work was supported by intas-00366 .
e. z. is gratefull to rfbr for the grant 0001 - 00617 .
the explicit expressions for the lowest order impact factors in the unpolarized case are @xmath72 to simplify the calculation of the traces we write down here the useful relations @xmath73 which follow from the on mass shell conditions .
explicit expressions for @xmath38 and @xmath39 are @xmath74\big[\beta^2m^2+({\mathbf{p}}-{\mathbf{k_1 } } \beta)^2\big ] } , { \nonumber}\\\phi_{02}{\mathbf{k_1}})&=\frac{2{\mathbf{k_1}}^2\big [ { \mathbf{k_1}{\lower-.2em\hbox{}^{2}}}+(m_2-m)^2\big]}{\big({\mathbf{q}}^2+m_2 ^ 2-m^2\big)^2}\end{aligned}\ ] ]
the lowest order contribution to the impact factor corresponding to the polarized proton with 4-momentum @xmath75 is @xmath76 the quantity @xmath77 has a form @xmath78 the sudakov decomposition of the 4-vectors @xmath79 is given in ( a.2 ) .
the exchanged gluon expansions are @xmath80 the quantity @xmath81 is equal @xmath82 with @xmath83 given in ( [ eq:11 ] ) and @xmath84.\ ] ] the quantity @xmath85 has a form @xmath86 with @xmath87\ ] ] and with the same expression for @xmath79 . the relevant representation for the exchanged gluon 4momenta is following @xmath88 and the same expression for @xmath79 .
impact factor for the unpolarized proton @xmath55 has a form @xmath89 for then calculation of the trace for @xmath55 we used sudakov representation @xmath90
the impact factors for the case of unpolarized protons are @xmath91 where @xmath92
let us now give the expressions for the impact factors in the case of rc to the polarized cross sections .
for the @xmath93 in ( [ eq:21 ] ) we have @xmath94 with @xmath95 and @xmath96 with the substitutions similar to ones for @xmath77 from appendix b for the momenta including @xmath97 and the substitutions similar to ones for @xmath85 for @xmath98 . besides @xmath99 the impact factor of unpolarized proton in this case have a form @xmath100 with @xmath101 and @xmath102
here we give some details used in performing the loop momenta integration . when calculating the relevant trace we use the shouthen identity @xmath103 this identity permits to express all conversions with levi - chivita tensor in the standard form .
for instance @xmath104 the second term in the right side of this equation can be expressed through the @xmath105 using the relation @xmath106 | single - spin asymmetry appears due to the interference of single and double gluon exchange between protons .
a heavy fermion model is used to describe the jet production in the interaction of gluon with the proton implying the further averaging over its mass . as usually in one - spin correlations ,
the imaginary part of the double gluon exchange amplitude play the relevant role .
the asymmetry in the inclusive set - up with the pion tagged in the fragmentation region of the polarized proton does not depend on the center of mass energy in the limits of its large values .
the lowest order radiative corrections to the polarized and unpolarized contributions to the differential cross sections are calculated in the leading logarithmical approximation . in general , a coefficient at logarithm of the ratio of cms energy to the pion mass depends on transversal momentum of the pion .
this ratio of the lowest order contribution to the asymmetry may be interpreted as the partial contribution to the odderon intercept .
the ratio of the relevant contributions in the unpolarized case can be associated with the partial contribution to the pomeron intercept .
the numerical results given for the model describe the jet as a heavy fermion decay fragments .
@xmath0 _ join institute for nuclear research , 141980 dubna , russia , + @xmath1 _ leningrad institute for nuclear physics , gatchina , russia _ _ |
this note concerns completely - integrable systems of yang - mills - higgs equations , and in particular those which may be viewed as higher - dimensional generalizations of the two - dimensional hitchin equations ( the self - duality equations on a riemann surface ) .
let us begin by briefly setting out the notation .
we denote local coordinates on @xmath3 by @xmath4 with @xmath5 . for simplicity
we take the gauge group to be su(2 ) throughout .
a gauge potential @xmath6 takes values in the lie algebra @xmath7 , so each of @xmath8 is an anti - hermitian @xmath9 matrix .
the curvature ( gauge field ) is @xmath10 $ ] .
a higgs field @xmath11 takes values in the lie algebra , or , if complex , in the complexified lie algebra @xmath12 .
its covariant derivative is @xmath13 $ ] , and gauge transformations act by @xmath14 .
the prototype system is the simplest 2-dimensional reduction @xcite of the 4-dimensional anti - self - dual yang - mills equations @xmath15 this reduction can be written as a conformally - invariant system on the complex plane @xmath16 , or more generally on a riemann surface @xcite , and is effected as follows .
if we take all the fields to depend only on the coordinates @xmath17 , and we define a complex coordinate @xmath18 and a complex higgs field @xmath19 , then ( [ sdym4 ] ) reduces to the hitchin equations @xmath20=0.\ ] ] several higher - dimensional generalizations of ( [ h2 ] ) have been introduced and studied over the years . but most such generalizations lack a notable property of the original system ( [ h2 ] ) , namely its complete integrability .
the purpose of this note is to describe some features , and some solutions , of an integrable @xmath21-dimensional generalization of ( [ h2 ] ) .
let us focus specifically on generalizations to @xmath22 real ( or @xmath0 complex ) dimensions which involve @xmath22 real ( or @xmath0 complex ) higgs fields .
such systems may naturally be viewed as dimensional reductions of pure - gauge systems in @xmath1 dimensions , satisfying linear relations on curvature such as ( [ sdym4 ] ) .
of greatest interest are those that have the eigenvalue form @xcite @xmath23 where @xmath24 is totally - skew , because the bianchi identities then imply that the gauge field satisfies the second - order yang - mills equations .
perhaps the best - known example is the octonionic system of @xcite , which has @xmath2 .
this may be written @xmath25 whereas the prototype ( [ sdym4 ] ) is essentially based on the quaternions , this system ( [ octonion ] ) is based on the octonions : the components of @xmath24 are constructed from the cayley numbers .
it is invariant under the group spin(7 ) , and its 7-dimensional reduction is invariant under @xmath26 .
we now reduce to four dimensions by requiring the fields to depend only on the variables @xmath27 , defining two complex variables and two complex higgs fields by @xmath28 then the reduction of ( [ octonion ] ) is @xmath29 + { \tfrac{1}{4}}[\phi_2,\phi_2^ * ] & = 0 , \nonumber \\
f_{12}-{\tfrac{1}{4}}[\phi_1,\phi_2 ] & = 0 , \nonumber \\ d_{{\bar{1}}}\phi_1-d_2\phi_2^*=0 , \,\ , d_{{\bar{2}}}\phi_1+d_1\phi_2^ * & = 0 .
\label{oct4}\end{aligned}\ ] ] here the subscript @xmath30 in @xmath31 and @xmath32 refers to @xmath33 , whereas @xmath34 refers to the complex conjugate variable @xmath35 .
the equations ( [ oct4 ] ) are more familiar in the @xmath36 ( real ) form @xmath37\right)^+=0 , \quad \left(d\phi\right)^-=0 , \quad d*\phi=0,\ ] ] where @xmath38 is a lie - algebra - valued 1-form formed from the four real higgs fields .
the ` plus ' superscript denotes the self - dual part of a 2-form , and the ` minus ' superscript the anti - self - dual part .
this system has appeared in several contexts over the years @xcite , and has variously been referred to as the non - abelian seiberg - witten equations or the kapustin - witten equations .
known solutions include several obtained using a generalized t hooft ansatz @xcite .
a different generalization of ( [ h2 ] ) , defined on any khler manifold , is one attributed to simpson @xcite . in @xmath0 complex dimensions , with complex coordinates @xmath39 , @xmath40
, it takes the form @xmath41+\ldots+ { \tfrac{1}{4}}[\phi_k,\phi_k^ * ] & = 0 , \nonumber \\
f_{ab}=0 , \,\ , [ \phi_a,\phi_b]=0 , \,\ , d_{{\bar{a}}}\phi_b=0 .
\label{simp}\end{aligned}\ ] ] note that for @xmath42 , this system reduces to the prototype ( [ h2 ] )
. for @xmath2 , it clearly it implies ( [ oct4 ] ) .
the converse is not true in general , but it is if one imposes appropriate global conditions : in particular for smooth fields on a compact khler surface , it has recently been shown that ( [ simp ] ) and ( [ oct4 ] ) are equivalent @xcite .
another approach to generalizing the basic 4-dimensional system ( [ sdym4 ] ) is to look for higher - dimensional versions which are completely - integrable @xcite . for simplicity , we begin with the case @xmath2 .
an integrable 8-dimensional yang - mills system is @xmath43 which clearly implies the octonionic equations ( [ octonion ] ) .
the system ( [ a4 ] ) has the symmetry group @xmath44/{{\mathbb{z}}}_2\subset{\rm so(8)}$ ] , which corresponds to a quaternionic khler structure @xcite . the adhm construction of instantons @xcite generalizes to this case @xcite .
consider now the reduction to four dimensions , with the same complex variables ( [ cxform ] ) as before .
then ( [ a4 ] ) reduces to @xmath45=0 , \,\ , [ \phi_a,\phi_b]=0 , \,\ , f_{ab}=0 , \,\ , d_{[a } \phi_{b]}=0,\ ] ] where @xmath46 .
this system is even more overdetermined than ( [ simp ] ) .
so we have a string of implications , where ( [ a4red ] ) implies ( [ simp ] ) implies ( [ oct4 ] ) implies the four - dimensional yang - mills - higgs equations ( the reduction of pure yang - mills from eight dimensions ) . generalizing ( [ a4red ] ) to @xmath0 complex dimensions is straightforward : we simply allow the indices @xmath47 to range from @xmath30 to @xmath0 .
the system ( [ a4red ] ) has a very large symmetry group , since it involves only the holomorphic structure of the underlying complex manifold .
this becomes clearer if we define @xmath48 as a @xmath49-form with values in the complexified lie algebra : then ( [ a4red ] ) can be written @xmath50=0 , \,\ , [ \phi\wedge\phi]=0 , \,\ , f^{2,0}=0,\ ] ] where @xmath51 now denotes the covariant exterior derivative . by contrast , the less - overdetermined systems ( [ simp ] ) and ( [ oct4 ] ) depend on an underlying geometric structure , and have less symmetry .
the system ( [ a4red ] ) is completely - integrable by virtue of being the consistency condition for a lax @xmath21-tet , namely @xmath52 where @xmath53 is a complex parameter .
the integrability conditions @xmath54=0=[\eth_a,\eth_{{\bar{b}}}]\ ] ] for all @xmath53 are equivalent to the equations ( [ a4red ] ) .
the aim now is to describe some solutions of ( [ a4red ] ) ; these will therefore also be solutions of the other systems ( [ simp ] ) , and ( [ kw ] ) in the @xmath2 case . the equations ( [ a4red ] ) or ( [ a4red2 ] ) are defined on any @xmath0-dimensional complex manifold , and in general one may also allow singularities .
for example , in the @xmath42 case on a compact riemann surface of genus @xmath55 , smooth solutions of ( [ h2 ] ) exist only when @xmath56 ; on the 2-sphere and the 2-torus , solutions necessarily have singularities @xcite .
note that the functions @xmath57 are holomorphic , by virtue of the equations ( [ a4red2 ] ) . in what follows
, we look for solutions which are smooth on @xmath58 , and for which @xmath59 is a polynomial in @xmath39 .
so they may also be viewed as being defined on the projective plane @xmath60 , with a singularity on the line at infinity . to illustrate ,
let us first consider the abelian case , with the fields being diagonal , namely @xmath61 , where @xmath62 .
then the equations ( [ a4red2 ] ) are easily solved .
the gauge field vanishes , and therefore we may take the gauge potential to vanish as well .
the remaining equations give @xmath63 , where @xmath64 is an arbitrary polynomial on @xmath58 .
this is the general abelian solution .
for the non - abelian su(2 ) case , we adopt a simplifying ansatz which is familiar from the lower - dimensional version @xcite .
namely let us assume that the gauge potential is diagonal : in other words , @xmath65 .
( it should be emphasized that there are solutions for which this assumption does not hold . )
then the general local solution is determined by a holomorphic function @xmath64 , plus a solution @xmath66 of the elliptic sinh - gordon equation @xmath67 in terms of these , the higgs fields are given by @xmath68 and the functions determining the gauge potential are @xmath69 note that one solution of ( [ sgeqn ] ) is @xmath70 , but this is effectively the abelian case of the previous paragraph . in order to get genuine non - abelian fields , we choose @xmath64 to have branch singularities , and then to get smooth fields one needs @xmath71 .
the simplest such fields are embeddings of solutions of ( [ h2 ] ) on @xmath16 into @xmath58 , depending on @xmath39 only via a fixed linear combination @xmath72 . for example , @xmath73 gives an embedding of the one - lump solution on @xmath16 @xcite .
some simple solutions that are not of this embedded type are as follows .
let @xmath74 be a polynomial of degree at least two , and take @xmath75 .
this gives higgs fields of the form @xmath76 where @xmath77 satisfies @xmath78 we now need a smooth solution of ( [ sgeqn2 ] ) satisfying the boundary condition @xmath79 as @xmath80 .
there exists a unique such solution , which is essentially a painlev - iii function @xcite .
in fact , if we define @xmath81 , where @xmath82 , then ( [ sgeqn2 ] ) becomes an equation of painlev - iii type , namely @xmath83 this has a unique solution with the required asymptotics .
the upshot is that any polynomial @xmath74 gives a solution of ( [ a4red2 ] ) which is smooth on @xmath58 and has @xmath84 it appears ( see for example the figure below ) that the gauge field @xmath85 is concentrated around the zero - set of @xmath86 . in the general @xmath0-complex - dimensional case ,
one expects the gauge field to be concentrated around a submanifold of complex codimension 1 , and for the field to be approximately abelian elsewhere .
the simplest case has @xmath86 quadratic , so that @xmath87 is a conic . for @xmath88 , with @xmath89 and @xmath90 respectively . ]
figure 1 is a plot of the norm @xmath91 of the gauge field , on the real slice @xmath92 , for the solutions corresponding to the choices @xmath89 ( on the left ) , and @xmath90 ( on the right ) . here
@xmath91 is computed using the metric @xmath93 on @xmath58 , which leads to the formula @xmath94 the figures were generated by solving ( [ piii ] ) numerically to get @xmath95 , and then using this formula ( [ modf ] ) .
clearly @xmath91 is concentrated around the conic @xmath87 .
the right - hand case corresponds to a degenerate conic , and is the reduced version of what was called instantons at angles@xcite for solutions of ( [ a4 ] ) .
there are some compact complex manifolds @xmath96 on which smooth solutions of ( [ a4red2 ] ) exist . as a trivial example
, one could take @xmath96 to be a product @xmath97 , where @xmath98 is a riemann surface of genus at least two , and @xmath99 is any other manifold : then a solution of ( [ h2 ] ) on @xmath98 is also a solution of ( [ a4red2 ] ) on @xmath97 .
the moduli space of solutions on any compact manifold , if it is non - empty , has a natural @xmath100 metric , which on general grounds one expects to be hyperkhler .
even more generally , one could allow singularities of a specified type , or equivalently for the ambient space to be non - compact . in this latter case
, some of the parameters in the solution space may have @xmath100 variation , giving rise to a moduli space with a well - defined metric . analysing the possible moduli space geometries which arise in this way
would be worthwhile , although a considerable task . in this note
, we have focused on a particular type of reduction of the integrable system ( [ a4 ] ) , and of its @xmath101-dimensional generalization .
there are several other dimensional reductions of the octonionic system ( [ octonion ] ) which are of interest : see , for example , reference @xcite . in each case
, the appropriate reduction of ( [ a4 ] ) gives an integrable sub - system , and hence a source of solutions .
* acknowledgments .
* this work was prompted by a communication from sergey cherkis .
the author acknowledges support from the uk particle science and technology facilities council , through the consolidated grant st / l000407/1 . | this letter describes a completely - integrable system of yang - mills - higgs equations which generalizes the hitchin equations on a riemann surface to arbitrary @xmath0-dimensional complex manifolds .
the system arises as a dimensional reduction of a set of integrable yang - mills equations in @xmath1 real dimensions .
our integrable system implies other generalizations such as the simpson equations and the non - abelian seiberg - witten equations .
some simple solutions in the @xmath2 case are described . * msc : * 81t13 , 53c26 . *
keywords : * gauge theory , higgs , integrable system . |
in models of galaxy formation by hierarchical clustering , the disks of spiral galaxies form by late accretion of gas which cools from an extended reservoir around the galaxy . in these models disks
are still expected to be growing at the present day .
such models have been very successful in accounting for properties of galaxies at optical and infrared wavelengths ( see , for example , kauffmann et al .
1993 , cole et al .
1994 , baugh , cole & frenk 1996 , kauffmann et al . 1999 ) .
the proposed disk formation mechanism should produce a signature at x - ray wavelengths which , in principle , is detectable with the present generation of satellites . at the virial temperature of galaxy halos , the dominant cooling mechanism is x - ray bremsstrahlung .
if the cooling rate is significant , this flux may be visible as a component of diffuse x - ray emission extending well beyond the galaxy s optical radius . in this paper
, we set out to test this disk formation model by looking for x - ray emission from the hypothesized reservoir of cooling gas around three large , late - type galaxies , ngc 2841 , ngc 4594 and ngc 5529 .
if we assume that the mass of the present disk has been built up by a constant rate of accretion over the age of the universe ( @xmath2 ) , a simple estimate of the x - ray luminosity can be made from the binding energy of the disk .
@xmath3 where @xmath4 is the present mass of the disk , @xmath5 ^{1/2}$ ] is the velocity required by material at the edge of the disk to escape the halo ( assuming a flat rotation curve ) and @xmath6 is the measured circular velocity of the disk . adopting parameters suitable for a large spiral galaxy such as ngc 2841 ( ie . , @xmath7 , and @xmath8 )
, this suggests that the halo should emit a bolometric luminosity of at least @xmath9 , @xmath10 , @xmath11 , @xmath12 and @xmath13 .
if supernovae and other feedback processes are effective in heating halo gas then even more energy must be radiated to assemble the observed disk .
note that eqn .
( [ eq : be ] ) predicts a minimum @xmath14 , and that the temperature of the cooling gas should scale as the halo virial temperature , @xmath15 .
thus , it is important to search for emission in the most massive systems available . the picture of hierarchical galaxy formation developed by white & frenk ( 1991 , wf91 ;
see also kauffmann et al .
1993 , cole et al . , 1994 ) predicts halo x - ray luminosities more directly by computing the cooling rate of gas as a function of time . at the present epoch ,
such models ( which are tuned to reproduce the observed optical properties of galaxies ) predict an x - ray luminosity of about @xmath16 for galaxies like ngc 2841 . [ cols= " < , > , > , > , > , > , > , > " , ]
we have searched for large - scale diffuse x - ray emission from three nearby , high rotation velocity spiral galaxies . in each case , we find that the observed emission is almost _ an order of magnitude _ below that required to build the disk at a constant rate within the age of the universe .
the accretion rate is similarly discordant with theoretical estimates of the cooling rate of gas in an isothermal or nfw - profile halo .
although extremely suggestive , our results are not definitive .
the reasons for this are : * strong constraints are derived only for the two nearer galaxies .
we may , by chance , have selected two systems that have anomalous cooling rates .
this is plausible for the bulge dominated ngc 4594 galaxy ( ` the sombrero ' ) .
ngc 2841 appears to be more typical of spiral galaxies , but its proximity makes accurate background subtraction difficult , and our results are dependent on the profile assumed for the x - ray emission .
by contrast , the third galaxy meeting our selection criteria is too distant for strong limits on the x - ray flux to be set .
we note , however , that the x - ray limits established for these galaxies are compatible with the emission that is thought to occur in the halo of our own galaxy ( eg .
, moore & davis , 1994 , wolfire , et al .
* there are a number of parameters that have been assumed in order to predict the x - ray flux from the properties of the halo .
no parameter , or combination of parameters , can simultaneously give a low x - ray luminosity and build the observed disk . to reconcile the predictions with the observed luminosities we must assume that either : ( a ) disks were built at high redshift ; or , ( b ) accretion onto disks occurs without radiating the binding energy of the gas at x - ray wavelengths ( e.g. in a multiphase flow ) .
our results encourage further work .
the current upper limits are set by our ability to subtract the background count - rate in the images .
this is exacerbated in the rosat data because the flux level at which individual point sources may be subtracted varies across the image on scales comparable to the spiral galaxy haloes .
future work requires large - scale imaging with a detector that has a near - uniform detection threshold , such as the epic camera aboard the xmm satellite ( lumb et al .
such a survey would need to target a greater fraction of the nearby galaxies , setting firmer limits on the diffuse x - ray emission and eliminating variations in galaxy properties as a major source of uncertainty .
if such work continues to fail to make detections , then we may be forced to conclude that spiral disks are no longer growing at the present epoch .
this would agree with traditional models of galaxy formation in which galaxies accrete their mass at early times and evolve as more or less closed systems to the present - day .
clearly , there is much to learn from future studies of the x - ray haloes of spiral galaxies .
the authors would like to acknowledge shaun cole and julio navarro for providing rotation curve modelling code and best fit parameters for ngc 2841 respectively .
we thank enzo branchini for making available to us his peculiar velocity field models prior to publication and alan smith for his work on a pilot version of this project as part of his undergraduate studies at the university of durham .
we have made use of the nasa extragalactic database ( ned ) and the leicester database and archive service ( ledas ) .
this project was carried out using computing facilities supplied by the starlink project , and was supported by a pparc rolling grant for `` extragalactic astronomy and cosmology at durham '' and by the european community s tmr network for galaxy formation and evolution .
ajb and csf acknowledge receipt of a pparc studentship and senior research fellowship respectively .
csf also acknowledges a leverhume fellowship .
abadi , m. , bower , r. g. , navarro , j. , 1999 , astro - ph/9904364 ( submitted to mnras ) baugh c. m. , cole s. , frenk c. s. , 1996 , mnras , 283 , 1361 baugh c. m. , cole s. , frenk c. s. , benson a. j. , lacey c. g. , 1998 , in p. carral and j. cepa eds . ,
star formation in early - type galaxies , asp conference series , vol .
163 , astron .
, san francisco , p. 227 .
bertschinger , e. , 1989 , apj , 340 , 666 blitz l. , spergel d. n. , teuben p. j. , hartmann d. , burton w. b. , 1999 , apj , 514 , 818 branchini e. , teodoro l. , frenk c. s. , schmoldt i. , efstathiou g. , white s. d. m. , saunders w. , rowan - robinson m. , keeble o. , tadros h. , maddox s. , oliver s. , sutherland w. , 1999 , mnras , 308 , 1 brighenti f. , matthews h. , 1999 , aj , 117 , 1056 burstein d. , jones c. , forman w. , marston a. p. , marzke r. o. , 1997 , apjs , 111 , 163 cole , s. , lacey , c. , 1993 , mnras , 262 , 627 cole s. , aragn - salamanca a. , frenk c. s. , navarro j. f. , zepf s. e. , 1994 , mnras , 271 , 781 cole s. , baugh c. m. , frenk c. s. , lacey c. g. , 1999 , in preparation de vaucoulers g. , 1979 , apj , 227 , 729 dickey j. m. , lockman f. j. , 1990 , araa , 28 , 215 fabbiano g. , juda j. z. , 1997 , apj , 476 , 666 forman w. , jones c. , tucker w. , 1985 , apj , 293 , 102 young , j. s. , allen , l. , kenny , j. d. p. , lesser , a. , rownd , b. , 1996 , aj , 112 , 1903 kauffmann g. , white s. d. m. , guideroni b. , 1993 , mnras , 264 , 201 kauffmannn g. , colberg j. m. , diaferio a. , white s. d. m. , 1999 , mnras , 303 , 188 kent s. m. , 1987 , aj , 93 , 816 kim d .- w . , fabbiano g. , trinchieri g. , 1992 , apjs , 80 , 645 lumb , d. , eggel , k. , laine , r. , peacock , a. , 1996 , proceedings of the international society for optical engineering ( spie ) , 2808 , 326 mittaz j. p. d. , carrera f. j. , romero - colmenero e. , mason k. o. , hasinger g. , mcmahon r. , andernach h. , bower r. , burgos - martin j. , gonzalez - serrano j. i. , wonnacott d. , 1999 , mnras , 308 , 233 mathews , w. g. , brighenti , f. , 1998 , apj , 503 , 15 moore , b. , davis , m. , 1994 , mnras , 270 , 209 mulchaey , j. s. , davis , d. s. , mushotzky , r. , f. , burstein , d. , 1996 , apj , 456 , 80 mushotzky r. f. , szymkowiak a. e. , 1988 , in `` cooling flows in clusters and galaxies '' , ed .
a. c. fabian , kluwer academic publishers , the netherlands , p. 53 .
navarro j. f. , 1998 , astro - ph/9807084 ( submitted to apj ) ( n98 ) navarro j. f. , frenk c. s. , white s. d. m. , 1996 , apj , 462 , 563 ( nfw ) nulsen p. e. j. , fabian a. c. , 1995 , mnras , 277 , 561 nulsen p. e. j. , fabian a. c. , 1997 , mnras , 291 , 425 pelligrini s. , ciotti l. , 1998 , a&a , 333 , 433 ponman t.j . , allan d. j. , jones l. r. , merrifield m. , mchardy i. m. , lehto h. j. , luppino g. a. , 1994 , nature , 369 , 462 read , a. m. , ponman , t. j. , 1998 , mnras , 297 , 143 read , a. m. , ponman , t. j. , strickland , d. k. , 1997 , mnras , 286 , 626 snowden , s.l .
, mccammon , d. , burrows , d.n . , mendenhall , j.a .
, 1994 , apj , 424 , 714 thomas p. a. , nulsen p. e. j. , fabian a. c. , 1987 , mnras , 228 , 973 trinchieri g. , kim d. w. , fabbiano g. , 1994 , apj . , 428 , 555 trinchieri g. , fabbiano g. , kim d. w. , 1997 , a&a , 318 , 361 tully r. b. , fouqu p. , 1985 ,
apjs , 58 , 67 white s. d. m. , frenk c. s. , 1991 , apj , 379 , 52 ( wf91 ) wolfire , m. g. , mckee , c. f. , hollenbach , d. , tielens , a. g. g. m. , 1995 , apj , 453 , 673 wu , k. k. s. , nulsen p. e. j. , fabian a. c. , 1999 , mnras , submitted .
zaritsky d. , smith r. , frenk c. s. , white s. d. m. , 1997 , apj , 487 , 39
following wf91 , we assume that cooling occurs within a static potential , and that the cooling gas remains isothermal as it flows to the centre .
the assumption of isothermality allows considerable simplification in the determination of the x - ray luminosity of the flow since it is directly related to the drop in potential energy ( ie . , the internal energy of the gas remains constant ) .
detailed , fully self - consistent flow models , such as those of bertschinger ( 1989 ) and abadi et al .
( 1999 ) , show that these assumptions are reasonable over the range of scales over which we wish to determine the flow solution .
the x - ray luminosity emitted by a unit mass of gas as it moves within the potential , @xmath17 , is in models such as those of kauffmann et al .
( 1993 ) and cole et al .
( 1994 ) gas is assumed to flow from the cooling radius towards the centre of the halo .
no mass is assumed to fall out of the flow until it reaches the central galaxy .
we find the total energy emitted by unit mass of gas as it flows inwards and then integrate from @xmath19 to @xmath20 to find the total energy emitted by the mass of gas , if we now make the assumption that mass flows to the centre more quickly than the structure of the flow changes , then the structure of the flow at a snapshot in time can be represented by the trajectory of a single mass unit .
hence below we consider two possibilities for the structure of the dark matter halo . in [ isocoolsec ] we consider an isothermal halo and in [ nfwcool ] we consider the dark matter profile described by nfw , in each case assuming that the gas initially traces the dark matter .
this potential was considered by wf91 .
the dark matter density profile is @xmath23 , where @xmath24 is radial distance in the halo .
the cooling radius , @xmath19 , in the halo is defined as the radius at which the cooling time , @xmath25 , of the gas equals the age of the universe , @xmath2 .
ideally , we would use the time since the halo last doubled in mass ( cole & lacey , 1993 ) , but this information is not available for an individual galaxy .
since mass cooling rates decline with time , this leads to a conservative comparison with our observational results .
( white & frenk 1991 ) , where @xmath27 is the baryon fraction , @xmath28 is the cooling function of the gas , @xmath29 is the fraction of the initial baryon density remaining in gaseous form ( which we normally assume to be 1 ) , @xmath30 is the gravitational constant , @xmath31 is the mass of a proton , @xmath32 is the mean molecular weight of the gas , @xmath6 is the circular velocity of the halo and @xmath33 is boltzmann s constant .
nfw provide a fitting formula for the density profiles of dark matter haloes in n - body simulations of hierarchical clustering .
this profile is characterised by two parameters : @xmath35 , a characteristic density , and @xmath36 , a characteristic radial scale in the halo .
nfw find that @xmath35 and @xmath36 are correlated in any given cosmology so that the nfw profile provides a universal , one - parameter model of the halo .
this formula provides a good approximation to the density profile of haloes of all masses over at least two orders of magnitude in radius and its shape is independent of the values of the cosmological parameters .
appropriate parameters for the galaxy haloes in any given cosmology are determined by requiring that the measured circular velocity of the galaxy corresponds to the maximum in the nfw rotation curve .
this must be fitted in an iterative fashion since the ratio between @xmath37 and @xmath38 itself depends on @xmath39 . | current theories of galaxy formation predict that spiral galaxies are embedded in a reservoir of hot gas .
this gas is able to cool onto the galaxy replenishing cold gas that is consumed by star formation .
estimates of the x - ray luminosity emitted in the cooling region suggest a bolometric luminosity of order @xmath0 in massive systems .
we have used rosat pspc data to search for extended x - ray emission from the haloes of three nearby , massive , late - type galaxies : ngc 2841 , ngc 4594 and ngc 5529 .
we infer 95% upper limits on the bolometric x - ray luminosities of the haloes of ngc 2841 , ngc 4594 and ngc 5529 of 0.4 , 1.2 and 3.8 @xmath1 respectively .
thus , the true luminosity lies well below the straightforward theoretical prediction .
we discuss this discrepancy and suggest a number of ways in which the theoretical model might be brought into agreement with the observational results .
a possible solution is that the gravitational potentials of these galaxies dark matter haloes are weaker than assumed in the current model .
alternatively , the present day accretion may be substantially less than is required on average to build the disk over the hubble time .
our results are , however , based on only three galaxies , none of which is ideal for this kind of study . a larger data set is required to explore this important problem further
. cooling flows - x - rays : galaxies - galaxies : formation - galaxies : ngc 2841 , ngc 4594 , ngc 5529 |
magnetic materials are essential for modern technology . up to date ,
most used magnetic materials involve elements with partially filled d or f - orbitals , like transitional metals , rare earth metals and their compounds .
magnetic ordering in these compounds originates from the localization of the electrons so that the kinetic energy can be minimized .
this is the result of competitions between the kinetic energy and the exchange splitting .
the former favors nonmagnetic state while the latter is the source of spin polarization .
magnetism from p - states rather than the above conventional d or f states has been a hot topic in recent years because of technical importance and theoretical interests as well .
there are several ways to introduce spin polarization to the p - states .
several alkali and alkaline earth compounds with elements from group iva ( carbides and silicides),@xcite va(nitrides , pnictides , and arsenides)@xcite , and via ( selenlides)@xcite are predicted to be magnetically ordered .
oxides are not so usual , but doping with c and n , or introducing cation vacancies can be sources of magnetism.@xcite another large category comes from nanostructures of graphene , graphite and c@xmath2@xcite or their modification such as doping or absorption of other atoms .
the origin of the spin polarization is explained mostly from the stoner s theory where localization of the p - orbitals triggers the instability of the nonmagnetic state to magnetic state@xcite .
bulk carbon compounds are rarely found to be spin polarized .
the most distinct feature of these compounds with p - orbital magnetism is that in most cases they are half metals , where in the one electron picture , one spin channel in the kohn - sham band structure has a finite density of states(dos ) , while the other is null .
this is very important and promising for its possible applications in spintronics.@xcite monocarbides mc , where m may be alkali metal and alkaline earth metal show the above interesting magnetic properties , namely half metallic ferromagnetism , due to there unique electronic structures as predicted by several theoretical works.@xcite in our previous work,@xcite ferromagnetic ordering and its electronic origin were investigated in alkaline earth carbides with the rocksalt structure .
it was found that the magnetism is from the polarization of the carbon p - orbitals , while the alkaline earth metals , due to their large atomic radii , guarantee the sufficient large distance between the carbons , which localizes the p - orbitals . at the same time
, they are also electron provider because of the smaller electronegativity than that of c. in this case , we proposed that the electronic structure and the spin magnetic moment can be nicely tuned , which gives us freedom to tailor their properties when used as a spin filter , or spin injector . in this work ,
we show how the electronic structure can be tailored by doping ( a@xmath3b@xmath4c ) and by hydrostatic pressures .
it turns out that the fundamental gap as well as the spin flip gap can be tuned and induce the transition from a semiconductor to metal .
the calculations were performed by the full - potential local - orbital code@xcite in the version fplo9.00 - 33 with the default basis settings .
all calculations were done within the scalar relativistic approximation .
the general gradient approximation of exchange - correlation functional was chosen to be that parameterized by perdew , burke , and ernzerhof.@xcite the number of k - points in the whole brillouin zone was set to @xmath5 in order to ensure the convergency of the results .
convergency of the total energy was set to be better than @xmath6 hartree together with that of the electron density better than @xmath7 in the internal unit of the codes .
the fcc unit cell ( space group fm@xmath8 m ) was used to represent the rocksalt structure with wyckoff positions assigned to metal ( m ) and c as m(0,0,0 ) and c(0.5,0.5,0.5 ) , respectively .
doping was modeled in a supercell with the total atom number up to 32 .
as found in our previous work,@xcite the dos around the fermi level are mainly contributed by the 2p - electrons of c. we constructed 3 carbon centered wannier functions(wfs ) .
the resulting wfs are p - like . in the basis of these wfs ,
if the small interactions below 0.001 ev are ignored , the elements of the tight - binding(tb ) non - magnetic hamiltonian @xmath9_{3\times3}$ ] can be written as , @xmath10 + 4t_4\cos(\frac{a}{2}k_j)\cos(\frac{a}{2}k_k)+2t_5\cos(ak_i)\ ] ] and @xmath11 where @xmath12 is the lattice constant , and subscripts @xmath13and@xmath14 and @xmath15 , respectively .
the hoping constants @xmath16 is onsite contribution , @xmath17 , @xmath18 and @xmath19 are from the nearest neighbor interactions , and @xmath20 is from the next - nearest neighbor interactions as shown in fig .
[ fig : tpp ] .
the values are @xmath21 ev , @xmath22 ev , @xmath23 ev , @xmath24 ev and @xmath25 ev .
we see that the strongest interactions come from neighboring carbons with orthogonal p - orbitals .
the bands from this tb hamiltonian are shown in fig .
[ fig : wfbands ] together with those from the full potential calculations .
the agreement of the two results is quite satisfactory .
( color online ) cartoon representation of the interactions between different p - like wfs centered on the carbons . ]
( color online ) comparison of the bands from tb model ( tb dashed curves ) and full potential calculations(fp solid curves ) . ] in order to illustrate the binding in the real space , the charge density contour of the compounds is shown in fig .
[ fig : edensity ] .
spherical distribution of the electrons around the ions is a typical case of ionic bonding between the metals and carbon . in this scenario ,
the metallic ions determines the distances between the carbons .
the characteristics of electrons around the fermi level are mostly determined by the interaction of the 2p - orbitals from c. exchange interaction is thus direct exchange between the carbons like in zinc - blende m@xmath26c@xcite .
this specific electronic structure gives us a large degree of freedom to tune the properties of the material as will be discussed in the following sections .
the contour lines of the electron charge density of kc at equilibrium lattice constant . ]
as changes of the volume can result in changes of the bandwidth , the dos of these compounds are strongly dependent on the pressure which results in the dependence of magnetism on the volume ( pressure ) .
the dependence of the spin magnetic moment on the volume of m@xmath27c is illustrated in fig .
[ fig : ssia ] .
dependence of the spin magnetic moment per primitive cell on the lattice constant . the corresponding pressure at the different lattice constant is also shown .
the equilibrium lattice constant @xmath28 of the compounds are marked with arrows . ]
it can be clearly seen that shrinking of the lattice constant can decrease the spin magnetic moment in all the compounds .
this is the results of the increased kinetic energy of the electrons under pressure as expected .
the lattice constant at which the spin polarization is set in is proportional to that at equilibrium , except that of nac which is close to the lattice constant at equilibrium .
this is because nac is metallic at equilibrium as shown in our previous work @xcite .
however , the processes of the magnetic moment approaching zero are different : the spin moment of kc , rbc and csc show an abrupt drop to zero at certain critical lattice constant while that of nac drops first to an intermediate value of 1.5 @xmath29 and then to zero .
the difference comes from the different dos at the fermi level . by simple consideration from the rigid band model ,
the susceptibility of the magnetic moment to external parameters is proportional to the dos at the fermi level and its derivative .
the dos of nac and kc in the vicinity of the transitional region are shown in fig .
[ fig : doskc ] and fig .
[ fig : dosnac ] in order to illustrate this point .
comparing the dos of kc at lattice constant of 4.5 and 4.6 in fig .
[ fig : doskc ] , we can see that a rigid shift of the states of spin up and spin down is obvious in the spin polarized state . in kc , the onset of spin polarization at 4.6
set the fermi level at the high dos about 2 ev@xmath30 in the spin up channel .
the fermi level is between two saddle points . in nac
, the fermi level is also between the two saddle points when the compound is at the intermediate magnetic moment state ( e.g. 1.2 @xmath29 when @xmath31 ) as shown in fig .
[ fig : dosnac ] .
the spin polarization at this lattice constant is because of the high dos at the fermi level in the nonmagnetic state ( not shown ) which triggers the magnetic instability according to the stoner s model .
increase of the lattice constant enhances the exchange splitting , and the fermi level passes through the upper saddle point as shown in fig .
[ fig : dosnac](b ) . in this case
the magnetic moment increase quickly to a higher value .
( color online ) band of kc at lattice constant @xmath32 ( a ) , and dos of kc at lattice constant @xmath33 and 4.6 ( b ) .
the minus sign in the dos indicates that it is of spin down channel , while the numbers without any signs indicate that they are of the spin up channel .
the following dos figures follow this convention if not specified . ]
( color online ) band of nac at lattice constant @xmath31 ( a ) , and dos of nac at lattice constant @xmath33 , 5.0 and 5.5 ( b ) . ]
the critical pressure of phase transition from the ferromagnetic ( fm ) state to the nonmagnetic ( nm ) state of csc and nac is approximately 28.2 gpa and 13.8 gpa .
the fm state of kc is so stable and begins to decrease at about 30 gpa . under high pressure over 80 gpa
we see the transition of the compound transforms into the nm state . not only the magnetic order , but also the band gap can be tuned by pressure . when the compounds are semiconductor with a finite gap , the bandwidth starts to broaden under pressure .
in this case the bandgap decreases continuously . normally insulating states transforms into metallic states .
an example from kc is shown in fig . [
fig : dos_kc ] . at zero pressure
, corresponding to the lattice constant @xmath34 the gap is 1.0 ev formed between the two spin states as in fig .
[ fig : dos_kc](a ) .
it is called the spin flip gap as noted by capelle @xcite .
the fundamental gaps are the same as the spin flip gap . from the band structure
, we also find that the gap is an indirect gap between the @xmath35 and @xmath36point in the brillouin zone . under pressure ,
the gap comes to closure as shown in fig .
[ fig : dos_kc](c ) .
the compound transforms from magnetic insulator to magnetic metal . at the some critical pressure ,
the highest occupied spin up states just touches the lowest unoccupied spin down states as in fig.[fig : dos_kc](b ) .
it reaches an interesting situation where no energy is required to flip the spin .
this spin - gapless semiconductor has some important features including that the electron excitation from valence bands to conduction bands needs no energy and the carriers are fully spin - polarized .
as the carriers are fully spin - polarized , they can be easily separated by hall effect .
the features are useful in some novel applications as recently reviewed by wang@xcite _
et al_. dos of kc at different lattice constant : ( a ) 6.21 , ( b ) 5.18 and ( c ) 4.76 respectively . ] dos of k@xmath37ba@xmath38c at different doping levels indicated by @xmath1 . ]
as was shown above , the electronic states near the fermi level are from the p - orbitals of carbon .
the alkali metals serve as electron provider which determines the fermi level of the system .
we already obtained that bac and kc have magnetic moment of 2.0 and 3.0 @xmath29 , respectively . in these two compounds , the spin up channel are fully occupied . in this case , the change of the electron number will only change the occupation of the spin down channel . if we substitute the ia elements by iia ones , forming compounds m@xmath3k@xmath4c , the magnetic moment @xmath39 varies as @xmath40 .
the variation of the dos and the magnetic moment at different doping levels when k is substituted by ba in kc are shown in fig .
[ fig : dos_kbac ] .
starting from kc , the spin up states are fully occupied while the spin down states are empty , because there are 3 valence electrons and 3 p - states from c are available .
doped with the iia element ba , for instance , the extra electrons can only go to the spin down channel .
then the metallic state is produced and the electrons at the the fermi level is fully spin polarized .
thus , a half metal is formed .
the magnetic moment decreases according to @xmath40 . at the same time , the half metallicity preserves .
thus a tunable half metal is obtained .
the decrease of the magnetic moment can benefit devices where magnetic static interferences can be reduced .
in summary , we performed electronic structure calculations of the mc compounds where @xmath39 is one of the ia elements ( na , k , rb , and cs ) based on dft .
the results show that there is direct exchange of p - electrons of c which is the source of magnetism .
the alkali metals serve as an electron source .
the states near the fermi level are contributed mainly from carbon .
all the compounds are ferromagnetic insulator at equilibrium with spin magnetic moment 3.0 @xmath29 , except that nac is metallic .
nac can be insulator by slight expansion of the lattice constant . the electronic structure and magnetism of the compounds
can be tuned by pressure and doping with alkaline earth metals .
the variation of the physical parameters can be understood by the rigid band model .
the compounds can have potential applications in spintronics and photoelectronics dure to its half metallicity and spin gapless feature .
this work was financially supported by international science & technology cooperation program of china ( 2012dfa51430 ) and the fundamental research funds for the central universities ( zygx2011j021 ) .
verma , mohini , p.s .
bisht , p. jensen , semicond .
sci . technol . * 25 * , 105002 ( 2010 ) .
gao , k.l .
yao , z.l .
liu , y. min , j. zhang , s.w .
fan , d.h .
zhang , j. phys . : condens .
matter * 21 * , 275502 ( 2009 ) .
k. zdoan , e. aioiu , i. galanakis , j. appl .
phys . , * 111 * , 113918 ( 2012 ) .
m. sieberer , j. redinger , s. khmelevskyi , and p. mohn , phys .
b , * 73 * , 024404(2006 ) .
e. yan , physica b , * 407 * , 879 ( 2012 ) .
gao , k.l .
yao , m.h .
song and z.l .
liu , j. magn .
, * 323 * , 2652(2011 ) .
m. yogeswari , g. kalpana , computational materials science * 54 * , 219 ( 2012 ) .
v. pardo , w.e .
pickett , phys .
b * 78 * , 134427 ( 2008 ) .
yazyev , rep .
* 73 * , 056501 ( 2010 ) .
i. @xmath41uti , j. fabian , and s. das sarma , rev .
* 76 * , 323 ( 2004 ) .
o. volnianska , p. boguslawski , j. phys . : condens
* 22 * , 073202 ( 2010 ) .
gao , k.l .
yao , appl .
lett . * 91 * , 082512 ( 2007 ) .
gao , k.l .
yao , e. aiolu , l.m .
sandratskii , z.l .
liu , and j.l .
jiang , phys .
b , * 75 * , 174442 ( 2007 ) .
ch.w . zhang , j. phys .
phys . * 41 * , 085006 ( 2008 ) .
w.x . zhang , z.d .
song , b. peng , w.l .
zhang , * 112 * , 043902 ( 2012 ) .
s.j . dong , h. zhao , appl .
. lett . * 98 * , 182501 ( 2011 ) .
k. koepernik and h. eschrig , phys .
b * 59 * , 1743 ( 1999 ) .
perdew , k. burke , and m. ernzerhof , phys .
lett . * 77 * , 3865 ( 1996 ) .
k. capelle , g. vignale , c.a .
ullrich , phys .
b * 81 * , 125114 ( 2010 ) .
wang , s.x .
zhang , npg asia materials * 2 * , 31 ( 2010 ) . | electronic structures of carbides with the rocksalt structure were calculated by full potential electronic codes solving the kohn - sham equation .
bonding characters were analyzed by constructing tight - binding hamiltonian based on maximally - localized wannier functions .
it was found that the cations in these compounds act as an electron provider and the frame is formed by the carbon atoms .
the electronic states in the vicinity of the fermi level are mainly from the p - orbitals of c. pressure and doping are two efficient ways to tune the magnetic and electronic properties of these compounds .
it turns out that a spin gapless semiconductor can be obtained by applying hydrostatic pressure up to tens of gigapascal .
higher pressure induced an insulator to metal transition because of band broadening .
compounds of ia group ( na , k , rb , cs ) were magnetic semiconductor at ambient conditions .
alloying with iia elements decrease the magnetic moment according to the law of @xmath0 , where @xmath1 is the relative atomic ratio of the iia elements to the ia ones .
the behaviors of the compounds under the pressure and the doping effects can be understood by a rigid band model . |
graphical models , such as factor graphs and normal factor graphs @xcite @xcite , can provide a concise description of the probabilistic assumption of an inference problem and they are indispensable for analyzing message passing inference algorithms such as bp ( belief propagation ) .
for example , the relationship between `` codes and graphs '' is one of key concepts of modern coding theory .
al - bashabsheh and mao @xcite recently shed a new light to normal factor graphs .
they clearly showed that _ holographic transformations _ to normal factor graphs are versatile tools for deriving non - trivial identities on the partition function of a normal factor graph @xcite .
a holographic transformation is a local graphical transformation that preserves the partition function .
it should be remarked that the holographic transformation has been used in many research fields .
the prominent example is the class of holographic algorithms invented by valiant @xcite .
he showed that several combinatorial enumeration problems defined on planer graphs can be transformed into perfect matching problems via appropriate holographic transformations .
such a planar perfect matching problem is solvable in polynomial time .
another example is duality theorems @xcite @xcite @xcite for codes defined on graphs .
the main contribution of this paper is a non - trivial expression , that is called _ dual expression _ , of the posterior values for a non - adaptive group testing problem .
the derivation is based on a holographic transformation to the normal factor graph representing a group testing inference problem .
the derivation process has similarities to the proof of macwilliams identity @xcite and a bitwise map decoding algorithm by hartmann and rudolph @xcite for binary linear codes .
however , in our case , we can not rely on the standard fourier ( _ i.e. , _ hadamard ) transformation because we need to treat or constraints instead of even parity constraints . a local linear transformation matched to or constraints plays a key role for the following discussion .
the research of group testing started from the celebrated work by dorfman @xcite and has been extensively studied @xcite .
we here suppose the following simplest setting for a non - adaptive group testing .
assume that we have @xmath0-_objects _ and some groups of these objects .
each object can take value 1 ( positive ) or 0 ( negative ) according to the prior probability for each object .
test _ can be applied for each predetermined group .
the result of a test is positive if the group contains a positive object ; otherwise the test result becomes negative .
our inference problem is to evaluate the posterior probabilities for objects from @xmath1-disjunctive test results and from our knowledge on the prior probabilities .
development of fast inference algorithms evaluating posterior probabilities ( or their estimates ) is an active area of research ; for example , see approximate inference algorithms based on bp @xcite @xcite .
let @xmath5)$ ] be a binary ( zero or one ) independent random variable representing the state of the @xmath6-th object ( _ i.e. , _ negative or positive ) .
the notation @xmath7 $ ] represents consecutive integers from @xmath8 to @xmath9 .
the vector of the random variables @xmath10 is thus distributed according to the joint distribution : @xmath11 where @xmath12 .
we suppose that an inference algorithm perfectly knows these prior probabilities @xmath13 .
assume that an undirected bipartite graph @xmath14 , called a _ pooling graph _
, is given where @xmath15 and @xmath16 are sets of vertices , and @xmath17 is the set of edges connecting a vertex in @xmath18 and a vertex in @xmath19 , namely @xmath20 .
the set @xmath21 is defined by @xmath22 \mid ( v^{(2)}_i , v^{(1)}_{j } ) \in e \}\ ] ] for @xmath23 $ ] .
the boolean function @xmath24 is just the logical or function with @xmath25-inputs defined as @xmath26 , x_k=1 ] .\ ] ] the indicator function @xmath27 $ ] takes the value one if the condition is true ; otherwise it takes the value zero . a binary random variable @xmath28)$ ] representing a test result is defined by @xmath29 , $ ] where the notation @xmath30 represents a sequence of variables @xmath31 .
the vector composed from @xmath28)$ ] is denoted by @xmath32 .
it is evident that there is one - to - one correspondence between @xmath33 and @xmath34 and also between @xmath35 and @xmath36 .
the index set @xmath21 represents a group corresponding to the @xmath6-th test result .
assume that we observed @xmath37 as a realization of @xmath38 .
our goal is to evaluate the log posterior probability ratio defined by @xmath39.\ ] ] the probability @xmath40 is the posterior probability on @xmath41-th object . from @xmath42
, we can obtain an estimate vector @xmath43 defined by @xmath44 ( \ell \in [ 1,n ] ) $ ] where this estimation rule can be seen as the bitwise map estimation rule .
it may be reasonable to consider the bitwise map estimation for this group testing problem because bitwise map estimation minimizes the bitwise estimation error probability . by using bayes theorem ,
the posterior probability can be rewritten as @xmath45,\ ] ] where @xmath46 is just a normalization constant and @xmath47 . as a simplified notation ,
if the domain of the variable is missing in a summation , all the possible values in the domain is taken to evaluate the sum .
as in many similar bitwise map estimation problems , naive evaluation according to ( [ exponential ] ) requires exponential time with the number of variables @xmath0 to marginalize all the variables @xmath48 ; namely computation time is @xmath2 if the maximum size of @xmath21 is bounded by a constant .
this prohibitive time complexity is the high burden to exploit the bitwise map estimation on this problem .
although it is still exponential time , the exponent of computation time can be greatly reduced if we are aware of the following simple fact . if @xmath49 , then we have @xmath50 for any @xmath51 . proof : if @xmath49 , then @xmath52 should be 0 for any @xmath51 because of the relation @xmath53 .
this means that @xmath54 is exactly one . in other words ,
the lemma states that all the objects in the group @xmath21 have the value zero only if @xmath6-th test result @xmath55 is zero .
this trivial but useful lemma can significantly reduce the problem size if the number of negative objects are small .
therefore , it might be better to redefine the reduced size problem for a given observation vector @xmath56 as follows .
let @xmath57 be the induced subgraph of @xmath58 where the vertices of @xmath59 and @xmath60 are given by @xmath61 : t_i = 0}\ { v_k^{(1 ) } \in v_1 \mid k \in \sigma(i ) \ }
\right ) , \\ v_2^ * & { \stackrel{\triangle}{= } } & \{v^{(2)}_i \in v_2 \mid t_i = 1 \}.\end{aligned}\ ] ] in other words , we can exclude the groups whose test result is zero in @xmath19 and its incident nodes in @xmath18 for evaluating the posterior probabilities .
for the following analysis , it would be convenient to rename the vertices in @xmath59 and @xmath60 as @xmath62 and @xmath63 .
the random variable corresponding to @xmath64 and @xmath65 are denoted by @xmath66)$ ] and @xmath67)$ ] , respectively .
figure [ ex1 ] illustrates an example of a pair @xmath58 and @xmath68 .
the original pooling graph is depicted in fig . [ ex1](a ) . in this case
, we have the test result @xmath69 which defines the induced subgraph @xmath68 illustrated in fig . [
ex1](b ) . and @xmath68 ] as in the cases of @xmath71
, we introduce similar notation such as @xmath72 , @xmath73 . in this problem setting , our goal can be recast as the evaluation of the log posterior probability ratio @xmath74 $ ] for given @xmath68 .
the symbol @xmath75 represents the vector that all its components are ones . in this subsection
, we will rewrite the posterior probabilities in sum - product form which is the foundation of the following discussion . as in the derivation of ( [
exponential ] ) , the posterior probabilities @xmath76 can be expressed as @xmath77 \\
\nonumber & = & \frac{1}{z ' } \sum_{x_1,\ldots , x_n } \left(\prod_{i=1}^m or(x_k|_{k \in \alpha(i ) } ) \right ) \\ &
\times & \left(\prod_{j=1}^n p_{x_j}(x_j ) \right ) \bbb i[x_\ell = b ] \end{aligned}\ ] ] for @xmath78 , @xmath79 $ ] and @xmath80 .
the symbol @xmath81 represents a normalization constant which is independent of the value of @xmath9 .
note that the two sets , @xmath82)$ ] and @xmath83)$ ] , are defined by @xmath84 \mid ( z_i^{(2 ) } , z_j^{(1 ) } ) \in e^ * \ } , \\
\beta(j ) & { \stackrel{\triangle}{= } } & \{i \in [ 1,m ] \mid ( z_i^{(2 ) } , z_j^{(1 ) } ) \in e^ * \ } , \end{aligned}\ ] ] respectively . for the following argument ,
it is useful to define the quantity @xmath85)$ ] by @xmath86,\end{aligned}\ ] ] that is called a
_ posterior value_. by using these posterior values , the log posterior probability ratio @xmath87 can be evaluated by taking the ratio between the posterior values for zero and one : @xmath88 we will further decompose @xmath89 in ( [ posterior ] ) into a finer sum - product form which will be more suitable for a normal factor graph representation to be described in the next section . as building blocks of the finer representation , we here introduce @xmath90 , and @xmath91 functions as follows .
the boolean equality function @xmath92 with @xmath25-inputs is defined by @xmath93.\ ] ] the weight function @xmath94 , \ell \in [ 1,n ] , b \in \{0,1\})$ ] is given by @xmath95 , & j = \ell .
\end{array } \right.\ ] ] by using these set of functions , @xmath89 can be represented as @xmath96 where @xmath97 is the set of new binary variables defined as @xmath98 , j \in [ 1,n ] , ( z_i^{(2)},z_j^{(1 ) } ) \in e^ * \}\ ] ] and @xmath99 are also binary variables .
the main contribution of this paper is the next theorem which gives another expression of the posterior value @xmath89 .
it will be the foundation of a novel map algorithm described later .
[ dualexpression ] the posterior value @xmath89 can be expressed as @xmath100 where @xmath79 , b \in \{0,1\}$ ] and variables @xmath101 are binary variables .
the notation @xmath102 represents the cardinality of @xmath103 .
the function @xmath104 , b \in \{0,1\})$ ] is defined by @xmath105 if @xmath106 .
if @xmath107 , then @xmath108 is defined as @xmath109,\hfill \mbox{otherwise}. \end{array } \right.\ ] ] in the expression of the posterior value ( [ finerrep ] ) , the indicator variables @xmath99 corresponding to @xmath90 nodes take all the possible binary @xmath110-tuples in the summation . on the other hand , in the expression of posterior values in theorem [ dualexpression ] , the indicator variables @xmath101 appeared in the summation correspond to @xmath111 nodes .
we therefore call this expression the dual expression .
the proof of this theorem heavily relies on a holographic transformation to the normal factor graph of the posterior value @xmath89 . in the next section
, we will discuss an appropriate holographic transformation to derive the dual expression . the normal factor graph ( nfg )
is a graphical representation of a function composed from a product of many functions .
the precise definition of the nfg can be found in @xcite @xcite but we here introduce a simplified definition enough for this paper .
the nfg of a sum - product form is an undirected graph with vertices corresponding to factor functions and edges corresponding to the variables .
the nfg of the posterior value ( [ finerrep ] ) , denoted by @xmath112 , is defined as follows . for each factor of ( [ finerrep ] ) such as @xmath113 a _ factor node _ is associated . in the following ,
we do not strictly distinguish a factor function from the corresponding factor node if there are no fear of confusion . in a similar way ,
the variables in ( [ finerrep ] ) such as @xmath114 and @xmath99 are assigned to edges .
the rule for the edge connections is simple ; if and only if a variable @xmath115 ( resp .
@xmath116 ) is an argument of a factor function @xmath117 , the edge @xmath115 ( resp .
@xmath116 ) is connected to the factor node @xmath117 . in other words , if and only if @xmath117 depends on @xmath115 ( resp .
@xmath116 ) , the factor node @xmath117 connects to the edge @xmath115 ( resp .
@xmath116 ) . according to the semantics for nfgs introduced by al - bashabsheh and mao @xcite , all the edge variables are assumed to be marginalized .
figure [ nfg1 ] illustrates the nfg for the posterior value in ( [ finerrep ] ) .
we will refer the factor nodes corresponding to the or ( resp .
eq ) function as _ or ( resp .
eq ) nodes_. in our context , an nfg corresponds to a posterior value in sum - product form .
a holographic transformation is a transformation of an nfg that preserves the marginal generating function . in the following discussion
, we will insert a pair of _ dual factor nodes _ into each edge connecting an or node and an eq node ( _ i.e. , _
@xmath115 ) .
the pair of factor nodes is carefully designed not to change the posterior value .
figure [ nfg3 ] is our blueprint that shows how we will proceed in this subsection . in fig.[nfg3 ] ( left ) , for each edge @xmath115 , a pair of dual nodes , @xmath118 and @xmath119 , is inserted
. these function nodes @xmath118 and @xmath119 are designed to satisfy the duality condition described later . due to the duality condition on @xmath118 and @xmath119 ,
the posterior value of this transformed nfg is the same as that of the original nfg . by grouping an eq node and @xmath119 nodes connected to it , a new factor node @xmath120
is created ( fig.[nfg3 ] ( right ) ) . in a similar way , combining an or node and its incident @xmath118-nodes , we obtain a new factor nodes that is called a _ skewed eq ( seq ) function_. it should be emphasized that seq function has almost the same truth table as that of eq function .
this fact is important to reduce the computational complexity to evaluate the posterior values . in the following subsections
, we will follow this blueprint and present details of dual factor nodes and new factor nodes
. these will be the basis of the proof of theorem [ dualexpression ] .
let us define @xmath121 by @xmath122 and @xmath123 by @xmath124 it is trivial to check that these two functions @xmath118 and @xmath119 satisfies the duality condition @xmath125 this condition guarantees that the posterior values of the nfg are unchanged if we inserted these function nodes into an edge corresponding to @xmath115 in fig.[nfg1 ] @xcite .
this is because a pair of function nodes is equivalent to an eq function which does not affect the consequence of the marginalization .
the next lemma tells that a sum - product form of an or function and @xmath118 functions produces an seq function .
[ orlemma ] for any @xmath126 , the following equality @xmath127 holds . proof : from the definition of @xmath118 , the right - hand side of ( [ thetaeq ] ) can be evaluated as @xmath128_{y_1,\ldots , y_r } \\
\label{ttt } & = & \left [ ( 1,0)^{\otimes r}- ( 0,-1)^{\otimes r } \right]_{y_1,\ldots , y_r}. \end{aligned}\ ] ] note that @xmath129 represents the tensor power ( kronecker power ) of a matrix @xmath130 .
the row vector @xmath131 represents the truth table of @xmath111 function as a row vector .
the notation @xmath132_{a_1,\ldots , a_r}$ ] denotes the @xmath9-th component of row ( or column ) vector @xmath133 where @xmath134 .
if @xmath25 is odd , then the right - hand side of ( [ ttt ] ) equals @xmath135_{y_1,\ldots , y_r}. $ ] in this case , the claim of the lemma holds because @xmath135_{y_1,\ldots , y_r } $ ] is the truth table of @xmath90 function .
if @xmath25 is even , the right - hand side of ( [ ttt ] ) becomes @xmath136_{y_1,\ldots , y_r } , $ ] which is equivalent to the right - hand side of ( [ thetaeq ] ) .
it can be seen that only simple tensor calculations are required to show the main claim of this lemma .
the right - hand side of ( [ thetaeq ] ) , @xmath137 , is referred to as the skewed eq function that is denoted by @xmath138 .
the next lemma plays an crucial role for grouping factor nodes around an eq node .
[ eqlemma ] the function @xmath139 can be expressed as @xmath140 for any @xmath141 , \ell \in [ 1,n ] , b \in \{0,1\}$ ] . proof : the truth table of @xmath90 function of @xmath142-inputs is given by the column vector @xmath143 from the definition of the function @xmath119 , we have the following tensor expression : @xmath144_{u , y_1,\ldots , y_r}. \end{aligned}\ ] ] we here define @xmath145 as @xmath146 from the definition of @xmath147 , if @xmath106 , we have @xmath148 otherwise ( _ i.e. , _
@xmath107 ) , the equality @xmath149 , \mbox{otherwise}. \end{array } \right.\ ] ] is obtained .
it is clear that @xmath150 holds .
we are now ready to prove theorem [ dualexpression ] .
as described in subsection [ holographic ] , the nfg illustrated in fig .
[ nfg3 ] ( left ) corresponds to the original posterior value due to the duality condition on @xmath118 and @xmath119 . by lemmas
[ orlemma ] and [ eqlemma ] , the posterior value @xmath89 can be rewritten as @xmath151 where @xmath152 is a set of variables defined by @xmath153 , j
\in [ 1,n ] , ( z_i^{(2)},z_j^{(1 ) } ) \in e^ * \}\ ] ] and @xmath154 note that the expression ( [ otherrep ] ) follows the new factor node grouping described in fig .
[ nfg3 ] ( right ) . in the non - vanishing summand of ( [ otherrep ] )
, eq function enforces that @xmath155 takes the same value for any @xmath6 . in other words , all the edges emitted from a skewed eq factor node should take the same value if the product in ( [ otherrep ] ) is not zero .
this observation leads to the claim of theorem [ dualexpression ] .
the next example presents how this dual expression works .
assume that @xmath156 and the pooling graph defined by @xmath157 ( see .
[ g1fig ] ) .
the prior probabilities of objects are given as @xmath158)$ ] .
table [ g1 ] shows the prior probabilities @xmath159 and
the indicator values @xmath160 \bbb i[or(x_3,x_3)=1]$ ] .
we here focus on the case where @xmath161 . from table
[ g1 ] , it is straightforward to evaluate the posterior values as @xmath162 .prior probabilities and indicator value [ cols="^,^,^,^,^",options="header " , ]
from engineering point of view , the primal advantage of theorem [ dualexpression ] is that it provides an efficient bitwise map estimation algorithm .
if the degrees of a node in a pooling graph is bounded by a constant , exhaustive evaluation of posterior values based on theorem [ dualexpression ] requires @xmath163-time ( a simple implementation trick can reduce this time complexity down to @xmath164 ) . if @xmath165 , this bitwise map algorithm achieves exponential speedup compared with a naive bitwise map algorithm based on ( [ posterior ] ) with time complexity @xmath166 . in a typical use of a non - adaptive group testing ,
the number of tests is much smaller than the number of objects ; _ i.e. _ , @xmath4 .
this implies that a situation satisfying @xmath165 is fairly common .
the authors would like to express their sincere appreciation to ryuhei mori for directing our interest towards holographic transformation .
this work was supported by jsps grant - in - aid for scientific research ( b ) grant number 25289114 .
d. sejdinovic and o. johnson , `` note on noisy group testing : asymptotic bounds and belief propagation reconstruction , '' _ forty - eight annual allerton conference on communication , control , and computing _ , pp .
998 - 1003 , 2010 . | in this paper , an exact bitwise map ( maximum a posteriori ) estimation algorithm for group testing problems is presented .
we assume a simplest non - adaptive group testing scenario including @xmath0-objects with binary status and @xmath1-disjunctive tests .
if a group contains a positive object , the test result for the group is assumed to be one ; otherwise , the test result becomes zero .
our inference problem is to evaluate the posterior probabilities of the objects from the observation of @xmath1-test results and from our knowledge on the prior probabilities for objects .
if the size of each group is bounded by a constant , a naive inference algorithm requires @xmath2-time for computing the posterior probabilities for objects .
our algorithm runs with @xmath3-time , which is exponentially faster than the naive inference algorithm under a common situation with @xmath4 .
the heart of the algorithm is the _ dual expression _ of the posterior values .
the derivation of the dual expression can be naturally described based on a holographic transformation to the normal factor graph ( nfg ) representing the inference problem . in order to handle or constraints in the nfg
, we introduce a novel holographic transformation that converts an or function to a function similar to an equal function . |
comets are probably the least - altered bodies in the solar system @xcite . as such
, they can provide key insights into physical and chemical processes occurring during its origin and earliest evolutionary epochs , including the origin of long - period and short period comets @xcite and the formation and composition of planets @xcite . by studying comets from different reservoirs we can probe the different environments in which they formed and
also better understand their role in initiating prebiotic chemistry on the early earth through delivery of water and organic matter by cometary impacts @xcite .
comets are primarily located in two distinct reservoirs of the solar system : the oort cloud and the kuiper belt .
the oort cloud is a source of long - period comets and ( probably ) halley - type comets @xcite .
the kuiper belt is a source of jupiter family comets @xcite . to explore the connections among the formation environment , subsequent chemical evolution , and present composition of comets
, efforts have been made to develop chemical taxonomies @xcite .
+ important cosmogonic quantities for the investigation of primordial constitutions ( such as the @xmath10 and d / h ratios ) , can be obtained by observations of comets from ground - based , space - based and airborne observatories @xcite , as well as from laboratory analyses of cometary ( interplanetary ) dust particles collected in the stratosphere @xcite .
these data provide measurements of chemical composition , isotopic ratios and molecular spin ratios , which are important to help establish the contribution of the solar system s natal molecular cloud to the compositions of primitive materials and comets .
in addition , isotopic ratios such as d / h can be compared with the value for the earth s ocean water to investigate the possible importance of cometary water delivery .
comet rendezvous space missions to individual comets present unique scientific opportunities .
@xmath15 @xmath16 spacecraft observations of several comets have allowed their nuclei to be imaged and the gas and dust properties of their comae to be studied in great detail .
these include : 1p / halley @xcite ; 81p / wild 2 @xcite ; 9p / tempel @xcite ; 103p / hartley 2 @xcite ; and the current rosetta mission to 67p / churyumov - gerasimenko @xcite .
ground - based observing campaigns have also played an important role in support of space missions @xcite and , in the case of 81p / wild 2 , laboratory analyses of the returned stardust samples have yielded unique insights into the nature of cometary dust @xcite .
comet 103p / hartley 2 ( hereafter 103p ) is a jupiter - family comet , with a short orbital period ( 6.5 years ) , and a low orbital inclination .
it was the target of the nasa epoxi mission @xcite .
the comet passed perihelion on 2010 october 28 at @xmath17 = 1.059 au and on 2010 october 21 made an exceptionally close approach to earth at @xmath18 = 0.12 au . in support of the epoxi mission , we conducted observations at radio and submillimeter wavelengths of comet 103p in the three weeks preceding the spacecraft rendezvous on ut 2010 november 4.58 . here
we report detections of hcn , @xmath0 , cs , and oh and upper limits for hnc and dcn , using the 12 m arizona radio observatory kitt peak ( 12 m ) and the 10 m submillimeter telescope ( smt ) , as well as the 15 m james clerk maxwell telescope ( jcmt ) and the 100 m robert c. byrd greenbank telescope ( gbt ) . we present the observational results and determine physical parameters such as column densities , and production rates .
finally , the @xmath10 ratio has been derived from @xmath0 , and an upper limit on the d / h ratio was obtained from dcn and hcn measurements .
comet 103p / hartley 2 was discovered on 1984 june 4 by malcolm hartley at the siding spring observatory @xcite .
103p has been frequently observed over the 30 years following its discovery , both by ground - based and space telescopes .
observations provided information about the gas production rate @xcite , and the mean radius @xcite . all the observations presented here adopted the same jpl / horizons ( jet propulsion laboratory ) ephemeris number 183 .
observations of comet 103p were made between 2010 october 13 and october 31 for monitoring emission from the oh : j = 3/2 , @xmath19=3/2 , f=@xmath20-@xmath21 ( 1667.3 mhz/18 cm ) lines at l - band ( program i d gbt/10c-059 ) .
a barycentric velocity reference frame was assumed for all observations .
the line of sight radial velocity on each day recorded by the jpl ephemeris was then subtracted from the data to set each spectral feature to a 0 km @xmath4 cometocentric velocity .
the gbt spectrometer was configured in its eight intermediate - frequency ( if ) , 12.5 mhz three - level mode , which enabled observing four 12.5 mhz frequency bands simultaneously in two polarizations through the use of offset oscillators in the if .
antenna temperatures were recorded on the @xmath22 scale @xcite with estimated 20@xmath6 uncertainties .
data were taken in the off - on position - switching mode , with the off position 60 east in azimuth with respect to the on - source position .
a single scan consisted of 2 minutes in the off - source position followed by 2 minutes in the on - source position .
automatically - updated dynamic pointing corrections were employed based on real - time temperature measurements of the structure input to a thermal model of the gbt .
zero points were adjusted at the beginning of each observing run by pointing on a nearby quasar .
the pointing accuracy of the gbt is 5 ( rms ) for blind pointing and 7 ( rms ) for offset pointing .
the gbt beam size at this frequency is @xmath11 8. observations of hcn : j = 3 - 2 ( 265.8864 ghz ) , @xmath23-@xmath0 : @xmath24 = @xmath25-@xmath26 ( 225.6978ghz ) , hcn : j = 1 - 0 ( 88.6318ghz ) , hcn : j = 2 - 1 ( 177.2612ghz ) , and cs : j = 4 - 3 ( 97.9809ghz ) , toward comet 103p were taken between 2010 october 22 and november 4 using the facilities of the arizona radio observatory ( aro ) : the 12 m telescope on kitt peak , arizona and the smt on mount graham , arizona .
the 1 mm observations were carried out at the smt with a dual - polarization alma band 6 receiver system , employing sideband - separating mixers with an image rejection of typically 1520 db .
the backends employed were a 2048 channel 1 mhz filter bank used in parallel ( 2 @xmath2 1024 ) mode , and 250 khz filter also in parallel ( 2 @xmath2 250 ) .
the temperature scale at the smt is @xmath22 ; radiation temperature is then defined as @xmath27 = @xmath22/@xmath28 , where @xmath28 is the main beam efficiency .
the 2 and 3 mm observations were conducted at the aro 12 m using dual - polarization sis mixers , operated in single - sideband mode with the image rejection @xmath29 20 db .
filter banks with 512 channels of 100 and 250 khz resolutions were used simultaneously in parallel mode for the measurements , along with an autocorrelator with 782 khz resolution .
the intensity scale of the 12 m is the chopper - wheel corrected antenna temperature , @xmath30 , including forward spillover losses , which is converted to radiation temperature by @xmath27 = @xmath30/@xmath31 , where @xmath31 is the corrected beam efficiency .
data at both facilities were taken in position - switching mode with an off position 30 west in azimuth .
the jpl ephemeris was used to determine the cometary position using the orbital elements .
the pointing accuracy is estimated at 1 rms for the smt and 5 for the aro 12 m . focus and positional accuracy were checked periodically on nearby planets or masers .
sub - millimeter spectroscopic observations of 103p ( programme
i d m10bu13 ) were made from the jcmt , located at the 4000 m level on mauna kea , hawaii , on ut 2010 october 21 , 22 , 23 , 25 and november 02 , 03 , 04 , using the harp heterodyne array .
harp is a 4 @xmath2 4 array receiver , but was used here in a single - receptor mode .
harp is remotely and quickly tunable to frequencies in the 325 - 375ghz ranges and utilizes an image sideband suppressor .
the output from the harp receiver is fed to the ` acsis ' digital autocorrelation spectrometer . for this work
, acsis was configured at its highest frequency resolutions ( 31mhz ) .
observations of hcn : j = 4 - 3 ( 354.5055ghz ) , hcn : j = 3 - 2 ( 265.8864ghz ) , @xmath32-@xmath0 : @xmath24 = @xmath33-@xmath34 ( 362.7360ghz ) , @xmath23-@xmath0 : @xmath24=@xmath35-@xmath36 ( 351.7686ghz ) , dcn : j = 5 - 4 ( 362.0465ghz ) and hnc : j = 4 - 3 ( 362.6303ghz ) were measured at the jcmt . at these frequencies , the jcmt has a gaussian beam of size @xmath11 15 ( full width at half power ) .
the harp receptors are separated by about 30 , and so the emission from the comet is anticipated to be encompassed by a single receptor the ` pointing ' receptor .
pointing of the telescope is achieved by doing spectral - line ` five - points ' on nearby line sources , and pointing accuracy is estimated at 2 " rms in each of two orthogonal coordinates ( azimuth and elevation ) .
a pointing check was performed approximately every hour .
telescope focus is similarly maintained throughout the night by measures of bright line - sources .
calibration of the ( @xmath37 ) brightness scale is achieved by making measures at standard frequencies ( eg co : j = 3 - 2 , 345ghz ) of astronomical sources used as reference calibrators .
the opacity of the sky above jcmt was measured by a water vapor meter ( wvm ) mounted so as to measure along the telescope line - of - sight .
opacity is expressed as if measured at the zenith at 225ghz .
the opacity ( in nepers ) on ut oct 21 , 22 , 23 , 25 was , on average , 0.10 , 0.05 , 0.05 , 0.05 and on nov 02 , 03 , 04 was 0.09 , 0.11 , 0.15 .
the latter conditions made observations difficult at these harp ( b - band ) frequencies , but conditions otherwise were most suitable for this program .
observed lines parameters including ut dates , observing frequencies ( @xmath38 ) , beam size ( @xmath39 ) , diameter of the projected beam size on the comet ( d ) , beam efficiency ( @xmath31 or @xmath28 ) , temperature ( @xmath40 or @xmath41 ) , fwhm line width ( @xmath42 ) , integrated intensity ( @xmath43 @xmath44 ) , heliocentric distance ( @xmath17 ) and comet distance at the times of measurements ( @xmath18 ) are listed in table [ obs_parameters ] .
the uncertainties of the line width and the integrated line intensity are determined from a gaussian fit and are reported at 1@xmath12 in the table [ obs_parameters ] .
representative spectra are shown in figures [ gbt_oh ] [ h2co_data ] from all facilities .
all the spectra are plotted in the cometocentric velocity frame .
the ut date of each observed transition is located in the figure .
all data are included as supplemental information .
figure [ gbt_oh ] displays the oh detection toward comet 103p / hartley 2 from the gbt taken on 2010 october 13 and 19 .
the gbt data were measured in a monitoring sequence every 1 , 2 , 3 or 5 days . the oh line inversion and gas production rates on these days
are reported in table [ oh_results ] .
0.274 km @xmath4 .
additional spectra from the observing campaign for oh is available in appendix [ appendix_oh ] . ]
figure [ hcn_data ] displays the j = 4 - 3 and the j = 3 - 2 transitions of hnc on 2010 october 22 toward comet 103p / hartley 2 from the jcmt and smt , respectively .
we detected four transitions of hcn with three telescopes ( smt , 12 m and jcmt ) .
gas production rates from these transition are reported in table [ qgas ] .
figure [ hcn_data ] shows two transitions observed on the same day .
furthermore , the line is mostly resolved with the jcmt and partially with the smt .
0.232 km @xmath4 and @xmath11 0.280 km @xmath4 for the jcmt and the smt , respectively .
( for interpretation of the references to color in this figure legend , the reader is referred to the web version of this article . ) .
the detections of hcn from all facilities and dates are available in appendix [ appendix_hcn ] . ]
figure [ cs_12 m ] displays the j = 2 - 1 transitions of cs on 2010 october 22 toward comet 103p / hartley 2 from the 12 m .
gas production rates from this transition are reported in table [ qgas ] .
these data were collected with a resolution of @xmath11 0.310 km @xmath4 and show a narrow line width . the line profile is not resolved here and only provides details on abundance and not molecular origin within the coma .
0.310 km @xmath4 . ]
figure [ h2co_data ] displays respectively the @xmath45 = @xmath33-@xmath34 and @xmath24 = @xmath35-@xmath36 transitions of @xmath32-@xmath0 and @xmath23-@xmath0 on 2010 november 2 toward comet 103p / hartley 2 from the jcmt .
we observed @xmath32-@xmath0 and @xmath23-@xmath0 the same day only with the jcmt .
a detailed discussion on the @xmath10 ratio can be found in section [ section_opr ] . taken on 2010 november 2 with the jcmt .
the spectral resolution are @xmath11 0.909 km @xmath4 and @xmath11 0.468 km @xmath4 for @xmath32-@xmath0 and @xmath23-@xmath0 , respectively .
the detection of @xmath0 from all facilities and dates are available in appendix [ appendix_h2co ] . ]
lllllllllllll oh & gbt&multiple lines & 2010 oct 13.04 & 1667.3 & 452.8 & 43348 & 0.95 & -0.13 @xmath5 0.01 & 1.84 @xmath5 0.13 & -0.24 @xmath5 0.02 & 1.079 & 0.132 + & & & 2010 oct 15.04 & & & 41706 & & -0.14 @xmath5 0.01 & 2.05 @xmath5 0.14 & -0.28 @xmath5 0.02 & 1.074 & 0.127 + & & & 2010 oct 16.40 & & & 40721 & & -0.10 @xmath5 0.01 & 1.74 @xmath5 0.15 & -0.18 @xmath5 0.03 & 1.071 & 0.124 + & & & 2010 oct 19.44 & & & 39736 & & -0.07 @xmath5 0.01 & 1.65 @xmath5 0.21 & -0.12 @xmath5 0.02 & 1.065 & 0.121 + & & & 2010 oct 22.31 & & & 39736 & & -0.03 @xmath5 0.01 & 2.09 @xmath5 0.28 & @xmath7 -0.07 & 1.062 & 0.121 + & & & 2010 oct 23.46 & & & 40064 & & -0.04 @xmath5 0.01 & 1.85 @xmath5 0.39 & @xmath7 -0.07 & 1.061 & 0.122 + & & & 2010 oct 24.46 & & & 40393 & & -0.03 @xmath5 0.01 & 1.08 @xmath5 0.32 & @xmath7 -0.03 & 1.060 & 0.123 + & & & 2010 oct 26.44 & & & 41706 & & 0.01 @xmath5 0.01 & 1.86 @xmath5 0.75 & @xmath7 0.03 & 1.059 & 0.127 + & & & 2010 oct 28.18 & & & 43020 & & 0.03 @xmath5 0.01 & 1.44 @xmath5 0.34 & @xmath7 0.04 & 1.059 & 0.131 + & & & 2010 oct 31.57 & & & 46304 & & 0.03 @xmath5 0.01 & 2.29 @xmath5 0.79 & @xmath7 0.07 & 1.060 & 0.141 + hcn&smt & 3 - 2 & 2010 oct 22.54 & 265886.4 & 28.4 & 2492 & 0.74 & 0.24 @xmath5 0.01 & 1.20 @xmath5 0.30 & 0.23 @xmath5 0.03 & 1.062 & 0.121 + & & & 2010 oct 29.35 & & & 2750 & & 0.19 @xmath5 0.05 & 1.06 @xmath5 0.30 & 0.21 @xmath5 0.01 & 1.059 & 0.134 + & & & 2010 nov 1.32 & & & 2955 & & 0.14
@xmath5 0.03 & 1.17 @xmath5 0.10 & 0.16 @xmath5 0.03 & 1.060 & 0.144 + & & & 2010 nov 1.68 & & & 2976 & & 0.30 @xmath5 0.04 & 1.21 @xmath5 0.10 & 0.36 @xmath5 0.04 & 1.060 & 0.144 + & & & 2010 nov 4.65 & & & 3212 & & 0.37 @xmath5 0.01 & 1.40 @xmath5 0.30 & 0.38 @xmath5 0.03 & 1.063 &
0.156 + @xmath23-@xmath0 & smt&@xmath46 & 2010 oct 29.55 & 225697.8 & 33.3 & 3264 & & 0.01 @xmath5 0.03 & 1.96 @xmath5 0.30&0.03 @xmath5 0.01 & 1.059 & 0.134 + hcn&12 m & 1 - 0&2010 oct 22.31 & 88631.8 & 70.7 & 6208 & 0.95 & 0.05 @xmath5 0.01 & 1.50 @xmath5 0.30 & 0.07 @xmath5 0.01 & 1.062 & 0.121 + & & & 2010 oct 23.33 & & & 6260 & & 0.08 @xmath5 0.02 & 0.50 @xmath5 0.20&0.04 @xmath5 0.01 & 1.061 & 0.122 + & & & 2010 oct 24.31 & & & 6311 & & 0.07 @xmath5 0.02 & 0.90 @xmath5 0.20 & 0.07 @xmath5 0.01 & 1.060 & 0.123 + & & & 2010 oct 28.3 & & & 6732 & & 0.05 @xmath5 0.03 & 1.62 @xmath5 0.30 & 0.08 @xmath5 0.05 & 1.059 & 0.131 + hcn&12 m & 2 - 1 & 2010 oct 23.54 & 177261.2 & 35.4 & 3130 & 0.65 & 0.35 @xmath5 0.02 & 1.00 @xmath5 0.40 & 0.34 @xmath5 0.07 & 1.061 & 0.122 + cs & 12 m & 2 - 1 & 2010 oct 22.53 & 97980.9 & 64.0 & 5616 & 0.88 & 0.07 @xmath5 0.02 & 0.40 @xmath5 0.20 & 0.03 @xmath5 0.01 & 1.062 & 0.121 + hcn & jcmt & 4 - 3 & 2010 oct 21.48 & 354505.5 & 14.1 & 1242 & 0.63 & 0.61 @xmath5 0.03 & 1.16 @xmath5 0.06 & 0.75 @xmath5 0.05 & 1.063 & 0.121 + & & & 2010 oct 22.43 & & & 1242 & & 0.59 @xmath5 0.06 & 1.20 @xmath5 0.20 & 0.79 @xmath5 0.13&1.062 & 0.121 + & & & 2010 oct 22.78 & & & 1252 & & 0.97 @xmath5 0.08 & 1.30 @xmath5 0.20 & 1.31 @xmath5 0.17&1.061 & 0.122 + & & & 2010 oct 23.41 & & & 1252 & & 0.57 @xmath5 0.04 & 1.20 @xmath5 0.10 & 0.73 @xmath5 0.07&1.061 & 0.122 + & & & 2010 oct 23.77 & & & 1262 & & 0.55 @xmath5 0.04 & 1.20 @xmath5 0.10 & 0.67 @xmath5 0.08&1.061 & 0.123 + & & & 2010 nov 2.44 & & & 1519 & & 0.79 @xmath5 0.05 & 1.20 @xmath5 0.10 & 1.03 @xmath5 0.10&1.061 & 0.148 + & & & 2010 nov 2.77 & & & 1529 & & 0.37 @xmath5 0.05 & 1.40 @xmath50.20 & 0.55 @xmath5 0.11&1.061 & 0.149 + & & & 2010 nov 3.62 & & & 1560 & & 0.30 @xmath5 0.04 & 1.10 @xmath5 0.20 & 0.33 @xmath5 0.07&1.062 & 0.152 + & & & 2010 nov 4.44 & & & 1601 & & 0.48 @xmath5 0.05 & 1.30 @xmath5 0.20 & 0.67 @xmath5 0.11&1.063 & 0.156 + & & 3 - 2 & 2010 oct 25.56 & 265886.4 & 18.9 & 1716&0.69 & 0.33 @xmath5 0.05 & 1.20 @xmath5 0.20 & 0.42 @xmath5 0.09&1.059 & 0.125 + @xmath32-@xmath0 & jcmt&@xmath47 & 2010 oct 21.56 & 362736.0 & 13.8 & 1214 & & 0.31 @xmath5 0.01 & 0.03 @xmath5 0.50&0.008 @xmath5 0.002 & 1.063 & 0.121 + & & & 2010 oct 23.57 & & 13.8 & 1234 & & 0.04 @xmath5 0.01 & 0.39 @xmath5 0.12 & 0.016 @xmath5 0.007&1.061 & 0.123 + & & & 2010 nov 2.69 & & 13.8 & 1494 & & 0.05 @xmath5 0.08 & 1.00 @xmath5 0.50 & 0.049 @xmath5 0.008&1.061 &
0.149 + @xmath23-@xmath0 & jcmt&@xmath48 & 2010 nov 2.58 & 351768.6 & 14.2 & 1531 & & 0.05 @xmath5 0.01 & 1.60 @xmath5 0.30 & 0.079 @xmath5 0.018 & 1.061 & 0.148 + dcn & jcmt & 5 - 4 & 2010 oct 22.58 & 362046.5 & 13.8 & 1216 & & - & -&@xmath7 0.01 & 1.062 & 0.121 + & & & 2010 oct 23.57 & & 13.8 & 1236 & & - & - & @xmath7 0.01&1.061 & 0.123 + & & & 2010 nov 2.69 & & 13.8 & 1497 & & - & - & @xmath7 0.03&1.061 & 0.149 + & & & 2010 nov 4.67 & & 13.8 & 1568 & & - & - & @xmath7 0.04&1.063 & 0.156 + hnc & jcmt & 4 - 3 & 2010 nov 2.69 & 362630.3 & 13.8 & 1495 & & - & - & @xmath7 0.03 & 1.061 & 0.149 + & & & 2010 nov 4.67 & & 13.8 & 1565 & & - & - & @xmath7 0.04&1.063 & 0.156 +
the production rate for water was indirectly measured from the observation of its photodissociation product oh with the gbt .
we used the equation of @xcite to determine the total oh production rate ( @xmath49 ) for each observation : @xmath50 where @xmath51 is the inversion of the ground state , @xmath52 ( k ) is the background temperature , @xmath53 ( s ) is the lifetime of the oh molecule , @xmath18 ( au ) is the earth - comet distance and @xmath54 ( mjy km @xmath4 ) is the integrated flux .
the dominant source of error in this formalism comes from the accuracy of the integrated flux measurement , as well as the inversion parameter @xmath51 , although the latter is negligible for these data .
given that the values of the heliocentric velocity ( @xmath55 ) of the comet ranged from -5 to + 2 ( table [ oh_results ] ) , a polynomial function was fit to the data presented in table 5 of @xcite between these values in order to find the inversion at the comet s heliocentric velocity on the date and ut time of the observations .
this inversion value is given in table [ oh_results].the oh photodissociation lifetime ( @xmath53 ) is 1.2@xmath2 @xmath56 @xmath57 s ( taken from @xcite ) , and is used in the final production rate calculation , and we assume @xmath52 @xmath58 2.7 k. the measured intensity and line width of each transition is given in table [ obs_parameters ] . the @xmath49 value is then corrected by a factor @xmath59 , which is defined as the fraction of molecules in the coma that are contained within the gbt beam .
we computed @xmath59 using a monte - carlo haser model @xcite for daughters molecules , assuming @xmath60 = 8.0 @xmath2 @xmath61 @xmath57 ( s ) @xcite , @xmath62 = 0.95 km @xmath63 and @xmath64 = 0.80 km @xmath63 @xcite .
the final production rates for oh on each date are given in table [ oh_results ] .
@xmath49 shows large diversity with time
. the average measured production rate of oh over this monitoring campaign is @xmath49 = ( 7.03 @xmath5 0.44 ) @xmath2 @xmath65 @xmath4 .
the conversion to @xmath66 is made by dividing the @xmath49 by the branching ratio of water dissociation to oh . the photodissociation of the @xmath67 molecule into h and oh is the most important process accounting for 85.5@xmath6 of all water molecules that are dissociated @xcite .
the mean water production rate was therefore @xmath66 = ( 8.22 @xmath5 0.51 ) @xmath2 @xmath65 @xmath4 .
[ cols= " < , < , < , < , < , < , < , < " , ] relevant deuterium fractionation ratios have been measured for both comets and the interstellar medium .
@xcite , @xcite , @xcite , @xcite and @xcite have presented upper limits for the production rates of several cometary molecules .
d / h isotope ratios have been measured for two cometary molecules in 103p : water and hydrogen cyanide . compared to models of interstellar chemistry ,
both the measured hdo/@xmath67 and dcn / hcn ratios are compatible with ion - molecule chemistry in gas at about 30 - 35 k. with this work we measured an upper limit for the dcn / hcn ratio of @xmath70.01 .
our value is consistent with previous measurements [ e.g. , d / h = 0.002 in comet hale - bopp ; @xcite , and 1.6 @xmath2 @xmath68 in comet 103p ; @xcite ] , although it does not place any new constraints . by comparison , a range of ( 0.4 - 7.0 ) @xmath2 @xmath69 for the dcn / hcn ratio was determined in the interstellar medium @xcite .
molecules that contain multiple h atoms can be spectroscopically distinguished based on the orientation of the spins of their h nuclei , either aligned in parallel or anti - parallel states ( e.g. @xmath70 , @xmath0 , @xmath67 ) , giving rise to @xmath8 and @xmath9 forms .
quantum - mechanically , transitions between the different forms are strictly forbidden giving rise to two distinct spectra and @xmath10 ratios ( oprs ) , which define the spin temperature ( @xmath13 ) of that species .
this means that the relative abundances of nuclear spin isomers are expected to remain invariant over time , and the opr has therefore been long thought to reveal information about the temperature of formation of cometary ices .
however , some processes are hypothesized to change the opr without breaking molecular bonds ( e.g. * ? ? ?
* ) , and phase transitions are effective in bringing the molecules into an equilibrated opr ( e.g. * ? ? ?
* ) . in the study of water molecules trapped in low temperature ( @xmath71 k ) solid ar matrices
, @xcite found that upon desorption at t @xmath14 260 k , fast @xmath72 conversion occurred .
in addition , theoretical investigations @xcite showed that photodesorption of water ice at 10 k would typically lead to bond breaking ( with the subsequent loss of opr information ) .
of the processes investigated by @xcite , only the ` kick - out ' mechanism would preserve the opr during desorption at low temperatures . on the other hand , it is not straightforward to extrapolate the high - temperature desorption and ar matrix - ice studies by @xcite to the release of volatiles from cometary ices , and which process dominates the desorption of ices at low temperatures .
the cosmogonic significance of opr and spin temperature in comets is therefore not clear , and the lack of opr measurements for molecules with smaller @xmath72 energy deficiencies ( e.g. , @xmath0 and @xmath73 ) has further limited our understanding of this indicator in comets .
the opr of h@xmath74co for hartley 2 was obtained from consecutive observations of the @xmath8-@xmath0 ( @xmath47 ) and @xmath9-@xmath0 ( @xmath48 ) lines on november 2 from data obtained at the jcmt ( see table [ obs_parameters ] ) .
the periodicity of the comet presents some uncertainty in abundances , though these two observations were in close enough proximity ( @xmath7 3 hrs ) .
therefore , the variation in rotation temperature is likely not a major factor .
other observational uncertainties are factored out in the ratio .
for these data we obtain an opr @xmath11 2.12 @xmath5 0.59 ( 1@xmath12 ) , or @xmath13
@xmath11 13.5@xmath75 k ( 1@xmath12 ) , although at only 1.5@xmath12 our opr measurement can not distinguish from equilibrium ( @xmath13 @xmath14 9 k at 1.5@xmath12 ) .
similarly , @xcite have measured @xmath13 @xmath14 18 k for @xmath73 in comet c/2001 a2 ( linear ) , which is consistent with that in hale - bopp @xmath13 @xmath14 15 k @xcite .
our results are also consistent ( within 1@xmath12 ) with those obtained for water by @xcite of opr @xmath11 2.79 @xmath5 0.13 ( mean of five measurements ) . the complexity of obtaining simultaneous observations of @xmath8 and @xmath9-@xmath0 , and the difficulty in obtaining a good signal - to - noise ratio for these data enforces the future need for sensitive , broad - band telescope receivers where these can be readily measured towards any comet .
such measurements will provide the necessary context to better understand the cosmogonic significance ( or lack of ) of our opr measurement of @xmath11 2.12 @xmath5 0.59 ( 1@xmath12 ) .
knowledge of the volatile chemistry of jupiter - family comets ( kuiper belt ) is very limited from both infrared and radio observations because jupiter - family comets have typically lower activity .
@xcite show the compositional diversity among jupiter - family comets from radio observations . before the observation of 103p , about a dozen jupiter - family comets
have been observed by radio techniques . by comparing the relative abundance of hcn among jupiter - family comets ,
@xcite reported that 2p / encke , 9p / tempel 1 and 22p / kopff are organics - normal , like 103p , while 19p / borrelly , 21p / giacobini - zinner and 73p / schwassmann - wachmann 3 are organics - depleted .
this suggests distinct processing histories for organics - depleted , organics - normal and organics - enriched comets . @xcite and
@xcite compared the abundances ratios among jupiter - family sampled from infrared observations .
the limited sample of few jupiter - family comets shows again that diversity is notable .
for example , @xcite concluded than 73p / schassmann - wachmann 3 and 103p are extremely different in their chemical composition .
correlations between short - term temporal variations in production rates of primary volatiles with nucleus rotation have been suggested only for two oort cloud comets .
@xcite reported a periodic variation of 40@xmath6 in the water production rate in c/2001 q4 ( neat ) , with a period of 19.58 @xmath5 0.1 hrs .
@xcite observed a variation in production rates for @xmath67 , co , @xmath0 , and @xmath73 in comet c/2002 t7 ( linear ) with a period of 2.32 days .
comet 103p is the third comet for which periodic variation in production of primary volatiles has been demonstrated , and the first one for which unambiguous association with nucleus rotation can be made through imaging @xcite . to date , oprs have been measured in a few species , namely @xmath76 , @xmath67 , and @xmath77 , for several comets by ground - based observations ( see * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
@xcite reported from @xmath67 , @xmath13 @xmath14 30 k , @xmath13 @xmath58 @xmath78 k , @xmath13 @xmath58 @xmath79 k respectively in comets c/1999 h1 ( lee ) , c/1999 s4 ( linear ) , and c/2001 a2 ( linear ) . @xcite and @xcite derived an opr of ammonia equal to 1.17 @xmath5 0.04 and 1.12 @xmath5 0.03 ( or @xmath13 @xmath58 @xmath80 k ) respectively in comet c/1999 s4 ( linear ) and c/2001 a2 ( linear ) .
some of the opr radios imply formation temperatures for cometary ice species of 25 @xmath81 35 k , similar to the formation temperature implied by the d / h ratios in @xmath67 and hcn , but a growing sample of comets also show equilibrated opr
( @xmath13 @xmath14 35 k , e.g. * ? ? ?
* ; * ? ? ?
for example , @xcite reported a water opr @xmath58 3.2 @xmath5 0.2 for c/1986 p1 ( wilson ) ( @xmath13 @xmath14 50 k ) .
we conducted observations of comet 103p / hartley 2 at both perihelion and at the time of the epoxi flyby .
we report detections of hcn , @xmath0 , cs , and oh and upper limits of hnc and dcn , using the aro kitt peak 12 m and smt , jcmt and the gbt towards comet 103p . we derived column densities , production rates , and relative abundances toward comet 103p .
we concluded that 103p is normal in hcn and depleted in @xmath0 which is in good agreement with other studies .
we obtained an upper limit the @xmath82 ratio of @xmath7 0.01 and an @xmath10 ratio of 2.2 @xmath5 0.59 ( 1@xmath12 ) has been derived from @xmath0 .
_ acknowledgements _ : + the national radio astronomy observatory is a facility of the national science foundation operated under cooperative agreement by associated universities , inc . the kitt peak 12 m telescope and the submillimeter telescope
are currently operated by the arizona observatory ( aro ) , steward observatory , university of arizona , with partial funding from the research corporation .
the james clerk maxwell telescope is operated by the joint astronomy centre on behalf of the science and technology facilities council of the united kingdom , the netherlands organisation for scientific research , and the national research council of canada .
this work was supported by the goddard center for astrobiology , by nasa s planetary astronomy and planetary atmospheres programs .
yjk acknowledges support from nsc grants 100 - 2119-m-003 - 001-my3 and 102 - 2119- m-003 - 008- for this work . , m. f. , millis , r. l. , schleicher , d. g. , et al . 1995 , icarus , 118 , 223 ahearn , m. f. , belton , m. j. s. , delamere , w. a. , et al .
2005 , science , 310 , 258 ahearn , m. f. , belton , m. j. s. , delamere , w. a. , et al .
2011 , science , 332 , 1396 , h. & von dishoeck , e. f. 2008 , a&a , 491 , 907 , jr .
, w. m. 2010 , in phd thesis , the catholic university of america , j. e. , lahuis , f. , van dishoeck , e. f. , et al .
2013 , a&a , 551 , a118 , n. , bockele - morvan , d. , crovisier , j. , et al .
2007 , p&ss , 55 , 1058 , n. , bockele - morvan , d. , colom , p. , et al .
2009 , a&a , 501 , 359 , n. , bockelee - morvan , d. , crovisier , j. , et al .
2011 , epsc - dps joint meeting 2011 , 938 , d. , crovisier , j. , mumma , et al .
2004 , in comets ii , ed m. c. festou , h. u. keller , and h. a. weaver ( tucson , az : university of arizona press ) , 391 , d. & de bergh , c. 2005 , p&ss , 53 , 1203 , j. , bockele - morvan , d. , biver , n. , et al .
2014 , icarus , 228 , 197 , b. p. , mumma , m. j. , gibb , e. l.,et al .
2009 , apj , 699 , 1563 , b. p. , villanueva , g. l. , paganini , l. , et al .
2013 , icarus , 222 , 740 , d. , tsou , p. , alon , j. , et al .
2006 , science , 314 , 1711 , c. f. , thomas , p. j. , brookshaw , l. , et al .
1990 , science , 249 , 366 , s. j. , sandford , s. a. , nakamura - messenger , k. , et al .
2010 , meteoritics and planetary science , 45 , 701 , m. r. , bertaux , j .- l . , qumerais , e. , et al .
2011 , apjl , 734 , l6 , l. , epifani , e. , brucato , j. r. , et al . 1999 ,
a&a , 343 , 87 , j. , leech , k. , bockelee - morvan , d. , et al .
1997 , science , 275 , 1904 , j. , encrenaz , t. , lellouch , e. , et al . 1999 , the universe as seen by iso , ed .
p. cox & m. f. kessler ( esa - sp ) , 427 , 161 , j. , bockele - morvan , d. , colom , p. , et al . 2004 , a&a , 418 , 1141 , j. , biver , n. , bockele - morvan , d. , et al 2009 , p&ss , 57 , 1162 , j. , colom , p. , biver , n. , et al . 2013 , icarus , 222 , 679 , n. , bonev , b. p. , disanti , et al . 2005 , apj , 621 , 537 , n. , vervack , jr . , r. j. , lisse , c. m. , et al . 2011 , apjl , 734 , l8 , n. , vervack , r. j. , weaver , h. a. , et al . 2013 , icarus , 222 , 707 , j. e. & irvine , w. m. 1999 , apj , 518 . 733 , m. , levison , h. , and dones , l. 2004 , in comets ii , ed m. c. festou , h. u. keller , and h. a. weaver ( tucson , az : university of arizona press ) , 193 , s. e. , willacy , k. , bodenheimer , p. , et al . 2009 , icarus , 200 , 672 , m. , jewitt , d. , guilbert - lepoutre , a. , et al . 2012 , apj , 756 , 80 , p. d. , cochran , a. l. , and combi , m. r. 2005 , in comets ii , ed m. c. festou , h. u. keller , and h. a. weaver ( tucson , az : university of arizona press ) , 425 , m. c. , keller , h. u. , and weaver , h. a. 2004 , in comets ii , ed m. c. festou , h. u. keller , and h. a. weaver ( tucson , az : university of arizona press ) , 3 , u. 2009 , icarus , 201,311 , e. l. , bonev , b. p. , villanueva , g. , et al . 2012 , apj , 750 , 102 , t. & watanabe , n. 2013 , chrv , 113 , 8783 , j. k. , nolan , m. c. , howell , e. s. , et al .
2010 , iau circ . , 9179 , 1 , j. k. , nolan , m. c. , howell , e. s. , et al .
2011 , apjl , 734 , l2 , w. m. , scherb , f. , mierkiewicz , e. , et al .
2002 , apj , 578 , 996 , m. 1986 , iau circ . 4197 , 1 , p. , lis , d. c. , bockele - morvan , d. , et al .
2011 , nature , 478 , 218 , l. 1957 , in bulletin de la class des sciences de lacadmie royale de belgique , 43 , 740 , d. c. , videen , g. , zubko , e. , et al .
2014 , apjl , 780 , l32 , w. f. , keady , j. j. , and lyon , s. p .. 1992 , ap&ss , 1995 , 289 , j. k. , schier , f. l. , and van dishoeck , e. f. 2004 , a&a , 416 , 603 , h. , watanabe , j .-
i . , ando , h. et al .
2001 , science , 294 , 1089 , h. & watanabe , j. 2002 , 34th cospar scientific assembly , 34 , h. , watanabe , j .-
i . , furusho , r. , et al .
2005 , apjl , 623 , l49 , h. & kobayashi , h. 2009 , apjl , 693 , 388 , h. , kobayashi , h. , dello russo , n. , et al .
2013 , icarus , 222 , 723 , m. m. , walsh , k. j. , ahearn , m. f. , et al .
2007 , icarus , 187 , 199 , m. m. & schleicher , d. g. 2012 , icarus , 222 , 691 , h. f. & duncan , m. j. 1997 , icarus , 127 , 13 , h. f. , dones , l. , and duncan , m. j. 2001 , aj , 121 , 2253 limbach , h .- h . , buntkowsky , g. , matthes , j. , et al .
2006 , physical chemistry chemical physics , 7 , 551 , d. c. , bockelee - morvan , d. , biver , n. , et al .
2010 , iau circ .
9185 , c. m. , fernandez , y. r. , reach , w. t. , et al .
2009 , publications of the astronomical society of the pacific , 121 , 968 , k. j. , alon , j. , bradley , j. et al .
2006 , science , 314 , 1724 , k. j. , ageorges , n. , ahearn , m. f. , et al .
2005 , science , 310 , 265 , k. j. , ahearn , m. f. , adams , j. a. , et al . 2011 , apjl , 734 , l1 , r. , owen , t. c. , jewitt , d. c. , et al .
1998 , science , 279 , 1707 , s. n. , savage , c. , ziurys , l. m. , et al .
2004 , apj , 615 , 1054 , s. n. , remijan , a. j. , womack , m. , et al .
2006 , apj , 649 , 1169 , m. j. , weissman , p. r. , and stern , s. a. 1993 , in protostars and planets iii , 1177 , m. j. , bonev , b. p. , villanueva , g. l. , et al .
2011 , apjl , 734 , l7 , m. j. & charnley , s. b. 2011 , annual review of astronomy and astrophysics , 49 , 471 , jr .
, r. l. , neugebauer , m. , and rahe , j. 1991 , iau colloq .
116 : comets in the post - halley era , 167 , c. , crovisier , j. , bockele - morvan , d. , et al .
2007 , sf2a-2007 : proceedings of the annual meeting of the french society of astronomy and astrophysics , 421 , h. , fuller , g. a. , millar , t. j. , et al .
2002 , p&ss , 50 , 1173 , w. t. , kelley , m. s. , and vaubaillon , j. 2013 , icarus , 226 , 777 , d. g. , ahearn , m. f. 1988 , apj , 331 , 1058 , r. 2012 , p&ss , 66 , 1 , r. , gish , m. , and vilesov , a. f. 2011 , jphch , 115 , 9682 , n. a. & franchi , i. a. 2013 , geochimica et cosmochimica acta , 105 , 73 , l. e. , schloerb , f. p. , and claussen , m. j. 1990 , apj , 364 , 672 , b. l. & haas , r. w. 1976 , astrophysical , suppl .
30 , 247 , g. l. , mumma , m. j. , bonev , b. p. , et al .
2012a , jqsrt , 113,202 , g. l. , disanti , m. a. mumma , m. j. , et al . 2012b , apjl , 747,37 , g. l. , magee - sauer , k. , mumma , m. j. , et al .
2013 , jqsrt 129,158
0.274 km @xmath4 .
spectra are plotted in a cometocentric velocity frame .
the red - dashed line denotes the j=3/2 , @xmath19=3/2 , f=@xmath20-@xmath83 transition at the comet velocity .
other features present in spectra are attributed to interstellar features at the time of observations . ]
[ appendix_hcn ] 0.280 km @xmath4 .
spectra are plotted in a cometocentric velocity frame . ]
0.340 km @xmath4 .
spectra are plotted in a cometocentric velocity frame .
relative intensities of the hyperfine lines are plotted below on 2010 oct .
@xmath40 was multiplied by a factor 3 on oct .
28.30 . ]
0.170 km @xmath4 .
spectra is plotted in a cometocentric velocity frame . ]
0.232 km @xmath4 and @xmath11 0.310 km @xmath4 for the 4 - 3 and 3 - 2 transition , respectively .
spectra is plotted in a cometocentric velocity frame .
the observed asymmetry in the line profile observed can be attributed to localized outgassing as mentioned by @xcite ] .
[ hcn_jcmt ]
taken on 29 october 2010 with the smt .
the spectral resolution is @xmath11 0.332 km @xmath4 .
spectra are plotted in a cometocentric velocity frame . ] taken for each day of observations with the jcmt . the spectral resolution
are @xmath11 0.984 km @xmath4 , @xmath11 0.252 km @xmath4 , @xmath11 0.909 km @xmath4 and @xmath11 0.468 km @xmath4 on 21 october , 23 october , 2.69 november and 2.58 november , respectively .
spectra are plotted in a cometocentric velocity frame . ]
0.232 km @xmath4 and @xmath11 0.025 km @xmath4 for hcn and dcn , respectively .
spectra are plotted in a cometocentric velocity frame .
the d / h ratio obtained here is @xmath7 0.01 ] | the jupiter - family comet 103p / hartley 2 ( 103p ) was the target of the nasa epoxi mission .
in support of this mission , we conducted observations from radio to submillimeter wavelengths of comet 103p in the three weeks preceding the spacecraft rendezvous on ut 2010 november 4.58 .
this time period included the passage at perihelion and the closest approach of the comet to the earth .
here we report detections of hcn , @xmath0 , cs , and oh and upper limits for hnc and dcn towards 103p , using the arizona radio observatory kitt peak 12 m telescope ( aro 12 m ) and submillimeter telescope ( smt ) , the james clerk maxwell telescope ( jcmt ) and the greenbank telescope ( gbt ) . the water production rate , @xmath1 = ( 0.67 - 1.07 ) @xmath2 @xmath3 @xmath4 , was determined from the gbt oh data . from the average abundance ratios of hcn and @xmath0 relative to water ( 0.13 @xmath5 0.03 @xmath6 and 0.14 @xmath5 0.03 @xmath6 , respectively ) , we conclude that @xmath0 is depleted and hcn is normal with respect to typically - observed cometary mixing ratios .
however , the abundance ratio of hcn with water shows a large diversity with time . using the jcmt data we measured an upper limit for the dcn / hcn ratio @xmath70.01 .
consecutive observations of @xmath8-@xmath0 and @xmath9-@xmath0 on november 2 ( from data obtained at the jcmt ) , allowed us to derive an @xmath10 ratio ( opr ) @xmath11 2.12 @xmath5 0.59 ( 1@xmath12 ) , corresponding to @xmath13 @xmath14 8 k ( 2@xmath12 ) . |
l dwarfs are a class of objects with spectral types cooler than the latest m dwarfs @xcite .
their population consists of both substellar ( i.e. brown dwarfs ) and stellar mass objects , with the brown dwarfs being identified by the presence of li absorption at 6708 in their spectra .
kelu-1 was one of the first such brown dwarfs discovered .
it was found at a trigonometric distance of 18.7 pc @xcite as part of a proper motion survey @xcite . since kelu-1 s discovery , more than 400 l dwarfs have been classified in large surveys such as the deep near - infrared survey ( denis ) , the 2-micron all - sky survey ( 2mass ) , and the sloan digital sky survey ( sdss ) .
a spectral class sequence was established @xcite , and kelu-1 was subsequently assigned a spectral type of l2 in the optical @xcite and l3 in the near - ir @xcite .
these other discoveries provided a baseline against which to compare kelu-1 s other characteristics .
for example , it was quickly realized that kelu-1 was much brighter than other l2 dwarfs , causing @xcite to first suggest kelu-1 a binary .
this super - luminosity has been seen in other studies @xcite where the main source of this over - brightness was attributed to either an unresolved binary or a young age .
@xcite observed kelu-1 on 1998 august 14 with the nicmos camera on hst and no companions were found for separations greater than 300 mas and magnitude differences @xmath76.7 mag ( f165 m filter ) .
@xcite observed kelu-1 in a non - ao , seeing - limited program on keck i using the nirc instrument on 1998 february 14 and 1999 february 9 .
because kelu-1 had a fairly circular point spread function during these observations ( psf [email protected] ) , it was used as a psf star for the other companion search targets of their project . while these observations did not discount kelu-1 as being a very close , unresolved binary
, they did warrant a look at the young age hypothesis .
based on the strength of the li i absorption feature , @xcite constrained the age of kelu - i to be 0.3 - 1 gyr . with this age range and kelu-1 s brightness , it should have an effective temperature of 21002350 k , about 400 k higher than objects with similar spectral types @xcite . for it to have an effective temperature consistent with other l3 dwarfs ,
@xcite indicate kelu-1 must have an age of @xmath010 myr , more than an order of magnitude lower than the li -established lower age limit .
these discrepancies mean that a young age is unlikely to be the cause of the over - luminosity . aside from the luminosity , there is other evidence that supports the binarity of kelu-1 .
it is reported to be photometrically and spectroscopically variable with a period of 1.8 hrs @xcite .
if this period corresponds to an orbital period , then the system would likely be circularized and the orbital radius of the system would be less than a solar radius @xcite .
in addition , @xcite claim that the @xmath0120 km s@xmath9 radial velocity produced by a very close binary such as this could account for the large line broadening ( 60@xmath15 km s@xmath9 ) measured by @xcite .
we are conducting an adaptive optics ( ao ) survey of l and t dwarfs at keck observatory in an effort to determine binarity in these objects ( an article describing the full details of this project is in preparation ) . because of the intrinsic faintness of l and t dwarfs ( magnitudes for kelu-1 are @xmath10=22.1 and @xmath11=19.2 ; * ? ? ? * ) , it is not possible to use them as natural guide stars ( ngs ) where the magnitude limit is about @xmath10=14 .
instead , we utilize the newly commissioned laser guide star adaptive optics system ( lgs ao ) where the wavefront corrections are performed on a layer of sodium atoms 80 km high in the earth s mesosphere that are excited by a 10 - 14 watt laser tuned to 598 nm .
the projected laser and returned light propagate along the same path , requiring the use of a natural star ( or other point - like astronomical source ) for tip - tilt measurements .
the tip - tilt reference can be as faint as @xmath11=18.5 , but must be within 60 of the science target .
thus , the area of sky reachable with ao is greatly increased by the less stringent requirements for a reference star provided by the used of the lgs , and many l and t dwarfs binaries can be observed with high resolution imaging from the ground .
we included kelu-1 in our survey in an effort to resolve the controversy over its binarity and over - luminosity and to search for any low - mass companions . in this article
we discuss the detection of a companion to kelu-1 , first reported in @xcite , and show the first evidence of orbital motion of the companion .
observations of kelu-1 were obtained with nirc2 ( k. matthews et al . in preparation ) behind the lgs ao system on 2005 march 4 and 2005 april 30 during shared - risk time . a suitable tip - tilt reference star ( id=0643 - 0289866 , @xmath11=14.3 , separation=27 ) was identified in the usno - b 1.0 catalog @xcite .
skies were photometric on both nights and the ao system provided very good corrections on our targets .
kelu-1 was observed with the @xmath2 filter during both epochs , with additional exposures in the @xmath3 filter during follow - up observations in april .
these filters were chosen to maximize the strehl ratio in our images .
the data were reduced using custom and public idl scripts . for each night , two sets of reduced images were produced from the data : a photometric set and an astrometric set . the only difference in these sets is that an image distortion correction was performed on the astrometric data set .
the distortion in the narrow camera is negligible within a region a few hundred pixels on a side around the center of the chip and gradually increases to as much as 4 - 5 pixels near the edge .
while the correction fixes the distortion with residuals of 0.81 and 0.62 pixels in the x and y directions , respectively , it does not preserve flux information .
consequently , its use in images for photometric analysis is not suitable .
the plate scale for both nights was determined using astrometric images of the core of the globular cluster m5 .
this field was observed with wfpc2 on hst , with the stars in common with our observations present in the planetary camera .
we imaged this field with a 4 ( in april ) or 5 ( in march ) position dither pattern in the narrow camera ( @xmath810@xmath12 field - of - view ) .
from the wcs information in the hst image headers , we determined the sky positions of several stars present in our nirc2 images .
these positions were then used to calculate the plate scale and rotation angle for each position in the dither .
the average of these values is the resultant scale and rotation for that night .
the errors in these quantities are equal to the standard deviation of the measurements .
table [ scale ] presents the plate scale for these nights . in our march 4 images of the kelu-1 field
, we saw what appeared to be a binary object with a separation @xmath0300 mas and a differential @xmath2 magnitude of @xmath00.4 . in order to obtain proper motion confirmation of the companion ,
we re - observed this system on april 30 . on both epochs , we determined the separation and position angle of the candidate companion by using an average of measurements on the individual reduced astrometric images .
the errors are the standard deviations of these measurements .
table [ astrom ] shows these results .
the moderately large proper motion of kelu-1 made it possible to confidently confirm physical companionship based on observations separated by only two months .
the measured positions of the companion were about 15@xmath13 away from its predicted background position , clearly demonstrating that these two objects have a common proper motion and are physically associated .
our binary conclusion firmly supports the recent announcement of kelu-1 s binarity from keck lgs ao observations obtained in may 2005 @xcite . for both nights ,
all of the photometric images were offset such that northeast object was in the same location in the overlap region .
this image stack was then median averaged to obtain the final image ( figure [ nirc2 ] ) , on which the photometry was performed .
the components were well separated and no psf subtraction routines were needed .
no photometric standards were observed on either night , so only differential photometry is shown in table [ photom ] .
the northeast object is the brighter component in both the @xmath3 and @xmath2 .
it will henceforth be called kelu-1a and the fainter southwest component kelu-1b .
our measured astrometry and differential photometry are consistent with those of @xcite .
we re - examined e. martn s 1998 hst nicmos images using the psf modeling technique described in @xcite . in short ,
this technique involves the sub - pixel shifting of two model psfs and iteratively changing the relative brightness of the psfs to obtain the minimum residual for each shifted pair .
the lowest minimum residual for all shifted pairs represents the best fit for the observed psf .
our analysis discovered that the observed psfs in all three filters ( f110 m , f145 m , and f165 m ) were better fit by double psf solutions than they were by single psf solutions .
the derived astrometry of the components is presented in table [ astrom ] . because the objects are unresolved , a 180@xmath5 ambiguity exists in the value of the position angle .
it is also possible to estimate the magnitudes of the two components ( listed in table [ proptable ] ) using this technique . because the two components are unresolved , the determination of the individual magnitudes is subject to systematic uncertainties in the fitting process .
therefore , it is difficult to quantify the uncertainty in these magnitudes .
consequently , although these calculations may not determine the actual difference in magnitudes between the two components , they do indicate the reliability of a binary solution .
furthermore , the computed magnitudes are also correlated , so if one object is actually fainter than computed , the other component would be brighter , thus conserving the total flux in the system .
the spectral types of the components can be estimated using the photometry and absolute magnitude - spectral type relation in @xcite and the differential photometry presented here . for this exercise
we assume that the @xmath14@xmath3 and @xmath14@xmath2 measurements in this study are equivalent to @xmath14@xmath3 and @xmath14@xmath15 in the cit photometric _ system _ as employed by @xcite while we know that this is not the case , our interest is only in _ approximately _ determining the spectral types of the components ; the _ only _ way to determine spectral types for these or any other objects is to obtain properly calibrated spectra . based on the multi - system , near - ir photometry study of @xcite , we conservatively impose an additional photometric error of 0.15 mag for using this assumption .
this error is added in quadrature to our measured photometric errors .
the @xmath16 and @xmath17 apparent magnitudes of kelu-1ab are [email protected] and [email protected] , respectively @xcite . coupled with our measured @xmath14@[email protected] and @xmath14@[email protected] ,
the apparent magnitudes of the resolved components become @xmath3[a]=12.99 , @xmath15[a]=12.37 , @xmath3[b]=13.49 , and @xmath15[b]=12.76 ( table [ proptable ] ) . converting these magnitudes from apparent to absolute and applying equations ( 2 ) and ( 3 ) from @xcite yield a spectral type of @xmath0l2@xmath11 for kelu-1a and @xmath0l3.5@xmath11 for kelu-1b .
these estimates are consistent with the optical and near - ir spectral types of the composite system and with the spectral types derived by @xcite . with this photometry and spectral type estimates
, we can place these objects on magnitude - spectral type diagram .
figure [ magvsst ] presents a modified version of figure 4 from @xcite . in agreement with @xcite
, it is quite clear that the over - luminosity of kelu-1ab was caused by the unresolved binary and not due to an unusually young effective temperature - based age @xcite .
finally , we can estimate the effective temperature from the photometry presented here .
@xcite calculate a @xmath15-band bolometric correction ( bc@xmath18 ) for kelu-1ab of [email protected] . while this calculation is independent of the spectral type of kelu-1ab ,
the spectral type used for kelu-1ab and all of the other objects in that study were based on near - ir spectra .
the spectral types for kelu-1a and b presented here are based on a relation derived from optical spectral types .
the near - ir spectral type of kelu-1ab is l3@xmath11 @xcite , whereas the optical spectral type is [email protected] @xcite . as discussed in @xcite
, it is neither uncommon nor unexpected for l and t dwarfs to be classified differently in the optical and near - ir .
consequently , it is unclear what value of bc@xmath18 should be used from @xcite .
fortunately , for early l dwarfs such as kelu-1 a and b , the range of bc@xmath18 is quite small .
the average value of bc@xmath18 for objects with near - ir spectral types l0l5 , inclusive , is 3.32 with a standard deviation of 0.07 .
since this average value is consistent with the value of bc@xmath18 computed for the unresolved kelu-1ab , we choose the unresolved value as the bolometric correction for the individual components .
we note , however , that the uncertainty in the effective temperature determination is largely dominated by the unknown age of kelu-1 and , given the constraints on the bolometric corrections of early l dwarfs , the exact value of bc@xmath18 is unimportant . based on our photometry , the absolute @xmath15 magnitudes of kelu-1a and b are 11.02 and 11.41 , respectively . assuming a @xmath15-band bolometric correction of 3.31 for each component yields bolometric magnitudes of 14.33 and 14.72 and luminosities ( log@xmath19 @xmath20 ) of -3.83 and -3.99 .
finally , using an age range of 0.3 - 1 gyr @xcite and the @xcite dusty atmosphere models , we determine an effective temperature range for kelu-1a of 19002100 k and for kelu-1b 17001900 k ( table [ proptable ] ) .
@xcite computed effective temperatures for kelu-1a and b ( 2020 k and 1840 k ) by scaling the effective temperature of kelu-1ab @xcite by their luminosity calculations . despite the differences between our method and theirs , our results are completely consistent .
it is not known if the li i absorption at 6708 detected by @xcite and others is from one or both components .
@xcite suggest that the presence of any lithium in the unresolved spectrum indicates both components are substellar and , at the very least , kelu-1b bears lithium in its atmosphere .
consequently , the masses of both components must be @xmath210.065 m@xmath22 @xcite . to estimate component masses
, we use the photometry derived here along with the age estimate of @xcite to place the system components on the dusty atmosphere model evolutionary tracks @xcite . from the combined mass estimates presented in figure [ dusty ]
, we estimate the mass of kelu-1a to be [email protected] m@xmath22 and that of kelu-1b to be [email protected] m@xmath22 , in good agreement with the mass estimates derived by @xcite . even though the conservative errors on these masses do include stellar masses , as mentioned above the true masses are likely to not be greater than about 0.065 m@xmath22 . of course , by following the orbital motion of the companion over time , it will be possible to derive dynamical masses , which will provide much needed constraints on the models
. the computation of these dynamical masses will be some time in coming .
table [ astrom ] presents all of the known astrometry of the kelu-1ab system as of 2006 january .
it is quite clear from this astrometry that kelu-1b has exhibited orbital motion from the time it was first resolved ( our 2005 march 4 observation ) to the time of its last observation ( a public archival hst observation on 2005 july 31 obtained by w. brandner ) .
in fact , the displayed motion suggests that the companion is still moving toward apoapsis . during this nearly 5 month period
the rate of change in the separation has not changed significantly and there is insufficient phase coverage to determine a reliable orbit .
nonetheless , we can make some crude estimations of the orbit based on the available data .
given so few points in the orbit , it is impossible to determine how much of the nearly linear motion is caused by a high inclination , very high eccentricity , or both . in order to simplify the discussion , we assume that the orbit is circular and thus , is being viewed nearly edge - on .
the most recently measured 2005 positions and the 1998 psf - modeled position provide some tight constraints on the inclination of the orbit .
although the position angle of the 4 resolved measurements is not significantly changing , it is not possible to extend a line connecting all four of these points directly back to kelu-1a .
therefore , the inclination of the orbit must be @xmath790@xmath5 .
we can get a rough estimate of the minimum inclination by assuming the maximum extent perpendicular to the 2005 july position is the separation of the modeled psfs in 1998 , i.e. 45 mas .
this assumption results in an inclination @xmath481@xmath5 and is much more constrained than the minimum inclination suggested by @xcite based on the size of the nicmos psf . finally ,
since it is quite clear that kelu-1b is still moving away from kelu-1a , this minimum inclination estimate will only increase with time .
the period of the binary can be estimated using the 2005 july 31 separation ( 298 mas or 5.4 au at the distance of kelu-1 ) and the masses derived above . with the caveat that the most recent separation is not the maximum orbital separation ,
the calculated period is @xmath23 years .
this period is important for two reasons .
first , it means that it is possible to obtain dynamical masses for the components in a reasonable time - frame , something that has been done for only a few other binary brown dwarf systems @xcite .
second , this period means that the 1.8 hour period ( or 3.6 hours if the variations are ellipsoidal in nature ) reported by @xcite is _ not _ due to this binary .
perhaps one of the components is itself a very close binary , or perhaps the shorter period is simply the rotation period for one of the components .
in addition , the 60 km s@xmath9 rotation velocity measured by @xcite on 1997 june 2 is from a composite spectrum .
the maximum orbital velocity for kelu-1ab is only @xmath03 km s@xmath9 , which translates to a displacement of @xmath10.09 and @xmath10.08 for the cs i and rb i atomics lines used by @xcite this wavelength shift can not account for the large equivalent widths of these lines ( 1.7 and 2.54 for cs i and rb i , respectively ) , so some other mechanism must be at work . in time it might be possible to separate the lines from each component using high resolution spectroscopy and resolve some of these issues .
we have used the lgs ao system on the keck ii telescope to observe the binary brown dwarf kelu-1ab .
we have also re - examined the 1998 hst observations of kelu-1 and show that the psfs in those images are best fit by a binary object solution .
images from multiple epochs confirm the pair as having a common proper motion and demonstrate that kelu-1b is still heading towards apoapsis .
we constrain the inclination of a circular orbit to @xmath481@xmath5 .
while this companion detection does rectify the over - luminosity `` problem , '' it can not account for the 1.8 hour photometric period or the 60 km s@xmath9 rotation velocity .
periodic monitoring of this system with high resolution imaging and spectroscopy is needed to ensure the maximum extent of the orbit is observed and measured , allowing for a more robust determination of the physical orbit . , c. c. , harris , h. c. , vrba , f. j. , guetter , h. h. , canzian , b. , henden , a. a. , levine , s. e. , luginbuhl , c. b. , monet , a. k. b. , monet , d. g. , pier , j. r. , stone , r. c. , walker , r. l. , burgasser , a. j. , gizis , j. e. , kirkpatrick , j. d. , liebert , j. , & reid , i. n. 2002 , , 124 , 1170 , d. a. , leggett , s. k. , marley , m. s. , fan , x. , geballe , t. r. , knapp , g. r. , vrba , f. j. , henden , a. a. , luginbuhl , c. b. , guetter , h. h. , munn , j. a. , canzian , b. , zheng , w. , tsvetanov , z. i. , chiu , k. , glazebrook , k. , hoversten , e. a. , schneider , d. p. , & brinkmann , j. 2004 , , 127 , 3516 , g. r. , leggett , s. k. , fan , x. , marley , m. s. , geballe , t. r. , golimowski , d. a. , finkbeiner , d. , gunn , j. e. , hennawi , j. , ivezi ' c , z. , lupton , r. h. , schlegel , d. j. , strauss , m. a. , tsvetanov , z. i. , chiu , k. , hoversten , e. a. , glazebrook , k. , zheng , w. , hendrickson , m. , williams , c. c. , uomoto , a. , vrba , f. j. , henden , a. a. , luginbuhl , c. b. , guetter , h. h. , munn , j. a. , canzian , b. , schneider , d. p. , & brinkmann , j. 2004 , , 127 , 3553 , d. g. , levine , s. e. , canzian , b. , ables , h. d. , bird , a. r. , dahn , c. c. , guetter , h. h. , harris , h. c. , henden , a. a. , leggett , s. k. , levison , h. f. , luginbuhl , c. b. , martini , j. , monet , a. k. b. , munn , j. a. , pier , j. r. , rhodes , a. r. , riepe , b. , sell , s. , stone , r. c. , vrba , f. j. , walker , r. l. , westerhout , g. , brucato , r. j. , reid , i. n. , schoening , w. , hartley , m. , read , m. a. , & tritton , s. b. 2003 , , 125 , 984 , f. j. , henden , a. a. , luginbuhl , c. b. , guetter , h. h. , munn , j. a. , canzian , b. , burgasser , a. j. , kirkpatrick , j. d. , fan , x. , geballe , t. r. , golimowski , d. a. , knapp , g. r. , leggett , s. k. , schneider , d. p. , & brinkmann , j. 2004 , , 127 , 2948 lccc 1998 aug 14 & 45@xmath118 & [ 38.0 or 218.0]@xmath111.9 & this work + 2005 mar 4 & [email protected] & [email protected] & this work + 2005 apr 30 & [email protected] & [email protected] & this work + 2005 may 1 & 291@xmath12 & [email protected] & @xcite + 2005 jul 31 & 298@xmath13 & [email protected] & this work lcc spectral type & l2@xmath11 & l3.5@xmath11 + @xmath3 magnitude & [email protected] & [email protected] + @xmath15 magnitude & [email protected] & [email protected] + f110 m magnitude & [email protected] & [email protected] + f145 m magnitude & [email protected] & [email protected] + f165 m magnitude & [email protected] & [email protected] + mass ( m@xmath22 ) & [email protected] & [email protected] + t@xmath24 ( k ) & 19002100 & 17001900 + | we have resolved kelu-1 into a binary system with a separation of @xmath0290 mas using the laser guide star adaptive optics system on the keck ii telescope . we have also re - analyzed a 1998 hst observation of kelu-1 and find that the observed psf is best fit by a binary object separated by 45 mas .
observations on multiple epochs confirm the two objects share a common proper motion and clearly demonstrate the first evidence of orbital motion .
kelu-1b is fainter than kelu-1a by [email protected] magnitudes in the @xmath2 filter and [email protected] magnitudes in the @xmath3 filter .
we derive spectral types of l2@xmath11 and l3.5@xmath11 for kelu-1a and b , respectively .
the separation of flux into the two components rectifies kelu-1 s over - luminosity problem that has been known for quite some time . given the available data we are able to constrain the inclination of the system to @xmath481@xmath5 and the orbital period to @xmath640 years . |
a typical plasma consisting of electrons and positive ions essentially causes temporal as well as spatial variations of collective phenomena due to large - mass difference between these particles .
this asymmetric diversity of collective plasma phenomena can , however , be nullified in pair - ion plasmas consisting of positive and negative ions with equal mass . in the latter , the space - time parity can be maintained because of the same mobility of the particles under electromagnetic forces .
such pair - ion plasmas have been generated in the laboratory , and three kinds of electrostatic modes have been experimentally observed to propagate along magnetic - field lines in paired fullerene - ion plasmas @xcite .
furthermore , in many industries such as integrated - circuit fabrication , there requires a plasma source having no energetic electrons in the plasma , since the deposited film is strongly damaged by a high - energy electron .
for this purpose , a radio - frequency plasma source has also been developed @xcite .
on the other hand , _ in situ _ measurements of charged particles in the polar mesosphere under nighttime conditions revealed that there exist positively charged nanoparticles @xcite .
such particles have been observed in a region dominated by both positive and negative ions , and very few percentage of electrons .
it was also clarified that the positive charge of these dust particles is due to the dominant charging effects of lighter positive ions compared to the heavier negative ions , and the presence of a very small number of electrons .
furthermore , in an experiment , it has been investigated that dust particles injected in pair - ion plasmas can become positively charged when the number density of negative ions greatly exceeds ( @xmath3 ) that of the electrons @xcite . in space environments ,
the possible role of negative ions has been discussed , and it has been found that dusts can be positively charged if there is a sufficient number density of heavy negative ions ( with mass @xmath4 amu ) @xcite .
thus , in laboratory and space environments , pair - ion plasmas with positively charged dusts and no high - energy electrons may not be unubiquitous .
so , collective plasma oscillations and nonlinear properties of electrostatic as well as electromagnetic waves in these pair - ion plasmas are worth investigating . to mention few , the nonlinear propagation of solitary waves ( sws ) and shocks in dusty plasmas
have been widely studied for understanding the electrostatic disturbances in space plasma environments as well as in laboratory plasma devices @xcite .
it has been shown that charged dust grains can drastically modify the existing response of electrostatic wave spectra in plasmas depending upon whether the charged dusts are considered to be static or mobile @xcite .
recently , there has been a growing interest in investigating the properties of electrostatic waves in pair - ion plasmas ( see , e.g. , refs .
however , to our knowledge , no detailed theory has been made to study the electrostatic small- as well as large - amplitude waves in dusty pair - ion plasmas with positively charged dusts .
so , our purpose is to investigate the propagation characteristics of electrostatic large- as well as small - amplitude waves in unmagnetized collisionless dusty pair - ion plasma with positively charged dusts .
we show that there exist two modes : fast " and slow " waves , the properties of which are studied .
we use the pseudopotential approach to study the properties of large - amplitude waves , whereas the reductive perturbation technique is used to investigate korteweg - de vries ( kdv ) solitons . the present work thus generalizes and extends the previous works , e.g. ref .
@xcite to include the effects of positively charged dusts , and different mass as well as thermal pressures of ions .
it is shown that in dusty pair - ion plasmas , sws exist with only the negative potential .
we consider the one - dimensional propagation of electrostatic waves in an unmagnetized collisionless dusty plasma consisting of singly charged adiabatic positive and negative ions , and immobile positively charged dusts .
we do not consider the dynamics of charged dusts as they are too heavy to move on the time scale of the ion - acoustic waves .
however , dusts can affect the wave dispersion and nonlinearity .
furthermore , we assume that the negative ions are heavier than the positive ions , and negative ion number density is much larger than that of electrons so that dusts become positively charged @xcite .
thus , the dominant higher mobility species in the plasma are the positive ions .
the immobile dust particles carry some charges so as to maintain the overall charge neutrality condition given by @xmath5 where @xmath6 is the unperturbed number density of charged species @xmath7 ( @xmath8 , @xmath9 , @xmath10 , respectively , stand for positive ions , negative ions and static positively charged dusts ) , @xmath11 @xmath12 is the unperturbed dust charge state .
the condition , in dimensionless form , becomes @xmath13 where @xmath14 and @xmath15 are the density ratios .
the basic equations to describe the dynamics of positive and negative ions in one space dimension are @xmath16 @xmath17 @xmath18 where @xmath19 , @xmath20 , and @xmath21 , respectively , denote the number density , velocity , and mass of @xmath7-species particles .
furthermore , @xmath22 , where @xmath23 is the elementary charge . also , @xmath24 is the electrostatic potential , @xmath25 is the boltzmann constant , and @xmath26 is the particle s thermodynamic temperature . in eq .
, we have used the adiabatic pressure law : @xmath27 with @xmath28 for each ion - species @xmath29 .
the adiabatic index @xmath30 @xmath31 , @xmath32 being the number of degrees of freedom ] is used for one - dimensional geometry @xmath33 of the system .
we will later see that in the long - wavelength limit , the phase velocity of the fast wave is much greater than the positive - ion thermal velocity .
this certainly justifies our assumption of adiabaticity that thermal conduction can not keep up with the moving wave front .
next , we normalize the physical quantities according to @xmath34 , @xmath35 , @xmath36 , where @xmath37 is the ion - acoustic speed with @xmath38 and @xmath39 denoting , respectively , the negative - ion plasma frequency and the debye length .
the space and time variables are normalized by @xmath40 and @xmath41 respectively .
thus , from eqs . - , the normalized set of equations is obtained as : @xmath42 @xmath43 @xmath44 where the upper ( lower ) sign of @xmath45 on the right - hand side of eq .
corresponds to positive ( negative ) ions . for convenience and to be used later , we denote @xmath46 and @xmath47 .
in order to figure out the number of linear wave modes and to study their properties , we assume the perturbed quantities to vary as @xmath48 , i.e. , in the form of oscillations with wave frequency @xmath49 and wave number @xmath50 .
thus , fourier analyzing the eqs . - , we obtain the following dispersion relation : @xmath51 in which @xmath49 and @xmath50 are normalized by @xmath52 and @xmath53 respectively .
the first and second terms on the left - hand side of eq .
are , respectively , due to the presence of positive- and negative - ion species , whereas the constant term @xmath54 on the right - hand side appears due to the effect of charge separation of the species ( deviation from quasineutrality ) .
the dispersion eq . admits two solutions of @xmath55 given by @xmath56 where @xmath57 and @xmath58k^2}$ ] . in order that these solutions are real
, the discriminant @xmath59 must be nonnegative .
we find that the discriminant is positive for all @xmath50 , when @xmath60 , @xmath61 , and @xmath62 are satisfied . in eq .
, the positive and negative signs , respectively , correspond to fast " and slow " wave modes in pair - ion plasmas .
recalling eq .
in dimensional form , we find that the effect of dispersion due to charge separation can not be neglected , because otherwise , for the fast " waves of which the phase speed is much larger than the ion - thermal speed , the sum of the two positive terms becomes zero , which is inadmissible .
furthermore , in order to avoid the wave damping due to the resonance with either the positive or negative ions , the phase speed is to be much larger than the negative ion thermal speed and much lower than the positive ion thermal speed . in particular , in absence of the charged dusts , and if ions , having the same mass , are in isothermal equilibrium , one can easily recover the dispersion equation ( 6 ) in ref .
furthermore , in particular , for @xmath63 , i.e. , in dust - free plasmas in which ions have the same mass and temperature , the two ion - modes , corresponding to positive and negative signs in eq . , reduce , in the long - wavelength limit @xmath64 , to @xmath65 and @xmath66 , or , in dimensional form , @xmath67 and @xmath68 .
the former is the usual ion - plasma wave , which propagates with a frequency greater than @xmath52 ( even in absence of the thermal pressure ) , whereas , the latter propagates forward in the manner of a sound wave in which the electrostatic oscillation disappears .
it has been observed in experiments that slow - wave modes do not favor the formation of solitons , whereas the fast modes may propagate as solitary waves due to nice balance of the dispersive and nonlinear effects @xcite . in the next sections , our attempt will be to study whether this ion - plasma wave ( fast mode ) propagates as solitary waves .
however , before going into the detail investigation for the nonlinear propagation , we first analyze numerically the properties of the fast " and slow " wave modes as follows : figures 1 to 3 show the plots of the frequencies of the fast- ( upper panel ) and slow - wave ( lower panel ) modes with the variations of the wave number @xmath50 for different sets of parameter values of @xmath69 , @xmath70 and @xmath71 as indicated in the figures . from fig .
1 , it is evident that both the fast- and slow - wave frequencies increase with @xmath50 as well as with higher values of the mass ratio @xmath69 . in contrast to the fast mode , the slow - wave frequency remains almost unaltered for values of @xmath72 , as well as in the limits of the long - wavelength , @xmath73 and short - wavelength , @xmath74 . in the former limit , though the frequency of the fast mode bears almost a constant value , the slow - wave frequency approaches to zero .
in all the figures 1 to 3 , the wave frequency is found to increase with the wave number @xmath50 implying that the wave phase speed exceeds that of the ion - sound wave .
figure 2 shows the similar profiles as in fig . 1 , but for different values of the temperature ratio @xmath70 .
it is found that @xmath70 , however , does not affect the frequency of the fast - wave mode ( see the upper panel of fig .
2 ) , whereas the slow - wave frequency increases with increasing values of @xmath70 .
in the latter , @xmath49 becomes larger with higher values of @xmath75 .
figure 3 shows that in contrast to the slow - wave modes , the effect of positively charged dusts is to decrease the wave frequency of the fast - wave modes .
it is also seen that for @xmath76 , charged dusts have very small effect on the wave modes .
here , we assume the perturbations to vary in a moving frame of reference @xmath77 , where @xmath78 is the nonlinear wave speed normalized by @xmath79 .
we also use the boundary conditions , namely , @xmath80 , @xmath81 , @xmath82 as @xmath83 . then from the continuity equation @xmath84 next , from eq . and
after eliminating @xmath20 , we obtain bi - quadratic equations for the densities @xmath19 which have the following solutions : @xmath85^{1/2 } , \label{density - p}\ ] ] @xmath86^{1/2 } , \label{density - n}\ ] ] where @xmath87 are the values of the number densities @xmath19 corresponding to the @xmath45 signs .
also , @xmath88 , @xmath89,@xmath90 , and @xmath91 . equations and give four possible combinations for the number densities , namely , @xmath92 , @xmath93 ; @xmath94 , @xmath95 , @xmath92 , @xmath95 and @xmath94 , @xmath93 .
we will later see that only one of them will favor the propagation of large - amplitude solitary waves . from eqs . and
it is clear that for real values of the densities , the conditions @xmath96 and @xmath97 must be satisfied .
these lead to the following bounds for the electrostatic potential : @xmath98 where @xmath99 and @xmath100 , implying that @xmath24 has the range from negative to positive values .
next , from the poisson equation , after integrating with respect to @xmath101 and using the above boundary conditions , we obtain the following energy - like equation for an oscillating particle of unit mass at the pseudoposition @xmath24 and pseudotime @xmath101 : @xmath102 where the pseudopotential @xmath103 is given by @xmath104 \notag\\ -\frac{2}{3\sqrt{3t}}\left[g_{n\pm}(\phi)-g_{n\pm}(0)\right],\label{v - pseudo}\end{aligned}\ ] ] with @xmath105 and @xmath106 . here , @xmath107 is the value of @xmath108 at @xmath109 . the pseudopotential @xmath103 and
hence the relevant expressions , to be shown shortly , can have four different forms depending on the consideration of the densities @xmath87 given by eqs . and .
however , we will show that only one of them corresponds to the existence of solitary waves or double layers .
the energy - like equation , which describes the evolution of arbitrary amplitude electrostatic perturbations , can also be obtained following refs .
the conditions for which the perturbations may propagate as solitary waves or double layers can be discussed as follows : + + * condition 1 : * @xmath110 at @xmath109 .
this can be easily verified from eq . by substituting @xmath109 .
also , @xmath111 for @xmath112 .
this condition will be examined numerically later
. + + * condition 2 : * @xmath113 at @xmath109 . from eq .
we have @xmath114 where @xmath115 and @xmath116 in which the first and second terms in the parentheses are corresponding to the positive and negative signs of @xmath45 .
thus , from eqs . and
we find that * if @xmath117 , then the condition @xmath113 at @xmath109 is satisfied for the expression [ eq .
] corresponding to @xmath118 ( i.e. , corresponding to the number densities @xmath119 ) .
however , for the expression corresponding to @xmath120 ( i.e. , corresponding to the number densities @xmath121 ) , the same condition is satisfied when @xmath122 . typically , for laboratory and space plasmas@xcite , @xmath123 and @xmath124 .
so , @xmath125 , and @xmath126 holds for @xmath127 together with @xmath62 for positively charged dusts . *
if @xmath128 , i.e. , if @xmath129 for @xmath127 , then @xmath113 at @xmath109 is satisfied for the expression corresponding to @xmath120 , i.e. , corresponding to the densities @xmath121 . however , for @xmath118 or for the densities
@xmath119 , the same is satisfied for @xmath130 .
but , in this case , @xmath131 , otherwise , we would have @xmath132 , which contradicts the above consideration @xmath127 . * when @xmath133 holds for @xmath127 together with @xmath62 , @xmath123 and @xmath124 , the condition @xmath113 at @xmath109
is satisfied corresponding to @xmath134 and @xmath135 , i.e. , for the number densities @xmath92 and @xmath95 . the other cases may not be of interest for the parameter regimes mentioned above .
thus , depending on the consideration of the solutions of the number densities @xmath87 , and hence the expressions for @xmath136 and its derivatives , the condition @xmath113 at @xmath109 may be satisfied for a certain range of values of @xmath78 , namely , @xmath137 , @xmath129 or @xmath133 . + + * condition 3 : *
@xmath138 at @xmath109 . from eq .
, we obtain @xmath139 where @xmath140 signs appear due to the expressions corresponding to @xmath141 . from eq . , we find that ( i ) for @xmath137 , @xmath142 at @xmath109 is always negative when @xmath142 takes the form corresponding to @xmath118 ; ( ii ) for @xmath129 , @xmath142 at @xmath109 is always positive [ when @xmath142 has the expression corresponding to @xmath120 ] ; ( iii ) in the range @xmath133 , @xmath142 at @xmath109 [ corresponding to @xmath134 and @xmath144 , i.e. , for the number densities @xmath92 and @xmath95 ] is negative for @xmath146 .
the latter is admissible for @xmath60 . in this case
, the range of @xmath78 is precisely @xmath147 .
+ + * condition 4 : * for a nonzero @xmath148 , the relations @xmath149 and @xmath150 are to be satisfied according to whether the solitary waves are compressive ( with @xmath151 ) or rarefactive ( with @xmath152 ) . here , @xmath148 @xmath153 represents the amplitude of the solitary waves .
however , in addition , to the above conditions , if @xmath154 vanishes instead of @xmath150 , the perturbations may develop into double layers .
these conditions will also be examined numerically . from the above discussions we find that there may be three possible regions , for which all the above conditions for the existence of large amplitude solitary waves or double layers are satisfied , these are + + ( i ) * region 1 : * @xmath155 together with @xmath156 , @xmath157 , @xmath62 .
the expression for @xmath103 and relevant others are corresponding to the number densities @xmath119 ; + + ( ii ) * region 2 : * @xmath129 together with @xmath156 , @xmath157 , @xmath62 . here , the expression for @xmath103 and relevant others are corresponding to @xmath121 ; + + ( iii ) * region 3 : * @xmath147 together with @xmath156 , @xmath157 , @xmath62 . in this case , the expression for @xmath103 and relevant others are corresponding to @xmath92 and @xmath95 . in the following numerical investigation , we will see that the existence of large - amplitude solitary waves may be possible only for the parameters in ` region 3 ' .
it will also be shown that @xmath158 except at @xmath109 , i.e. , the double layers may not exist in the plasmas .
next , we numerically investigate the conditions 1 to 4 stated above for the existence sws or double layers .
figures 4 to 7 show the profiles of the pseudopotential @xmath103 ( upper panel ) and its derivative @xmath159 ( lower panel ) for different values of the parameters @xmath69 , @xmath78 , @xmath70 and @xmath71 .
we find that for certain parameter values , @xmath103 crosses the @xmath24-axis and @xmath160 at only negative values of @xmath24 in @xmath112 , implying the existence of only negative sws .
these negative values of @xmath24 represent the heights or amplitudes @xmath148 of the large - amplitude solitons .
the latter can be obtained by numerically solving eq . .
from figs . 4 to 7 , it is clear that there is no common value of @xmath24 @xmath153 for which @xmath110 and @xmath161 are satisfied for the existence of positive sws .
notice , however , that there are also some parameter regimes corresponding to @xmath69 , @xmath78 , @xmath70 and @xmath71 for which either @xmath103 @xmath162 does not cross the @xmath24-axis or @xmath163 , and hence no potential well is formed for particle trapping . for example , in fig . 4 , @xmath164 for @xmath165 . in this case
, the large - amplitude solitary profiles may exist for @xmath166 with a fixed set of parameters @xmath167 and @xmath168 .
the similar ranges for @xmath78 , @xmath70 and @xmath71 ( see figs . 5 to 7 ) for the existence of large - amplitude solitons , respectively , are @xmath169 with fixed @xmath170 and @xmath168 ; @xmath171 with fixed @xmath170 , @xmath167 , and @xmath172 ; @xmath173 with a fixed set @xmath170 , @xmath167 , and @xmath174 .
we find that the absolute value of the depth of the potential well decreases with increasing values of @xmath69 ( fig .
4 ) , @xmath70 ( fig . 6 ) and @xmath71 ( fig .
7 ) , which may lead to the enhancement of the width and detraction of the amplitude @xmath175 of the sws .
this width ( amplitude ) may , however , be decreased ( increased ) with increasing values of the mach number @xmath78 as in fig .
our next attempt is to numerically solve eq . to exhibit the profiles of the large - amplitude solitons .
these are shown in fig . 8 for different sets of parameters @xmath69 , @xmath78 , @xmath70 and @xmath71 .
evidently , the amplitude of the large - amplitude soliton decreases while its width increases with increasing values of @xmath69 and @xmath71 . on the other hand , as the values of @xmath78 @xmath1 increase , the soliton amplitude increases ( decreases ) while its width decreases .
it may be interesting to examine the profiles of the sws whose amplitudes tend to zero .
such small - amplitude solutions of the sws can be obtained by expanding @xmath103 in powers of @xmath24 up to @xmath176 around the origin , i.e. , @xmath177 , where @xmath178 and @xmath179 , and using the boundary conditions , namely @xmath82 , @xmath180 as @xmath83 , as : @xmath181 here , @xmath182 is the amplitude and @xmath183 is the width of the soliton .
alternatively , one can follow a perturbation technique to investigate the dynamical evolution as well as the properties of kdv solitons .
this will be done in the next section .
we consider the nonlinear propagation of small but finite amplitude electrostatic waves in dusty pair - ion plasmas . in the standard reductive perturbation technique ,
the stretched coordinates are considered as @xmath184 and @xmath185 , where @xmath186 is a small parameter @xmath187 measuring the weakness of perturbations and @xmath188 is the wave speed .
the dynamical variables are expanded as @xmath189 we then substitute these expansions and the stretched coordinates into eqs . - , and equate different powers of @xmath186 . in the lowest order of @xmath186 ( i.e. , @xmath190 ) we obtain the following relations for the first - order perturbations : @xmath191 where @xmath192 and @xmath193 .
the expression for the wave speed in the moving frame of reference is given by @xmath194 we note that since @xmath60 ( for @xmath61 and @xmath195 ) , we have @xmath196 ( in contrast to @xmath137 for large - amplitude sws ) and @xmath188 increases with increasing values of @xmath69 and @xmath70 . proceeding to the next order of @xmath186 ( i.e. , @xmath197 ) , the details are omitted for simplicity , we obtain a set of equations for the second - order perturbed quantities .
the latter are then eliminated to obtain , after few steps , the following kdv equation .
@xmath198 where @xmath199 .
the coefficients of nonlinearity and dispersion are , respectively , given by @xmath200}{2v_0(1+m\mu)},\label{nonlinear - coeff}\ ] ] @xmath201 inspecting on the coefficients @xmath202 and @xmath203 [ eqs . and ] , we find that @xmath203 is always positive . also , @xmath202 is always negative , because @xmath204 for @xmath205 . these imply that the small - amplitude sws exist with only the negative potential .
the stationary soliton solution of the kdv equation can be obtained by applying a transformation @xmath206 , where @xmath207 is the constant phase speed normalized by @xmath79 , and imposing the boundary conditions for localized perturbations , namely , @xmath24 , @xmath208 , @xmath209 as @xmath210 as : @xmath211 where @xmath212 is the amplitude and @xmath213 is the width of the soliton .
note that the qualitative behaviors , i.e. , the properties of the amplitude and width of the small - amplitude solitons given by eqs .
[ obtained from the energy - like equation with an approximation ] and [ obtained from the kdv equation ] for different values of @xmath69 and @xmath70 will remain the same .
however , we analyze only the properties of the kdv soliton .
figures 9 and 10 show the profiles of the soliton amplitude ( upper panel ) and width ( lower panel ) [ eq . ] with the variation of @xmath214 for different values of @xmath69 and @xmath70 .
figure 9 shows that the absolute value of the amplitude decreases with @xmath214 and also with increasing values of @xmath69 , whereas the width decreases with increasing @xmath214 ( except for a fixed @xmath215 at which the width approaches a constant value ) and @xmath69 . on the other hand ,
the variations of the amplitude and width with @xmath214 for different values of @xmath70 show almost opposite features . in the latter , the amplitude ( absolute value ) increases but the width decreases with increasing values of @xmath70 . in both the figures 9 and 10
, the soliton amplitude is seen to decrease with @xmath214 and approaches more or less a constant value as @xmath214 approaches @xmath54 . for the dynamical evolution of the soliton
, we numerically solve eq . by runge - kutta scheme with an initial condition
@xmath216 $ ] .
the development of the pulse at different times @xmath217 ( upper left ) , @xmath218 ( upper right ) , @xmath219 ( lower left ) and @xmath220 ( lower right ) is shown in fig . 11 .
the parameter values are considered as @xmath221 , @xmath222 and @xmath172 .
it is seen that the leading part of the initial pulse steepens due to positive nonlinearity , and as time goes on , the pulse separates into solitons and a residue due to the wave dispersion .
furthermore , once the solitons are formed and separated , they propagate without changing their shape due to the nice balance of the nonlinearity and dispersion ( see the plot for @xmath220 ) .
we have investigated the propagation characteristics of electrostatic waves in an unmagnetized collisionless pair - ion plasma with immobile positively charged dusts .
the latter become positively charged when the number density of negative ions exceeds that of positive ions and is much larger than that of electrons so that electron contribution can be neglected @xcite . in the linear regime
, there appear two ion modes , namely fast " and slow " waves in dusty pair - ion plasmas . due to the higher values of the phase velocity of the fast ion wave than the ion thermal speeds , the landau damping effect is negligibly small .
we find that in the long - wavelength limit @xmath64 , the fast ( slow ) mode propagates with a frequency greater ( lower ) than the frequency of negative - ion oscillations .
furthermore , in the limit @xmath73 , the frequency of the fast modes almost assumes a constant value .
we also find that the thermal pressure has very small effect on the fast - wave modes , whereas it can modify the frequency of the slow - wave modes .
in contrast to the effects of different mass of the ions , the effect of positively charged dusts is to decrease the wave frequency of the fast waves . in the nonlinear regime , we have studied the propagation of large- as well as small - amplitude perturbations in dusty pair - ion plasmas .
we show that perturbations can develop into solitary waves , and no double - layer solution can be formed . using the pseudopotential approach ,
we derive a energy - like equation , which ( along with the pseudopotential ) is numerically analyzed to study the properties of the large - amplitude solitons for different values of the system parameters @xmath69 , @xmath70 and @xmath214 or @xmath71 which satisfy @xmath61 , @xmath195 and @xmath62 .
it is found that large - amplitude sws exist for the mach number satisfying @xmath147 . in our numerical investigation ,
the parameter regimes for the existence of large - amplitude sws are obtained as ( i ) @xmath166 with fixed @xmath167 and @xmath168 , ( ii ) @xmath169 with a fixed set @xmath170 and @xmath168 , ( iii ) @xmath171 with fixed @xmath170 , @xmath167 , and @xmath172 , and ( iv ) @xmath173 with a fixed set @xmath170 , @xmath167 , and @xmath174 .
it is shown that the effects of different masses @xmath0 , different temperatures @xmath1 of ions as well as the charged dust impurity @xmath2 are to diminish the soliton amplitudes .
the latter can be enhanced by increasing the nonlinear wave speed @xmath78 .
furthermore , the widths of the large - amplitude solitons can be increased ( decreased ) by the effects of @xmath69 and @xmath71 ( @xmath78 and @xmath70 ) . for small - amplitude waves ,
we derive a kdvb equation which is numerically solved to present the dynamical evolution of solitons .
furthermore , we have numerically studied the properties of the amplitude and width of the kdv solitons for different values of the system parameters .
we find that both the large- and small - amplitude waves propagate with only the negative potential in pair - ion plasmas with positively charged dusts .
the theoretical results may be useful for the observation of electrostatic waves in space plasmas , e.g. , a dusty meteor trail region in the upper atmosphere , in industrial electron - free pair - ion plasmas as well as for the experimental verification of the excitation of ion - acoustic waves in laboratory dusty pair - ion plasmas .
this work was partially supported by the sap - drs ( phase - ii ) , ugc , new delhi , through sanction letter no
. f.510/4/drs/2009 ( sap - i ) dated 13 oct .
, 2009 , and by the visva - bharati university , santiniketan-731 235 , through memo no .
aca - r-6.12/921/2011 - 2012 dated 14 feb . , 2012 .
50 w. oohara , d. date , and r. hatakeyama , phys .
lett . * 95 * , 175003 ( 2005 ) .
e. yabe and k. takahashi , appl .
phys . lett . *
65 * , 694 ( 1994 ) .
m. rapp , j. hedin , and i. strelnikova , m. friedrich , j. gumbel , and f .- j .
lbken , geophys .
32 * , l23821 ( 2005 ) .
kim and r. l. merlino , phys .
plasmas * 13 * , 052118 ( 2006 ) .
e. grn , g. e. morfill and d. a. mendis , planetary rings ( eds r. greenberg , and a. brahic ) , ( univ .
of arizona press , tucson , 1984 ) . c. k. goertz , rev .
* 27 * , 271 ( 1989 ) .
a. barkan , n. d. angelo , and r. l. merlino , phys .
lett . * 73 * , 3093 ( 1994 ) .
n. c. adhikary , m. k. deka , and h. bailung , phys .
plasmas * 16 * , 063701 ( 2009 ) . y. nakamura and a. sarma , phys .
plasmas * 8 * , 3921 ( 2001 )
. n. n. rao , p. k. shukla , and m. y. yu , planet .
space sci . * 38 * , 543 ( 1990 ) .
p. k. shukla and v. p. silin , phys .
scr . * 45 * , 508 ( 1992 ) .
a. p. misra , a. r. chowdhury , and k. r. chowdhury , phys .
a * 323 * , 110 ( 2004 ) .
a. p. misra , k. r. chowdhury , and a. r. chowdhury , phys .
plasmas * 14 * , 012110 ( 2007 ) . n. s. saini and i. kourakis , phys .
plasmas * 15 * , 123701 ( 2008 ) .
a. a. mamun , phys .
a * 372 * , 4610 ( 2008 ) .
a. a. mamun , r. a. cairns , and p. k. shukla , phys .
a * 373 * , 2355 ( 2009 ) .
n. c. adhikary , phys .
a * 376 * , 1460 ( 2012 ) .
h. u. rehman , chin .
* 29 * , 65201 ( 2012 ) .
s. ghosh , n. chakrabarti , m. khan , and m. r. gupta , pramana - j . phys . * 80 * , 283 ( 2013 ) .
s. mahmood , h. u. rehman , and h. saleem , phys .
scr . * 80 * , 035502 ( 2009 ) .
y. nakamura , j. l. ferreira , and g. o. ludwig , j. plasma phys . * 33 * , 237 ( 1985 ) .
a. y. wong , introduction to experimental plasma physics , vol . 1 ,
springer ( 1977 ) .
a. p. misra , n. c. adhikary , and p. k. shukla , phys .
e * 86 * , 056406 ( 2012 ) . | the propagation of electrostatic waves in an unmagnetized collisionless pair - ion plasma with immobile positively charged dusts is studied for both large- and small - amplitude perturbations . using a two - fluid model for pair - ions
, it is shown that there appear two linear ion modes , namely the
fast " and slow " waves in dusty pair - ion plasmas .
the properties of these wave modes are studied with different mass @xmath0 and temperature @xmath1 ratios of negative to positive ions , as well as the effects of immobile charged dusts @xmath2 . for large - amplitude waves ,
the pseudopotential approach is performed , whereas the standard reductive perturbation technique ( rpt ) is used to study the small - amplitude korteweg - de vries ( kdv ) solitons .
the profiles of the pseudopotential , the large amplitude solitons as well as the dynamical evolution of kdv solitons are numerically studied with the system parameters as above .
it is found that the pair - ion plasmas with positively charged dusts support the propagation of solitary waves ( sws ) with only the negative potential .
the results may be useful for the excitation of sws in laboratory dusty pair - ion plasmas , electron - free industrial plasmas as well as for observation in space plasmas where electron density is negligibly small compared to that of negative ions . |
there are two general approaches to the study of the complex effects of galaxy collisions .
the first is statistical , such as the study of particular properties of a reasonable sample of interacting galaxy systems , or a grid of numerical models covering some range of initial conditions .
the second approach is the case study , the detailed investigation of a particular collisional system .
the first approach is the path more frequently followed , in the literature .
a great deal can be learned from observations of the global properties of many systems , which are easier to acquire ( individually ) , than the high sensitivity , high resolution , multi - wavelength data needed for a good case study . in the realm of simulations , the statistical approach also has many attractions .
for example , in many regions of parameter space the model outcomes are quite sensitive to collision parameters , so converging on the correct parameters can be a prolonged task .
( at the same time , such sensitivities can go a long way toward guaranteeing the uniqueness of a successful model . )
one of the greatest successes of the statistical approach ( spurred by iras results ) in the last two decades is an understanding of how gas is redistributed in major mergers , and how ultra - luminous , super - starbursts result .
insights obtained from numerical models ( e.g. , barnes and hernquist 1972 ) also played a crucial role .
statistical studies also provide important inputs to the topics of mass transfer and induced star formation ( sf ) in galaxy collisions . even aside from studies of luminous major merger remnants , there has been much work on the questions of whether and by how much sf is enhanced by interactions .
the early color analysis of larson and tinsley ( 1978 ) , suggested color dispersion enhancement , rather than bluer colors .
the nuclear spectrophotometry by keel et al .
( 1985 , also kennicutt et al .
1987 ) suggested enhancement of nuclear sfrs .
more recently , bergvall et al . ( 2003 and earlier work cited therein ) find no difference in the broadband colors of sample of 59 interacting systems relative to a control sample of 38 galaxies .
they do find a moderate increase in central sf and far - infrared emission in the interacting sample .
they argue that earlier work , claiming greater enhancements was biased towards ir luminous merger remnants , in contrast to their sample .
barton , geller & kenyon ( 2003a , and barton gillespie , geller & kenyon 2003b ) obtained b and r band photometry and optical spectroscopy of 190 galaxies in pairs and compact groups .
they also found enhanced core sf , and a number of post - starburst , as well as starburst cores .
in addition they found an anti - correlation between galaxy separation and sf , and suggested starbursts were preferentially triggered at closest approach , decaying thereafter .
these statistical studies shed the most light on nuclear sf , suggesting that it has a short duty cycle , and perhaps requires a relatively strong disturbance to funnel the gas fuel inward ( keel 1993 ) . in this symposium
we have seen some beautiful observations of sf in tidal structures .
the statistical studies suggest that such sf does not add up to a very large amount .
this echos the result of the statistical study of schombert , wallin & struck - marcell ( 1990 ) on the colors of tidal bridges and tails . yet , in tails it is the nature , rather than the quantity of sf that is of interest . case studies with more details on the sf history , as well as instantaneous rates in individual sf regions are needed to address the many unanswered questions .
the processes of tidal or splash mass transfer are very hard to study either individually , or statistically .
we can observe the amount of gas between or outside the interacting galaxies , but how much has been transferred already , and how much will yet be pulled out ?
the statistical question can be addressed with models , but there have been few such studies .
wallin & stuart ( 1992 ) gave us a survey and analysis of 1000 restricted 3-body encounters with mass transfer from a particle disk around the primary .
these models evenly sampled a number of collision parameters , and quantified dependences , such as inclination , some of which had been known since toomre and toomre 1972 ( see struck 1999 , sec .
howard et al .
( 1993 ) published a nice atlas of 86 n - body simulations ( with rigid halos ) of the effects of encounters on ( 2-d ) disks .
i produced a small model grid to study the hydrodynamics of colliding gas disks ( struck 1997 ) .
statistical studies show us the big picture , and give the average answer to big picture questions , but often leave us wondering about the specific mechanisms .
case studies provide detailed answers to some of those questions , but they require a large amount of high quality observational data , and a substantial modeling effort to interpret it . as yet , not many have been published .
however , we can now optically resolve individual star clusters in nearby interacting systems , map tenuous gas distributions , and we have the computer power to do the modeling , so it is a great time for case studies .
i will justify that statement with two examples .
bev smith and i have been working for some years on this beautiful system ( see figure 1 , and smith & wallin 1992 , smith et al .
1997 , struck & smith 2003 ) .
it is an asymmetric collisional ring galaxy , with a substantial bridge connecting it to a nearly edge - on companion , and it also contains some distinctive tidal tails .
it has a variety of features that make it an especially interesting case for the study of induced sf and mass transfer .
the primary has a prototypical starburst nucleus , but their is little evidence for recent sf in the ring wave , in contrast to most gas - rich collisional ring systems ( see appleton and struck - marcell 1996 ) .
the companion has a post - starburst spectrum .
there are intriguing knots of recent sf in one of the countertails , and roughly in a line on the north side of the bridge .
more than half of the gas mass is located outside the main disks , in the bridge and tails ( smith et al .
1997 ) . in the bridge
, there is an offset between the center line of the old stars , and that of the gas .
lanon et al .
( 2001 ) have published detailed spectral evolutionary models of hst observations , so we have a good picture of the sf history of the nucleus of ngc 7714 . in a word , this history seems to be characterized by repetitive bursts .
unfortunately , high resolution hst spectra are not available for extra - nuclear regions , or for the companion .
( a. lanon is working to obtain vlt data , however . )
we have recently published n - body hydrodynamic ( sph ) models of this system that can account for nearly all of the observed morphological and kinematic features ( smith & struck 2003 ) .
successful models require a high inclination collision , which is somewhat prograde for galaxies .
such an orbit allows the simultaneous production of a complete ring , and the tails . in terms of mass transfer we found that we could indeed fling a large fraction of the gas out of the two disks in such model collisions , and without unusual initial conditions , like especially extended gas disks .
specifically , our models put a somewhat large gas fraction than observed into the great hi loop to the north , somewhat less than observed in the bridge , and a good deal in the companion countertail .
this last feature is not isolated in the observations , except as a slight eastern extension of the companion disk .
the models suggest that it lies behind both the companion disk and the bridge , and can not be easily distinguished .
interestingly , we found that the bridge consists of several superposed , but dynamically distinct components .
in addition to the two usual tidal components stretching from the near side of each galaxy towards its companion , there is also the superposed countertail of ngc 7715 , and the remnants of an old tail winding around from the far - side of ngc 7715 to the primary ( ngc 7714 ) .
mass transfer onto the primary from this old bridge started before closest approach .
some of the early transfer material was incorporated into the primary disk , or in a plane very close to it , and subsequently flung out ( again ! ) as the inner sw countertail .
this feature , and the fact that the location and mass of some of the bridge components are very sensitive to initial collision parameters , illustrate the intricacies of mass transfer that can be found in detailed case studies .
our models include simple feedback prescriptions that can tell us something about induced sf .
the gas compression history can also suggest sf phenomena even when we do nt really have enough particles to fully resolve them .
examples of the latter include the result that the ( star - forming ) inner sw countertail of ngc 7714 may consist of mass transferred material that has been compressed in the primary disk .
similarly , the models show that the old bridge component is overrun and shocked by a newer bridge component .
this might account for the line of young star clusters on the north side .
the feedback model further suggests that the companion experienced a starburst at closest approach , consistent with its post - starburst spectrum .
it also suggests multiple bursts have occurred in the core of the primary , in agreement with the spectral synthesis results
. however , the exact timing and number of these bursts are not completely predictable .
the models suggest that they could be driven by compressions from the ( m=0 ) ring wave component of the disturbance , and fed by inward mass transfer resulting from the spiral component . on the other hand ,
repetitive starbursts in the gas - rich core are driven by the feedback terms even without an external disturbance .
( this result derives from control runs and other unpublished modeling . )
it is likely that the gas density in the core or its susceptibility to sf was less than in the models .
i have been working on this galaxy with the ocular galaxy collaboration ( see elmegreen et al .
1991 , 1995a , b , 2000 , 2001 , kaufman et al .
the hubble heritage image of this graceful pair has been often reproduced .
the models suggest that it has been one of the gentlest close interactions known .
they further suggest that the companion approached from slightly above the primary ( in the west ) , interacting with its outer edge , and then moving slightly below and to the north and then east of the primary .
the encounter is retrograde relative to the primary , so its optical disk is not greatly perturbed , though its larger hi disk is .
the encounter is prograde and nearly in - plane relative to the companion , which is estimated to be about 60 - 80% of the mass of the primary .
encounters that strongly torque the disk produce long tidal tails , and also the transient ocular ( eye - shaped ) morphology from material that loses angular momentum ( elmegreen et al .
this system is a prototypical example of that process of disk rearrangement .
in addition , the models suggest that there is some moderate mass transfer from the companion to the primary , both presently and earlier in a glancing interaction on the west side .
they also suggest that the collisional perturbation did not produce the spirals in the primary ; they were almost certainly pre - existing .
observationally , the recent star formation in this system is mostly extended ( and beautiful ! ) with no excess in the galaxy cores .
specifically , the young star clusters are concentrated in the long spiral arms . in the companion ,
the young clusters are concentrated on the rim of the ocular .
the tidal tail is younger and shorter than many we have seen at this symposium and is not presently the site of active sf .
the models , projected into the future , indicate that it will grow !
the feedback models yield a bit more sf in the core of the primary than observed , but otherwise with the same widely distributed sf in the primary , and concentration of sf in the ocular and tidal tail of the companion .
the models suggest some interesting patterns in the sf in the primary over the course of the interaction , but these must be further analyzed .
what have we learned from these and other published case studies , and how do they complement the statistical studies ? in the area of mass transfer , the generally good agreement between model and observational gas distributions in two very different cases - arp 284 ( extreme gas removal from the disks ) and ngc 2207 ( little perturbation of the primary disk ) - is very encouraging .
these general results also agree with expectations derived from exploratory model grids .
the more detailed comparisons to observation in these two systems give us confidence that hydrodynamic models can quite accurately reproduce details of collisional morphology and kinematics on scales of about a few kpc .
this could motivate further checks of model predictions , e.g. , a metallicity study of the countertails of ngc 7714 to see if there is evidence of differences that might be expected if the inner tail gas came from the companion .
such specific predictions are not possible without detailed modeling of individual systems . in terms of sf ,
case studies are required to model modes of sf that are either unique to a specific system , or nearly so .
possible examples from the cases above include the inner sw tail of ngc 7714 and , the line of sf regions on the northern edge of the arp 284 bridge .
less unique examples , but still with system specific characteristics include the absence of sf in the ngc 7714 ring , sf in the rim of the ic 2163 ocular , and the scattered sf in the ngc 2207 disk .
the bulk of interaction induced sf in the universe occurs in merger remnants and the cores of unmerged collision partners .
statistical studies are ideal for studying the mean characteristics of this type of sf , and case studies would contribute little if this sf has a large stochastic component
. however , understanding anomalous sf , like the examples of the previous paragraph , may be essential to understanding the formation of globular clusters and tidal dwarfs .
these are minority populations , but still very interesting .
in fact , answering the questions of how tidal dwarfs and globulars form , and also understanding the sytematics of wave induced sf in disks will require many detailed case studies .
appleton , p. n. , & struck , c. 1996 , fund .
phys . , 16 , 111 barnes , j. e. , & hernquist , l. 1992 , , 30 , 705 barton , e. j. , geller , m. j. , & kenyon , s. j. 2000 , , 530 , 660 barton gillespie , e. , geller , m. j. , & kenyon , s. j. 2003 , , 582 , 668 bervall , n. , laurikainen , e. , & aalto , s. 2003 , astro - ph 0304384 ( a&a in press ) elmegreen , b. g. , et al .
1995a , , 453 , 139 elmegreen , b. g. , et al .
2000 , , 120 , 630 elmegreen , d. m. , et al . 1995b , , 453 , 100 elmegreen , d. m. , et al .
2001 , , 121 , 182 elmegreen , d. m. , et al .
1991 , , 244 , 52 howard , s. , keel , w. c. , byrd , g. , & burkey , j. 1993 , , 417 , 502 kaufman , m. , et al .
1997 , , 114 , 2323 keel , w. c. 1993 , , 106 , 1771 keel , w. c. , et al .
1985 , , 90 , 708 kennicutt , jr . , r. c. , et al .
1987 , , 93 , 1011 lanon , a. , et al .
2001 , , 552 , 150 larson , r. b. , & tinsley , b. m. 1978 , , 219 , 46 schombert , j. m. , wallin , j. f. , & struck - marcell , c. 1990 , , 99 , 497 smith , b. j. , struck , c. , & pogge , r. w. 1997 , , 483 , 754 smith , b. j. , & wallin , j. f. 1992 , , 393 , 544 struck , c. 1997 , , 113 , 269 struck , c. 1999 , physics reports , 321 , 1 struck , c. , & smith , b. j. 2003 , , 589 , 157 toomre , a. , & toomre , j. 1972 , , 178 , 623 wallin , j. f. , & stuart , b. v. 1992 , , 399 , 29 | the amount , timing and ultimate location of mass transfer and induced star formation in galaxy collisions are sensitive functions of orbital and galaxy structural parameters .
i discuss the role of detailed case studies and describe the results for two systems , arp 284 and ngc 2207/ic 2163 , that have been studied with both multiwaveband observations , and detailed dynamical models .
the models yield the mass transfer and compressional histories of the encounters and the `` probable causes '' or triggers of individual star - forming regions .
# 1_#1 _ # 1_#1 _ = # 1 1.25 in .125 in .25 in |
for almost four decades , since the discovery of quasars , mounting observational evidence has accumulated that black holes indeed exist in nature .
recent observations ( wilms et al .
2001 ) of the steep emissivity of seyfert 1 galaxy mcg6 - 30 - 15 , indicating strong photon emission at radii near the event horizon ; and observations of the lack of evidence of the expected ion `` dusty '' torus of m87 ( perlman et al .
2001 ) , have prompted astrophysicists to suggest a new energy source .
however , it is hardly a new energy source to relativists , i.e. , those who study einstein s theory of general relativity .
they knew for sometime , at least theoretically , what black holes were capable of doing ( williams 1991 , 1995 ) . williams ( 1995 , 2003 , 2002 ) shows , through theoretical and numerical ( monte carlo ) _ n - particle _ calculations of penrose ( 1969 ) processes , occurring at radii inside the ergosphere of a rotating black hole near the event horizon : including the `` plunging '' regimes ( bardeen , press , & teukolsky 1972 ; williams 1995 ; krolik 1999 ) , that the black hole can yield escaping particles with energies up to @xmath3 gev .
these particles escape in the form of collimated , symmetrical and asymmetrical jets about the polar axis , confirming the existence of intrinsically collimated vortical jets , found theoretical by de felice & calvani ( 1972 ) ; de felice & curir ( 1992 ) ; de felice & carlotto ( 1997 ) ; de felice & zanotti ( 2000 ) : from geometrical studies of particle trajectories in a kerr ( 1963 ) metric .
the kerr metric in general describes the spacetime separation of events in the gravitational field of a rotating compact massive object . in light of the above observational surprises , particularly the steep emissivity of x - rays producing the broad fe k@xmath4 emission line at @xmath5 kev in mcg6 - 30 - 15 ( wilms et al .
2001 ) and similar agns , it appears that gravity has triumphed over proposed forms of electromagnetic energy extraction from a black hole , as will be described in this paper .
this should be of no surprise near the event horizon , where the gravitational forces are so strong that electromagnetic radiation itself becomes trapped .
overall , energy extraction from black holes and the production of their associated jets have been the most poorly understood phenomena of today .
it is clear that gravitational accretion and magnetic fields play a role , but how ?
has been the mystery .
we observe these jets in quasars and microquasars due to supermassive and stellar size black holes , respectively .
therefore , we know that any effective model must have the commonality to explain jets in both systems . at present there are two popular trains of thought associated with energy extraction and the production of jets in black holes : one is that the jets are inherent properties of geodesic trajectories in the kerr metric of a rotating black hole , and thus , can be described by einstein s general theory of relativity ; and the other is that the accretion disk and its magnetic field through magnetohydrodynamics ( mhd ) are producing the jets .
perhaps it could be a combination of the two , with gravity controlling the flow near the event horizon ( williams 2004 ) , and mhd controlling the flow at distances farther away .
the observations of the jet of m87 suggest this may be the case ( junor , biretta , & livio 1999 ; perlman et al .
2001 ) . there are some proposed mhd model calculations using a general relativistic accretion disk that involve having the magnetic field lines of the disk `` anchor '' to conductive ionized particles of the disk , inside the ergospheric region , extracting rotational energy from a kerr black hole , by way of a poynting flux of electromagnetic energy , out to infinity .
such models have been proposed to explain recent observations of possible direct evidence for the extraction of energy from a rotating black hole ( wilms et al .
2001 ) . in this paper , however , i point out problems with such models that make these models highly improbable to be at work , i.e. , extracting the energy needed to be consistent with general observations of sources powered by black holes .
it is agreed by the author that some form of the penrose mechanism is employed , but it is argued below that electromagnetic energy extraction is not an effective way to use this mechanism .
associated problems with such models are described in detail in the appendix . in a classical paper by bardeen et al .
( 1972 ) , astrophysical implausible penrose processes are discussed concerning the breakup of subrelativistic objects in the ergosphere
. however , i point out that the `` penrose - williams '' mechanism , described by williams ( 1995 ) , involves relativistic scattering processes : such processes can be very efficient ( piran , shaham , & katz 1975 ; williams 1995 ) , and do not fall under the category of being implausible due to hydrodynamical constraints ( bardeen et al .
1972 ) , since the incident and target particles in the collisions are already relativistic , having speeds @xmath6 .
the penrose mechanism as described here ( williams 1995 ) has a `` one - on - one '' consistent relationship with accretion disk particles .
for example , particles from the accretion disk can populate the high energy gravitationally blueshifted trapped orbits ( or so - called plunging orbits ) at @xmath7 , the marginal stable orbit ( bardeen et al .
particles in these now populated orbits can undergo penrose processes with lower soft x - ray energy infalling accretion disk photons : penrose compton scattering ( pcs ) produces copious distributions of high energy x - rays and soft @xmath8-rays , and penrose pair production ( ppp ) ( @xmath0 ) produces copious distributions of relativistic electron - positron ( @xmath9 ) pairs , with up to @xmath10% of the particles escaping along vortical orbits ( [ sec : collimation ] ) that circle the polar axis of the kbh many times , as spacetime itself is dragged around because of gravity ( williams 2004 ) .
the particles escape to infinity along well defined four - momentum trajectories , with some intersecting the disk ( i.e. , returning to be reprocessed and/or escaping to infinity ) .
this scenario is particularly consistent with recent observations of mcg6 - 30 - 15 ( [ sec : mcg ] ) , and other black hole sources ( [ sec : observation ] ) . importantly ,
in these penrose processes we do not need the magnetic field of the accretion disk to `` communicate '' between the accretion disk and the black hole .
therefore , there is no need for the blandford and znajek ( bz ) ( 1977 ) proposed type models ( and their many associated problems ) in the direct role of energy extraction from a spinning black hole .
however , their presence appears to be need once particles are on escaping orbits , serving the same effects they do in the jets of protostars , i.e. , appearing to have a dominant role on a large scale , within the weak field limit , at distances outside the strong effects of general relativity . as for producing the observed synchrotron radiation , indicating the present of a magnetic field near the core region , this radiation
could very well be produced by the intrinsically self - induced magnetic field due to the dynamo - like action of the escaping penrose produced @xmath9 pairs , escaping on vortical , coil - like trajectories concentric the polar axis , in the form of a swirling `` current '' plasma .
this , therefore , adds more to the unimportance of the accretion disk magnetic field near the event horizon
. moreover , although suggested to be evidence of rotational magnetic energy extraction from the seyfert 1 galaxy mcg6 - 30 - 15 ( wilms et al . 2001 ) , it appears , as we shall see in this paper , that it is gravitational energy momentum being extracted , in the form of a relativistic particle flux via penrose processes , as described by williams ( 1995 ) , and _ not _ the poynting flux of electromagnetic energy suggested : produced by magnetic field lines torquing the black hole or plunging accretion disk material , as described by the bz - type models . in the penrose - williams mechanism ,
the steep emissivity profile [ @xmath11 of x - ray photons observed ( wilms et al .
2001 ) , requiring a x - ray source that is both powerful and very centrally concentrated ( which can not be explained by standard accretion disk models ) , is consistent with energy being extracted by penrose compton scattering processes , occurring at radii between the marginally bound and marginally stable orbits , @xmath12 and @xmath13 , respectively ( williams 1995 ) .
this black hole source mcg6 - 30 - 15 will be discussed further in [
sec : mcg ] .
nevertheless , once these penrose processes have occurred and particles are on escaping trajectories , they can then interact , say with the expected large scale structure disk magnetic field , at some effective radius @xmath14 where this field becomes important in jet collimation , probably similar to a relative radius existing for the collimated bi - polar jets of protostars , which , as mentioned above , appear to be undergoing some type of bz effect the direct effect is still somewhat unclear .
it appears that the magnetic field of the accretion disk serves to aid in collimating into jets gravitational binding energy release due to gravitational accretion , in both protostars and agns ( or microquasars ) ; however , in the latter it appears that the jets are superimposed with collimated particles from penrose processes .
so , overall , in this paper , an analysis of the penrose mechanism is presented to describe gravitational - particle interactions close to the event horizon at radii @xmath15 and down to the _ photon orbit _ , @xmath16 , for a canonical kbh with @xmath17 ( thorne 1974 ) , where @xmath18 is the angular momentum per unit mass parameter .
in this fully general relativistic description , polar jets of relativistic particles of photons and @xmath9 pairs are produced and collimated by gravity alone , without the necessity of the external magnetic field of the accretion disk .
this theoretical and numerical model of penrose processes can apply to any size black hole , and suggests a complete theory for the extraction of energy momentum from a rotating black hole . in [ sec : model ] a summary of the general formalism of the model is presented . in [ sec : results ] , results of theoretical and numerical calculated luminosities and energies are presented , along with discussion of the escaping particles space momentum trajectories : featuring asymmetrical polar distributions and vortical orbits . also in [ sec : results ] , agreement with observations of specific sources are presented . finally , in
[ sec : conclusions ] a summary and conclusions are presented .
the primary model ( williams 1995 ) consists of a supermassive @xmath19 rotating kerr black hole plus particles from an assumed relativistic bistable thin disk / ion corona [ or torus , i.e. , advection - dominated accretion flow ( adaf ) ] , two - temperature [ separate temperatures for protons ( @xmath20 k ) and electrons ( @xmath21 k ) ] accretion ( williams 2003 ; novikov & thorne 1973 ; eardley & lightman 1975 ; eilek 1980 ; eilek & kafatos 1983 ) . the bistable accretion disk can exist either in the thin disk phase and/or the ion corona ( adaf ) phase , or oscillate between the two ( see williams 2003 and references therein ) in various degrees which could be responsible for the observed variability . the penrose effect as employed here can operated in either phase .
the penrose mechanism is used to extract rotational energy momentum by scattering processes inside the ergosphere ( @xmath22 , in the equatorial plane for @xmath17 ) .
see williams ( 1995 ) for a detailed description of the model .
the `` quasi - penrose '' ( williams 1991 , 1995 ) processes investigated are ( a ) penrose compton scattering ( pcs ) of equatorial low energy radially infalling photons by equatorially confined ( @xmath23 ) and nonequatorially confined ( @xmath24 ) orbiting target electrons , at radii between the marginally bound ( @xmath25 ) and marginally stable ( @xmath26 ) orbits , where @xmath27 is the so - called carter constant ( carter 1968 ) , referred to as the @xmath28 value ( williams 1995 ) ; ( b ) ppp ( @xmath29 ) at @xmath12 ; and ( c ) ppp ( @xmath0 ) by equatorial low energy radially infalling photons and high energy gravitationally blueshifted ( by factor @xmath30 ) nonequatorially confined @xmath8-rays at the photon orbit ( @xmath16 ) , where @xmath31 is the `` blueshift '' factor given by the @xmath32 component of the kerr metric ( see williams 1995 ) .
although in the scattering plane the incident angle of the infalling photon relative to the target particle is expected in general to be at least between @xmath33 ( due to the bending of light and/or inertial frame dragging ) , maximum energy is extracted in the process when the incident angle is @xmath34 , as it is for radially infalling photons ( @xmath35 , where @xmath36 is the azimuthal coordinate angular momentum ) .
note , the target particles are initially in bound ( marginally stable or unstable ) trapped orbits , trapped in the sense of possibly having no other way of escaping save these penrose processes ( bardeen et al .
1972 ; williams 1995 ) .
note also that , as the nonequatorially confined target particle , whose orbital trajectory is derived by williams ( 1991 , 1995 ; see also williams 2002 ) , passes through the equatorial plane , in its bound circular orbit at constant radius , the @xmath28 value , a constant of motion as measured by an observer at infinity ( carter 1968 ; williams 1995 ) , equals @xmath37 , where @xmath38 is the polar coordinate momentum of the particle .
setting @xmath39 , in the carter constant expression for the orbital @xmath28 value , gives the maximum and minimum latitudinal angles of the trajectories about the equatorial plane for wilkins ( 1972 ) `` spherical - like '' nonequatorially confined orbits ( see williams 2003 ) .
these unstable , bound or marginally bound orbits ( equatorial , nonequatorial ) of the target particles are assumed to be populated by accretion disk instability processes and prior penrose processes .
such particles must satisfy conditions to have a turning point at the scattering radius ( note , a bound stable orbit is considered to have a `` perpetual '' turning point ) .
these conditions depend on the orbital conserved parameters of the particle : @xmath40 , the energy , and @xmath36 , or @xmath28 . in williams ( 1995 , 2002 )
such conditions are discussed in detail ; see also the possible scenario discussion in [ sec : asymmetry ] of this present manuscript .
in addition , the `` instability phase , '' during which the target particle orbits are presumed to be populated , could very well be related to the timescale of the prominent observed variabilities of the source .
radial infalling equatorially confined incident photons are assumed , not only for maximum energy extraction but for the simplicity of their geodesics as well , since it appears that an infalling equatorially confined photon will not acquire gravitationally blueshifted orbital energy momentum as measured by an observer at infinity , only frame dragging blueshifted energy ( eq .
[ 2.8@xmath41 of williams 1995 ) .
this is because the @xmath28 value of such photons is zero ( see eq . [ 2.27 ] of williams 1995 ) .
the incoming photons , however , need not be confined to the equatorial plane . in these calculations if equatorially confined infalling photons were not desired , @xmath42 of the initial photon would not be set equal zero .
that is , the model calculation is set up such that one can change the initial energy - momentum four vector components ( or four - momenta ) of the incident and target particles to accommodate any 3-space dimensional geometrical disk configuration .
moreover , in an adaf ( including the ion corona ) , during the infall of particles , through the ergosphere , some of the particles are expected to become trapped in nonequatorial `` spherical - like orbits '' ( wilkins 1972 ) : such orbits would past through the equatorial plane : here is where the scattering takes place in these calculations .
note , the target photons at the photon orbit can only exist in nonequatorially confined orbits ( williams 1995 ) ; this is also pointed out by bardeen ( 1973 ) .
monte carlo _
n - particle _ computer simulations of up to @xmath43 scattering events of infalling accretion disk photons ( normalized to a power - law distribution ) are executed for each computed penrose produced luminosity spectrum ( williams 2003 ) .
energy and momentum ( i.e. , four - momentum ) spectra of escaping particles ( @xmath8-rays , @xmath9 pairs ) , as measured by an observer at infinity , are obtained per each 2000 scattering events per monochromatic infalling photon distribution .
the following constituents are used ( williams 1995 ) : ( 1 ) general relativity is used [ the kerr metric spacetime geometry yields equatorially and nonequatorially confined spherical - like ( wilkins 1972 ) particle orbits and escape conditions , conserved energy and angular momentum parameters , and transformations from the boyer - lindquist coordinate frame ( blf ) to the local nonrotating frame ( lnrf ) ] .
note , blf is the observer at infinity ( boyer & lindquist 1967 ) ; lnrf is the local minkowski ( flat ) spacetime . ( 2
) special relativity is used [ in the lnrf , physical processes ( i.e. , the scatterings ) are done ; lorentz transformations between inertial frames are performed ; and lorentz invariant laws are applied ] .
( 3 ) cross sections are used [ application of the monte carlo method to the cross sections , in the electron rest frame for pcs , in the proton rest frame for ppp(@xmath29 ) , and in the center of momentum frame for ppp(@xmath0 ) , give the distributions of scattering angles and final energies ] .
in general , energies attained using the proposed accretion disk model are the following ( williams 1995 ) : i. _ pcs_.for the input photon energy range @xmath44 kev to 0.15 mev , the corresponding output energy range is @xmath45 3 kev to 7 mev . the input photon range covers the range of photons in a thin disk ( @xmath46 kev ) , thin disk / ion corona ( @xmath46 kev , @xmath47 kev ) , and adaf ( @xmath47 kev ) for a @xmath19 kbh ( williams 2003 , 1995 ) .
the input luminosity spectra are based on observations , consistent with a power - law distribution in the x - ray , and accretion disk theory .
typical output luminosity spectral distributions from pcs are displayed in figure 1@xmath18 , the curves passing through numbers @xmath48 ( as will be described below in the discussion of the model produced luminosity ) .
( @xmath29).there are no escaping pairs for radially infalling equatorially confined @xmath8-rays ( @xmath4940 mev ) and no energy boost : implying that the assumption : negligible recoil energy given to the proton , made in the conventional cross section , and perhaps the geometry of the scattering must be modified .
it had been predicted ( leiter & kafatos 1978 ) that pairs with energies ( @xmath491 gev ) could escape .
see williams ( 1995 ) for further details of this ppp process .
( @xmath0).an input photon energy range @xmath45 3.5 kev to 200 mev yields output ( @xmath9 ) energy range @xmath45 1 mev to 10 gev ( for a proton maxwell - boltzmann distribution ) , and higher up to @xmath45 54 gev ( for a proton power - law distribution , with input photon energy @xmath50 gev ) , where maxwell - boltzman and power - law distributions are for the accretion disk protons : undergoing nuclear proton - proton scatterings , which yield neutral pion decays @xmath51 ( eilek 1980 ; eilek & kafatos 1983 ; mahadevan , narayan , & krolik 1997 ) that can possibly populate the photon orbit .
below , i refer to such decay produced photons and subsequent @xmath9 pair production ( from such photons ) , which can occur in adafs , as eilek s particles ( eilek 1980 ) .
specific disk model correlations are the following [ see williams ( 1995 , 2003 ) for further details ] : \1 . without instabilities [ implying the classical thin disk ( novikov & thorne 1973 ) ] : * pcs can convert infalling ( incident ) soft x - rays [email protected] kev to moderate x - rays , escaping with energies in the range @xmath45 3@xmath52262 kev . the upper and lower bounds on the energy of the outgoing photons are set by the initial four - momentum conditions of the target orbiting electron ( with @xmath53 mev at @xmath12 ) and the incident photon ( with @xmath54 kev ) undergoing pcs
. these initial four - momenta are consistent in general with the following : theoretical accretion disk models , the threshold energy values for the scattering process to occur , and what brings about the most `` efficient '' energy extraction process [ see williams ( 1995 ) for details defining the efficiencies ] .
these initial momenta are substituted into appropriate theoretical analytically derived model equations ( describing the penrose scattering process in the ergosphere of a kerr black hole ; see williams 1995 ) , and the equations are computed .
the output energy range presented above gives the lowest and highest energy values obtained by the escaping penrose compton scattered ( pcs ) photons , for the given input energy range of the incident photons . *
inwardly directed pcs photons that have an appropriate turning point ( see williams 2002 ) can serve as seed @xmath8-rays for the ppp ( @xmath0 ) at the photon orbit . specifically , the pcs photons that satisfy conditions to have a turning point , acquire gravitationally blueshifted energies as high as @xmath55 mev .
* ppp ( @xmath0 ) can convert infalling soft x - ray photons to relativistic @xmath9 pairs , escaping with energies in the range @xmath45 2@xmath526 mev , i.e. , infalling photons can pair produce at the photon orbit with photons populated by prior pcs . \2 . with instabilities ( implying the thin disk / ion corona or adaf ) : * pcs can convert infalling x - rays [email protected] mev to escaping photon energies in the range @xmath45 0.4@xmath527 mev .
the differences in the calculations of the above case in item 1 and the present case of item 2 are the energies of the infalling incident photons and the target electrons ( @xmath56 mev ) , including nonequatorially confined targets , with @xmath57 mev , for @xmath58 , corresponding to adafs with @xmath59 , respectively , after being gravitationally blueshifted at @xmath12 by factor @xmath60 , where @xmath61 is the polar coordinate momentum component as measured by an observer at infinity ( i.e. , in the blf ; see [ sec : model ] ) .
[ note , the conserved energy @xmath62 and azimuthal angular momentum @xmath63 of the target nonequatorially confined test particle orbits are given by analytically derived expressions presented in williams ( 2004 , 2002 , 1995 ) . ] the accretion disk model , used , is discussed in detail in williams ( 2003 ; see figure 1 and table 1 of that reference ) . in general , the target electron orbits are assumed to be populated during instability phases , more or less , in both the thin disk and thin disk / ion corona ( or adaf ) . in [ sec : asymmetry ] in the discussion of a possible scenario for `` jet reversal , '' a brief description is included on populating the target electron orbits from the inner region of a thin disk ( novikov & thorne 1973 ) . * pcs photons that can populate the photon orbit ( as in item 1.@xmath64 ) have gravitationally blueshifted energies specifically in the range @xmath65 mev , where @xmath66 is the energy at the photon orbit due to prior pcs ( williams 2002 ) .
the lower limits of @xmath66 are due to pcs by equatorially [ @xmath53 mev , @xmath67 and nonequatorially [ @xmath68 mev , @xmath69 confined electron targets , respectively : note , as in item 2.@xmath18 , these nonequatorially confined target electrons are assumed to come from an adaf ( @xmath70 kev ) .
* ppp ( @xmath0 ) can convert infalling soft x - ray photons to relativistic @xmath9 pairs , escaping with energies in the range @xmath71 mev .
note , the `` stability '' of a turning point being perpetual ( i.e. , bound ) at the photon orbit decreases with increasing energy of the incoming incident photons undergoing pcs by equatorially confined target electrons .
these calculations show that the most stable orbits ( or turning points ) appear to be the ones in which the infalling incident photons and the orbiting target electrons are self - consistent , i.e. , of the same accretion disk phase ( e.g. , thin disk or adaf ) .
note , whenever the thin disk is present , the processes described above in items 2.@xmath18@xmath522.@xmath72 will occur in addition to those described in items 1.@xmath18@xmath521.@xmath72 .
\3 . with instabilities [ implying the thin disk / ion corona model or adaf plus eilek s particles ( eilek 1980 ; eilek & kafatos 1983 ) to populate the target particle orbits , of electrons and photons , particularly the large @xmath28-value orbits ; see williams ( 1995 ) ] : * pcs can convert infalling photons [email protected] mev to escaping energies in the range @xmath45 6@xmath5214 mev
this is in addition to the energy distribution of escaping pcs photons given in item 2.@xmath18 .
eilek s particles contribute to the ion corona , nonequatorially confined @xmath9 pairs with energies peak around @xmath73 mev . at the peak , such electrons with inward trajectories ( @xmath74 )
would have to satisfy conditions to have a turning point at @xmath75 ( williams 2004 ) , requiring , say for the scattering radius @xmath12 , @xmath76 and/or @xmath77 ; these electrons do not appear to be important in the pcs process .
observations suggest , however , that pcs by eilek s nonequatorially confined @xmath9 pairs with @xmath78 mev , yielding escaping energies in the range given above , might be important , requiring , for turning points to exist at @xmath12 , @xmath79 and/or @xmath80 , respectively ( compare fig . 1@xmath18 ; see also williams 2003 ) .
[ note , eilek s electrons @xmath81 mev may be an important source of synchrotron emission into the ir , for a magnetic field strength @xmath82 g ( see [ sec : asymmetry ] ) . ] * ppp ( @xmath0 ) can convert infalling soft x - rays to relativistic @xmath9 pairs , escaping with energies ranging from @xmath45 300 mev to as high @xmath83 gev [ for a proton maxwell - boltzman distribution ( see item iii above ) ] , with input photon energy @xmath84 mev from @xmath85 decays ( eilek & kafatos 1983 ) .
that is , the input ( target ) photons are gravitationally blueshifted ( by factor @xmath86 ) at the photons orbit to energies @xmath87 , and are assumed to have a turning point at ( or near ) this scattering radius , with @xmath88 , respectively , where the subscript @xmath89 represents the orbiting target photon .
these ppp ( @xmath0 ) processes occur in additions to those given in item 2.@xmath72
. the exact range of the ppp electrons will depend on which of the inwardly directed photons , after being gravitationally blueshifted , satisfy conditions to have a turning point at or near the photon orbit [ see williams ( 2002 ) for details ] . for completeness , if protons of eilek s particles have a power - law distribution ( mahadevan et al .
1997 ) as mentioned in item iii , the maximum energy attained , using the penrose - williams model , for the escaping ppp ( @xmath0 ) electrons , is @xmath3 gev , for input photon energy @xmath50 gev , after being gravitationally blueshifted to @xmath90 gev , with @xmath91 .
note that , there will be a slight time delay between pcs and ppp ( @xmath0 ) in items 1.@xmath18@xmath521.@xmath72 and 2.@xmath18@xmath522.@xmath72 that might be consistent with the time offset ( @xmath92 min ) between x - ray and ir flares observed in microquasar grs 1915 + 105 , indicating that these flares are produced by the same event : the x - ray flares occur with the apparent disappearance of the inner x - ray emitting region of the accretion disk ; and the subsequent ir flares are proposed to be due to synchrotron emitting ejecta of relativistic plasma into the polar direction ( eikenberry et al .
1999a ; eikenberry et al .
1999b ) .
before discussing the luminosity spectra produced by these penrose processes , we first discuss the `` characteristic voids , '' existing , in general , in observed spectra of all agns , more or less ( compare fig .
1@xmath18 ) , and how these penrose processes suggest an explanation for them .
these observed voids appear to be caused by the `` transitional energy regime '' between thin disk ( @xmath93 kev ) and ion corona ( or adaf ) ( @xmath94 kev ) states ( see table 1 of williams 2003 ) : therefore , we expect the penrose process to be void of participating particles with energy in the range @xmath95 kev , i.e. , if we assume such particles are sufficiently short - lived , save the infalling disk electrons with @xmath96 kev that can satisfy conditions to populate the equatorially confined target electron orbits ( @xmath23 ) at radii @xmath97 ( see [ sec : asymmetry ] and paragraph below ) , where @xmath98 indicates a general particle energy , predicted theoretically by a particular phase of the accretion disk .
the pcs photon energies @xmath99 produced by incident and target particles in the transitional energy regime , and the subsequent gravitationally blueshifted energy @xmath66 at the photon orbit , of incoming pcs photons satisfying conditions to have a turning point there ( williams 2002 ) , expected to undergo ppp ( @xmath0 ) , are found to give characteristic voids in the following regimes : * for thin disk / ion corona ( @xmath70 kev ) : @xmath100 mev and @xmath101 mev , where the upper limit originates from nonequatorially confined target electrons , consistent with the electron temperature in the ion corona ( see williams 2003 ) .
compare with item 2.@xmath64 above .
* for thin disk / ion corona ( @xmath102 kev ) : @xmath103 mev and @xmath104 mev . * for thin disk / ion corona ( @xmath105 kev ) : @xmath106 mev and @xmath107 mev .
note , the pcs processes considered above are those occurring at or near @xmath12 : since the highest energy will be extracted from this scattering radius , and it appears that the orbits at this radius will be the first to be populated , as the disk temperature increases ( [ sec : asymmetry ] ) , i.e. , because of the larger energy blueshift factor acquired , and the smaller @xmath27 needed , relative to these parameters at @xmath13 .
the disk electron energies @xmath108 kev but @xmath109 , being gravitationally blueshifted by a factor @xmath110 ( see [ sec : model ] ) , satisfying appropriate turning point conditions ( williams 2004 ) , with @xmath111 , are assumed to populate the nonequatorially confined target particle orbits for pcs ( see also [ sec : asymmetry ] ) , where @xmath112 mev is the rest mass energy of an electron .
a relativistic four - momentum treatment of disk particle processes in thin disk / ion corona accretion , inside the ergosphere , appears to be needed to theoretically validate this plausible assumption : at present , however , we do not have such a model ; therefore , we must rely on what observations convey to us .
moreover , pcs by equatorially confined electron targets ( @xmath23 ) , assuming to originate from `` mild '' instabilities in the thin disk ( that would cause the electron energy to increase to @xmath113 kev , however , while still predominantly in the thin disk phase ) and radially infalling photons confined along the equatorial plane , originating from the ion corona ( @xmath114 kev ) , are not included in the above consideration of the characteristic voids .
the reason for the exclusion is that observations suggest such pcs may not be important ( compare fig . 1@xmath18 , curve between points 6 and 7 ) , which could mean that these target orbits are depopulated while the disk is in the thin disk or transient phase , and therefore not available for pcs in the ion corona ( adaf ) phase .
further , for the subsequent gravitationally blueshift of such inward directed pcs photons , with @xmath115 mev , @xmath116 mev , @xmath117 mev , corresponding to potential turning point energies : @xmath118 mev , @xmath119 mev , @xmath120 mev , at the photon orbit , for the ion coronas in items ( @xmath18)@xmath52(@xmath72 ) , respectively , we find that most of these potential turning points , it seems , are `` highly '' uncertain ( based on whether @xmath121 for @xmath122 ) .
nevertheless , as the energy of the trapped target electron is increased ( @xmath24 ) , consistent with the general ion corona electron temperature , the uncertainty of the turning point orbit , being true , decreases ; then @xmath123 and @xmath124 are as given above in items ( @xmath18)@xmath52(@xmath72 ) . finally , in the above characteristic voids , pcs involving thin disks with energies less than @xmath125 kev , and pcs involving eilek s nonequatorially confined @xmath9 pairs , possibly occurring in the ion corona or adaf ( particularly @xmath126 mev , the range , based on observations , that appears to be the regime satisfying turning point conditions ) , are not included .
inclusion of these would slightly affect the voids , yet the distinctive characteristics would remain .
we will return to this discussion of the characteristic voids later in this section .
the luminosity spectrum due to penrose processes for the specific case of quasar 3c 273 is plotted in figure 1@xmath18 , along with the observed spectrum for comparison ( heavy solid curves superimposed with squares or dots ) . the outgoing ( escaping ) luminosity spectrum produced by the penrose scattered particles
is given by ( williams 2003 ) @xmath127 where @xmath128 is the cosmological distance of the black hole source ; @xmath129 is the flux of escaping photons ; @xmath130 and @xmath131 are the emittance of incoming and captured photons , respectively ; the @xmath132 values define the total fraction of the particles that undergoes scattering [ @xmath133 for pcs and @xmath134 for ppp ( @xmath0 ) ] .
the values of @xmath135 are the fitting factors , which can make the penrose calculated luminosities agree with observations for the specific case of 3c 273 , to account for in general our letting every particle scatter in the model calculations , since in a realistic situation every particle will not scatter . in short , the @xmath132 values , defined as somewhat free parameters , are probabilities , which are @xmath136 , but @xmath137 ; they are dependent on the cross sections for pcs and ppp ( @xmath138 , @xmath139 ) , the fraction of the luminosity from the disk intersecting the scattering radii ( @xmath140 , @xmath141 ) , and the expansion rate of the jet ( @xmath142 ) .
note , from equation ( [ eq : lum ] ) we obtain the model calculated continuum emission given by the top curves on figure 1@xmath18 ( labeled with numbers for specific cases of target and incident particles ; see below ) if we allow @xmath143 , and set the remaining @xmath132 s equal 1 , where we are assuming that the polar angle subtending the bandwidth , @xmath144 , straddling the equatorial plane , impinged by the luminosity , is @xmath145 at @xmath12 and @xmath146 at @xmath147 ) .
see williams ( 2003 ) for further details and complete definitions of the @xmath132 values .
the spectrum resulting from the ppp ( @xmath0 ) is produced by letting the escaping pairs undergo `` secondary penrose compton scattering '' ( spcs ) with low energy ( @xmath148 mev ) radially infalling equatorial accretion disk photons ( @xmath149 ) .
tables 1 and 2 give model parameters corresponding to some of the numbers on figure 1@xmath18 [ see williams ( 2003 ) for other numbers ] . on these tables
the parameters are defined as follows : @xmath14 is the scattering radius ; @xmath150 is the target electron energy for pcs ; @xmath151 is the energy value where most of the ppp ( @xmath0 ) electrons , used as targets for the spcs , are created ; @xmath152 is the initial infalling incident photon frequency ; @xmath153 and @xmath154 correspond to the points ( solid squares or dots superimposed on the small - dotted or dashed curve , respectively ) which give the continuum luminosity resulting from several distributions of pcs or spcs events ( each distribution has 2000 scattering events ) ; @xmath155 is the observed luminosity at @xmath153 ( the average frequency of the interval @xmath156 where most of pcs or spcs photons are emitted per 2000 scattering events ) .
each distribution of 2000 infalling photons have monochromatic energies normalized to the power - law distribution for 3c 273 based on observations .
the @xmath132 values given in the brackets are values used to fit the general model spectra to agree with specific observations .
overall , to produce the calculated penrose luminosity spectra of figure 1@xmath18 , 74,000 infalling photon scattering events are used .
thus , as one can see from figure 1@xmath18 , the penrose - williams mechanism can generate the necessary luminosity observed , and the three model calculated regions of emission [ due to pcs by equatorially confined targets ( curve passing through nos .
@xmath157 ) , by nonequatorially confined targets that cross the equatorial plane ( curve passing up from 6 through nos .
@xmath158 ) , and ppp ( @xmath0 ) ( curve passing through nos .
@xmath159 ) ] are consistent with the three major regions of emission in all quasars and agns . moreover , taking into consideration the characteristic voids , discussed earlier , proposed to be produced by the different phases of the accretion disk , the lack of participating particles for pcs would cause a void between points 5 and 7 ( @xmath160 mev ) on figure 1@xmath18 , suggesting a transitional energy regime between a thin disk ( @xmath161 kev ) and ion corona ( @xmath162 kev ) . comparing the observed spectrum of 3c 273 to the model calculated spectrum , it appears that this quasar has a similar accretion disk structure .
this suggests that a second void should occur between @xmath163 mev , as it does , agreeing strikingly well with observations between the energies of points @xmath164 and @xmath165 .
further , comparing figures 8@xmath18 and 8@xmath64 of williams ( 2003 ) , where figure 8@xmath64 is the same as figure 1@xmath18 of this present manuscript , figure 8@xmath18 ( quasar 3c 379 ) does not appear to have an appreciable inner region thin disk to effectively populate and depopulate equatorially confined target electron orbits for pcs , which would give energies up to @xmath166 kev , like that of figure 8@xmath64 .
the lack of populating the equatorially confined target orbits suggests that @xmath24 for the disk electrons , implying the presence of an ion corona or adaf , and leads to speculation that perhaps the equatorially confined ( @xmath23 ) particles were `` lost '' in a prior thermal , yet cooler , instability phase .
this interpretation for the accretion is consistent with observations of 3c 279 , displayed in figure 8@xmath18 : 3c 279 appears to have an effective radiating ion corona up to @xmath167 kev , similar to the case of item ( @xmath64 ) , in identifying the voids , but it lacks appreciable evidence of an inner region thin disk , at least at this time of observation i.e .
, since 3c 279 is classified as an optically violently variable ( ovv ) quasar ( see williams 2003 ) .
such adaf phase , presumed for 3c 279 , could produce particles to populate the nonequatorially confined target electron orbits for pcs and subsequent ppp ( @xmath0 ; see items 2.@xmath18@xmath522.@xmath72 above ) and could satisfy the conditions to produce eilek s high energy particles ( see items 3.@xmath18 and 3.@xmath64 above ; compare also table 1 of williams 2003 ) , giving rise to a model calculated spectrum consistent with the observed spectrum of 3c 279 ( compare fig .
8@xmath18 ; see williams 2003 for further details ; see also [ sec:3c273 ] ) .
the observed spectra of microquasars ( or galactic black holes ) , in general , appear not to have pcs emission by the nonequatorially confined target electrons , neither the highest energy @xmath8-ray emission due to ppp ( @xmath0 ) , suggesting that these sources may not have an ion corona ( or adaf ) , which would be need to populate the orbits to generate such emission , at least in the highest energy regime ( compare figure 1@xmath18 ) .
general calculated spectra resulting from a self - consistent thin disk penrose process model for stellar mass black holes ( @xmath168 ) appear like a scaled - down figure 1@xmath18 ( with photon luminosity @xmath169 for total energy range @xmath170 kev@xmath528 mev ) , without any appreciable curve labeled between points 6@xmath5213 ( williams & hjellming 2002 ) ; see [ sec : cyg ] .
this is consistent with observations of galactic black holes ( liang 1998 ) .
the gravitomagnetic ( gm ) force field is the gravitational analog of a magnetic field .
it is the additional gravitational force that a rotating mass produces on a test particle .
the gm force is produced by the gradient of @xmath171 , where @xmath172 is the frame dragging velocity ( bardeen et al .
1972 ) and @xmath173 is the gm potential ( thorne , price , & macdonald 1986 ) .
analysis of the equations governing the trajectories of the penrose process particles shows that the gm force , which acts proportional to the momentum of a particle , alters the incoming and outgoing momentum parameters of the incident and scattered particles , resulting in asymmetrical polar distributions , and thus , appearing to break the reflection symmetry of the kerr metric , above and below the equatorial plane ( williams 2004 , 2002 , 2003 , 1999 ) .
effects of the gm force acting on the ppp ( @xmath0 ) process can be discerned from comparing figures 1@xmath128 and 1@xmath174 . when half of the 2000 target photons are allowed to have initial polar coordinate momentum @xmath175 and the other half @xmath176 , of equal absolute values , with increasing @xmath177 , the @xmath9 `` jet ( @xmath178 ) to counter - jet ( @xmath179 ) '' ratio @xmath180 achieves a maximum @xmath181 , favoring @xmath182 ( williams 2002 ) , as seen in figure 1@xmath174 ( compare fig .
1@xmath128 ) .
the corresponding polar angles of escape for cases of figures 1@xmath128 and 1@xmath174 are displayed in figures 2@xmath18 and 2@xmath64 , respectively .
polar coordinate momentum distributions , @xmath183 , for escaping pcs photons are displayed in figure 3 , where the primes indicate final conditions .
the corresponding polar angles of escape for the cases of figure 3 are given in figure 4 .
notice the effects of the gm force field causing the ( photon jet to counter - jet ) ratio @xmath184 to vary from nearly symmetric to asymmetric for the different cases shown . of these cases
the largest ratio achieved is @xmath185 ( figs .
3@xmath72 and 4@xmath72 ) . the direct cause of the asymmetry in the polar direction appears to be due to the severe inertial frame dragging in the ergosphere in which the gm field lines are spacetime dragged in the direction that the black hole is rotating [ see williams ( 2002 ) for details ;
see also williams 2004 ] .
the resulting gm force acting on the particles produces the asymmetry . in most cases ,
the distribution favors the @xmath186 direction ( see figs . 3 and 5@xmath128 ) ; however , at particularly low energies , the asymmetry appears to reverse .
for example , in the case of ppp ( @xmath0 ) at the low initial energies @xmath187 kev and @xmath188 mev for the infalling and orbiting photons , respectively , producing escaping @xmath9 pairs with energies peak around @xmath189 mev , @xmath190 per 2000 events ( fig .
5@xmath18 ) , and after undergoing spcs ( williams 2003 ) per 2000 infalling disk photons ( @xmath191 kev ) , the asymmetry in the final photon polar distribution , for the spcs , is reverse , with the inverse of the number of particles scattered in the positive polar direction to that in the negative direction @xmath192^{-1}=402/165\simeq 2.44 $ ] , favoring the @xmath179 direction ( fig .
5@xmath64 ) .
this would make the @xmath179 jet appear more energetic and , thus , brighter , since the ppp @xmath9 polar jets , in this case , are nearly symmetrical , as can be seen in fig .
such behavior is consistent with hjellming and rupen s ( 1995 ) observations of gro j1655 - 40 .
these authors concluded that the jets themselves must be intrinsically asymmetric , and the sense of the asymmetry must change from event to event .
moreover , they found that the jets lie almost in the plane of the sky , so relativistic beaming can not explain the observed brightness ratios .
[ note , the potential for `` jet reversal '' due to the gm force field can be seen in eq . ( 47 ) of williams ( 2002 ) and eq .
( 8) of williams ( 2004 ) : occurring for particle distributions with relatively large @xmath193 and/or relatively small @xmath194 ( corresponding to small @xmath195 ) . ]
also , the jet space velocity lorentz factor found by these authors ( @xmath196^{-1/2}\approx 2.5{\rightarrow}e\sim 1.3 $ ] mev ) is consistent with the target electron energy , of the spcs , we have found here , displaying the jet reversal ( compare @xmath197 stated above and figs .
5@xmath18 and 5@xmath64 ) , where we are assuming that the bulk velocity of a `` blob '' is @xmath198 `` average '' space velocity of the individual ppp electrons per bulk distribution , i.e. , assuming @xmath199 , valid at least in the case of the small scale , fast varying galactic black holes .
the model calculated space velocities of the ppp @xmath9 jet particles for initial energies consistent with observed microquasars ( with @xmath200 ) are in the range @xmath201 , for @xmath202 mev , implying @xmath203 , respectively ; compare fig .
thus , the consistency of apparent jet reversal , of these penrose processes , with observations , gives more compelling evidence that it is probably the penrose - williams mechanism at work , close to the event horizon , within @xmath13 , extracting rotational gravitational energy momentum : in the form of a particle flux , as opposed to the so - called bz - type models , proposed to extract energy and momentum : in the form of electromagnetic poynting flux and alfvn waves , respectively ( with the major problem still existing of converting to the necessary particle flux to fuel the observed jets ) .
note , a specific possible scenario for the jet reversal in the case of a @xmath205 microquasar , similar to that of gro j1655 - 40 ( hjellming & rupen 1995 ) , for a classical thin relativistic accretion disk ( novikov & thorne 1973 ) , is the following : as secular density and thermal instabilities begin occurring in the inner region of a time dependent accretion disk commonly referred to as the `` lightman instabilities '' ( lightman 1974a , 1974b ; williams 1995 , 2003 ) , @xmath206 increases to a `` reasonable '' maximum @xmath207 kev , being consistent with observations .
the infalling disk particle electrons with energies @xmath208 , and satisfying conditions for a turning point to exist at specific radii ( williams 1995 ) between @xmath209 , respectively , will be gravitational blueshifted according to the blueshift factor : @xmath210 , respectively ( recall discussion in
[ sec : model ] , first paragraph ) , populating the equatorially confined ( @xmath211 ) target electron orbits with @xmath212 mev , respectively , for pcs ( williams 1995 ) .
this appears to be the catalyst to `` turn on '' the self - consistent penrose - williams mechanism .
[ note , the above reasonable maximum energy means before the critical surface density @xmath213 is reached ( lightman 1974a , 1974b ) , which causes the ion coronal / torus two - temperature phase to set in , or before the inner `` hot '' region ( novikov & thorne 1973 ) extends to @xmath214 ( eardley & lightman 1975 ) , for @xmath215 , @xmath216 , and @xmath217 , where @xmath4 is the viscosity parameter , @xmath218 the kompaneets parameter , and @xmath219 the sub - eddington accretion rate ; see williams ( 2003 ) . ]
the subsequent escaping pcs x - ray emission becomes more and more asymmetric , favoring the @xmath186 direction , as the infalling initial photon energy is increased , say due to disk instabilities ( compare figs .
3@xmath18 and 3@xmath64 ) .
as pcs of infalling disk photons ( @xmath191 kev ) depopulates the equatorially confined target electron orbits , some of the photons with @xmath220 , @xmath221 , and @xmath222 , after being gravitationally blueshifted by factor @xmath86 , satisfy conditions to have a turning point at the photon orbit ( williams 1995 , 2002 ) , populating , and thus supplying target photons for ppp ( @xmath0 ) in the range of @xmath223 , respectively , for the range of @xmath99 above , where @xmath224 . for @xmath225 mev , @xmath226 , as given by the analytical derived expressions of the conserved energy @xmath40 and angular momentum @xmath36 of nonequatorially confined particle trajectories ( see williams 1995 , 2002 , 2004 ) , the subsequent ppp ( @xmath0 ) with infalling disk photons @xmath227 ( assuming negligible electrons are left in the equatorially confined orbits between @xmath13 , @xmath12 ) will produce slightly asymmetrical jets ( favoring the @xmath186 direction ; compare fig .
5@xmath18 ) .
the total energetics due to pcs and ppp ( @xmath0 ) at this phase will favor @xmath178 , therefore , producing a `` brighter '' jet in this polar direction .
however , when some of these ppp electrons subsequently interact with infalling disk photons through spcs , the final emitted escaping photon jets undergo apparent reversal ( favoring @xmath179 ; compare figs .
5@xmath64 and 5@xmath72 ) : thus , the total energetics will now favor the @xmath228 direction .
compare figures 3@xmath18 , 3@xmath64 , and 5@xmath18@xmath525@xmath72 , considering the observed time delays between outbursts ( hjellming & rupen 1995 ; eikenberry et al .
1999a ) and those expected between the different penrose processes : pcs , ppp ( @xmath0 ) , spcs ; and the synchrotron emission ( @xmath229 ghz ) by the escaping ppp electrons ( of fig .
5@xmath18 ) : due to , perhaps , their expected intrinsic magnetic field ( or an external accretion disk magnetic field ) , according to @xmath230 ( burbidge , jones , & odell 1974 ) , for @xmath82 g ( williams 2003 ; this assumed value , although consistent with observations in many instances , needs further investigation ) .
in addition , some of the ppp electrons ( of fig .
5@xmath18 ) will be created with @xmath231 mev , @xmath232 , @xmath233 , and @xmath234 ( recall that @xmath235 ) , satisfying the condition to have a turning point at the iso - energy orbit @xmath236 ( circular orbit of equal energy at constant radius @xmath237 ; see williams 1995 , 2004 ) , with @xmath238 and @xmath239 at radii @xmath240 ( the last bound orbit for a material particle , deep within the ergosphere ) , before escaping to infinity along vortical orbits ( [ sec : collimation ] ) , satisfying ( williams 1995 ) @xmath241 or @xmath242 , implying no turning point in @xmath243 , i.e. , @xmath244 , yet @xmath245 .
note , this satisfying of the condition to have a turning point at @xmath240 , before escaping to infinity along vortical trajectories , is also true for the supermassive kbh ( williams 2004 ) .
observations of gro j1655 - 40 ( hjellming & rupen 1995 ) suggest that after the jet outbursts : due to lightman instabilities , inner region disk depletion , penrose processes , and plunging orbit ( bardeen et al .
1972 ) population - depopulation processes , the disk settles back down to its low , `` initial '' state , to prepare once again to repeat the total _ disk instability - penrose emission cycle _ , as described above , indefinitely ( i.e. , as long as there exists available matter to accrete ) .
moreover , the disk instabilities are expected to change the accretion rate , thereby causing the penrose processes to vary .
so , in conclusion of this section , it appears that once the initial requirement has been met : of populating the equatorially confined target electron bound , unstable orbits , inside the ergosphere , between @xmath75 [ at @xmath246 for maximum pcs energy extraction ( williams 1995 ) ] , the kbh operates as a self - consistent system , emitting @xmath9 and photon jets , relying only on the accretion disk to supply the incident infalling photons , and to populate the initial equatorially confined electron target orbits [ i.e. , due to disk instabilities ( kafatos & leiter 1979)]indicating the beginning of the `` cycle . '' and within this cycle for particularly low particle initial energies , the gm field can cause the jet brightness asymmetry to reverse .
[ note , see williams ( 2002 ) for a complete description of the relations between the gm field and the space momenta displayed in the figures shown here . ] in addition , in the case of quasars - type agns ( williams 1995 , 2003 ) , it appears that an adaf is needed to populate the relatively high energy nonequatorially confined target electron orbits for pcs ; and to populate the highest energy photons at the photon orbit for ppp ( @xmath0 ) , yielding maximum escaping energies @xmath247 gev ( as discussed in [ sec : energy - luminosity ] ) .
it is found that the penrose scattered particles escape along vortical trajectories collimated about the polar axis ( williams 1995 , 2000 , 2003 , 2004 ) .
these distributions are fluxes of coil - like trajectories of relativistic jet - type particles , escaping out from the equatorial plane at the scattering radius @xmath248 , concentric the polar axis .
the highest energy particles have the largest @xmath249 values ( compare fig . 1@xmath72 ; compare also figs .
3@xmath64 and 4@xmath64 of williams 1995 ) .
note , @xmath250 is negative ( inward toward the polar axis ) for many of the pcs photons ( williams 2002 ) , and positive for all of the @xmath9 pairs ( compare fig .
1@xmath64 ) .
the helical angle of escape @xmath251 , of particle type @xmath252 , relative to the equatorial plane , for the highest energy scattered particles ranges from @xmath253 to @xmath2 for pcs ( compare fig . 4 ) and @xmath254 to @xmath255 for the @xmath9 pairs ( compare fig .
2 ) ; compare also figs . 6 , 7 , and 9 of williams ( 2002 ) .
the above characteristics of the escaping particles , along with their @xmath256 values ( compare figs .
1@xmath128 , 1@xmath174 , 3 , and 5 ) , imply strong collimation about the polar axis , giving rise to relativistic jets with particle velocities up to @xmath6 ( compare fig .
1@xmath204 ) .
note , such vortical trajectories and collimation are consistent with the findings of de felice et al .
( de felice & curir 1992 , de felice & carlotto 1997 , de felice & zanotti 2000 ) , from spacetime geometrical studies of general particle geodesics in a kerr metric .
moreover , the gm force field , discussed in the last section , responsible for the inertial frame dragging and the asymmetrical jets , also serves to boost the jets into opposite polar directions ( williams 2002 ) .
in addition to statements made in [ sec : energy - luminosity ] concerning the model calculated spectra of 3c 273 and 3c 279 , below i summarize some of the important features resulting from application of the penrose - williams mechanism to observations of both 3c 273 and 3c 279 ( see fig .
8@xmath18 of williams 2003 ) .
the observed spectra of both these sources can be explained by these penrose processes and the assumed accretion model : specified in [ sec : model ] [ see williams ( 1995 , 2003 ) for further details ] . as we can see from figure 1@xmath18
, there is a striking similarity between the energy range of the observed spectrum of 3c 273 and the model spectra produced by these penrose processes . upon comparing the spectra of radio - loud quasars 3c 273 and 3c 279 , based on these penrose processes ,
we find the following ( williams 2003 ) : the shape of the observed spectrum of 3c 273 looks like the `` enhanced '' ( i.e. , the highest observed energetic state ) spectrum of 3c 279 , except for the higher luminosities in 3c 279 and the radio tail in 3c 273 . the higher luminosity and the apparent lack of a radio tail in 3c 279
is probably , largely , due to the radiation of 3c 279 being beamed more in the direction of the observer than the radiation of 3c 273 .
therefore , the spectrum of 3c 279 has been doppler blueshifted to an _ observed _ higher energy interval ; and the _ apparent _ luminosity has been increased .
this is consistent with radio observations which detect more superluminal motion ( or relativistic beaming near the line of sight of the observer ) in 3c 279 than in 3c 273 ( porcas 1987 ) . on the other hand
, it seems that 3c 273 has a `` hotter '' inner accretion disk and is in a predominantly bimodal quasi - stable state : appearing to be in the most effective or `` extreme '' thin disk / ion corona state as opposed to 3c 279 : which appears to oscillates in a highly variable fashion between the thin disk and ion corona phases for this reason 3c 279 is classified as an ovv quasar .
the hotter state of the accretion disk ( ion corona ) , which is heated by a runaway thermal instability ( shapiro , lightman , & eardley 1976 ) , would result in enhanced penrose processes [ pcs and ppp ( @xmath0 ) ] , and enhanced synchrotron radiation due to the presence of more relativistic electrons , particularly if eilek s ( eilek 1980 ; eilek & kafatos 1983 ) particle reactions ( @xmath257 ) occur , hence contributing to the prominent observed radio tail of 3c 273 .
this ion corona / adaf state , existing in conjunction with the thin disk , appears to be the case always in the continuum emission of 3c 273 and sometimes in the emission spectrum of 3c 279 , with 3c 279 not quite achieving the full `` hot '' ion corona / adaf status of 3c 273 ( williams 2003 ) , neither achieving the full `` cool '' @xmath258 kev thin disk phase , where more penrose processes would occur to liberate trapped energy .
thus in summary , the differences in the spectra of 3c 279 and 3c 273 are probably due to ( 1 ) the more beaming effect in 3c 279 , and ( 2 ) the predominantly extreme hot , cool phases of 3c 273 .
now , again , based on the `` characteristic voids '' discussed in
[ sec : energy - luminosity ] , 3c 279 appears to be similar to case ( b ) and 3c 273 to that of case ( c ) .
compare models @xmath259 and @xmath260 on table 1 of williams ( 2003 ) : models @xmath259 are similar to 3c 273 , and model @xmath260 is similar to 3c 279 .
recent observations of the bright seyfert 1 galaxy mcg6 - 30 - 15 [ particularly of the broad fe k@xmath4 emission line at @xmath5 kev , believed to be originating from the inner accretion disk plasma ( wilms et al .
2001 ) ] , and other such type agns , are consistent with these model calculations .
a qualitative model calculated scenario to explain the observed spectral observations of mcg6 - 30 - 15 , by these penrose processes , is as follows .
if we assume that the plunging orbits of the target electron , inside the ergosphere , have been populated by accretion disk instabilities ( as described in
[ sec : asymmetry ] ) , self - consistent computer simulations of these penrose processes consistent with mcg6 - 30 - 15 have model parameters for radial infalling photons ( @xmath261 kev ) from a thin disk ( novikov & thorne 1973 ) , that either undergo pcs by equatorially confined orbiting target electrons ( @xmath53 mev ) at @xmath12 , or ppp ( @xmath262 ) at @xmath147 .
the energies ( due to frame dragging ) attained by the @xmath263 up to @xmath264 escaping particles , returning to the disk to be reprocessed and/or escaping to infinity , are the following : for pcs photons , @xmath265 kev for equatorially confined orbiting target electrons , with relative incoming and outgoing photon luminosities @xmath266 , respectively , where @xmath267 . and
for the relativistic ppp electrons ( with @xmath268 and @xmath269 mev ) , @xmath270 mev [ consistent with synchrotron radiation into the radio regime for @xmath271 g , and inverse compton scattering ( spcs of disk photons ) into the x - ray / soft @xmath8-ray regime with relative incoming and outgoing photon luminosities @xmath272 , for @xmath273 , at @xmath274 mev , respectively ] , suggesting relatively weak , less powerful and less prominent radio jets , i.e. , a radio quiet agn , like a seyfert galaxy ( compare figs .
1@xmath18 , 5@xmath18 and 5@xmath64 for similarities and dissimilarities ) .
note , for self - consistency , @xmath177 is assumed based on prior pcs photons with @xmath220 that satisfy conditions for the existence of a turning point at the photon orbit ( williams 2002 ) .
note also that , at these low energies for @xmath275 and @xmath177 , the spcs polar jets appear to `` flip , '' undergoing brightness jet reversal ( as discussed in [ sec : asymmetry ] ) , differing by a factor @xmath276 , in particle numbers , favoring @xmath277 ( compare fig . 3 and figs .
5@xmath278 ) , whereas the initial ppp target electron polar jets , differ by a factor @xmath50 favoring @xmath279 ( compare fig .
5@xmath18 ) . the pcs photon distribution in the range such as @xmath99 above , emitted from @xmath280 , with the highest energy photons concentrated in the equatorial plane ( compare fig . 4 ) ,
is expected to be consistent with the observed extremely steep emissivity profile @xmath281 of wilms et al .
( 2001 ) , indicating that most of the fe k@xmath4 line emission originates from the inner region of a relativistic accretion disk .
specific details of the emissivity @xmath282 of these penrose processes , particularly of the pcs , will be presented in a future paper by the author .
recent radio observations of active galaxy m87 ( junor et al . 1999 ) suggest that electromagnetic collimation becomes important at radii @xmath283 , wherein the initial `` open angle '' of the jet @xmath284 ( at radii @xmath285 ) is made smaller to @xmath286 by the electromagnetic field , where @xmath287 ( @xmath288 , the radius of the ergosphere at the equator ) .
this is consistent with the penrose mechanism providing ( in addition to the relativistic particles ) the initial collimation at radii ( @xmath285 ) , i.e. , closer to the black hole .
since m87 is a giant elliptical galaxy , this could mean that its geometric configuration is possibly helping to maintain the initial collimation by the black hole ( williams 2003 ) : which begins at @xmath289 , and must extend out to at least @xmath290i.e .
, until , it appears , electromagnetic collimation takes over .
however , before one can say for certain of the electromagnetic processes occurring , a time dependent mhd evolution of the penrose escaping particle plasma must be performed ( presently under investigation by the author ) .
it should not be ruled out that the intrinsic collimation due to the black hole , of the escaping relativistic plasma : and any associated `` dynamo '' generated magnetic field , may be sufficient to maintain collimation .
further , concerning m87 , its observed spectrum in general can be explain by the penrose mechanism presented in this paper .
some observational properties of m87 are the following ( eilek 1997 ) : @xmath291 ; striking comparisons of radio ( very large array ) and optical ( hubble space telescope ) images of the jet ; optical and possibly x - ray emission believed to be of synchrotron origin ; and more recently , the mid - ir observations ( perlman et al . 2001 ) showing that the nuclear ir emission is entirely consistent with synchrotron radiation , and there is no evidence for thermal emission from a dusty nuclear torus .
based on these properties the following scenario can be devised according to the penrose - williams mechanism .
the jet is no doubt beamed , since observed superluminal motions give apparent velocities up to @xmath292 , implying line - of - sight angle @xmath293 , bulk lorentz factor @xmath294 , jet doppler factor @xmath295 , and jet brightness boost @xmath296 , respectively ( biretta , sparks , & macchetto 1999 ; see also williams 2003 ) .
m87 is probably an evolve blazar - type agn ( ovv quasar and bl lac object ) .
its luminosity spectrum ( although less powerful , less energetic ) most likely resembles that of 3c 279 ( [ sec:3c273 ] ; see also williams 2003 ) .
the most noticeable change in the spectrum from times past is probably the lack of high energy @xmath8-rays : due to the lack of the availability of infalling low energy ( soft x - ray ) disk photons , or the lack of high energy ppp ( @xmath0 ) electrons , to undergo effective spcs , that would result in escaping trajectories for the scattered @xmath8-rays .
since the jet of m87 is still seen prominently in the radio / optical / x - ray , an optically thin hot ion torus , pcs , ppp ( @xmath0 ) , and subsequently synchrotron radiation of the ppp electrons ( particularly into the optical : implying @xmath297 mev for @xmath298 g , respectively ) , are consistent with the observations .
the parenthetical statement above suggests that the magnetic field producing the synchrotron radiation may be that of the escaping penrose plasma rather than that of the popular proposed large scale dipolar - like field of the accretion disk ( since large - scale , large - strength dipolar accretion disk fields are in practice difficult to create ) ; this however requires an investigation .
moreover , besides coming from the inner region of a relativistic thin disk ( novikov & thorne 1973 ) , there are two possibilities for producing the observed soft x - ray emission , within the confinements of the penrose - williams mechanism : ( 1 ) a synchrotron origin requires electron energies @xmath113 gev ( @xmath299 ) for @xmath50 kev emission at @xmath271 g , and could very well be produced by the ppp ( @xmath0 ) , at least for ultrarelativistic @xmath9 pairs up to @xmath3 gev ( williams 2003 ) for @xmath300 as low as @xmath83 g ; and ( 2 ) the jet is beamed , and self - compton scattering of lower energy radio and ir synchrotron photons by the escaping , intrinsically polar collimated ppp electrons is occurring : the observed energies of the inverse / self - compton scattered photons are blueshifted due to doppler boosting into the optical and x - ray regimes , respectively , according to @xmath301 ( dermer , schlickeiser , & mastichiadis 1992 ) for @xmath302 , @xmath303 .
now , both items above could equally occur , more or less ; however , since superluminal motion appears to be important in m87 , item ( 2 ) is most likely the dominant , somewhat ruling out the other item .
if this dominance is true , then the energies of the jet electrons need only be as high as @xmath304 mev for @xmath271 g. this is consistent with the penrose processes described here , in the presence of a thin disk / ion corona accretion without the need of eilek s @xmath85
decays to populate the photon orbit ( see [ sec : energy - luminosity ] , items @xmath305 ) .
note , such ion coronas or tori are poor radiators , and expected to be of relatively low density , with @xmath306 kev ; this may account for the lack of evidence for an inner `` dusty '' torus emitting thermal radiation in the mid - ir observations by perlman et al .
however , it still seems unlikely that such a low electron energy and particle density ion torus ( or adaf ) can be jet fuel for the bz - type models near the event horizon , inside the ergosphere , as required by such models ( blandford & begelman 1999 ) .
[ see williams ( 2003 ) for a complete description of the accretion disk model consistent with these penrose processes and observations .
] nevertheless , such bz - type models ( e.g. , punsly 1991 ; koide et al .
2000 ) might be important at @xmath307 , as suggested by observations ( junor et al . 1999 ) , particularly if the penrose - williams particles are used as fuel .
the penrose - williams model presented here applies to all mass size kbhs , with the stellar mass black hole appearing as a scaled - down supermassive hole .
when the parameters are expressed in gravitational units ( @xmath308 ) , the penrose process emission energy - momentum spectra ( @xmath309 vs. @xmath40 ; @xmath38 vs. @xmath40 ; @xmath310 vs. @xmath40 ) over the range of masses are approximately identical .
the luminosity spectra of these penrose processes for the different masses , in general , span over a range @xmath311 ( compare figs .
1@xmath18 and 6 ) . in general
, the differences of the penrose process output luminosities between supermassive kbhs and `` micro - massive '' kbhs are determined by the bolometric luminosity of the incoming photons ( eilek 1980 ; williams 2003 ) , directly dependent on the accretion rate , which is governed by the surrounding accretion disk environment .
for example , the observations of the classical stellar / galactic black - hole candidate cygnus x-1 ( liang 1998 ) can be explained by these penrose processes : processes consistent with cyg x-1 have model parameters for radial infalling photons ( @xmath191 kev ) from a thin disk ( novikov & thorne 1973 ) , that either undergo pcs by equatorially confined orbiting target electrons ( @xmath53 mev ) at @xmath12 or ppp ( @xmath262 ) at @xmath147 .
the energies ( due to frame dragging ) attained by the @xmath312 up to @xmath313 escaping particles , returning to the disk to be reprocessed and/or escaping to infinity , are the following : for the pcs photons , @xmath314 kev , with relative incoming and outgoing photon luminosities @xmath315 , respectively , where @xmath316 . and
for the relativistic ppp electrons ( with @xmath317 and @xmath318 mev ) , @xmath319 mev [ consistent with synchrotron radiation into the radio regime for @xmath271 g , and inverse compton scattering ( spcs of disk photons ) into the hard x - rays / soft @xmath8-ray regime with relative incoming and outgoing photon luminosities @xmath320 , for @xmath321 , between @xmath322 kev@xmath323 mev ] ; compare figure 6 ; see williams & hjellming ( 2002 ) for further details .
note , for self - consistency , @xmath177 is assumed based on prior pcs photons with @xmath220 that satisfy conditions for the existence of a turning point at the photon orbit ( williams 2002 ) .
note also that , as in the cases of gro j1655 - 40 ( [ sec : asymmetry ] ) and mcg6 - 30 - 15 ( [ sec : mcg ] ) , at these low energies for @xmath275 and @xmath177 , the spcs polar jets undergo slight so - called jet reversal ( as discussed in [ sec : asymmetry ] ) , differing by a factor @xmath324 favoring @xmath277 , whereas the initial ppp target electron polar jets , differ by a factor @xmath325 favoring @xmath279 ( compare figs . 3@xmath18 , 3@xmath64 , and 5 ) . in the above model for cyg x-1 ,
the ppp electron energy @xmath197 can increase to @xmath326 mev , as the infalling thin disk photon energy for pcs by equatorially confined target electrons is increased to @xmath327 kev ( williams & hjellming 2002 ) , say due to disk instabilities ( compare [ sec : asymmetry ] ) .
this appears to be the case for cyg x-1 when in its `` high '' state ( mcconnell et al .
1989 ) , and to explain the persistent power - law @xmath8-ray tail up to @xmath328 mev ( mcconnell et al .
1994 ) .
now , concerning @xmath329 quasi - periodic oscillations ( qpos ) observed in galactic black holes ( strohmayer 2001 ; remillard et al .
2002 ; abramowicz et al .
2002 ) , such qpos can be predicted from the penrose scattering processes described here .
the `` qpos '' of , say , a given local distribution of neighboring target electrons , responsible for pcs into the x - ray / soft @xmath8-ray regime , emitting from geodesic orbits at radii between @xmath13 and @xmath12 , can be obtained from ( bardeen et al .
1972 ) @xmath330 where @xmath331 is the coordinate angular velocity of a circular orbit ; @xmath332 is the inverse of the blueshift factor ( see [ sec : model ] ) , commonly referred to as the `` redshift '' factor ; @xmath333 is the orbital velocity in the azimuthal direction of the target particles relative to the lnrf , i.e. , as measured by a general observer at rest relative to this frame ( bardeen et al .
1972 ; see also williams 1995 ) ; @xmath334 is the radius of the circumference about the axis of symmetry ( thorne et al .
so , with the frame dragging _ angular _ velocity given by @xmath335 , we find ( from eq . [ [ eq : qpo ] ] ) the predicted range to be between : @xmath336 hz , @xmath337 hz at @xmath13 , @xmath12 , respectively , corresponding to periods @xmath50 ms , as measured by an observer at infinity .
note , the counterpart qpos for a supermassive ( @xmath338 ) kbh are @xmath339 hz ; this relatively low frequency is probably the reason these counterpart qpos have yet to be detected in sources harboring such massive kbhs ( see miller et al . 2002 and references therein ) .
these calculations suggest that @xmath49 khz qpos may also be due to the inertial frame dragging of the nonequatorially confined target particles orbital `` ring '' at scattering radius @xmath14 ( williams 1995 ) , particular of the nodes ( points at which the orbit , in going between negative and positive latitudes , intersects the equatorial plane)which happens to be where the most effective penrose scattering processes would occur , as resulting emitting regions of neighboring particles sweep across the line of sight of the observer . in this case
, the observed oscillation frequencies , given by @xmath172 ( bardeen et al . 1972 ) , might be slightly smaller ( between @xmath340 hz , @xmath341 hz at @xmath13 , @xmath12 , respectively ) and appear twice as fast as those given above or in pairs .
see williams ( 2002 ) for a discussion of the nonequatorially confined spherical - like orbits , first proposed by wilkins ( 1972 ) .
the above findings are consistent with the qpos proposed to originate from orbits within the radius of the marginal stable orbit @xmath13 ( zhang , shrohmayer , & swank 1997 ) , and the suggestion that the energy distribution of the energetic electrons must be oscillating at the qpo frequency ( morgan , remillard , & greiner 1997 ) . note , in the above qualitative , yet self - consistent models , for the radio quiet seyfert galaxy ( @xmath342 ; [ sec : mcg ] ) and the galactic black hole cyg x-1 ( @xmath343 ; present section ) , for the initial conditions used based on properties of the accretion disk , the main differences in the emitted spectra are the number of penrose produced @xmath9 pairs escaping , and the range of @xmath197 : for the seyfert galaxy @xmath197 is in the narrow range @xmath344 mev , and for the galactic black hole , @xmath345 mev . in both cases , most ( if not all ) of the ppp electrons have turning points in the nonequatorially confined ( spherical - like ) electron orbits at @xmath246 , as discussed in
[ sec : asymmetry ] , indicating that these electrons escape along vortical trajectories collimated about the polar axis , without interacting appreciably with the inner edge of the bound stable accretion disk ( located at @xmath346 ; williams 2004 ) .
from the penrose - williams model presented here , to extract energy momentum from a rotating black hole , we can conclude the following : pcs is an effective way to boost soft x - rays to hard x - rays and @xmath8-rays up to @xmath347 mev .
ppp ( @xmath0 ) is an effective way to produce relativistic @xmath9 pairs up to @xmath348 gev : this is the probable mechanism producing the fluxes of relativistic pairs emerging from cores of agns ; and when relativistic beaming is included , apparent energies @xmath45 tev can be achieved ( williams 2003 ) .
these penrose processes can operate for any size rotating black hole , from quasars to microquasars ( i.e. , galactic black holes ) .
overall , the main features of quasars : ( a ) high energy particles ( x - rays , @xmath9 pairs , @xmath8-rays ) coming from the central source ; ( b ) large luminosities ; ( c ) collimated jets ; ( d ) one - sided ( or uneven ) polar jets which under certain conditions the asymmetry brightness appears to `` flip , '' can all be explained by these penrose processes . moreover , it is shown here that the geodesic treatment of individual particle processes close to the event horizon , as governed by the black hole , is sufficient to described the motion of the particles .
this is consistent with mhd that the behavior of such individual particles on geometry ( or gravity)-induced trajectories is also that of the bulk of fluid elements in the guiding center approximation ( de felice & zanotti 2000 ) . in light of this , with some ease , mhd should be incorporated into these calculations , particularly to describe the flow of the penrose escaping particles away from the black hole , to perhaps further collimate and accelerate these jet particles out to the observed distances .
importantly , we can conclude that , the difference between quasars , radio quiet and radio loud galaxies , and microquasars , appears to be the presence or the lack of a two - temperature adaf : with or without nuclear reactions ( @xmath349 ) in the inner region of the accretion disk ( see eilek 1980 ; eilek & kafatos 1983 ) .
quasars appear to have thin disk / ion corona ( adaf ) with nuclear reactions . in the case of the radio quiet and radio loud galaxies
the adaf may no longer be `` nuclear reactive , '' however just hot , and in some cases the disk may have evolved back to its cool thin disk phase , including the associated thermal - cycle lightman instabilities ( lightman 1974a , 1974b ) .
the microquasars , on the other hand , appear in general not to satisfy conditions for the existence of an adaf , which is determined by the accretion rate ( williams & hjellming 2002 ) , but do appear to satisfy conditions to have a soft x - ray inner region and an apparent thermal - cycle instability , with disk temperature up to @xmath102 kev . finally , what makes the penrose mechanism described here so admirable is that it allows one to relate the macroscopic conditions , i.e. , of the global gravitational field of the kbh , to the microscopic world of particle physics .
this description , which is progressively being proven by observations , to be the correct description , allows us to see directly how energy is extracted from a black hole .
the physics used in this penrose analysis is that of special and general relativity . from this analysis and its consistency with observations ,
we arrive at the following conclusion : close to the event horizon , gravity and particle - particle interactions , in the ergosphere , of highly curved spacetime ( where the effect of the external accretion disk magnetic field is apparently negligible ) , are sufficient to described energy - momentum extraction from a rotating black hole .
i first thank god for his thoughts and for making this research possible .
next , i thank dr .
fernando de felice and dr .
henry kandrup for their helpful comments and discussions .
also , i thank dr .
roger penrose for his continual encouragement .
i am grateful to the late dr .
robert ( bob ) hjellming for his helpful discussions and cherished collaboration
. part of this work was done at the aspen center for physics .
this work was supported in part by a grant from nsf at nrao and an aas small research grant .
associated problems with popular mhd models are described below : \1 . in order to explain observations of the seyfert 1 galaxy mcg6 - 30 - 15 , that copious photons are been extracted from the black hole from radii less than the marginal stable orbit @xmath13 ( @xmath350 , in gravitational units with @xmath235 , where @xmath351 is the mass of the black hole )
, it has been claimed that the force lines of the disk magnetic field @xmath352 couple with matter deep within the `` plunging '' region @xmath353 , thereby extracting rotational energy in the form of electromagnetic energy ( wilms et al .
2001 ; krolik 2000 ) .
however , the first detailed numerical relativistic time - dependent mhd calculations in a kerr metric ( koide et al . 2000 ; meier & koide 2000 ; meier , koide , & uchida 2001 ) show that in order for magnetic field lines to extend inward to the numerical limited radius @xmath354being frozen to the plasma , of keplerian velocity , the disk material must be initially counter rotating : opposite the direction that the black hole is rotating .
this appears inconsistent with the observations of zhang , cui , & chen ( 1997 ) and in general the physics occurring inside the ergosphere in which inertial frames are dragged in the direction that the black hole is rotating .
even though we know that particles can have retrograde orbits inside the ergosphere , relative to an observer at infinity , it is highly improbable that the whole disk of matter will be counter rotating , at least in the general sense .
further , it appears that the net rotational energy being `` extracted '' in the numerical simulation of these authors ( koide et al .
2000 ) in the form of electromagnetic energy over and above the gravitational binding energy released due to the hydrodynamic energy transported into the black hole is merely the rotational energy from the nonphysical initial condition that the accretion disk plasma is counter rotating as it falls into the ergosphere . ) !
not only is this an untrue statement , but it is a violation of the laws of physics . the frame dragging circular velocity inside the ergosphere as measured by an observer at infinity is @xmath355 ( see bardeen et al . 1972 ; misner et al .
1973 ; thorne et al . 1986 ; williams 1995 ; see also [ sec : cyg ] ) . ] on the other hand , for a co - rotating disk these authors found that the inward limiting radius is even larger ( @xmath356 ) , attributed to a centrifugal barrier ( koide et al .
although this time - dependent mhd model is an excellent representation of subrelativistic ( @xmath357 ) jet formation in a kbh magnetosphere , the inconsistencies of this mhd model , as matter nears the event horizon ( @xmath358 ) , is probably an indication of the limitation , of such fluid dynamical models , in describing energy extraction from a rotating black hole : this being based on the guiding center approximation , wherein the single - particle approach is essential close to the black hole ( de felice & carlotto 1997 ; de felice & zanotti 2000 ) , i.e. , the behavior of individual particles is also that of the bulk of fluid elements .
this means that gravitational - particle interactions , such as the penrose processes describe here ( in this paper ) , are required .
note , in these penrose processes , which occur close to the event horizon , elementary electromagnetic and atomic forces dominate on the microscopic scale , while gravity is dominant on the macroscopic scale thus , as it should be in the strong gravitational potential well of the kbh ; but far away from @xmath359 electromagnetism appears to dominate macroscopically ( junor , biretta , & livio 1999 ) .
moreover , stability of the co - rotating disk , falling inward to the limiting radius @xmath356 , at the keplerian velocity , when magnetic field lines are coupled to the infalling plasma , with the jet formation similar to that of the schwarzchild black hole case ( koide , shibata , & kudoh 1999 ) , suggests that the large scale magnetic field plays a dominant role at large distances from @xmath359 , irrespective of whether or not the black hole is rotating .
in addition , these numerical inward limiting radii , at least in the case of the counter - rotating disk ( @xmath346 ) , may also be a display of the horizon being a `` vacuum infinity '' ( punsly & coroniti 1989 ; punsly 1991 ; williams 2003 ) : to the associated magnetic field charge neutral disk particle plasma , in accordance with the `` no - hair '' theorem ( carter 1973 ; misner , thorne , & wheeler 1973 ; williams 1995 ) , suggesting that the interaction of the disk magnetic field with particles in bound , trapped orbits at radii @xmath353 is negligible compared to the penrose gravitational - particle interactions described here .
therefore , it appears that electromagnetic energy can not be effectively extracted from the so - called plunging region : where gravitational - particle interactions will clearly dominate if the magnetic flux of an axisymmetric @xmath360 , as it does in general upon nearing the vacuum infinity horizon [ punsly & coroniti 1989 ; punsly 1991 ; williams 2003 ; see also bik ( 2000 ) and bik & ledvinka ( 2000 ) for a detailed general relativistic calculation showing this ] .
\2 . to convert the electromagnetic energy to particle energy at the event horizon , and to duplicate the observed luminosities from a poynting flux
, it requires a large - scale magnetic field strength @xmath361 g ( wilms et al . 2001 ; blandford & znajek 1977 ) . in order to create @xmath9 pairs along the field lines , as in the case of pulsars , a field strength of at least @xmath362 g is needed ( sturrock 1971 ; sturrock , petrosian , & turk 1975 ) .
( the mechanism , however , for the generation of the pairs in an electromagnetic field to date is a subject of debate . )
the first of the large strengths required above appears to be achieved for supermassive kbhs , at present
i.e . , with speculated assumptions .
but for galactic black holes ( microquasars ) with masses @xmath363 , @xmath364 g seems highly impossible to generate from , in most cases , a binary system accretion disk plasma flow .
an effective model for agns must also operate for microquasars as well .
moreover , according to electrodynamics , in general , to lift the particles `` frozen '' to the magnetic field lines , from a disk , accelerating them to relativistic speeds , there has to be an electric field component @xmath365 ( lovelace 1976 ) .
however , there exist problems in generating sufficient @xmath366 parallel to the polar direction ( @xmath367 axis ) ; none of the polar mhd models of this particular type adequately gets rid of this problem .
magnetic reconnection may be a solution to some degree .
\3 . to get around problems in items 1 and 2 (
specifically , the large strength field required and the vacuum infinity horizon ) it is assumed that a `` hot '' ion corona or torus - like accretion can provide the necessary jet particles : ( @xmath18 ) for the magnetosphere to act on , accelerating and collimating through centrifugal driving winds ( see below ) ; and ( @xmath64 ) to provide the hot ram pressure , to `` ram '' the magnetic field lines inward to the event horizon
. however , now there appears to be a problem concerning how to liberate particles from trapped orbits inside the ergosphere ( particularly in the plunging regime ) onto escaping orbits .
particles in the plunging regime , as defined by bardeen , press , and teukolsky ( 1972 ) , i.e. , massless and material particles ( with @xmath368 ) originating from infinity , can only escape , by being injected onto escaping orbits by some physically process near the black hole such as the penrose scattering processes described here since nothing can come out of the hole ( bardeen et al .
therefore , the bz - type models are faced with yet another problem , as the magnetic field is assumed to get closer to the kbh : where general relativistic effects must be considered , i.e. , how do we get the necessary escaping particles in numbers out of the ergospheric region ( @xmath353 ) into the jets by such models ? moreover , with observations showing m87 not having the expected large `` dusty '' thermal ir - emitting torus ( perlman et al . 2001 ) that could have possibly served as particle jet `` fuel '' for a bz - type model , the penrose mechanism to extract energy momentum , as described by williams ( 1995 ) , the so - called penrose - williams mechanism , appears to be the only possible , plausible way to power this agn , and thus , generate its jets ( [ sec : m87 ] ) .
so , in summary , in addition to the problems above associated with the bz - type mhd models , there still exists the historical problem : how does one convert from electromagnetic energy to the particle energies observed in the jets , emanating from the region where energy is observed to be extracted , i.e. , inside the ergosphere close to the event horizon ?
none of the existing bz - type mhd models thus far adequately solves this `` age - old '' problem .
\4 . in the centrifugal driven winds ( blandford & payne 1982 ) mentioned above
, the following is assumed : if the disk magnetic field lines subtends an angle of more than @xmath369 to the rotation axis , the gas will be flung away from the disk into collimated jets with speeds a few times the escape velocity at the magnetic footprint on the disk .
now , this may be true at @xmath370 , but near the event horizon @xmath359 , the escape conditions ( see williams 1995 ) must be adequately applied .
recently , a general relativistic mhd treatment of evolving tori that includes in some degree features of the blandford & payne ( 1982 ) type - models , which allow for centrifugal driven winds to power the jets ( hirose et al .
2004 ; de villiers , hawley , & krolik 2004 ; de villiers , hawley , & krolik 2003 ) , found no such appreciable relativistic winds emerging from the horizon , nor the so - called plunging region , nor the ergospheric accretion disk that could be tied directly to rotational energy extraction from the black hole , although it was found that the lorentz force inside the ergosphere increased due to inertial frame dragging .
the plunging region for @xmath371 lacked adequate resolution , suggesting perhaps the need for a general relativistic particle geodesic treatment , according to the guiding center approximation ( discussed in item 1 of this appendix ; see also williams 2003 ) .
these mhd calculations ( hirose et al .
2004 ; de villiers et al . 2004 ; de villiers et al .
2003 ) did , however , confirm the existence of the predicted funnel region ( rees et al .
1982 ; see williams 2003 ) , and are consistent with the evolved magnetic field configuration found by bik ( 2000 ) and bik & ledvinka ( 2000 ) , i.e. , that radial lines are expelled from the surrounding equatorial region ( @xmath372 ) , but at the poles @xmath373 .
this clearly shows that the classical bz - type models ( blandford & znajek 1997 ) , where magnetic field lines are proposed to anchor to the event horizon , thereby extracting rotational energy , could not be an important source , because @xmath372 in the region of importance for extracting rotational energy . also these mhd calculations seem to confirm the importance of magnetic fields on a large scale , in gravitational accretion processes , i.e. , in aiding mass outflows to large distances in the jets of black holes as well as those of protostars ( see williams 2004 ) .
\5 . finally , to clear up any confusion , the authors of the historical paper ( wilms et al .
2001 ) loosely called the bz - type models the penrose effect the very name for years that had distinguished williams
( 1991 , 1995 , 1999 , 2001 ) internationally known successful four - dimensional penrose model ( see also piran & shaham 1977 ; leiter & kafatos 1978 ; kafatos & leiter 1979 ; kafatos 1980 ; wagh & dadhich 1989 ) from the bz - type models .
strangely , these authors did not reference williams investigation .
nevertheless , to set the record straight , the penrose mechanism [ as summarized here and described in detail in williams ( 1995 ) ] , which involves gravitational extraction of energy from a spinning black hole , based on that visualized by penrose ( 1969 ) , and that of the so - called bz mechanism , which involves electromagnetic extraction of energy ( blandford & znajek 1977 ) , are two very different models .
so different that the statement made by the authors in wilms et al .
( 2001 ) , `` for parameters relevant to our discussion , the extra energy source is provided by the spin via the penrose effect occurring within the radius of marginal stability ( but outside of the stretched horizon ) , '' indeed requires a proper reference , since williams ( 1991 , 1995 ) model is popularly known as the only existing completely worked out model of the penrose mechanism : occurring within the radius of marginal stability @xmath13 . whatever the case may be , the recent observations of mcg6 - 30 - 15 ( wilms et al .
2001 ) and m87 ( perlman et al . 2001 ) introduce compelling evidence suggesting that perhaps it is the effects of williams black hole source model that is being observed ( as described in this paper ) , and hardly those of the bz - type models .
the evidence presented here strongly suggests that observed black hole sources have a central energy generation similar the mechanism described in this present paper .
so , to avoid any further confusion , it seems appropriate to refer to williams model as the penrose - williams mechanism , which i interchangeably refer to as just the penrose mechanism , out of respect for its originator penrose ( 1969 ) .
abramowicz , m. a. , bulik , t. , bursa , m. , & kluzniak , w. 2002 ( astro - ph/0206490 ) bardeen , j. m. 1973 , in black holes , ed .
c. dewitt & b. s. dewitt ( new york : gordon and breach science publishers ) , 215 bardeen , j. m. , press , w. h. , & teukolsky , s. a. 1972 , , 178 , 347 bik , j. 2000 , pramana , 55 , no . 4 , 481 ( gr - qc/0101091 ) bik , j. & ledvinka , t. 2000 , nuovo cimento , 115 b , 739 ( gr - qc/0012006 ) bini , d. , cherubin , c. , jantzen , r. t. , & mashhoon , b. 2003 , class .
quantum grav . , 2 , 457 ( gr - qc/0301055 ) biretta , j. a. , sparks , w. b. , & macchetto , f. 1999 , apj , 520 , 621 blandford , r. d. , & begelman , m. c. 1999 , mnras , 303 , l1 blandford , r. d. , & payne , d. g. 1982 , , 199 , 883 blandford , r. d. , & znajek , r. l. 1977 , , 179 , 433 boyer , r. h. , & lindquist , r. w. 1967 , j. math .
8 , 265 burbidge , g. r. , jones , t. w. , & odell , s. l. 1974 , apj , 193 , 43 carter , b. 1968 , phys .
, 174 , 1559 carter , b. 1973 , in black holes , ed . c. dewitt , & b. s. dewitt ( new york : gordon and breach science publishers ) , 57 cui , w. , zhang , s. n. , & chen , w. 1998 , apj , 492 , l53 de felice , f. , & calvani , m. 1972 , nuovo cimento , 10b , 447 de felice , f. , & carlotto , l. 1997 , , 481 , 116 de felice , f. , & curir , a. 1992 , class .
quantum grav . ,
9 , 1303 de felice , f. , & zanotti , o. 2000 , gen .
, 8 , no . 32 , 1449 ( astro - ph/9912413 ) dermer , c. d. , schlickeiser , r. , & mastichiadis , a. 1992 , a&a , 256 , l27 de villiers , j - p . , hawley , j. , & krolik , j. 2003 , apj , 599 , 1238 ( astro - ph/0307260 ) de villiers , j - p .
, hawley , j. , & krolik , j. 2004 , apj , submitted ( astro - ph/0407092 ) eardley , d. m. , & lightman , a. p. 1975 , apj , 200 , 187 eikenberry , s. s. , matthews , k. , morgan , e. h. , remillard , r. a. , & nelson , r. w. 1999a , apj , 494 , l61 ( astro - ph/9710374 ) eikenberry , s. s. , matthews , k. , murphy , t. w. , morgan , e. h. , remillard , r. a. , & muno , m. 1999a , apj , 506 , l31 ( astro - ph/9807204 ) eilek , j. a. 1980 , , 236 , 664 eilek , j. a. , & kafatos , m. 1983 , apj , 271 , 804 heinz , s. , choi , y .- y . , reynolds , c. s. , & begelman , c. 2002 , apj , in press ( astro - ph/0201107 ) hirose , s. , krolik , j. , de villiers , j - p . , & hawley , j. 2004 , apj , 606 , 1983 ( astro - ph/0311500 ) hjellming , r. m. , & rupen , m. p. 1995 , nature , 375 , 464 junor , w. , biretta , j. a. , & livio , m 1999 , nature , 401 , 891 kafatos , m. 1980 , apj , 236 , 99 kafatos , m. , & leiter , d. 1979 , apj , 229 , 46 kerr , r. p. 1963
, , 11 , 237 koide , s. , meier , d. l. , shibata , k. , & kudoh , t. 2000 , apj , 536 , 668 koide , s. , shibata , k. , & kudo , t. 1999 apj , 522 , 727 koide , s. , shibata , k. , kudo , t. , & meier , d. l. 2002 , science , 295 krolik , j. h. 1999 , apj , 515 , l73 krolik , j. h. 2000 , in explosive phenomena in astrophysical compact objects , proceedings of the 1st kias astrophysics workshop ( astro - ph/0008372 ) leiter , d. & kafatos , m. 1978 , , 226 , 32 liang , e. p. 1998 , phys . rep .
, 302 , 67 lightman , a. p. 1974a
, apj , 194 , 419 lightman , a. p. 1974b apj , 194 , 419 lovelace , r. v. e. 1976 , nature , 262 , 649 mahadevan , r. , narayan , r. , & krolik , j. 1997 , , 486 , 268 mcconnell , m. et al .
1989 , apj , 343 , 317 mcconnell , m. et al .
1994 , apj , 424 , 933 meier , d. l. , & koide , s. 2000 , private communications , at the aspen center for physics meier , d. l. , koide , s. , & uchida , y. 2001 , science , 291 , 84 miller , j. m. et al .
2001 , apj , 546 , 1055 misner , c. w. , thorne , k. s. , & wheeler , j. a. 1973 , gravitation ( new york : w. h. freeman and company ) morgan , e. h. , remillard , r. a. , & greiner , j. 1997 , apj , 482 , 993 novikov , i. d. , & thorne , k. s. 1973 , in black holes , ed .
c. dewitt & b. s. dewitt ( new york : gordon and breach science publishers ) , 343 penrose , r. 1969 , rivista del nuovo cimento : numero speciale , 1 , 252 perlman , e. s. , sparks , w. b. , radomski , j. , packham , c. , fisher , r. s. , pia , r. , & biretta , j. a. 2001 , apj , 561 , l51 piran , t. , & shaham , j. 1977 , phys .
d , 16 , no .
6 , 1615 piran , t. , shaham , j. , & katz , j. 1975 , apj , 196 , l107 porcas , r. w. 1987 , in superluminal radio sources , ed . j. a. zensus & t. j. pearson ( cambridge : cambridge univ .
press ) , 12 price , r. h. 1972 , , 5 , 2419 punsly , b. 1991 , apj , 372 , 424 punsly , b. , & coroniti , f. v. 1989 , , 40 , 3834 rees , m. j. , begelman , m. c. , blandford , r. d. , & phinney , e. s. 1982 , nature , 295 17 remillard , r. a. , muno , m. p. , mcclintock , j. e. , & orosz , j. a. 2002 , apj , 580 , 1030 ( astro - ph/0202305 ) shapiro , s. l. , lightman , a. p. , & eardley , d. m. 1976 , apj , 204 , 187 stella , l. , vietri , m. , & morsink , s. m. 1999 , 524 , l63 strohmayer , t. e. 2001 , apj , 552 , l49 sturrock , p. a. 1971 , apj , 164 , 529 sturrock , p. a. , petrosian , v. , & turk , j. s. 1975 , apj , 196 , 73 thorne , k. s. , price , r. h. , & macdonald , d. a. 1986 black holes : the membrane paradigm ( yale university press , new haven ) wagh , s. m. , & dadhich , n. 1989 , , 183 , 137 wilkins , d. c. 1972 , , 5 , 814 williams , r. k. 1991 , ph.d .
thesis , indiana univ .
williams , r. k. 1995 , , 51 , 5387 williams , r. k. 1999 , in proceedings of the eighth marcel grossmann meeting on general relativity , jerusalem , israel , ed . t. piran & r. ruffini ( singapore : world science ) , 416 williams , r. k. 2000 , preprint williams , r. k. 2001 , in relativistic astrophysics : 20@xmath374 texas symposium , austin , texas , ed . j. c. wheeler & h. martel ( new york : american institute of physics ) , 448 ( astro - ph/0111161 ) williams , r. k. 2003 , , submitted ( astro - ph/0306135 ) williams , r. k. 2002 , , submitted ( astro - ph/0203421 ) williams , r. k. 2004 , , 611 , in press ( astro - ph/0404135 ) williams , r. k. , & hjellming , r. m. ( posthumous ) 2004 , to be submitted , in preparation wilms , j. , reynolds , c. s. , begelman , m. c. , reeves , j. , molendi , s. , staubert , r. , & kendziorra , e. 2001 , , 328 , l27 ( astro - ph/0110520 ) zhang , s. n. , cui , w. , & chen , w. 1997 , apj , 482 , l155 zhang , w. , shrohmayer , t. e. , & swank , j. h. 1997 , apj , 482 , l167 crrrrrrrr & @xmath14 & @xmath150 & @xmath375 & @xmath376 & @xmath377&@xmath378 & @xmath379 & @xmath380 + case no .
& @xmath381&(mev)&(hz ) & ( hz ) & @xmath382&@xmath382 & & + @xmath383 ...... & @xmath384&@xmath385&@xmath386&@xmath387 & @xmath388&@xmath389 & @xmath390&@xmath391 + 2 ........ &@xmath392&@xmath393&@xmath394&@xmath395 & @xmath396&@xmath397&@xmath398&@xmath399 [ 0.632 ] + 5 ........ &@xmath384&@xmath400&@xmath401&@xmath402 & @xmath403&@xmath404 & @xmath390&@xmath399 [ 0.611 ] + 6 ........ &@xmath384&@xmath400&@xmath405&@xmath406 & @xmath407&@xmath389 & @xmath390&@xmath408 + 7 ........ &@xmath384&@xmath385&@xmath409&@xmath410 & @xmath411&@xmath389 & @xmath390&@xmath408 + @xmath412 ...... &@xmath384&@xmath413&@xmath405&@xmath414 & @xmath415&@xmath416 & @xmath390&@xmath399 [ 0.121 ] + 11 ........ &@xmath384&@xmath417&@xmath405&@xmath418 & @xmath419&@xmath420 & @xmath390&@xmath399 [ 0.005 ] + 13 ........ &@xmath384&@xmath421&@xmath405&@xmath422 & @xmath423&@xmath424 & @xmath390&@xmath399 [ 0.001 ] + crrrrrrrr @xmath425 & @xmath426&@xmath376 & @xmath377 & @xmath378 & @xmath427 & @xmath428&@xmath429 + case no . &
( mev ) & ( hz ) & @xmath430 & @xmath430 & & & + 14@xmath431 ........ & @xmath432 & @xmath433&@xmath434 & @xmath389 & @xmath435 & @xmath436&@xmath436 + 15 ........ &@xmath437&@xmath438&@xmath439 & @xmath440 & @xmath441&@xmath442 [ 0.8 ] & @xmath443 [ 0.859 ] + 17 ........ &@xmath444&@xmath445&@xmath446 & @xmath447 & @xmath441&@xmath443 [ 0.1 ] & @xmath443 [ 0.132 ] + 19 ........ &@xmath448&@xmath449&@xmath450 & @xmath451 & @xmath441&@xmath399 [ 0.05 ] & @xmath452 [ 0.069 ] + 22 ........ &@xmath453&@xmath454&@xmath455 & @xmath456 & @xmath441&@xmath457 [ 0.022]&@xmath458 [ 0.008 ] + 25 ........ &@xmath459&@xmath460&@xmath461 & @xmath462 & @xmath441&@xmath457 [ 0.008]&@xmath452 [ 0.001 ] + | in this paper , i present results from a theoretical and numerical ( monte carlo ) _
n - particle _ fully relativistic four - dimensional analysis of penrose scattering processes ( compton and @xmath0 ) in the ergosphere of a supermassive kerr ( rotating ) black hole .
these general relativistic model calculations surprisingly reveal that the observed high energies and luminosities of quasars and other active galactic nuclei , the collimated jets about the polar axis , and the asymmetrical jets ( which can be enhanced by relativistic doppler beaming effects ) _ all _ are inherent properties of rotating black holes . from this analysis
, it is shown that the penrose scattered escaping relativistic particles exhibit tightly wound coil - like cone distributions ( highly collimated vortical jet distributions ) about the polar axis , with helical polar angles of escape varying from @xmath1 to @xmath2 for the highest energy particles .
it is also shown that the gravitomagnetic ( gm ) field , which causes the dragging of inertial frames , exerts a force acting on the momentum vectors of the incident and scattered particles , causing the particle emission to be asymmetrical above and below the equatorial plane , thus appearing to break the equatorial reflection symmetry of the kerr metric .
when the accretion disk is assumed to be a two - temperature bistable thin disk / ion corona ( or torus , defining an advection - dominated accretion flow ) , energies as high as @xmath3 gev can be attained by these penrose processes alone ; and when relativistic beaming is included , energies in the tev range can be achieved , agreeing with observations of some bl lac objects . when this model is applied specifically to quasars 3c 279 and 3c 273 , and the seyfert 1 galaxy mcg6 - 30 - 15 , their observed high energy luminosity spectra in general can be explained .
this energy - momentum extraction model can be applied to any size black hole , irrespective of the mass , and therefore applies to microquasars as well . when applied to the classical galactic black hole source cygnus x-1 , the results are consistent with observations .
the consistency of these penrose model calculations with observations suggests that the external magnetic field of the accretion disk plays a negligible role in the extraction of energy momentum from a rotating black hole , inside the ergosphere , close to the event horizon where gravitational forces , and thus the dynamics of the black hole , appear to be dominant , as would be expected . |
in random matrix theory , a marchenko - pastur law describes the asymptotic behavior of the spectrum of the so - called wishart matrices @xcite . in free probability , a marchenko - pastur ( or free poisson ) law plays the role that a poisson distribution does in probability theory : it is the limiting distribution of @xmath7 when @xmath8 . for this reason
it is called a free poisson law in the context of free probability . on the other hand
, an arcsine law appears in probability theory as the law of the proportion of the time during which a wiener process is non - negative . in monotone probability
, an arcsine law plays the role of a gaussian law @xcite .
in particular , an arcsine law is a monotone stable law with stability index @xmath9 @xcite .
arizmendi et al .
@xcite found an interplay between marchenko - pastur and arcsine laws .
they introduced a class @xmath10 of freely infinitely divisible distributions whose lvy measures are mixtures of a symmetric arcsine law .
the building block of this class is a symmetric beta distribution @xmath11 the free lvy measure of @xmath12 coincides with an arcsine law .
moreover , @xmath12 is equal to the free multiplicative convolution of an arcsine law with a marchenko - pastur law , and hence , is freely infinitely divisible .
moreover , its cauchy - stieltjes transform ( or cauchy transform for short ) can be calculated explicitly as @xmath13 this paper studies a class of cauchy - stieltjes ( or cauchy for short ) transforms related to marchenko - pastur laws and monotone stable laws .
we deform the above cauchy transform ( [ eq0111 ] ) to introduce a family of probability measures which include the symmetric beta distribution @xmath12 , marchenko - pastur and some other beta distributions as special cases . more explicitly , for @xmath14
, we define @xmath15 the branches of powers have to be defined carefully and the precise definition is presented in section [ sec2 ] .
it can be shown that the function ( [ eq46 ] ) defines the cauchy transform of a probability measure @xmath0 for @xmath16 and @xmath17 satisfying what we call an admissible condition .
this condition is related to stable distributions .
the reciprocal cauchy transforms @xmath18 satisfy @xmath19we note that the same relation appears for probability measures introduced by motkowski @xcite .
this relation enables us to calculate the inverse map explicitly : @xmath20 the inverse map of the reciprocal cauchy transform , which is hard to calculate in general , is crucial to investigate free infinite divisibility .
therefore , the explicit form of @xmath21 is quite useful and we can prove the free infinite divisibility of @xmath22 for some parameters .
the probability measure @xmath1 turns out to be a free compound poisson distribution with lvy measure a monotone @xmath2-stable law @xmath23 . from proposition 4 of @xcite ,
if symmetric or positive , @xmath1 coincides with the free multiplicative convolution of a marchenko - pastur law @xmath24 and the monotone @xmath2-stable distribution @xmath23 : @xmath25moreover , @xmath26 is freely infinitely divisible for other parameters , not only for @xmath27 .
an interesting case of @xmath22 is @xmath28 which is a beta distribution with the density @xmath29 on @xmath30 .
we prove that this is freely infinitely divisible if and only if @xmath6 .
we also mention that , while an arcsine law is not freely infinitely divisible , some monotone stable laws are .
this fact was implicitly proved by biane in a different context ; see corollary 4.5 of @xcite .
in this paper , @xmath32 and @xmath33 respectively denote the upper half - plane and the lower half - plane of @xmath34 .
an additive free convolution @xmath35 of compactly supported probability measures @xmath36 and @xmath37 on @xmath38 is the probability distribution of @xmath39 , where @xmath40 and @xmath41 are self - adjoint free independent random variables with distributions @xmath36 and @xmath37 , respectively @xcite .
this convolution was extended to all borel probability measures in @xcite .
a probability measure @xmath36 on @xmath38 is said to be @xmath42-infinitely divisible if for any @xmath43 , there is @xmath44 such that @xmath45 . for a probability measure @xmath36 on @xmath38 ,
let us denote by @xmath46 the cauchy transform and by @xmath47 its reciprocal : @xmath48 and @xmath49 .
bercovici and voiculescu @xcite proved the existence of @xmath50 and @xmath51 such that @xmath47 is univalent in @xmath52 and @xmath53 .
the voiculescu transform @xmath54 is defined in @xmath55 to be @xmath56 .
the free convolution @xmath35 is characterized by @xmath57 in @xmath58 for some @xmath59 .
@xmath60 is called an @xmath31-transform . a probability measure
@xmath36 is @xmath42-infinitely divisible if and only if @xmath54 is the restriction of an analytic map from @xmath32 into @xmath61 @xcite .
this is also equivalent to the lvy - khintchine type representation suggested in @xcite @xmath62 for some @xmath63 and a non - negative measure @xmath37 satisfying @xmath64 and @xmath65 .
we call @xmath37 the lvy measure of @xmath36 .
the following is useful to calculate the lvy measure . for a @xmath42-infinitely divisible measure @xmath36
, its voiculescu transform can be written as @xmath66 for some @xmath67 and a non - negative finite measure @xmath68 @xcite .
the measure @xmath68 can be calculated , by using the stieltjes inversion formula @xcite , as @xmath69 for all continuity points @xmath70 of @xmath68 .
considering the relation @xmath71 and ( [ eq00 ] ) , we obtain @xmath72 and @xmath73 , where @xmath74 is the real number of ( [ eq00 ] ) .
in particular , if the functions @xmath75 converges uniformly to a continuous function @xmath76 ( @xmath77 ) on an interval @xmath78 $ ] , then @xmath68 is absolutely continuous in @xmath78 $ ] with density @xmath76 . hence @xmath37 is also absolutely continuous in @xmath78 $ ] with density @xmath79 regarding atoms , the following formula holds : @xmath80 for any @xmath81 .
multiplicative free convolution @xmath83 for probability measures on @xmath84 was investigated in @xcite .
this convolution corresponds to the probability distribution of @xmath85 , or equivalently @xmath86 , where @xmath40 and @xmath41 are positive free independent random variables .
this convolution is characterized by @xmath82-transforms defined as follows . for a probability measure @xmath36 on @xmath38 ,
we let @xmath87 . @xmath88 coincides with a moment generating function if @xmath36 has finite moments of all orders . in @xcite ,
@xmath88 was proved to be univalent in the left half - plane @xmath89 for a probability measure @xmath36 on @xmath84 with @xmath90 .
moreover , @xmath91 contains the interval @xmath92 .
then a map @xmath93 is defined by the inverse of @xmath88 .
the @xmath82-transform is defined as @xmath94 using the @xmath82-transform , @xmath95 is characterized as @xmath96 in a common domain including an interval of the form @xmath97 . more generally , a multiplicative convolution @xmath95 can be defined if @xmath36 or @xmath37 is supported on @xmath84 .
while ( [ eq5 ] ) is expected to hold also in this case , it is not known whether an @xmath82-transform can be defined for every probability measure .
it was shown in @xcite to hold for measures with bounded support and non - vanishing mean , while the bounded case when @xmath36 has vanishing mean was solved in @xcite .
for the unbounded case , as a partial solution , arizmendi and prez - abreu defined an @xmath82-transform of a symmetric probability measure as follows . for a symmetric distribution @xmath98
, there is a unique probability distribution @xmath99 on @xmath84 such that @xmath100 for @xmath101 . using a property of @xmath102
, we can conclude that @xmath88 is univalent in @xmath103 .
moreover , @xmath104 contains the interval @xmath92 .
therefore , we can define @xmath105 and @xmath106 .
then ( [ eq5 ] ) still holds if @xmath36 or @xmath37 is symmetric and the other is supported on @xmath84 .
finally we recall the analogues of compound poisson distributions , which will be important in this paper . a probability measure @xmath36
is said to be free compound poisson if @xmath107 for a probability measure @xmath37 with @xmath64 and a @xmath108 . in this case , @xmath109 coincides with the lvy measure of @xmath36 . the marchenko - pastur law @xmath24 with mean one belongs to the class of free compound poisson measures ; the pair @xmath110 is given by @xmath111 .
@xmath24 is also characterized by @xmath112 in terms of the @xmath82-transform .
let @xmath114 and @xmath115 . for any @xmath116 , we will find an @xmath117 such that the function @xmath118 is defined as an analytic map in @xmath119 . to make the definition precise ,
we take branches of powers @xmath120 , @xmath121 and @xmath122 as follows : * @xmath120 and @xmath122 are respectively defined as @xmath123 and @xmath124 in @xmath125 , where @xmath126 denotes a logarithm satisfying @xmath127 ; * @xmath121 is defined to be @xmath128 in @xmath129 $ ] , where @xmath130 is a logarithm satisfying @xmath131 .
we show that these branches enable us to define @xmath132 as an analytic function in @xmath119 for an @xmath117 depending on @xmath133 . under the definition ( 2 ) , the function @xmath134 is equal to the generalized binomial expansion @xmath135 for @xmath136 , where @xmath137 is the generalized binomial coefficient @xmath138 .
therefore , for @xmath101 with large @xmath139 , the function @xmath140 can be written as @xmath141 where @xmath142 is defined by @xmath143 . for any @xmath144
, there is an @xmath117 , independent of @xmath145 $ ] , such that the image of the map @xmath146 is contained in the sector @xmath147 .
therefore , we can take the power of ( [ eq403 ] ) by @xmath148 and @xmath149 is well - defined as an analytic map in @xmath119 .
we note that @xmath150 can be expanded in a series regarding @xmath151 : @xmath152 for some complex coefficients @xmath153 with @xmath154 . in the second line
, we used the formula @xmath155 .
this formula is valid in @xmath119 if the function @xmath156 is understood to be the generalized binomial expansion .
let us define @xmath157 for @xmath158 , where @xmath117 is large enough depending on @xmath159 .
then we have the following . [ thm001 ] let @xmath160 and @xmath115 .
then @xmath161 holds in @xmath162 for some @xmath117 .
we note that @xmath163 is equal to @xmath164 in @xmath119 with large @xmath117 . also , we note that @xmath165 for small @xmath166
. then @xmath167 for @xmath158 . under further conditions on @xmath168 ,
the function @xmath149 is well - defined in @xmath32 with values in @xmath33 , and therefore defines a probability measure .
[ thm1 ] suppose @xmath169 , @xmath14 and @xmath170 .
assume that either of the following conditions is satisfied : \(1 ) @xmath171 and @xmath172 ; \(2 ) @xmath173 and @xmath174 .
then @xmath132 is the cauchy transform of a probability measure , which we denote by @xmath113 .
moreover , @xmath175 is univalent in @xmath32 .
if @xmath176 satisfies ( 1 ) or ( 2 ) , it is said to be admissible .
let @xmath177 .
we can immediately check that @xmath178 as @xmath179 , non tangentially .
therefore , what needs to be proved is that @xmath180 analytically maps the upper half - plane to the lower half - plane .
we first focus on the case @xmath171 and @xmath181 $ ] .
then the image of the map @xmath182 in @xmath32 can be described as in fig .
[ dia8 ] after some steps described in figs .
[ dia5][dia7 ] .
we can see that the image of the map @xmath182 is contained in the sector @xmath183 .
this implies the desired conclusion .
in the case @xmath173 , we draw similar pictures ; see fig .
[ dia9][dia12 ] . in fig .
[ dia12 ] , the image of @xmath182 is contained in the sector @xmath183 .
therefore , the image of the map @xmath184 is contained in @xmath32 . in each step
described in the figures , a new univalent map is added , so that after all the steps , the map @xmath175 is also univalent in @xmath185 . under the map @xmath186 which can be obtained from the rotation and the translation of fig .
@xmath187 is tangent to @xmath188 and @xmath189 is tangent to the @xmath190 axis at @xmath191 .
, width=188 ] under the map @xmath186 which can be obtained from the rotation and the translation of fig .
@xmath187 is tangent to @xmath188 and @xmath189 is tangent to the @xmath190 axis at @xmath191 .
, width=188 ] under the map @xmath186 which can be obtained from the rotation and the translation of fig .
@xmath187 is tangent to @xmath188 and @xmath189 is tangent to the @xmath190 axis at @xmath191 . ,
width=188 ] under the map @xmath186 which can be obtained from the rotation and the translation of fig .
@xmath187 is tangent to @xmath188 and @xmath189 is tangent to the @xmath190 axis at @xmath191 .
, width=188 ] under the map @xmath192 .
@xmath193 and @xmath194 are half lines starting at @xmath195 . @xmath196 and @xmath197 are preimages of @xmath198 and @xmath199 of fig .
[ dia11 ] for the map @xmath200 , respectively.,width=188 ] under the map @xmath192 .
@xmath193 and @xmath194 are half lines starting at @xmath195 . @xmath196 and @xmath197 are preimages of @xmath198 and @xmath199 of fig .
[ dia11 ] for the map @xmath200 , respectively.,width=188 ] under the map @xmath186 which can be obtained from the rotation and the translation of fig . [ dia11 ] .
@xmath187 is tangent to the @xmath190 axis and @xmath189 is tangent to @xmath188 at @xmath191.,width=188 ] under the map @xmath186 which can be obtained from the rotation and the translation of fig .
[ dia11 ] .
@xmath187 is tangent to the @xmath190 axis and @xmath189 is tangent to @xmath188 at @xmath191.,width=188 ] [ rem2 ] \(i ) the admissible condition is related to monotone stable distributions as mentioned in the next section .
\(ii ) we have @xmath201 for any admissible @xmath17 .
therefore , the right inverse of @xmath202 can be calculated as @xmath203 from theorem [ thm001 ] .
\(iii ) from the relation @xmath204 , we can conclude that @xmath149 does not define a probability measure for @xmath205 and admissible @xmath17 .
the reason is as follows . if @xmath36 is a probability measure and not a point mass , then @xmath206 for any @xmath101 ; see corollary 5.3 of @xcite .
hence @xmath207 if @xmath208 and @xmath47 is univalent around @xmath209 .
therefore @xmath210 can not be written as @xmath211 for a probability measure @xmath37 on @xmath38 .
\(iv ) the measure @xmath113 satisfies self - similarity with respect to @xmath212 as follows .
if @xmath36 is a probability distribution of a random variable @xmath40 , then let @xmath213 denote the distribution of @xmath214 .
for @xmath215 , we have @xmath216
let @xmath217 be a monotone ( strictly ) @xmath2-stable distribution @xcite characterized by @xmath218 where @xmath17 satisfies the admissible condition .
@xmath219 is the centered arcsine law with variance @xmath220 and @xmath221 is a cauchy distribution or a delta measure .
the following properties are valuable to note here .
+ ( 1 ) @xmath217 is supported on @xmath84 if and only if @xmath171 and @xmath222 .
\(2 ) @xmath217 is symmetric if and only if @xmath223 .
\(3 ) both @xmath224 and @xmath225 are free @xmath226-stable distributions , but not strictly stable .
+ the proofs are as follows .
let @xmath227 , @xmath228 . from the stieltjes inversion formula ,
the density @xmath229 of @xmath230 is given by @xmath231}{\pi(|x|^{2\alpha}-2r|x|^\alpha \cos ( \alpha \pi + \theta ) + r^2)^{1/(2\alpha ) } } , & x>0 , \\
\frac{\sin[\frac{1}{\alpha}\arg(|x|^\alpha + re^{i(\pi -\theta)})]}{\pi(|x|^{2\alpha}-2r|x|^\alpha \cos \theta + r^2)^{1/(2\alpha ) } } , & x<0 , \end{cases}\ ] ] where @xmath232 is defined in @xmath233 so that it takes values in @xmath234 $ ] .
now the properties ( 1 ) and ( 2 ) can be proved easily .
\(3 ) it was proved in @xcite that free @xmath226-stable distributions are characterized in terms of the voiculescu transform @xmath235 , where @xmath236 and @xmath237 $ ] . moreover , strictly stable laws correspond to the case @xmath238 . since @xmath239 , we have @xmath240 , which for @xmath241 or
@xmath242 means that @xmath243 is free @xmath226-stable , but not strictly stable .
the main theorem of this section is the following .
@xmath244 is a free compound poisson distribution for any admissible @xmath17 .
moreover , its lvy measure @xmath245 is given by the monotone stable distribution @xmath246 .
thanks to proposition 4 of @xcite , it suffices to prove that @xmath247 , or equivalently , @xmath248 , in an open set of the form @xmath119 . as in ( [ eq7 ] ) , a naive relation @xmath249 may not be valid . to avoid this problem
, we understand that @xmath250 , appearing below , is defined by using the generalized binomial expansion @xmath251 for @xmath252 .
then for any @xmath253 , the following calculation is correct in @xmath119 with large @xmath117 : @xmath254 therefore , @xmath255 on the other hand , the voiculescu transform of @xmath244 is given as @xmath256 in @xmath119 .
therefore , we have proved @xmath248 . with proposition 4 of @xcite
, the above result implies @xmath257 if @xmath244 and @xmath246 are symmetric or supported on @xmath84 .
we do not know if this holds for any admissible pair @xmath17 since @xmath82-transforms are not defined for probability measures which are not symmetric or supported on @xmath84 .
[ thm22 ] let @xmath17 satisfy either of the following conditions : @xmath171 and @xmath258 ; @xmath173 and @xmath259 . then @xmath257 . in general , the density of @xmath260 is difficult to calculate . in some cases ,
however , the density is explicit as we show below .
\(1 ) let us consider @xmath261 .
then @xmath262 is the free multiplicative convolution of the marchenko - pastur law and a symmetric cauchy distribution . this is absolutely continuous with a strictly positive density on @xmath38 written as @xmath263 we mention that this probability measure belongs to a class proposed in @xcite .
\(2 ) let @xmath264 .
then the corresponding probability measure is supported on @xmath84 with a density @xmath265 \(3 ) as shown in @xcite , @xmath266 for @xmath267 is a symmetric beta distribution : @xmath268 in addition to @xmath244 , some monotone stable distributions are also @xmath42-infinitely divisible .
this property was essentially proved by biane @xcite .
@xmath230 is @xmath42-infinitely divisible if and only if @xmath17 satisfies either of the following conditions : \(1 ) @xmath269 and @xmath270 ; \(2 ) @xmath271 .
in fact , biane considered only special values for @xmath272 , but the same proof can be applied to the above result . finally , we note the @xmath82-transforms of @xmath260 and @xmath230 .
[ prop21 ] let @xmath17 satisfy either of the following conditions : @xmath171 and @xmath258 ; @xmath173 and @xmath259 .
then \(1 ) @xmath273 , @xmath274 , \(2 ) @xmath275 , @xmath274 .
the voiculescu transform @xmath276 can be calculated as @xmath277 .
let us define @xmath278 . then @xmath279 . since @xmath280 holds , the desired formula follows .
a similar calculation is possible for @xmath244 .
in the previous section , we proved that @xmath113 is @xmath42-infinitely divisible whenever @xmath281 . in this section
we will determine infinite divisibility for @xmath282 .
we found the general case is too difficult to treat , so that we only consider the problem for some parameters .
the main results of this section are the following . * if @xmath171 and @xmath6 , then @xmath113 is @xmath42-infinitely divisible . * if @xmath283 and @xmath284 , then @xmath113 is @xmath42-infinitely divisible .
* @xmath285 is @xmath42-infinitely divisible if and only if @xmath286 . *
if @xmath287 , there exists an @xmath288 such that @xmath113 is not @xmath42-infinitely divisible for @xmath289 .
we also show that some beta distributions are @xmath42-infinitely divisible , and some are not .
to prove the free infinite divisibility of @xmath113 , we introduce a subclass of @xmath42-infinitely divisible distributions . a probability measure @xmath36
is said to be in class @xmath290 stands for _ univalent inverse reciprocal cauchy transforms_. ] if @xmath47 is univalent in @xmath32 and , moreover , @xmath210 has an analytic continuation from @xmath291 to @xmath32 as a univalent function .
the following property was implicitly used in @xcite .
[ lem1 ] @xmath292 implies that @xmath36 is @xmath42-infinitely divisible .
the voiculescu transform @xmath54 has an analytic continuation to @xmath32 defined by @xmath293 . if there existed a point @xmath294 such that @xmath295 , then @xmath296 . since @xmath297 for @xmath298 , @xmath299 for @xmath300 .
therefore , @xmath301 never belongs to @xmath291 .
however , since @xmath302 is univalent in @xmath32 and @xmath303 , @xmath301 must satisfy @xmath304 , which contradicts the inequality @xmath305 .
therefore , @xmath54 maps @xmath32 into @xmath61 .
if @xmath36 is @xmath42-infinitely divisible , then @xmath47 is always univalent in @xmath32 .
this can be proved for instance by using the so - called subordination functions .
let @xmath36 be @xmath42-infinitely divisible and @xmath306 be the probability measure corresponding to the voiculescu transform @xmath307 .
for @xmath308 , an analytic function @xmath309 exists so that it satisfies @xmath310 .
@xmath311 is called a subordination function .
the reader is referred also to eq .
( 5.4 ) of @xcite , where the following replacements are required : @xmath36 by @xmath312 and @xmath313 by @xmath314 .
the relation @xmath310 is equivalent to @xmath315 moreover it is proved in theorem 4.6 of @xcite that @xmath316 taking the limit @xmath317 in ( [ eq111 ] ) , we get @xmath318 so that @xmath319 is univalent in @xmath32 . for instance , the normal law @xmath320 is in @xmath290 from the result of @xcite .
moreover , we can easily prove that wigner s semicircle law , the marchenko - pastur law and the cauchy distribution belong to @xmath290 .
@xmath290 is closed under the weak topology .
this is proved as follows .
the convergence of @xmath44 implies the local uniform convergence of the voiculescu transforms @xmath321 @xcite .
since @xmath322 converges locally uniformly , the limit function is univalent .
also @xmath323 itself converges to a univalent function .
therefore the limit measure belongs to the class @xmath290 .
we note that @xmath290 is a proper subset of all @xmath42-infinitely divisible distributions .
for instance , let @xmath36 be a probability measure characterized by the voiculescu transform @xmath324 . then @xmath325 .
we can find two distinct points @xmath326 such that @xmath327 with @xmath328 for small @xmath329 and @xmath330 near to @xmath331 .
this example also proves that @xmath290 is not closed under the free convolution , since the measures @xmath37 and @xmath332 , respectively defined by @xmath333 and @xmath334 , both belong to @xmath290 . from theorem [ thm1 ]
the map @xmath335 is univalent in @xmath32 for any admissible @xmath176 and @xmath177 , so that we only have to prove the inverse @xmath336 is univalent in @xmath32 .
[ thm21 ] let @xmath17 be an admissible pair .
then @xmath337 if either of the following conditions holds : \(1 ) @xmath171 and @xmath6 ; \(2 ) @xmath283 and @xmath284 . by remark [ rem2 ] ,
the explicit formula for @xmath338 is @xmath339 let us define @xmath340 and @xmath341 .
first we consider @xmath283 .
since the image of the function @xmath342 for @xmath101 is contained in the sector @xmath343 , one can easily see that @xmath344 contains a half line starting from @xmath191 .
in particular , @xmath345 is univalent in @xmath32 . therefore , we can take that line as a slit for the function @xmath346 , which then becomes univalent outside the slit .
let us focus on the case @xmath171 . if @xmath6 , one can prove that @xmath347 is contained in a sector with central angle @xmath348 and therefore @xmath345 is univalent in @xmath32 . since @xmath349 , the map @xmath346 can be defined as a univalent map in that sector .
we take @xmath350 and @xmath351 as a special case .
then @xmath352 is the beta distribution @xmath5 for @xmath353 : @xmath354 indeed , now we have @xmath355 .
it holds that @xmath356 if @xmath357 or @xmath358 and @xmath359 if @xmath360 .
the stieltjes inversion formula @xcite @xmath361 implies the conclusion .
a consequence of theorem [ thm21 ] is that the beta distribution @xmath5 is @xmath42-infinitely divisible for @xmath362 .
more strongly , we can prove the following .
the beta distribution @xmath5 @xmath363 is @xmath42-infinitely divisible if and only if @xmath362 . the lvy measure @xmath364 for @xmath365 can be calculated as @xmath366 by remark [ rem2](ii ) , @xmath367 is calculated as @xmath368 if @xmath369 , the function @xmath370 has a zero point in the upper half - plane , so that @xmath367 never be defined as an analytic function .
if @xmath371 , @xmath367 is analytic and univalent in the upper half - plane . for the lvy measure
, the voiculescu transform is @xmath372 .
it holds that @xmath373 as @xmath77 if @xmath374 or @xmath358 and that @xmath375 as @xmath77 . after some more calculations
, one can see @xmath376 where @xmath68 is the measure in ( [ eqtau ] ) .
@xmath68 does not have an atom since @xmath377 for any @xmath81 .
the lvy measure @xmath364 is equal to @xmath378 as explained in section [ sec22 ] .
if @xmath379 is not real , the support of @xmath380 is unbounded .
the density for large @xmath381 can be calculated as @xmath382 in particular , @xmath380 belongs to a class introduced in @xcite . in subsection [ subsec41 ] ,
the free infinite divisibility of @xmath22 was proved for some parameters in terms of the class @xmath290 . in section [ sec2 ]
, we succeeded in proving the free infinite divisibility of @xmath260 since the voiculescu transform had a quite explicit form .
for other parameters , it is difficult to investigate the free infinite divisibility .
a possible case is for @xmath271 and @xmath384 . in this case
, the voiculescu transform has a quite explicit form as in the case @xmath27 and @xmath42-infinite divisibility can be determined completely . indeed , the voiculescu transform is @xmath385 in contrast to the case @xmath27 , infinite divisibility depends on the parameter @xmath212 if @xmath384 .
let @xmath386 .
then @xmath285 is @xmath42-infinitely divisible if and only if @xmath286 . the lvy measure @xmath387 for @xmath388 can be calculated as @xmath389 because of remark [ rem2](iv ) , let us consider @xmath390 for simplicity .
after some calculations , we get @xmath391 therefore , if @xmath392 , we can find a point @xmath393 such that @xmath394 . if @xmath395 , we can calculate @xmath396}{|3x^2 + 3xs + s^2|^2},\ ] ] and therefore @xmath397 takes a positive value at a point . by symmetry ,
also @xmath398 can take a positive value .
therefore , @xmath399 is not @xmath42-infinitely divisible for @xmath400 . for @xmath401 , after some calculations , it holds that @xmath402 so that @xmath403 is @xmath42-infinitely divisible .
the measure @xmath68 in ( [ eqtau ] ) is absolutely continuous with respect to the lebesgue measure since @xmath404 locally uniformly in @xmath38 as @xmath77 .
the lvy measure is given by @xmath378 , where @xmath68 is defined in ( [ eqtau ] ) .
we prove the following . for @xmath287 and @xmath406 $ ] , there exists an @xmath288 such that @xmath113 is not @xmath42-infinitely divisible for @xmath289 .
let @xmath407 .
it is sufficient to find a zero point of the function @xmath341 .
the function @xmath342 maps @xmath32 to a shifted sector @xmath408 .
if @xmath409 , @xmath410 and the unit circle @xmath411 have intersection which is an arc with an end point @xmath195 .
let us denote by @xmath412 the angle of the other end point of that arc .
we can take @xmath413 to be @xmath414 .
this work was initiated during the authors visit to cimat , thanks to the hospitality of professor vctor prez - abreu , who also suggested many improvements of the paper .
o. arizmendi , o.e .
barndorff - nielsen and v. prez - abreu , on free and classical type @xmath415 distributions , braz .
j. probab .
* 24 * , no . 2 ( 2010 ) , 106127 . o. arizmendi and v. prez - abreu , the @xmath82-transform of symmetric probability measures with unbounded supports , proc .
amer . math .
* 137 * ( 2009 ) , 30573066 .
o. e. barndorff - nielsen and s. thorbjrnsen , classical and free infinite divisibility and lvy processes . in u. franz and m. sch rmann ( eds . ) : _ quantum independent increment processes ii .
quantum lvy processes , classical probability and applications to physics _ , _ _
_ _ springer , pp 33 - 160 , 2006 .
belinschi and h. bercovici , partially defined semigroups relative to multiplicative free convolution , int . math .
notices , no . 2 ( 2005 ) , 65101 .
belinschi , m. boejko , f. lehner and r. speicher , the normal distribution is @xmath42-infinitely divisible , adv .
math . * 226 * , no .
4 ( 2011 ) , 36773698 .
h. bercovici and v. pata , stable laws and domains of attraction in free probability theory ( with an appendix by philippe biane ) , ann . of math .
( 2 ) * 149 * , no . 3 ( 1999 ) , 10231060 . h. bercovici and d. voiculescu , free convolution of measures with unbounded support , indiana univ . math . j. * 42 * , no . 3 ( 1993 ) , 733773 .
biane , processes with free increments , math . z. * 227 * ( 1998 ) , 143174 .
t. hasebe , analytic continuations of fourier and stieltjes transforms and generalized moments of probability measures , j. theoret .
, to appear .
t. hasebe , monotone convolution and monotone infinite divisibility from complex analytic viewpoint , infin . dimens
. anal .
quantum probab .
* 13 * , no . 1 ( 2010 )
, 111131 .
v.a . marchenko and l.a .
pastur , distribution of eigenvalues for some sets of random matrices , mathematics sbornik * 72 * ( 1967 ) , 507536 .
w. motkowski , fuss - catalan numbers in noncommutative probability , doc .
15 * ( 2010 ) , 939955 .
n. muraki , monotonic independence , monotonic central limit theorem and monotonic law of small numbers , infin
. dimens .
quantum probab .
* 4 * ( 2001 ) , 3958 .
v. prez - abreu and n. sakuma , free infinite divisibility of free multiplicative mixtures of the wigner distributions , j. theoret .
, to appear .
rao , r. speicher , multiplication of free random variables and the s - transform : the case of vanishing mean elect .
comm . in probab .
* 12 * ( 2007 ) , 248258 .
r. speicher and r. woroudi , boolean convolution , in free probability theory , ed .
d. voiculescu , fields inst .
( amer . math .
soc . , 1997 ) , 267280 .
g. teschl , _ jacobi operators and completely integrable nonlinear lattices _ , mathematical surveys and monographs * 72 * , amer . math .
soc . , providence , 2000 .
d. voiculescu , addition of certain non - commutative random variables , j. funct .
66 * ( 1986 ) , 323346 .
d. voiculescu , multiplication of certain noncommuting random variables , j. operator theory * 18 * ( 1987 ) , 223235 . | we consider a class of probability measures @xmath0 which have explicit cauchy - stieltjes transforms .
this class includes a symmetric beta distribution , a free poisson law and some beta distributions as special cases .
also , we identify @xmath1 as a free compound poisson law with lvy measure a monotone @xmath2-stable law .
this implies the free infinite divisibility of @xmath3 .
moreover , when symmetric or positive , @xmath1 has a representation as the free multiplication of a free poisson law and a monotone @xmath2-stable law .
we also investigate the free infinite divisibility of @xmath0 for @xmath4 .
special cases include the beta distributions @xmath5 which are freely infinitely divisible if and only if @xmath6 .
mathematics subject classification 2010 : 46l54 ; 46l53 key words : free infinite divisibility , free poisson law , monotone stable law , beta distribution |
peter hall wrote many influential works on high - dimensional data , though notably he largely eschewed the notions of sparsity and penalised likelihood that have become so popular in recent years .
nevertheless , he was interested in variable selection , and wrote several papers that involved ranking variables in some way . perhaps his most well - known papers in this area , though , concern geometrical representations of high - dimensional data .
@xcite was one of the pioneering works in the early days of high - dimensional data analysis that tried to understand the properties of low - dimensional projections of a high - dimensional isotropic random vector @xmath0 in @xmath1 .
as motivation , let @xmath2 have @xmath3 and suppose that @xmath4 this condition says that the regression function of @xmath5 on @xmath6 is linear . then , using the isotropy of @xmath0 , @xmath7 moreover , @xmath8 and we conclude that @xmath9 , or equivalently , @xmath10 the left - hand side of is always non - negative , so can be used as a measure of the extent to which the condition holds .
remarkably , under very mild conditions on the distribution of @xmath0 , @xcite proved that if @xmath11 is drawn from the uniform distribution on the unit euclidean sphere in @xmath1 , then @xmath12 as @xmath13 .
this is equivalent to the statement @xmath14 as @xmath13 .
see also @xcite , who showed that under mild conditions , most low - dimensional projections of high - dimensional data are nearly normal . of course , when @xmath0 has a spherically symmetric distribution , holds for every @xmath2 with @xmath3 .
but the result of this paper shows that even without spherical symmetry , there is a good chance ( in the sense of random draws of @xmath11 as described above ) that holds , at least approximately , when @xmath15 is large . an important statistical consequence of this is that even if the relationship between a response @xmath16 and a high - dimensional predictor is non - linear , say @xmath17 for some unknown link function @xmath18 and error @xmath19 , standard linear regression procedures can often be expected to yield an approximately correct estimate of @xmath11 up to a constant of proportionality .
the generalisation of this result that replaces @xmath6 with @xmath20 , where @xmath21 is a random @xmath22 matrix with orthonormal columns , also plays an important role in justifying the use of sliced inverse regression for dimension reduction @xcite .
another seminal paper that articulated many of the key geometrical properties of high - dimensional data is @xcite .
this paper begins with the simple , yet remarkable , observation that if @xmath23 , then @xmath24 as @xmath13 .
thus , data drawn from this distribution tend to lie near the boundary of a large ball .
similarly , the pairwise distances between points are almost a deterministic distance apart , and the observations tend to be almost orthogonal .
in fact , the authors go on to explain that , under much weaker assumptions than gaussianity , the data lie approximately on the vertices of a regular simplex , and that the stochasticity in the data essentially appears as a random rotation of this simplex . as well as clarifying the relationship between support vector machines ( e.g. * ? ? ? * ) and distance weighted discrimination classifiers @xcite in high dimensions , the paper forced researchers to rewire their intuition about high - dimensional data , and precipitated a flood of subsequent papers on high - dimensional asymptotics .
the last 15 years or so have seen variable selection emerge as one of the most prominently - studied topics in statistics .
although peter s instinct was to think nonparametrically , he realised that he could contribute to a prominent line of research in the variable selection literature , namely marginal screening ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) , via the deep understanding he developed for rankings .
@xcite defined variable rankings through their generalised correlation with a response , while @xcite studied variable transformations prior to ranking based on correlation as a method for dealing with heavy - tailed data . for classification ,
@xcite proposed a cross - validation based criterion for assessing variable importance , while in the unsupervised setting , @xcite suggested ranking the importance of variables for clustering based on nonparametric tests of modality .
these works above were underpinned by peter s realisation that he could explain how perhaps his favourite tool of all , namely the bootstrap , could be used to quantify the authority of a ranking @xcite .
in fact , there are some subtle issues here , particularly surrounding the issue of ties .
peter developed an ingenious method for proving that even though the standard @xmath25-out - of-@xmath25 bootstrap does not handle this issue well , the @xmath26-out - of-@xmath25 bootstrap overcomes it in an elegant way .
i believe that peter may have become interested in classification problems in the early 2000s at least partly through ideas of bootstrap aggregating , or bagging @xcite .
indeed , in @xcite , a preprint of which was already available in early 2000 , peter had attempted to understand the effect of bagging in @xmath27-estimation problems .
this is a typical example of peter s extraordinary ability to explain empirically observed effects through asymptotic expansions .
one of the other interesting contributions of this work is that subsampling ( i.e. sampling without replacement ) half of the observations closely mimics ordinary @xmath25-out - of-@xmath25 bootstrap sampling , a very useful fact that has been observed and exploited in several other contexts , including stability selection for choosing variables in high - dimensional inference @xcite and stochastic search methods for semiparametric regression @xcite .
classification problems are ideally suited to bagging , because the discrete nature of the response variable means that small changes to the training data can often yield different outputs from a classifier ; in the terminology of @xcite , many classifiers are ` unstable ' .
suppose we are given training data @xmath28 , where each @xmath29 is a covariate taking values in a general normed space @xmath30 , and @xmath31 is a response taking values in @xmath32 .
assume further that we have access to a classifier @xmath33 constructed from the training data , so that @xmath34 is assigned to class @xmath35 . to form the bagged version @xmath36 of the classifier , we draw @xmath37 bootstrap resamples @xmath38 from @xmath39 , and set @xmath40 peter got me interested in bagging nearest neighbour classifiers .
ironically , the nearest neighbour classifier had been described by breiman as stable , since the nearest neighbour appears in more than half in fact , around @xmath41 of the bootstrap resamples ; thus the bagged nearest neighbour classifier is typically identical to the unbagged version . in @xcite
, however , we studied the effect of drawing resamples ( either with or without replacement ) of smaller size @xmath26 . naturally , this reduces the probability of including the nearest neighbour in the resample , and the bagged classifier is now well approximated by a weighted nearest neighbour classifier with geometrically decaying weights ; see also @xcite .
in order for bagging to yield any asymptotic improvement over the basic nearest neighbour classifier , we require @xmath42 ( when sampling without replacement ) and @xmath43 ( when sampling with replacement ) ; in order to converge to the theoretically - optimal bayes classifier , we require @xmath44 but @xmath45 . once classification problems had piqued his interest , peter set about trying to answer some of the key questions on rates of convergence and tuning parameter selection that would naturally have occurred to him given his earlier work on nonparametric inference .
@xcite studied the performance of classifiers constructed from kernel density estimates of the class conditional distributions on @xmath46 . a particularly curious discovery he made
there is that even in the simplest case where @xmath47 and where the class conditional densities @xmath48 and @xmath18 cross only at the single point @xmath49 , the rate of convergence and order of the asymptotically optimal bandwidth depends on the sign of @xmath50 . in @xcite , we considered similar problems in the context of @xmath51-nearest neighbour classification , obtaining an asymptotic expansion for the regret ( i.e. the difference between the risk of the @xmath51-nearest neighbour classifier and that of the bayes classifier ) which implied that the usual nonparametric error rate of order @xmath52 was attainable with @xmath51 chosen to be of order @xmath53 .
the form of the expansion made me realise that the limiting ratio of the regrets of the bagged nearest neighbour classifier and the @xmath51-nearest neighbour classifier ( with both the resample size @xmath26 and the number of neighbours @xmath51 chosen optimally ) depended only on @xmath54 , and not on the underlying distributions . to my great surprise , this limiting ratio was greater than 1 when @xmath47 , equal to 1 when @xmath55 and less than 1 for @xmath56 ( though approaching 1 for large @xmath54 ) .
it took me some years to explain this phenomenon in terms of the optimal weighting scheme @xcite . in more recent years
, peter turned his attention to a wealth of other important , though perhaps less well studied , issues in classification .
some of these were motivated by what he saw as drawbacks of existing classifiers .
for instance , in @xcite , he developed classifiers based on componentwise medians , to alleviate the difficulties of both computing and interpreting multivariate medians ; such methods can be highly effective for high - dimensional data that may have heavy tails . in @xcite , he studied robust versions of nearest neighbour classifiers for high - dimensional data that try to perform an initial variable selection step to reduce variability .
@xcite presented simple scale adjustments to make distance - based classifiers ( primarily designed to detect location differences ) less sensitive to scale variation between populations ; see also @xcite .
@xcite and @xcite concerned settings where one might want to incorporate the prior probabilities into a classifier , and where these prior probabilities may be significantly different from @xmath57 , respectively .
finally , @xcite discovered the phenomenon that estimating the risk of a classifier , and estimating the tuning parameters to minimise that risk , are two rather different problems , requiring the use of different methodologies .
i first met peter as a phd student when he visited cambridge in 2002 .
i spent an hour or so discussing a problem i was working on that involved using ideas of james stein estimation to find small confidence sets for the location parameter of a spherically symmetric distribution @xcite .
i was blown away at the speed with which he was able to understand where my difficulties lay , and make helpful suggestions . shortly afterwards
, he invited me to spend six weeks at the australian national university in canberra in july
august 2003 .
i arrived utterly exhausted after nearly 24 hours in the air , but peter was full of energy when he kindly picked me up from the bus station .
almost the first thing he said to me was : ` i ve got a problem i thought we could think about ... ' , and he proceeded to take out a pen and pad of paper ; one could nt help but be drawn along by his enthusiasm for research . everything with peter happened at breakneck speed , whether it was dashing around the supermarket , a driving tour through the rural australian capital territory or , of course , writing papers .
many of his collaborators will have experienced discussing a problem with peter one evening and returning to the office the following morning to find that he had typed up a draft manuscript that would form the basis of a joint paper .
his prose was always elegant , and he had a wonderful ability to see his way through technical asymptotic arguments , aided by almost physicist - like intuition for what ought to be true .
one of my favourite peter stories , which i initially heard second - hand but which he later confirmed was true , concerned a time when he d been asked to teach an elementary statistics course to students with really very little quantitative background . realising that he d lost some of the students along the way , and in order not to ruin their grades , peter had a cunning idea and spent the last class before the final going through the problems that he d set on the exam . to his horror
, however , the students still flunked the exam .
when peter bumped into one of the students and asked in bemusement ` what happened ?
i went through the questions in the last class ' , the student replied ` yes , but you did them in a different order ' !
peter had seemingly boundless energy and capacity to work , but he was also a very gentle individual in many ways .
he was extraordinarily generous to others , and particularly junior researchers for whom he did so much .
he was a remarkable person and i miss him very deeply .
99 biau , g. and devroye , l. ( 2010 ) on the layered nearest neighbour estimate , the bagged nearest neighbour estimate and the random forest method in regression and classification _ j. mult
_ , * 101 * , 24992518 .
dmbgen , l. , samworth , r. j. and schuhmacher , d. ( 2013 ) stochastic search for semiparametric linear regression models . in _ from probability to statistics and back : high - dimensional models and processes a festschrift in honor of jon a. wellner_. eds m. banerjee , f. bunea , j. huang , v. koltchinskii , m. h. maathuis , pp . | in this article , i summarise peter hall s contributions to high - dimensional data , including their geometric representations and variable selection methods based on ranking .
i also discuss his work on classification problems , concluding with some personal reflections on my own interactions with him . |
any respectable list of desert island card games must include war to pass the time ( the expected length of a game is infinite , after all @xcite ) .
but to maintain your remembrance of things past , it s hard to beat the game of memory .
memory is played with a deck of @xmath5 cards .
the cards are numbered from @xmath6 up to @xmath0 , with each number appearing twice . the deck is shuffled and then the cards are laid face down in a tableau .
a _ move _ consists of flipping first one card and then a second .
if the cards match , both are removed from play and the current player moves again .
otherwise , they are flipped back over and play passes to the next player .
play ends when all pairs have been removed ; the player who has removed the most pairs is the winner .
the _ length _ of a game is the total number of moves .
perhaps surprisingly , strategy plays a role in the two - player version of the game @xcite . in some situations it may not be to a player s advantage to acquire new information by flipping over a card that has not been flipped before .
the reason is that her opponent also acquires the same information . in this exploration
we consider the solitaire version of memory , in which strategy plays no part .
many games , such as poker and go fish , do not fair well as one - person games .
memory does , as long as one shifts the goal from collecting the largest number of pairs to collecting all pairs in the fewest moves . as an additional simplification ,
we assume that the player has perfect memory
. you may find it helpful to play a few games of solitaire memory on your own .
you can use some subset of an ordinary deck of cards ; play online at a number of websites , such as @xcite or @xcite ; or , as of this writing , find a computer version on the wall at the airport ( at least in houston or minneapolis ) . in the solitaire version of memory , acquiring information by flipping over an unknown card
is always helpful .
it is therefore not hard to show that an optimal strategy is to proceed as described in algorithm [ alg : strategy ] . throughout this paper , we assume that the player is using this optimal strategy .
we consider three questions regarding this one - person game of memory .
+ flip them over and remove them flip over an unknown card flip the mate and remove both flip over another unknown card ( and remove the pair if you are lucky enough to find the mate ) * 1 .
what is the expected length of a game ( i.e. , how many moves are required ) ? * + each card must be flipped over at least once .
if it is not matched the first time it is flipped over , then it is flipped again , and removed , once the location of its mate is known .
thus , each card is flipped either once or twice , and therefore the total number of card flips is between @xmath5 and @xmath7 .
since two cards are flipped over on each move , this means that the number of moves in any game will be between @xmath0 and @xmath5 .
analysis of the possibilities for the end of the game shows that the last card to be flipped over will only be flipped once , and therefore the length of the game can not be @xmath5 .
therefore , the length lies between @xmath0 and @xmath8 .
example [ ex : extremal ] ( in section [ sec : setup ] ) illustrates that both of these lengths are possible .
, with logarithmic @xmath9-scale . on the right we use a linear @xmath9-scale and a restricted domain to highlight the fact that over @xmath10 of games for @xmath11 will have a length lying between @xmath12 and @xmath13.,title="fig : " ] , with logarithmic @xmath9-scale . on the right
we use a linear @xmath9-scale and a restricted domain to highlight the fact that over @xmath10 of games for @xmath11 will have a length lying between @xmath12 and @xmath13.,title="fig : " ] computer investigations indicate that the expected length of a game is roughly @xmath14 . in sections [ sec :
explen ] and [ sec : limsums ] we will determine the exact length , and show that it is approximately @xmath15 , with the error in this approximation approaching 0 as @xmath16 .
figure [ fig : dist - percentage ] shows the distribution of game lengths for @xmath11 .
the expected length in this case is about 160.8589 , and our approximation gives about 160.8593 .
a _ lucky _ move is one in which the player flips two cards that have not been flipped before and they happen to match . note that a game of length @xmath0 is one in which every move is lucky .
the second question , to be addressed in section [ sec : sumdleven ] , is : + *
2 . what is the expected number of lucky moves in a game ?
* + early in the game , the player has little information about the locations of the various cards . in all likelihood , each card flipped over will be the first of its pair to be seen .
but certainly by the @xmath17th flip , the player will begin encountering the mates of cards she has already seen .
this brings us to our third and final question , to be addressed in section [ sec : repeat ] : + * 3 .
how many flips are required before the player should expect to have seen both cards of a pair ? *
a game of memory is not affected by how the cards are physically laid out . in light of this
, it will be convenient to assume that the cards are arranged in a single row . and
since the cards are shuffled before they are laid out , when our strategy calls for the player to flip an unknown card , there is no advantage to flipping one particular unknown card in preference to another .
we may therefore assume that unknown cards are simply flipped from left to right . with this convention , game play depends only on the order in which the cards are laid out .
[ ex : n=6ex ] suppose that @xmath18 and the cards are laid out in the order @xmath19 play will proceed as follows : 1 .
the player flips over the first two cards , which are a 1 and a 2 . since they do nt match , they are flipped back over .
the player flips over the next card , which is another 1 . remembering that the first card was also a 1
, the player flips the first card to get the match and removes the two 1s .
the player flips the next two cards , which are a 6 and a 2 .
they do nt match , so they are flipped back .
the player now knows the locations of both 2s .
she therefore flips them and removes them .
the player flips the next two cards , which are both 3s .
they match , so she removes them .
since these cards had not been flipped before , this is an example of a lucky move .
the player flips the next two cards : a 5 and a 4 .
they do nt match , so they are flipped back .
the player flips the next card , which is the second 4 . remembering that the previous card was also a 4 , she flips it and removes the two 4s .
the player flips the second 5 , flips the first 5 again , and removes the two 5s .
the player flips the second 6 , flips the first 6 again , and removes the two 6s . [ ex : extremal ] we can now give examples of the shortest and longest possible games . if the cards are laid out in the order @xmath20 then the game will consist of @xmath0 lucky moves ; this is the shortest possible game .
it is only slightly harder to illustrate the longest possible game .
we will let you verify that if the cards are laid out in the order @xmath21 then it will take @xmath8 moves to complete the game .
although we have presented all of our examples by specifying the order in which the cards are laid out , in fact game play depends on only the relative positions of the pairs .
these relative positions can be visualized using the notion of an _
interconnection network_. an interconnection network @xcite consists of @xmath5 points along with @xmath0 endpoint - disjoint arcs that connect pairs of points .
( equivalently , such a network encodes an involution without fixed points . )
figure [ fig : icnetwork ] illustrates two deals for @xmath18 along with the associated interconnection network , in which points labeled with the same number are connected by an arc .
( the first of these deals is the one in example [ ex : n=6ex ] above . )
the two deals in figure [ fig : icnetwork ] differ only in the labels we have assigned to each pair of cards , and therefore the first deal leads to essentially the same game as the second .
we wish to identify those games with the same underlying interconnection network . in general
, there is an @xmath22-to-@xmath6 map from size-@xmath5 deals to size-@xmath5 interconnection networks .
this map for @xmath23 is illustrated in figure [ fig : icn - n2 ] .
we prefer to remain in the realm of card games , so we designate a representative for each class of deals .
we call a deal _ standard _ if , when the game is played , the pairs are removed in order from 1 to @xmath0
. put another way , a deal is standard if the second occurrence of @xmath24 occurs to the left of the second occurrence of @xmath25 for each @xmath24 between 1 and @xmath0 .
the deals in examples [ ex : n=6ex ] and [ ex : extremal ] are all standard .
let @xmath26 denote the set of standard permutations of the multiset @xmath27 .
at this point , we have reduced a game of memory to an element of some @xmath26 .
we write these elements in one - line notation ( reminiscent of an actual row of cards ) .
for example , define @xmath28 by @xmath29 and @xmath30 .
the element @xmath31 represents the game @xmath32 , in which the first and fourth cards are 2s and the second and third are 1s . for completeness , we include a proof of the following formula for the number of inequivalent games of memory that is , the cardinality of @xmath26 . for @xmath33 , the number of standard games of memory on a deck of @xmath5 cards is @xmath34 the first equality involving the _
double factorial _ function is true by definition .
the second equality can be checked by simple algebra .
we prove by induction that the number of games is given by the above product of odd integers .
for @xmath35 the formula is trivially true .
now suppose that @xmath36 .
any standard deal of a deck of size @xmath5 can be constructed by adding two copies of @xmath0 to a standard deal of a deck of size @xmath37 . in order for the augmented deal to be standard
, the second occurrence of @xmath0 must be at the end of the row .
thus , there are @xmath38 possible locations for the first occurrence of @xmath0 .
the lemma therefore follows by the principle of mathematical induction .
fix @xmath39 . for any @xmath40 ,
let @xmath41 denote the smallest index @xmath42 such that @xmath43 for some @xmath44 .
in other words , @xmath41 is the position of the first card in @xmath31 that matches a previous card .
since we are assuming that the deal is standard , this first matching card will be a 1 .
let @xmath45 count the number of permutations @xmath46 for which @xmath47 .
notice that @xmath41 is always between 2 and @xmath48 , so @xmath49 if @xmath50 or @xmath51 .
then for a randomly chosen @xmath52 , the probability that @xmath53 is @xmath54 and therefore the expected value of @xmath41 is @xmath55 it turns out that there is a simple formula for the numerator of the last fraction : [ lem : expnum ] for every positive integer @xmath0 , @xmath56 we proceed by induction on @xmath0 . the equation is easily seen to be correct when @xmath35 .
now suppose that @xmath57 , and the equation holds for @xmath58 .
a permutation @xmath40 with @xmath59 can be constructed from a permutation in @xmath60 in one of two ways .
the first way is to start with a permutation @xmath61 with @xmath62 , insert an @xmath0 into @xmath63 somewhere after position @xmath42 , and then add a second @xmath0 at the end .
the number of permutations that can be constructed in this way is @xmath64 .
the second way is to start with a permutation @xmath65 with @xmath66 , insert an @xmath0 before position @xmath67 , and then add another @xmath0 at the end .
this can be done in @xmath68 ways .
thus , we have the recurrence @xmath69 we can now compute : @xmath70\\ & = \sum_{j=2}^{n+1 } j \cdot ( 2n-1-j ) \cdot a(n-1,j ) + \sum_{j=2}^{n+1 } j \cdot ( j-1 ) \cdot a(n-1,j-1)\\ & = \sum_{j=2}^{n+1 } j \cdot ( 2n-1-j ) \cdot a(n-1,j ) + \sum_{j=1}^n ( j+1 ) \cdot j \cdot a(n-1,j)\\ & = \sum_{j=2}^n 2nj \cdot a(n-1,j ) = 2n \cdot \sum_{j=2}^n j \cdot a(n-1,j)\\ & = 2n \cdot 2^{n-1 } ( n-1 )
! \qquad\qquad \text{(by the inductive hypothesis)}\\ & = 2^n n!. \qedhere\end{aligned}\ ] ] for @xmath33 , the expected position of the first match is @xmath71 using equation and lemma [ lem : expnum ] , we find that the expected position of the first match is @xmath72 the expected position of the first match grows as @xmath73 .
straightforward applications of stirling s approximation @xmath74 ( see , for example , @xcite ) yield @xmath75
our main goal in this section is to prove the following theorem : [ thm : main ] the expected length of a game with @xmath5 cards is @xmath76 , where @xmath77 .
consider any standard deal @xmath40 . in the game based on @xmath31
, the player will turn over unknown cards two at a time until she comes to the first matching card at position @xmath41 .
what happens at that point depends on whether @xmath41 is even or odd .
if @xmath41 is odd , then the card at position @xmath41 , which is the second 1 , will be flipped over at the beginning of a move .
since we are assuming that the player has perfect memory , she will then flip the first 1 and remove both 1s . the number of moves required to flip all the cards up to position @xmath41 and then remove the matching 1s is therefore @xmath78 . on the other hand
, if @xmath41 is even then the card at position @xmath41 will be the second card flipped in a move .
if the first card flipped in that move was also a 1 then the move is lucky and the two 1s are removed right away . if not , then it will take an additional move to flip the two 1s and remove them .
thus , if @xmath41 is even then the number of moves to flip the first @xmath41 cards and remove the two 1s is either @xmath79 , if the first 1 is at position @xmath80 , or @xmath81 , if not .
once the two 1s have been removed , the player continues by flipping the unknown card at position @xmath82 , and she again flips unknown cards two at a time until she reaches the next matching card , the second 2 . if the second 2 is at position @xmath83 , then as above the number of moves required to flip all of the cards from position @xmath82 to @xmath83 and remove the matching 2s is @xmath84 if @xmath85 is odd , @xmath86 if @xmath85 is even and the first 2 is at position @xmath87 , and @xmath88 otherwise .
once the 2s have been removed , the player flips unknown cards two at a time until she finds the second 3 , and she continues in this way until all the pairs have been matched and removed .
thus , we see that the second occurrences of each of the numbers from 1 to @xmath0 can be thought of as breaking the deal @xmath31 into @xmath0 blocks , with each block ending with a second occurrence .
the function @xmath41 that we studied in the last section is the length of the first block , but we will have to consider the lengths of all of the blocks to determine the number of moves in the game .
let the lengths of the blocks be @xmath89 , @xmath90 ,
, @xmath91 , and notice that since the entire deck contains @xmath5 cards , @xmath92 .
we will call a block _ lucky _ if the block has length at least 2 and the last two cards in the block match .
the analysis above shows that the number of moves required to flip all of the cards in the @xmath24th block and then remove the two cards labeled @xmath24 is @xmath93 , where @xmath94 is 1/2 if @xmath95 is odd , 0 if @xmath95 is even and the @xmath24th block is lucky , and 1 if @xmath95 is even and the @xmath24th block is not lucky .
thus , if @xmath96 is the number of blocks in @xmath31 of even length and @xmath97 is the number of these even - length blocks that are lucky , then the length of the game is @xmath98 for example , if @xmath18 and @xmath31 is the deal given in example [ ex : n=6ex ] , then the lengths of the blocks are 3 , 2 , 2 , 3 , 1 , 1 , and the third and fourth blocks are lucky .
therefore @xmath99 and @xmath100 , and the length of the game is @xmath101 which is in agreement with our analysis of this game in example [ ex : n=6ex ] . to prove theorem [ thm : main ] we will need to find the expected values of @xmath96 and @xmath97 : the expected number of even blocks and the expected number of lucky even blocks . to compute these
, we will keep track of the number of blocks of each kind and each length among all of the @xmath102 standard deals . for
any positive integer @xmath42 , let @xmath103 be the total number of blocks of length @xmath42 in all standard deals of a deck of @xmath5 cards , and for @xmath104 let @xmath105 be the total number of lucky blocks of length @xmath42 .
the greatest possible length for a block is @xmath48 , so @xmath106 for @xmath107 .
the expected number of blocks of length @xmath42 is then @xmath108 and similarly the expected number of lucky blocks of length @xmath42 is @xmath109 using this notation , we can write the expected value of @xmath96 as @xmath110 , and the expected value of @xmath97 is @xmath111 .
notice that , although we have written these expected values as infinite sums , in fact the sums are finite , since all but finitely many terms in each sum are 0 .
thus the expected length of a game is @xmath112 there are several steps required to pass from equation ( [ eq : lengthform ] ) to theorem [ thm : main ] .
the first is to replace the @xmath113 and @xmath114 in ( [ eq : lengthform ] ) with ( what turn out to be ) their asymptotic limits .
we will show that , for large @xmath0 , @xmath115 we therefore define @xmath116 equivalently , @xmath117 we wish to rewrite using the asymptotic limits of the @xmath113 and @xmath114 . to this end , we first compute @xmath118 although the sum on the left - hand side is finite , the two sums on the right - hand side are infinite . however , it is not hard to see that these two infinite sums converge .
indeed , we can evaluate the first one by using the partial fractions decomposition @xmath119 this gives us @xmath120 since the alternating harmonic series converges to @xmath3 .
therefore @xmath121 next , we compute @xmath122 for the first part we again use a partial fractions decomposition : @xmath123 this gives us @xmath124 so @xmath125 combining , , and , we see that the expected length of the game is @xmath126 in the next section , we will show that @xmath127 substituting these values in will complete the proof of theorem [ thm : main ] .
in order to determine the limits required to complete the proof of theorem [ thm : main ] , we need to establish some facts about the @xmath128 and @xmath129 . to establish these facts , we begin by finding effective recurrences that can be used to compute @xmath128 and @xmath129 .
first we determine recurrences for the @xmath103 . if @xmath35 , then the only standard deal is @xmath131 , which has a single block of length 2 .
so @xmath132 and @xmath133 for @xmath134 .
for @xmath135 , as we saw before , any standard deal @xmath40 can be constructed from a standard deal @xmath61 by inserting an @xmath0 in one of the @xmath8 possible positions in @xmath63 and then adding a second @xmath0 at the end .
now , what happens to the blocks of @xmath63 when this is done ?
a block of length @xmath42 in @xmath63 either becomes a block of length @xmath136 ( if the first @xmath0 is inserted into that block ) or length @xmath42 ( if the first @xmath0 is inserted somewhere else ) . also , there is a new block at the end of length either 2 ( if both copies of @xmath0 are at the end ) or 1 ( if not ) .
so , for @xmath137 : @xmath138 dividing by @xmath139 , we get corresponding recurrences for @xmath140 : @xmath141 , @xmath142 for @xmath134 , and for @xmath135 , @xmath143 finally , we use the definition of @xmath128 ( equation ) to convert these recurrences into recurrences for @xmath144 .
[ lem : dbrec ] the numbers @xmath128 satisfy the following equations : @xmath145 the derivation of these equations involves only routine algebra .
we illustrate by deriving the formula for @xmath146 : @xmath147 using these recurrences , we can now study the numbers @xmath128 .
we begin by determining these numbers for small values of @xmath42 .
[ lem : dbvalues ] 1 . for @xmath33 ,
@xmath148 2 .
for @xmath33 , @xmath149 3 . for @xmath137 , @xmath150
all three statements are proven by induction using lemma [ lem : dbrec ] , with the proof of each statement after the first also making use of the previous statement . for all @xmath0 and @xmath151 , @xmath152 by induction on @xmath0 . for @xmath35 , we interpret the products in the numerator and denominator of the last fraction in as empty products , which are equal to 1
. so the inequality to be proven is @xmath153 , which follows easily from the formulas in lemma [ lem : dbrec ] .
now suppose that @xmath135 , and the lemma holds for @xmath58 .
let @xmath154 then the inductive hypothesis is that for all @xmath151 , @xmath155 .
we now consider three cases : case 1 : @xmath156 .
then by lemma [ lem : dbvalues ] , @xmath157 case 2 : @xmath158 . then @xmath159 case 3 : @xmath160 . since @xmath135 , this implies that @xmath107 .
therefore @xmath161 , so @xmath162 thus @xmath163 to get a better idea of the size of the bound in the last lemma , note that @xmath164 now , it is not hard to see that the product @xmath165 increases as @xmath0 increases , and it is well known that as @xmath166 it converges to @xmath167 ( see , for example , @xcite ) .
therefore for @xmath104 we have @xmath168 in particular , it follows that @xmath169 as @xmath170 , which justifies our claim that gives the asymptotic limits for the @xmath113 . to prove theorem [ thm : main ] , for fixed @xmath0
, we need to find the sum of the numbers @xmath128 , for @xmath42 even . as a start on this , we compute the sum of all of the @xmath128 : @xmath171 the first sum on the right - hand side of equation is the expected number of blocks ( of all lengths ) . but for every standard deal the number of blocks is @xmath0 , so this expected number is @xmath0
. we can evaluate the second sum on the right - hand side of by using the partial fractions decomposition : @xmath172 thus , @xmath173 according to lemma [ lem : dbvalues ] , @xmath174 , so @xmath175 to separate out the contributions of the even- and odd - numbered terms in we will use the following fact .
[ lem : dbshape ] for @xmath39 , there exist numbers @xmath176 and @xmath177 such that @xmath178 and @xmath179 we use induction on @xmath0 .
it is easy to verify that the lemma holds for @xmath35 and @xmath23 ( with @xmath180 , @xmath181 , @xmath182 , and @xmath183 ) .
now suppose that @xmath184 , and the lemma holds for @xmath58 .
to prove that the lemma holds for @xmath0 , we will show that @xmath185 we can therefore set @xmath176 to be one of @xmath186 or @xmath187 , and set @xmath177 to be one of @xmath188 or @xmath189 . to prove ,
suppose that @xmath190 ; we must verify that @xmath191 .
if @xmath156 , this follows immediately from lemma [ lem : dbvalues ] .
now suppose @xmath192 , and note that @xmath193 .
then by the inductive hypothesis we have @xmath194 and therefore @xmath195 similar reasoning can be used to prove . to prove ,
suppose @xmath196 .
if @xmath197 then we can repeat the reasoning in to show that @xmath191 . if @xmath198 then @xmath107 , so @xmath199 , and therefore @xmath200 lemma [ lem : dbshape ] will be useful to us because of the following fact .
[ lem : evenodddiff ] let @xmath201 be a monotonic sequence with the absolute value of each term bounded by some constant @xmath202 .
write @xmath203 for the sum of the odd - indexed terms , i.e. , @xmath204 if @xmath85 is odd and @xmath205 if @xmath85 is even .
write @xmath206 , so that @xmath207 is the sum of the even - indexed terms
. then @xmath208 suppose the sequence is monotonically increasing .
if @xmath85 is even , then @xmath209 and @xmath210 this implies that @xmath211 , so @xmath212 . if @xmath85 is odd , then we obtain the inequality @xmath213 , which leads to the stronger bound of @xmath214 .
the argument is similar if the sequence is monotonically decreasing .
we are finally ready to evaluate the first limit needed to complete our proof of theorem [ thm : main ] : [ lem : dbsumeven ] @xmath215 by lemma [ lem : dbshape ] , for any positive integer @xmath216 the sequence @xmath217 consists of at most three monotonic subsequences .
we can therefore apply lemma [ lem : evenodddiff ] to each subsequence with the bound @xmath218 from to show that @xmath219 adding @xmath220 , we can rewrite this as @xmath221 letting @xmath222 and applying equation , we conclude that @xmath223 or in other words @xmath224 the lemma follows now by letting @xmath170 .
we now do similar calculations for the lucky blocks . since the steps are similar to those in the last subsection , we skip many of the details .
if @xmath35 , then the only block in the standard deal @xmath226 is lucky , so @xmath227 and @xmath228 for @xmath192 . for @xmath135 , the numbers @xmath105
satisfy the recurrences @xmath229 therefore @xmath230 , @xmath231 for @xmath192 , and for @xmath135 , @xmath232 finally , applying the definition of @xmath129 we get @xmath233 [ lem : dlbound ] for all @xmath0 and @xmath151 , @xmath234 also , for @xmath135 , @xmath235 by induction on @xmath0 .
[ lem : lsum ] for all @xmath0 , @xmath236 note that the first equation in lemma [ lem : lsum ] indicates that the expected number of adjacent matching pairs in a random deal is @xmath6 .
the first equation can be proven by induction on @xmath0 .
for the second , we begin with the definition of @xmath129 : @xmath237 now we use partial fractions to evaluate the last sum : @xmath238 thus @xmath239 [ lem : dlshape ] for @xmath240 , there exists a number @xmath241 such that @xmath242 and @xmath243 by induction on @xmath0 .
it is easy to check that @xmath244 and @xmath245 . to establish the lemma for @xmath184
, we show that @xmath246 the proofs are similar to those in the proof of lemma [ lem : dbshape ] .
[ lem : dlsumeven ] @xmath247 by lemma [ lem : dlshape ] , for any positive integer @xmath216 the sequence @xmath248 consists of at most two monotonic subsequences . as in the proof of lemma [ lem : dbsumeven ] , we apply lemma [ lem : evenodddiff ] with @xmath249 to show that @xmath250 adding @xmath251 , letting @xmath222 , and dividing by 2 , we conclude that @xmath252 and the lemma follows .
this completes the proof of theorem [ thm : main ] .
observe that the proofs of lemma [ lem : dbsumeven ] and [ lem : dlsumeven ] indicate that @xmath253 in theorem [ thm : main ] is @xmath254 . as a bonus
, our last lemma also gives us the answer to question 2 : the expected number of lucky moves in a game tends to @xmath3 as @xmath170 .
there is a lucky move for each lucky block whose length is even , so the expected number of lucky moves is @xmath111 .
the lemma now follows from equation and lemma [ lem : dlsumeven ] .
the authors would like to thank dan archdeacon for useful conversations leading to this paper . 9 e. alfthan , optimal strategy in the childrens [ sic ] game memory ( 2007 ) , available at http://www.math.kth.se/xcomb/x1.pdf .
classic memory game ( 2012 ) , available at http://classicmemorygame.com/. concentration memory game ( 2012 ) , available at http://www.mathsisfun.com/games/memory/index.html .
p. flajolet , r. sedgewick , _ analytic combinatorics _ , cambridge university press , cambridge , 2009 r. l. graham , d. e. knuth , o. patashnik , _ concrete mathematics _ , second edition , addison - wesley , reading , ma , 1994 .
j. c. lagarias , a. m. odlyzko , d. b. zagier , on the capacity of disjointly shared networks , _ comput .
networks isdn systems _ * 10 * no . 5 ( 1985 ) 275285 .
e. lakshtanov , v. roshchina , on finiteness in the card game of war , _ amer .
monthly _ * 119 * ( 2012 ) 318323 .
j. wstlund , an elementary proof of the wallis product formula for pi , _ amer .
monthly _ * 114 * ( 2007 ) 914917 . | the game of memory is played with a deck of @xmath0 pairs of cards .
the cards in each pair are identical . the deck is shuffled and the cards laid face down . a move consists of flipping over first one card then another .
the cards are removed from play if they match . otherwise , they are flipped back over and the next move commences .
a game ends when all pairs have been matched .
we determine that , when the game is played optimally , as @xmath1 : * the expected number of moves is @xmath2 .
* the expected number of times two matching cards are unwittingly flipped over is @xmath3 . *
the expected number of flips until two matching cards have been seen is @xmath4 . |
frobenius splittings of algebraic varieties appear prominently in tight closure theory , have emerged as a fundamental tool in the study of the representation theory of algebraic groups , and have tantalizing links to concepts in the minimal model program .
suppose that @xmath6 is a reduced ring of characteristic @xmath7 with normalization in its total ring of fractions @xmath8 .
it follows from ( * ? ? ?
* exercise 1.2.4(e ) ) that if @xmath9 is frobenius split then so is @xmath10 .
the goal of this paper is to study to what extent the converse holds .
of course , there are many non - frobenius split affine varieties whose normalizations are regular and thus frobenius split ( for example , the cusp @xmath11 in any characteristic ) and so we focus our attention not just on the ring , but on the ring and a choice of a potential frobenius splitting . by definition ,
frobenius splitting _ on a variety @xmath12 is an @xmath13-linear map @xmath14 which sends the section @xmath15 to @xmath15 .
when @xmath0 is affine , the existence of a frobenius splitting is equivalent to existence of a map @xmath14 such that @xmath16 for some section @xmath17 ( in other words , to require that @xmath4 is surjective ) . on a normal @xmath12 , surjective maps @xmath18
are very closely related to boundary divisors @xmath19 such that @xmath20 is log canonical , see @xcite and @xcite .
based upon our intuition with non - normal log canonical singularities ( which are usually called _ semi - log canonical _ ) one should be able to detect surjective @xmath18 by studying the normalization of @xmath0 .
in fact , this correspondence between log canonical pairs @xmath20 and surjective @xmath4 suggests that the following question should have an affirmative answer .
[ quest.motivation ] suppose that @xmath0 is an affine variety in characteristic @xmath7 with normalization @xmath21 , and set @xmath22 to be the @xmath23-iterated frobenius .
given a map @xmath24 , it always extends to a unique map @xmath25 .
if @xmath26 is surjective , does it follow that @xmath4 is also surjective ?
perhaps unfortunately , this question has a negative answer .
a counterexample is given by the scheme @xmath27/(x^2y - z^2)$ ] , ( * ? ? ?
* example 8.4 ) .
however , that counterexample possesses substantial inseparability : if @xmath28 is the non - normal locus of @xmath0 and @xmath29 is its pre - image inside @xmath30 , then the induced map @xmath31 is generically inseparable .
we show that if we can avoid this inseparability and also a certain variant of wild ramification , then the question has a positive answer .
in particular , we say that a ring possesses _ _ if it avoids these positive characteristic pathologies ( see definition [ def.hst ] for a precise definition ) .
our main theorem follows : [ thm : maintheorem ] [ theorem [ thm : main ] ] suppose that @xmath0 is a reduced @xmath5-finite affine scheme having .
further suppose that @xmath24 is an @xmath13-linear map .
if the unique extension @xmath25 is surjective , then @xmath4 is also surjective .
the main theorem should also be viewed as complementary to ( * ? ? ?
* theorem 6.26 ) . in the context of a finite surjective map @xmath32 between _ normal _ varieties , that result answers the same question for @xmath14 and its extension @xmath33 ( if it exists ) . also see @xcite where 1-dimensional non - normal @xmath5-pure rings are studied .
we now discuss in more detail the motivation for this theorem . in this context
, maps @xmath34 correspond to @xmath35-divisors @xmath19 on @xmath30 such that @xmath36 is @xmath35-cartier , see @xcite and section [ sec : prelim ] .
the condition @xmath4 being surjective ( _ i.e. _ the pair @xmath37 being @xmath5-pure ) corresponds to the pair @xmath37 having log canonical singularities @xcite , @xcite . on a characteristic zero variety @xmath0 that is s2 ( _ i.e. _ serre s second condition ) and g1 ( _ i.e. _ gorenstein in codimension @xmath15 ) with normalization @xmath38 and conductor divisor @xmath39 , if there exists a @xmath40-divisor @xmath41 on @xmath0 such that @xmath42 is @xmath35-cartier and furthermore that @xmath43 is log canonical , then @xmath44 is called _ semi - log canonical_. via the map-(@xmath4 ) divisor-(@xmath19 ) correspondence , our main theorem should be interpreted as saying : ll as an explicit example of this statement , and of the translation between maps @xmath45 and divisors , we have the following corollary of our main theorem , a criterion for certain non - normal algebraic varieties to be @xmath5-split .
* corollary [ cor.hstnormalizationqgorenstein ] .
* _ suppose that @xmath12 is an affine @xmath5-finite scheme satisfying and which is also s2 , g1 and @xmath40-gorenstein with index not divisible by the characteristic @xmath46 .
set @xmath30 to be the normalization of @xmath0 and set @xmath29 to be the divisor on @xmath30 corresponding to the conductor ideal , _ i.e. _ @xmath47 .
then @xmath0 is @xmath5-pure ( equivalently @xmath5-split ) if and only if @xmath48 is @xmath5-pure .
_ in the above corollary , if one replaces the two occurrences of the word @xmath5-pure by semi - log canonical and log canonical ( respectively ) , one obtains a well known criterion for a scheme having semi - log canonical singularities , see @xcite . more generally , the correspondence between log canonical and @xmath5-pure singularities is still conjectural , although an area of active research .
[ conj.fpurevslogcanonical ] suppose that @xmath49 is a normal variety of characteristic zero and that @xmath19 is a divisor on @xmath49 such that @xmath50 is @xmath35-cartier .
the pair @xmath51 is log canonical if and only if @xmath51 has dense @xmath5-pure type pair @xmath52 has dense @xmath5-pure type if the map corresponding to @xmath19 is surjective after reduction to characteristic @xmath7 for infinitely many @xmath7 .
see @xcite for more details . ] .
see @xcite and @xcite for connections between this conjecture and open conjectures in arithmetic geometry ( these connections will not be utilized in this paper ) .
the setting for our applications will necessarily deal with geometry on non - normal schemes , and as such we will be dealing with an extended notion of divisors , see @xcite . for us ,
the important divisors are the @xmath35-almost cartier divisor ( or simply _ @xmath35-ac divisors _ ) , and we will review the definitions in detail in section [ sec : divisors ] . our main result implies the following : * corollary [ cor.fpurevsslc ] .
* _ assume now that @xmath0 is an s2 , g1 and seminormal variety in characteristic zero .
further suppose that @xmath19 is an @xmath35-ac divisor ( _ i.e. _ @xmath35-weil sheaf ) on @xmath53 assuming that conjecture [ conj.fpurevslogcanonical ] holds , @xmath20 has semi - log canonical singularities if and only if it has dense @xmath5-pure type . _
another application of theorem [ thm : maintheorem ] is to give a statement of inversion of adjunction for divisors on characteristic @xmath7 schemes with that closely aligns with the characteristic @xmath54 picture , compare with @xcite .
inversion of adjunction for characteristic @xmath7 schemes was studied first in @xcite , see also takagi @xcite and @xcite .
however the direct analog of kawakita s result is not possible in characteristic @xmath7 ( * ? ? ?
* example 8.4 ) .
the culprit is the same conditions of inseparability and wild ramification which obstruct a positive answer to the motivating question . however : * corollary [ cor.invofadj ] . * _ suppose that @xmath0 is a normal scheme of characteristic @xmath7 , @xmath19 is a @xmath35-divisor on @xmath0 and @xmath55 is a reduced integral weil divisor on @xmath0 , with no common components with @xmath19 , such that @xmath56 is @xmath35-cartier with index not divisible by @xmath46 .
denote by @xmath57 the normalization of @xmath55 and @xmath58 the natural map .
there exists a canonically determined @xmath35-divisor @xmath59 on @xmath57 such that @xmath60 .
furthermore , if @xmath61 has , then @xmath62 is @xmath5-pure near @xmath55 if and only if @xmath63 is @xmath5-pure . _
-12pt_acknowledgements : _ the authors would like to thank bryden cais , neil epstein and kevin tucker for inspiring conversations related to this project .
the authors would also like to thank the referee , shunsuke takagi and kevin tucker for numerous useful comments on an earlier draft of this manuscript .
this work was initiated when the second author was visiting the university of utah during fall 2010 .
throughout this article , all rings will be commutative with unity , noetherian , excellent and , unless otherwise specified , of prime characteristic and all schemes will be separated and noetherian .
we frequently use the following notations and conditions on rings and schemes .
finally , in order to avoid confusion , we remark that we use the notation @xmath64 to denote the ring of integers @xmath65 localized at the prime ideal @xmath66 . for a reduced ring @xmath6 ( resp .
@xmath0 is a reduced scheme ) , we use @xmath67 ( resp .
@xmath68 ) to denote its total ring of fractions . furthermore
, we use @xmath8 ( resp .
@xmath30 ) to denote the normalization of @xmath6 in @xmath67 ( resp . the normalization of @xmath0 in @xmath68 ) .
a ring is called s2 if it satisfies serre s second condition , _
i.e. _ the localization at any prime in @xmath6 of height at least @xmath69 or @xmath15 has depth at least @xmath69 or @xmath15 respectively
. it is called g1 if it is gorenstein in codimension 1 .
a finite extension of reduced rings @xmath70 is called _ subintegral _ ( resp . _
weakly subintegral _ ) if it induces a bijection on prime spectra such that the residue field maps are all isomorphisms ( resp .
purely inseparable ) .
a reduced ring @xmath6 is called _ seminormal _ if it possesses no proper subintegral extensions inside its own field of fractions . any reduced ring @xmath6 has a _ seminormalization _ @xmath71 which is the unique largest subintegral extension of @xmath6 which is contained inside @xmath67 .
for an introduction to seminormalization , see @xcite or @xcite .
it is important to note that if @xmath6 is seminormal then the conductor ideal @xmath72 is a radical ideal in both @xmath6 and @xmath8 , see @xcite or @xcite . for rings of characteristic @xmath7 ,
we denote by @xmath73 the frobenius homomorphism sending @xmath74 to its @xmath46-th power . given any @xmath6-module , @xmath75
, one can view @xmath75 as an @xmath6-module via the frobenius map and obtain a new @xmath6-module @xmath76 , called the _ frobenius pushforward of @xmath75_. this is the @xmath6-module with the same underlying additive group as @xmath75 but scalar multiplication is defined by the rule @xmath77 for @xmath78 , @xmath79 .
one can iterate the frobenius and for each @xmath23 we get a frobenius push forward @xmath80 whose @xmath6-module structure is defined similarly . for a scheme @xmath0 over a field of characteristic @xmath46
, one can ask about the @xmath13-module structure of @xmath81 .
connections between this question and the geometry of @xmath0 has been a strong guiding force in the study of such characteristic @xmath7 schemes .
for example , @xmath0 is regular if and only if @xmath82 is a flat @xmath13-module @xcite .
we now review some of these relationships .
a scheme @xmath0 is _ f - finite _ provided @xmath81 is coherent , _
i.e. _ a finitely generated @xmath13-module for some ( equivalently all ) @xmath83 . [ cols= " <
, < " , ] the class of @xmath5-finite schemes is particularly nice because @xmath5-finite schemes are abundant , namely varieties over a perfect field are all @xmath5-finite .
@xmath5-finite schemes are always locally excellent @xcite and they always locally have dualizing complexes @xcite . as in the introduction ,
we say @xmath0 is _ @xmath5-split _ provided there is a map @xmath84 which sends @xmath15 to @xmath15 and we say @xmath0 is _ @xmath5-pure _ provided the frobenius map on @xmath13 is a pure morphism . for @xmath5-finite affine schemes , @xmath5-splitting and @xmath5-purity coincide , see for example @xcite and @xcite . to describe how these notions relate to characteristic @xmath7 geometry , we use the setting of pairs .
for a more detailed treatment see @xcite . a _ prime divisor _ on a normal irreducible scheme @xmath49 is a reduced irreducible subscheme of codimension 1 and a _ weil divisor _ is any element of the free abelian group generated by the prime divisors .
this abelian group is denoted by @xmath85 .
a @xmath35-divisor is an element of @xmath86 , _
i.e. _ a divisor with rational coefficients .
an element @xmath87 is called an _
integral divisor _ if it is contained within @xmath88 ( in other words , if it is a weil divisor , and we wish to emphasize that its coefficients are integral ) .
cartier divisor _ is a weil divisor that is locally principal and a @xmath35-cartier divisor is a @xmath40-divisor @xmath89 such that @xmath90 is integral and cartier for some @xmath91 . for a @xmath40-cartier divisor @xmath89 , the smallest positive integer @xmath92 such that @xmath90 integral and cartier
is called the _ index of @xmath89_. see @xcite for additional discussion of @xmath40-divisors in this context .
we now discuss divisors on non - normal s2 schemes .
one should note that our main theorem can be proven without appeal to these objects , and our main corollary [ cor.fpurevsslc ] is interesting even when the divisor @xmath93 .
thus the reader not already familiar with divisors on non - normal schemes may wish to skip to section [ sec.extendingnormal ] at this point .
@xcite , @xcite [ def.almostcartierdivisor ] for an s2 equidimensional reduced scheme @xmath0 , an _
ac divisor _ ( or _ almost cartier divisor _ ) is a coherent @xmath13-module @xmath94 satisfying the following properties * @xmath95 is s2 . * @xmath96 , abstractly , for all points @xmath97 of codimension 0 or 1 . as was pointed out by the referee ,
the terminology _ almost cartier _ is misleading since any weil divisor on a normal scheme is almost cartier ( the terminology is taken from @xcite ) .
therefore , in order to avoid this confusion , we instead call almost cartier divisors by the name ac divisors
. these divisors form an _ additive _ group via tensor product up to s2-ification , ( * ? ? ?
* part 2 , section 5.10 ) , which we denote by @xmath98 . given @xmath99
, we sometimes use @xmath100 to denote the sheaf @xmath94 defining @xmath89 . note that given any @xmath101 non - zero at any minimal prime , we use @xmath102 to denote the element of @xmath98 corresponding to @xmath103 . by a _
@xmath35-ac divisor _
_ @xmath64-ac divisor _ )
we mean an element of @xmath104 ( resp .
@xmath105 ) .
we say that a divisor @xmath94 of @xmath98 is _ effective _ if @xmath106 and we say that @xmath107 ( resp .
@xmath108 ) is _ effective _ if @xmath109 for some effective @xmath99 and some @xmath110 ( resp .
@xmath64 ) .
we say that two ac divisors @xmath111 and @xmath112 are _ linearly equivalent _ if there is @xmath113 in @xmath68 such that @xmath114 .
we call an ac divisor _ cartier _ if the sheaf @xmath95 is a line bundle .
we say that an element @xmath107 ( resp .
@xmath115 ) is _
@xmath40-cartier _ ( resp .
_ @xmath64-cartier _ ) if there exists @xmath116 ( resp .
@xmath117 ) such that @xmath118 for some cartier divisor @xmath28 in @xmath98 .
we say that two elements @xmath119 ( resp .
@xmath108 ) are _ @xmath40-linearly equivalent _ ( resp .
@xmath64-linearly equivalent ) , denoted @xmath120 ( resp .
@xmath121 ) , if there exists a non - zero integer @xmath122 ( resp .
@xmath123 ) such that @xmath124 and there exists an element @xmath101 such that @xmath125 .
[ rem.pathologiesfordivisorsonnonnormal ] one should note that two distinct ac divisors of @xmath98 can be identified in @xmath126 , see ( * ? ? ?
* page 172 ) , and in particular , the natural map @xmath127 is generally not injective unlike the case when @xmath0 is normal .
working with @xmath108 is not common .
however , this group behaves much better than @xmath126 in characteristic @xmath7 for our purposes , see theorem [ thmdivisormapcornonnormal ] .
when working on normal varieties , this distinction is less important because @xmath108 as a subset of @xmath126 is simply the set of divisors whose coefficients do not have @xmath46 in their denominators . for non - normal varieties however , the natural map from @xmath108 to @xmath126 is not generally an injection .
therefore , when working in characteristic @xmath7 on _ non - normal _ varieties , we will generally work with @xmath108 .
suppose that @xmath0 is a reduced equidimensional scheme which is s2 and g1 .
further suppose that @xmath0 possesses a _
canonical module @xmath128 _ , _ i.e. _ a module isomorphic to the first non - zero cohomology of a dualizing complex at each point . by a canonical divisor @xmath129
we mean any embedding of @xmath130 up to multiplication by a unit of @xmath131 , see @xcite and @xcite .
notice that the condition that @xmath0 is g1 is exactly the condition needed to guarantee definition [ def.almostcartierdivisor](ii ) . by a pair , we mean a tuple @xmath52 where @xmath49 is a s2 and g1 scheme and @xmath19 is an element of @xmath126 or @xmath108 .
a pair is called _ log @xmath35-gorenstein _
_ log @xmath64-gorenstein _ ) provided @xmath50 is @xmath35-cartier ( resp .
@xmath64-cartier ) . when @xmath93 the ` log ' is omitted and the scheme @xmath49 is just called @xmath35-gorenstein ( resp .
@xmath64-gorenstein ) .
now suppose that @xmath49 is normal and @xmath132 is a log resolution , which always exists in characteristic zero by @xcite .
decompose @xmath133 where @xmath134 is prime and @xmath135 is a canonical divisor that agrees with @xmath136 wherever @xmath137 is an isomorphism .
the pair @xmath52 is _ log canonical _
provided @xmath138 for all @xmath139 .
if @xmath0 is an s2 but not necessarily normal scheme , denote by @xmath21 its normalization and @xmath29 the divisor associated to the conductor @xmath140 in @xmath30 ; _ i.e. _ @xmath47 .
if additionally @xmath0 is g1 , we say @xmath141 is _ semi - log canonical _
provided @xmath42 is @xmath35-cartier and @xmath142 is log canonical
. a more detailed treatment of these concepts in characteristic @xmath54 can be found in ( * ? ? ?
* chapter 16 ) or ( * ? ? ?
* chapter 3 ) .
we call an additive map , @xmath144 , _
@xmath143-linear _ provided it locally satisfies @xmath145 .
these are identified with the set of @xmath13-module homomorphisms in @xmath146 .
such maps @xmath4 naturally correspond to effective @xmath35-divisors @xmath147 such that @xmath148 is @xmath35-cartier . in the normal
setting , variants of this correspondence have appeared in many places such as @xcite and @xcite , this correspondence was recently formalized in @xcite .
however , we need a version of this correspondence in the non - normal setting as well .
first suppose that @xmath12 is a s2 and g1 reduced and equidimensional affine scheme which is also the spectrum of a semi - local ring .
the usual arguments still imply that @xmath149 .
therefore any section @xmath150 which is non - zero at each generic point of @xmath0 , induces an effective ac divisor @xmath151 via ( * ? ? ?
* proposition 2.9 ) and ( * ? ? ?
* remark 2.9 ) ( for example , choose the embedding of @xmath152 into @xmath68 which sends @xmath4 to @xmath15 ) .
we define the divisor @xmath147 associated to @xmath4 to be the element @xmath153 .
conversely , suppose we are given an effective @xmath64-ac divisor @xmath19 such that @xmath154 for some integer @xmath155 which is not divisible by @xmath46 and for some @xmath99 .
additionally suppose that @xmath156 .
thus @xmath89 corresponds to a section @xmath157 , up to multiplication by a unit of @xmath131 . without loss of generality
, we may assume that @xmath158 for some @xmath159 and so we may view @xmath160 as an element @xmath161 .
these two observations lead us to the following correspondence .
[ thmdivisormapcornonnormal ] for @xmath6 a semilocal @xmath5-finite reduced , s2 , and g1 ring and set @xmath12 .
we have the following bijection of sets : @xmath162 here the equivalence relation @xmath163 on the right is generated by two equivalences .
* we say that @xmath164 are equivalent if there exists a unit @xmath165 in @xmath166 such that @xmath167 . *
we say that @xmath150 and @xmath168 in @xmath169 are equivalent .
the proof is more subtle than in the normal case : if @xmath170 and @xmath171 are elements of @xmath172 that determine the same @xmath64-divisor , it does not imply that one is a unit multiple of the other as is the case for normal rings .
the reason for this is that there can be torsion in @xmath98 .
we certainly know that every @xmath143-linear map @xmath4 as described above induces an effective @xmath19 in @xmath108 by the procedure described above . furthermore
, this procedure is surjective by ( * ? ? ?
* proposition 2.9 ) .
thus we simply have to show that the two equivalence relations described in ( i ) and ( ii ) above are sufficient to induce a bijection .
so suppose that @xmath173 and @xmath174 induce the same divisor @xmath19 . by using ( ii )
, we may assume that @xmath175 .
thus we have two sections @xmath176 inducing the same @xmath64-ac divisors in @xmath177 , say @xmath178 and @xmath179 respectively .
it follows that there is an integer @xmath155 not divisible by @xmath7 such that @xmath180 . by making @xmath92 bigger if necessary
, we may take @xmath181 for some @xmath182 . as a section
, we claim that @xmath183 induces the divisor @xmath184 in @xmath98 via ( * ? ? ?
* proposition 2.9 ) .
we now prove this claim .
since divisors are determined in codimension @xmath15 , we may assume that @xmath12 where @xmath6 is a @xmath15-dimensional gorenstein ring . because @xmath6 is gorenstein , by duality for a finite map
, @xmath185 is isomorphic to @xmath186 ( abstractly ) . choose @xmath187 to be an @xmath186-module generator of @xmath185 .
we may thus write @xmath188 for some @xmath189 .
notice @xmath190 which corresponds to the sheaf @xmath191 .
it follows that @xmath192 therefore , because @xmath193 generates @xmath194 as an @xmath195-module by ( * ? ? ?
f ) or ( * ? ? ?
* lemma 3.9 ) , @xmath196 which proves the claim . thus @xmath197 and @xmath198 agree up to multiplication by a unit and so @xmath170 and @xmath199 are indeed related by relations ( i ) and ( ii ) .
we also have to verify that if @xmath170 and @xmath199 are related by conditions ( i ) or ( ii ) , then they induce the same element of @xmath108 . certainly condition ( i )
is harmless . to check condition ( ii ) ,
one simply has to tensor equation ( [ eqndivisortopower ] ) by @xmath200 .
if @xmath6 is not semilocal but additionally there is a quasi - isomorphism @xmath201 ( which occurs for example , if @xmath6 is essentially of finite type over a field , or more generally a gorenstein local ring ) , then the theorem above still holds as long as one restricts the left - hand - side to @xmath202 of course , the @xmath203 direction of the correspondence in theorem [ thmdivisormapcornonnormal ] always exists .
alternately , see ( * ? ? ?
* remark 9.5 ) .
suppose that @xmath6 is a reduced @xmath5-finite local ring of characteristic @xmath7 .
fix a map @xmath204 .
the pair @xmath205 is called _ @xmath5-pure _ if @xmath4 is surjective .
suppose that @xmath0 is an s2 , g1 and reduced @xmath5-finite scheme and @xmath206 corresponds at each point @xmath207 to a map @xmath208 via theorem [ thmdivisormapcornonnormal ] . the pair @xmath44 is called _ @xmath5-pure _ if @xmath209 is @xmath5-pure for all @xmath207 . while there is substantial freedom in the choice of @xmath210 associated to @xmath41 at each point , it is straightforward to verify that if two maps @xmath210 and @xmath211 are related via the two conditions ( i ) and ( ii ) in theorem [ thmdivisormapcornonnormal ] , then @xmath210 is surjective if and only @xmath211 is surjective . for a finite inclusion of rings @xmath212 and an @xmath6-linear map
@xmath213 one can ask when there is a @xmath55-linear map @xmath214 so that @xmath215
. the study of such extensions of maps in the case where @xmath6 and @xmath55 are both normal domains is carefully laid out in @xcite . however , even in that case , an arbitrary @xmath143-linear map on @xmath6 need not extend to a @xmath143-linear map on @xmath55 .
* example 3.4 ) consider for example @xmath216 , @xmath217 and the inclusion @xmath218 \subset \mbf{f}_3[x ] = s$ ] .
an @xmath6-basis for @xmath219 is @xmath220 and an @xmath55-basis for @xmath221 is @xmath222 .
any map @xmath223 is defined by the images @xmath224 .
a map @xmath225 which extends @xmath4 must agree with @xmath4 on @xmath220 . however , @xmath226 since @xmath26 is @xmath143-linear and so @xmath227 .
thus when @xmath228 is not divisible by @xmath229 , @xmath4 can not extend .
generic separability is necessary to guarantee extensions of maps ( * ? ? ?
* proposition 5.1 ) .
when the inclusion of fraction fields is not generically separable then only the zero map extends .
this separability allows one to use the trace map as a vehicle for understanding such extensions in the normal case . in the next section , where we address out motivating question
, we will see that the non - normal setting is even more complicated .
for an affine variety @xmath12 denote by @xmath230 the normalization . if we set @xmath140 to be the conductor ( the largest ideal of @xmath6 that is also an ideal of @xmath8 ) the inclusion @xmath231 extends to the following commutative diagram where the inclusions are the obvious ones and the other maps are the natural surjections .
\(a ) at ( -1,1 ) @xmath8 ; ( b ) at ( -1,-1)@xmath6 ; ( c ) at ( 1,0)@xmath232 ; ( d ) at ( 1,-2)@xmath233 ; ( b ) edge ( a ) ; ( a ) edge ( c ) ; ( b ) edge ( d ) ; ( d ) edge ( c ) ; recall also the following lemma .
* exercise 1.2.4(e ) ) [ lem : extendingmapsundernormalization ] suppose that @xmath6 is a non - normal reduced ring , @xmath8 is its normalization and @xmath140 is the conductor ideal .
any @xmath6-linear map @xmath204 is compatible with @xmath140 ( _ i.e. _ @xmath234 ) and extends uniquely to an @xmath8-linear map @xmath235 .
a proof can be found in ( * ? ? ?
* propositions 7.10 , 7.11 ) .
suppose we are given a map @xmath204 as above .
lemma [ lem : extendingmapsundernormalization ] implies that there exists a map @xmath235 and furthermore that the induced map @xmath236 extends to an @xmath237-linear map @xmath238 . in order to prove our main theorem
, we will relate the surjectivity of @xmath4 to that of @xmath26 by studying the surjectivity of @xmath239 verses @xmath240 . following the ideas of @xcite , it is natural to attempt to apply the trace map ( for the inclusion @xmath241 ) to solve this problem .
more generally , for a finite inclusion @xmath242 of reduced rings @xmath243 and @xmath29 where each minimal prime of @xmath29 lies over a minimal prime of @xmath243 , one can ask whether the trace map @xmath244 restricts to a _ surjective _ map from @xmath29 to @xmath243 ( here @xmath244 is defined to be the sum of the individual field trace maps ) . if this were always the case in our setting , one would have a _ surjective _ trace map @xmath245 and one could use this show that surjectivity of @xmath4 directly .
however , the next examples show the trace map can fail to be surjective ( or even fail to induce a map from @xmath29 to @xmath243 ) for inclusions @xmath242 .
suppose that @xmath246 is a perfect field of characteristic @xmath69 and consider the ring @xmath247/(xz^2 - y^2 ) \cong k[a^2 , ab , b ] \subseteq k[a , b]$ ] .
the conductor ideal is @xmath248 and so @xmath249 \cong r/\goth{c } \subseteq r^{\textnormal{n}}/\goth{c } \cong k[a]$ ] is generically purely inseparable .
thus the trace map @xmath245 is the zero map ( and in particular , not surjective ) .
likewise , one can construct similar examples of galois extensions @xmath241 of dedekind domains which are generically separable ( and so the trace map is non - zero ) but which have wild ramification and so the trace map is not surjective . consider the extension @xmath250 \subseteq k[x ] = b$ ] where the characteristic of @xmath246 is _ not _ equal to @xmath69 . the trace map @xmath251 yields @xmath252 .
however , we also have the inclusion @xmath253 \subseteq k[x ] = b$ ] noting that @xmath254 $ ] and @xmath255 $ ] have the same fraction field , @xmath256
. therefore @xmath257 .
by using pushout diagrams of schemes ( as in @xcite ) one can construct a ring @xmath6 where @xmath258 .
[ xmp1 ] let @xmath246 be any field and consider rings @xmath259
\oplus k[y ] \mid \text{$s$ and $ t$ have the same constant term } \}\ ] ] and @xmath260 \oplus k[y ] \mid \text{$u$ and $ v$ have the same constant term } \}\ ] ] over @xmath246 both having a node at the origin .
consider the normalizations @xmath261 \oplus k[y]$ ] and @xmath262\oplus k[y]$ ] respectively .
one can see the conductor of @xmath243 in @xmath263 is the ideal made up of all pairs @xmath264 with zero constant term .
likewise the conductor of @xmath29 in @xmath265 is the ideal made up of all pairs @xmath266 with no constant term . in this example
, we consider the trace on each irreducible component , and then add the resulting trace maps .
somewhat abusively , we call this sum the `` trace '' also , and denote it by @xmath267 .
clearly this trace map sends the conductor to the conductor but @xmath268 is not contained in @xmath243 because @xmath269 .
compare with the question `` trace map attached to a finite homomorphism of noetherian rings '' of bryden cais on asked on december 3rd , 2009 . in light of these examples ,
the condition that @xmath270 is too restrictive .
however , the following ( recursive ) definition , which is substantially weaker , will be exactly what we want in order to answer our motivating question .
[ def.hst ] suppose that @xmath6 is a reduced local ring and @xmath12 .
define @xmath230 to be the normalization with conductor @xmath140 .
we set @xmath39 and @xmath271 to be the subschemes defined by @xmath140 and set @xmath272 and @xmath273 to be the associated reduced subschemes .
we say @xmath0 has _ hereditary surjective trace _ provided that there is some irreducible component @xmath274 of @xmath273 dominated by an irreducible component @xmath275 of @xmath272 such that : * the induced trace map @xmath276 is surjective and * @xmath274 also has ( this condition is vacuous if @xmath274 is normal ) . in the language of commutative algebra
, a ring having _ _ means that there exist minimal associated prime ideals of @xmath140 ( which is an ideal of both @xmath6 and @xmath8 ) , @xmath277 and @xmath278 such that @xmath279 satisfying ( i ) is surjective and ( ii ) @xmath280 also has .
we say that a ( non - local ) scheme @xmath12 has _ _ if it has at every point .
observe that the dimension of @xmath28 is strictly smaller than that of @xmath0 , and so the recursive process of definition [ def.hst ] will stop after finitely many steps .
[ rem.strongerhst ] it would also be natural to require conditions ( i ) and ( ii ) for _ every _ irreducible component @xmath274 of @xmath273 dominated by an irreducible component @xmath275 of @xmath272 .
however , we will not need this stronger condition .
we consider the following special case of our main result whose proof we feel is illuminating .
[ xmp : basecase ] recall that any complete one dimensional seminormal ring of characteristic @xmath7 with algebraically closed residue field @xmath246 is isomorphic to the completion of the coordinate ring of some set of coordinate axes in @xmath281 by @xcite .
suppose that @xmath6 is an @xmath5-finite complete two - dimensional s2 seminormal ring with algebraically closed residue field .
use @xmath8 to denote the normalization of @xmath6 and suppose that @xmath282 is a surjective map which is compatible with the conductor ideal @xmath140 . because @xmath6 is s2 , @xmath283 is equidimensional and reduced
. therefore , @xmath283 is also @xmath5-pure and thus seminormal and so it is isomorphic to a direct sum of completions of coordinate rings of coordinate axes in various @xmath284 . set @xmath28 to be the pull - back of the diagram @xmath285 in other words , @xmath28 is the ring @xmath286 notice that @xmath287 is subintegral by construction , see @xcite .
therefore , @xmath288 since @xmath6 is seminormal .
this implies @xmath289 and so @xmath290 is also isomorphic to coordinate axes . given any component @xmath291 of @xmath292 and a component @xmath293 of @xmath294 dominating it ,
suppose that @xmath295 $ ] .
provided @xmath46 does not divide @xmath92 we see that definition [ def.hst](i ) holds . part ( ii ) holds vacuously as the components here are normal .
recall a small lemma about extending maps in local rings .
* observation 5.1 ) [ lem : local ] let @xmath6 be a local ring and @xmath296 a proper ideal .
suppose there is a surjective @xmath6-linear map @xmath297 which is the restriction of an @xmath6-linear map @xmath298 .
then @xmath299 is surjective . finally , we prove our main theorem .
[ thm : main ] suppose that @xmath0 is a reduced affine @xmath5-finite scheme having .
further suppose that @xmath24 is an @xmath13-linear map .
if the unique extension @xmath25 is surjective , then @xmath4 is also surjective .
we proceed by induction , the case where @xmath0 is zero - dimensional is obvious since then @xmath300 . furthermore , the statement is local so we may assume that @xmath12 where @xmath6 is a local @xmath5-finite ring of characteristic @xmath7 and @xmath230 .
we aim to show that an @xmath6-linear map @xmath213 which extends to a surjective @xmath8-linear map @xmath301 is also surjective .
notice that @xmath140 is radical in @xmath8 ( and thus also in @xmath6 ) because @xmath26 is surjective and @xmath140 is @xmath26-compatible .
modulo @xmath140 , one has maps @xmath302 and @xmath303 by lemma [ lem : extendingmapsundernormalization ] .
\(a ) at ( -1,1 ) @xmath8 ; ( b ) at ( -1,-1)@xmath6 ; ( c ) at ( 1,0)@xmath232 ; ( d ) at ( 1,-2)@xmath233 ; ( fa ) at ( -3,1)@xmath304 ; ( fb ) at ( -3,-1)@xmath305 ; ( fc ) at ( 3,0)@xmath306 ; ( fd ) at ( 3,-2)@xmath307 ; ( b ) edge ( a ) ; ( a ) edge ( c ) ; ( b ) edge ( d ) ; ( d ) edge ( c ) ; ( fa ) edge node[above]@xmath308 ( a ) ; ( fb ) edge node[above]@xmath309 ( b ) ; ( fb ) edge ( fa ) ; ( fc ) edge node[above]@xmath310 ( c ) ; ( fd ) edge node[above]@xmath311 ( d ) ; ( fd ) edge ( fc ) ; we will examine the behavior of the maps @xmath312 .
when @xmath26 is surjective then so too is @xmath313 and one can ask whether the surjectivity of @xmath313 implies surjectivity of @xmath314 .
this is equivalent to the surjectivity of @xmath4 by lemma [ lem : local ] .
since @xmath6 has , there are components of @xmath292 and @xmath294 which have a surjective trace map between their normalizations .
specifically , let @xmath315 and @xmath316 be such that @xmath317 and @xmath318 are the components in question .
furthermore , note that @xmath280 is local and @xmath319 is semilocal .
we consider the normalization of these rings and , by our hypothesis , we know that the trace map @xmath320 is surjective .
there are induced @xmath6-linear maps @xmath321 since minimal primes of a ring are always compatible .
furthermore because @xmath313 is surjective so too is @xmath322 .
so we have the following diagram .
\(a ) at ( -1,1 ) @xmath232 ; ( b ) at ( -1,-1)@xmath233 ; ( c ) at ( 1,0)@xmath319 ; ( d ) at ( 1,-2)@xmath280 ; ( fa ) at ( -3,1)@xmath306 ; ( fb ) at ( -3,-1)@xmath307 ; ( fc ) at ( 3,0)@xmath323 ; ( fd ) at ( 3,-2)@xmath324 ; ( b ) edge ( a ) ; ( a ) edge ( c ) ; ( b ) edge ( d ) ; ( d ) edge ( c ) ; ( fa ) edge node[above]@xmath310 ( a ) ; ( fb ) edge node[above]@xmath311 ( b ) ; ( fb ) edge ( fa ) ; ( fc ) edge node[above]@xmath325 ( c ) ; ( fd ) edge node[above]@xmath326 ( d ) ; ( fd ) edge ( fc ) ; since @xmath233 is local , by lemma [ lem : local ] , @xmath314 is surjective if and only if @xmath327 is . by lemma [ lem : extendingmapsundernormalization ]
, @xmath328 extends to a map @xmath329 and @xmath330 extends to a map @xmath331 for clarity , the diagram for @xmath332 is below .
\(a ) at ( -2,2)@xmath333 ; ( b ) at ( -2,0)@xmath334 ; ( c ) at ( 2,2 ) @xmath335 ; ( d ) at ( 2,0)@xmath280 ; ( b ) edge ( a ) ; ( a ) edge node[above]@xmath336 ( c ) ; ( b ) edge node[above]@xmath326 ( d ) ; ( d ) edge ( c ) ; note that @xmath337 is surjective as @xmath322 is . by construction , @xmath280 has , and since it is lower dimensional , by our inductive hypothesis it is sufficient to show that @xmath332 is surjective .
now we are in a situation to use the surjective trace map @xmath320 .
the following commutative diagram shows that @xmath338 is surjective as the top square is commutative by ( * ? ? ?
* corollary 4.2 ) .
see also ( * ? ? ?
* proposition 4.1 , theorem 5.6 ) .
this completes the proof .
( fb ) at ( -1.5,0)@xmath339 ; ( fc ) at ( -1.5,1.5)@xmath333 ; \(b ) at ( 1.5,0)@xmath340 ; ( c ) at ( 1.5,1.5)@xmath335 ; \(b ) edge node[right]@xmath341(c ) ; ( fb ) edge node[left]@xmath342(fc ) ; ( fb ) edge node[below]@xmath343 ( b ) ; ( fc ) edge node[above]@xmath336 ( c ) ;
we use the map - divisor correspondence described in theorem [ thmdivisormapcornonnormal ] to study schemes with and semi - log canonical singularities .
while @xmath5-pure schemes are known to have log canonical singularities , there is only a conjectural converse , see conjecture [ conj.fpurevslogcanonical ] in the introduction .
we pause to prove a few needed lemmas .
we refer the reader to @xcite for a detailed description of the reduction to characteristic @xmath344 process .
we also acknowledge the following abuse of notation , by @xmath344 we technically are referring to an open and zariski dense set of maximal ideals in @xmath345 , a finitely generated @xmath65-algebra used in the reduction to characteristic @xmath344 process . again , see the aforementioned reference for more details . see ( * ? ? ?
* theorem 2.3.6 ) for the relevance of the algebraically closed base - field assumption .
suppose we are given components of the conductor subschemes @xmath346 and @xmath347 with an degree @xmath348 finite morphism @xmath349 .
performing reduction to characteristic @xmath344 ( in particular @xmath350 ) , the trace map @xmath351 is clearly surjective .
continuing recursively , we can require that @xmath352 for @xmath353 s determined by a finite set of varieties .
this is certainly enough to guarantee .
in fact , we can guarantee the stronger variant of found in remark [ rem.strongerhst ] .
[ lem.divofextensionsplusconductor ] suppose that @xmath12 is an s2 , g1 seminormal scheme with normalization @xmath21 .
further suppose that @xmath24 corresponds to a divisor @xmath147 .
denote by @xmath25 the extension of @xmath26 to @xmath30 .
the @xmath35-divisor on @xmath30 corresponding to @xmath26 , denoted @xmath354 , satisfies @xmath355 where @xmath29 is the divisor corresponding to the conductor on @xmath30 , _
i.e. _ @xmath47 ( the conductor is pure codimension 1 because @xmath0 is s2 ) . first we explain how to pull back divisors via @xmath160 .
this pull - back process is completely determined in codimension 1 ( which is reasonable , since the divisors are determined in codimension 1 ) . thus , suppose that we are given a @xmath356 . by construction , since @xmath95 is ac , it is easy to pull - back ( work outside a set of codimension 2 , or see @xcite ) .
we define @xmath357 .
it is straightforward to verify that this is well defined .
the statement of the lemma can also be checked in codimension 1 and so we assume that @xmath6 is 1-dimensional and local .
write @xmath358 for some @xmath359 ( we can do this because @xmath19 is @xmath64-ac ) .
the pullback of @xmath147 is then just @xmath360 .
we claim it is sufficient to check the statement when @xmath361 ( so that @xmath362 ) .
to see this claim , choose @xmath363 such that @xmath364 is zero ( which we can do since we have reduced to the case where @xmath6 is gorenstein ) .
this @xmath365 generates @xmath366 as an @xmath82-module and so @xmath367 for some @xmath368 .
it follows that @xmath369 .
thus @xmath370 .
therefore , if @xmath371 , we can add @xmath372 to both sides of the equation , which proves the claim . in this context , @xmath6 is gorenstein and local and @xmath8 is regular and semi - local .
it follows that @xmath373 is a free @xmath8-module .
fix @xmath374 to be a generator ( and assume it sends @xmath15 to some @xmath375 which generates @xmath140 as an @xmath8-module ) . also notice that the assumptions imply that @xmath4 generates @xmath185 as an @xmath186-module .
furthermore , for any @xmath376-module generator @xmath377 , we know that we have @xmath378 up to multiplication by a unit ( which we can then absorb into @xmath365 obtaining a true equality ) . at the level of the field of fractions , @xmath4 and @xmath26 are the same map and @xmath187 is multiplication by @xmath379 .
thus , since @xmath378 , we have @xmath380 again at the level of the field of fractions
. therefore @xmath381 .
this implies that @xmath354 is the divisor of @xmath140 as desired .
[ cor.hstnormalizationqgorenstein ] suppose that @xmath12 is an affine @xmath5-finite scheme satisfying and which is also s2 , g1 and @xmath64-gorenstein .
set @xmath30 to be the normalization of @xmath0 and set @xmath29 to be the divisor on @xmath30 corresponding to the conductor ideal , _ i.e. _ @xmath47
. then @xmath0 is @xmath5-pure if and only if @xmath48 is @xmath5-pure .
since @xmath0 is @xmath64-gorenstein , by working sufficiently locally , we may assume that the zero divisor on @xmath0 corresponds to a map @xmath382 .
therefore lemma [ lem.divofextensionsplusconductor ] implies that @xmath383 corresponds to the divisor @xmath29 .
an application of theorem [ thm : main ] completes the proof .
if @xmath12 is a curve singularity with a node at @xmath207 , then @xmath30 is smooth and the conductor ideal is simply the ideal of @xmath229 . in particular ,
if one takes a @xmath384-module generator @xmath385 and extends it to a map @xmath386 , the divisor @xmath147 is zero while the divisor @xmath354 is the divisor of the origin with coefficient @xmath15 .
however , now suppose that a curve @xmath12 has a cusp singularity at @xmath207 .
note @xmath30 is still smooth and fix @xmath387 to be the preimage of @xmath229 .
the conductor ideal is the _ square _ of the ideal of the point of @xmath388 .
in particular , if one takes a @xmath384-module generator @xmath385 and extends it to a map @xmath386 , the divisor @xmath147 is zero while the divisor @xmath354 is the divisor of the origin with coefficient @xmath69 .
as before , consider a local non - normal s2 affine reduced scheme @xmath12 and the natural map @xmath21 on the normalization .
write @xmath230 and let @xmath28 be the subscheme associated to the conductor in @xmath0 and @xmath29 the divisor associated to the conductor in @xmath30 , _ i.e. _ @xmath47 . [ cor.fpurevsslc ]
assume conjecture [ conj.fpurevslogcanonical ] holds .
suppose that @xmath0 is a seminormal , s2 and g1 pair of finite type over an algebraically closed field @xmath246 of characteristic zero and @xmath107 is such that @xmath389 is @xmath35-cartier .
the pair @xmath20 has semi - log canonical singularities if and only if it has dense @xmath5-pure type .
we now refer the reader to both @xcite and @xcite for a detailed description of the reduction to characteristic @xmath344 process in this context .
as before , we also acknowledge the following abuse of notation , by @xmath344 we technically are referring to an open and zariski dense set of maximal ideals in @xmath345 , a finitely generated @xmath65-algebra used in the reduction to characteristic @xmath344 process . again , see the aforementioned references for more details .
since the s2 property can be detected by examining the support of finitely many @xmath390 modules , the reductions to characteristic @xmath344 are also s2 . likewise because @xmath0 is g1 , there is a subset @xmath391 of codimension greater than @xmath15 , such that @xmath392 is gorenstein .
thus the same can be preserved after reduction to characteristic @xmath344 . finally , since @xmath0 is seminormal
, so are its reductions to characteristic @xmath344 , to see this use ( * ? ? ?
* corollary 2.7(vii ) ) .
set @xmath21 to be the normalization and fix @xmath271 and @xmath39 to be the subschemes defined by the conductor , respectively .
we can reduce these schemes and subschemes to characteristic @xmath344 as well .
assume that @xmath20 is semi - log canonical which implies that @xmath393 is log canonical .
for a zariski - dense set of characteristic @xmath344 , by assumption we have that @xmath394 is @xmath5-pure . by working on sufficiently small affine charts
, we also assume that @xmath395 and the same holds for @xmath396 in characteristic @xmath344 .
in fact , we may even represent @xmath397 and assume that it is @xmath64-cartier ( or @xmath64-linearly equivalent to zero ) .
it may also be helpful to the reader to notice that , under either hypothesis , @xmath398 has no common components with @xmath29 and so pathologies discussed in remark [ rem.pathologiesfordivisorsonnonnormal ] can be avoided by viewing @xmath19 as an element of @xmath399 , the @xmath40-weil divisorial sheaves are the @xmath400 which equal @xmath13 along the non - normal locus of @xmath0 .
associated @xmath40-divisors may be treated more like @xmath40-divisors on normal varieties ( in particular , the subgroup of weil divisorial sheaves has no torsion ) .
] , see @xcite . for @xmath344
, we may assume that the index of @xmath396 is not divisible by the characteristic @xmath7 .
thus @xmath401 induces a map @xmath402 , which extends to @xmath403 .
the divisor associated to @xmath26 is thus @xmath404 and so @xmath26 is surjective by lemma [ lem.divofextensionsplusconductor ] .
now @xmath405 has for @xmath344 , and so @xmath4 is surjective for a zariski - dense set of @xmath344 .
this proves the ( @xmath406 ) direction .
conversely , suppose that @xmath20 has dense @xmath5-pure type .
but again by lemma [ lem.divofextensionsplusconductor ] this implies that @xmath393 also has dense @xmath5-pure type , which implies that @xmath393 is log canonical by @xcite and so @xmath20 is semi - log canonical by definition .
we first review the inversion of adjunction statement we are concerned with .
fix a pair @xmath62 where @xmath0 is a normal scheme , @xmath55 a reduced integral weil divisor and @xmath19 an effective @xmath35-divisor , with no common components with @xmath55 , such that @xmath407 is @xmath40-cartier .
set @xmath408 to be the normalization of @xmath55 and recall that there is a canonically defined divisor @xmath59 called the _ different of @xmath19 on @xmath57 _ which satisfies @xmath409 .
in general , adjunction and inversion of adjunction is the comparison of the singularities of a pair @xmath410 with the singularities of @xmath63 .
the implication `` @xmath62 is log canonical @xmath406 @xmath411 is log canonical '' is called the _ adjunction direction_. the converse implication ( at least near @xmath55 ) is known as _ inversion of adjunction _ ; see @xcite , ( * ? ? ?
* chapter 17 ) .
the direct analog of the adjunction direction is known in characteristic @xmath7 ( * ? ? ?
8.2(iv ) ) .
in particular , in characteristic @xmath7 , if additionally the index of @xmath407 is not divisible by @xmath7 then there exists a canonically determined divisor @xmath59 , called the _
@xmath5-different _ , such that @xmath412 and furthermore if @xmath62 is @xmath5-pure , then so is @xmath63 .
the @xmath5-different @xmath59 is constructed as follows : we work locally and so may assume that @xmath12 is the spectrum of a local ring .
by hypothesis , there exists a map @xmath413 corresponding to @xmath414 .
the ideal @xmath415 is @xmath416-compatible ( see for example , ( * ? ? ?
* section 7 ) ) , and so there is an induced map @xmath417 .
now , @xmath55 is not necessarily normal ( or s2 or g1 ) so it is difficult interpret @xmath4 as a divisor .
however , by lemma [ lem : extendingmapsundernormalization ] , @xmath4 extends to a map @xmath418 . finally , we associate to this map the divisor @xmath419 . in ( * ? ?
* example 8.4 ) an example is produced where @xmath62 is not @xmath5-pure ( _ i.e. _ @xmath416 is not surjective ) but @xmath63 is @xmath5-pure ( _ i.e. _ @xmath26 is surjective ) . in other words ,
inversion of adjunction fails .
this counterexample is loaded with exactly the pathologies that are avoided by rings having and the following corollary is easy to prove .
[ cor.invofadj ] suppose that @xmath0 is a normal scheme of characteristic @xmath7 , @xmath19 is a @xmath35-divisor on @xmath0 and @xmath55 is a reduced integral weil divisor on @xmath0 , with no common components with @xmath19 , such that @xmath56 is @xmath35-cartier with index not divisible by @xmath46 .
denote by @xmath57 the normalization of @xmath55 and @xmath58 the natural map .
there exists a canonically determined @xmath35-divisor @xmath59 on @xmath57 such that @xmath420 .
furthermore , if @xmath55 has , then @xmath62 is @xmath5-pure near @xmath55 if and only if @xmath63 is @xmath5-pure . we need only prove the final statement . using the notation above
, it follows from lemma [ lem : local ] that @xmath416 is surjective if and only if @xmath4 is surjective . by our main theorem ,
@xmath4 is surjective if and only if @xmath26 is surjective because @xmath55 has . :
_ generalized divisors on gorenstein schemes _ , proceedings of conference on algebraic geometry and ring theory in honor of michael artin , part iii ( antwerp , 1992 ) , vol . 8 , 1994 , pp . 287339 .
mr1291023 ( 95k:14008 ) : _ flips and abundance for algebraic threefolds _ ,
socit mathmatique de france , paris , 1992 , papers from the second summer seminar on algebraic geometry held at the university of utah , salt lake city , utah , august 1991 , astrisque no . 211
mr1225842 ( 94f:14013 ) : _ positivity in algebraic geometry .
ii _ , ergebnisse der mathematik und ihrer grenzgebiete .
3 . folge . a series of modern surveys in mathematics [ results in mathematics and related areas .
3rd series .
a series of modern surveys in mathematics ] , vol .
49 , springer - verlag , berlin , 2004 , positivity for vector bundles , and multiplier ideals .
mr2095472 ( 2005k:14001b ) : _ gluing schemes and a scheme without closed points _ , proceedings of the 2002 john h. barrett memorial lectures conference on algebraic and arithmetic geometry ( p. t. y. kachi , s. mulay , ed . ) , contemporary mathematics , vol .
386 , american mathematical society , providence , ri , 2005 , pp . 157172 | if @xmath0 is frobenius split , then so is its normalization and we explore conditions which imply the converse . to do this , we recall that given an @xmath1-linear map @xmath2 , it always extends to a map @xmath3 on the normalization of @xmath0 . in this paper
, we study when the surjectivity of @xmath3 implies the surjectivity of @xmath4 . while this does nt occur generally , we show it always happens if certain tameness conditions are satisfied for the normalization map .
our result has geometric consequences including a connection between @xmath5-pure singularities and semi - log canonical singularities , and a more familiar version of the ( @xmath5-)inversion of adjunction formula .
frobenius map , frobenius splitting , f - purity , semi - log canonical , log canonical , inversion of adjunction , normalization , seminormal |
there are some sport tournaments where the ranking of the competitors is based on the results of games played against each other . in the world of sport
there is no consensus in using a particular ranking method .
various evaluation methods have been applied to different events taking into account the specifics of the particular sport activity .
further complication emerges when participants are teams , and the final ranking should also reflect individual results .
one characteristic example for that situation is a chess team championship . in this article
it is assumed that the final ranking of participants of a sport tournament is based on the outcome of the games that have been played against each other .
the result of a game is given according to the rules of the particular sport , however , reasonable data transformation is also allowed .
ranking of the participants will be made by applying the pairwise comparison models of multicriteria decision making methodology .
it has two main advantages : the obvious interpretation of the games played against each other and the ability to address the problem of intransitivity ( cyclical preferences regarding three alternatives : a is better than b , b is better than c , but c is better than a ) , a common feature of subjective evaluations by individuals , but a well - known phenomenon in objective sport results , too .
the alternatives to be compared are the participants of the tournament , their game results will be incorporated into a pairwise comparison matrix , and the final result will follow the ranking of the priority vector derived from the matrix with an estimation method .
it is worth mentioning that this approach can be applied not only for some real competitions ( national soccer , basketball , hockey , etc .
championships , chess tournaments ) , but for the ranking of teams or individuals having historic data of their rivalry ( tennis or chess players , athletes ) .
although the application of complete pairwise comparison matrices ( i.e. everybody met all the other competitors at least once ) can raise interesting questions to be analyzed ( comparison of the official rankings and those obtained by using the pairwise comparison method , for instance ) , the focus of this article is to use incomplete pairwise comparison matrices for the final ranking .
this is the reason why the analyzed case is a chess olympiad : it is an ideal example of the potential application of incomplete pairwise comparison matrices .
chess competitions are often organized as a swiss - system tournament .
all participants face each other for a determined number of games ( rounds are often organized at the same time ) , while there is no knockout phase .
it means that a loss in the first rounds does not make impossible for the contestant to win the championship , as well as all participants play the same number of matches .
the final rank of the players ( teams ) is determined mainly by the application of different lexicographical orders , but it lacks a well established solution for determining the final ranking taking into account the performance of opponents of a team is a central issue as not all possible matchings materialized , which poses a serious challenge .
it will be shown that an alternative way based on the results against each other could give a ranking which is in some sense intuitively better than the currently used methods .
the more detailed analysis will take the results of the 39th chess olympiad open tournament as a basis .
the competition took place from september 20th to october 3rd , 2010 in khanty - mansiysk , russia .
this choice was supported by the following arguments : * it was an important sport event recently ; * results were easy to collect and to adapt to the pairwise comparison method ; * not only the winner , but the position of other participants were of interest ; * the reciprocity of the pairwise comparison matrix was a reasonable assumption ; * participants were interested in the size of their win or lose ; * the official ranking method is debated .
all of these issues will be thoroughly discussed .
the article is organized as follows . in section
[ theory ] the applied theory and some methods for the consistent approximation of incomplete pairwise comparison matrices will be described .
section [ adapt ] deals with the representation of the tournament results in a pairwise comparison format . special features of the chess olympiad which make the necessary assumptions reasonable will also be presented here .
section [ rankings ] analyzes different rankings with a focus on comparing the official result with those of calculated from pairwise comparisons .
for this purpose a distance measure will be introduced .
section [ characteristics ] highlights some interesting properties of the example .
section [ visualization ] uses multidimensional scaling ( mds ) to draw the rankings on a two - dimensional map based on the distance defined in section [ rankings ] .
it reveals that the different rankings obtained from pairwise comparisons are close to each other . in section [ conclusion ] the summary of the results and the outline of further research
are given .
pairwise comparisons have been widely used in decision making since saaty published the ahp method @xcite . in a multicriteria decision making problem
the @xmath0 real matrix @xmath1 is a pairwise comparison matrix if it is positive and reciprocal that is @xmath2 and @xmath3 for all @xmath4 .
the reciprocity condition also means that @xmath5 for all @xmath6 .
matrix element @xmath7 is the numerical answer to the question how many times is the @xmath8th alternative more important / better / favourable than the @xmath9th? the final aim of the use of pairwise comparisons is to determine a weight vector @xmath10 for the alternatives , where @xmath11 somehow approximates the pairwise comparison @xmath7 . the solution is obvious if matrix @xmath12 is consistent , namely @xmath13 for all @xmath14 , because there exists a positive @xmath15-dimensional vector @xmath16 such that @xmath17 for all @xmath4 . in the inconsistent case the real values of the decision maker
can only be estimated .
saaty proposed the eigenvector method ( em ) for this purpose , which is based on the perron theorem @xcite , as a positive matrix has a dominant eigenvalue with multiplicity one and an associated strictly positive eigenvector .
distance - minimization techniques , like logarithmic least squares method ( llsm ) @xcite minimize the function @xmath18 where @xmath19 is the proper difference of @xmath7 from its approximation @xmath11 ( in case of llsm @xmath20 ) .
it may happen that a subset of pairwise comparisons are unknown due to the lack of available data , uncertain evaluations , or other problems .
incomplete pairwise comparison matrices were introduced in @xcite , for example ( missing elements are denoted by @xmath21 ) : @xmath22 in order to handle incomplete pairwise comparison matrices , introduce the variables @xmath23 for the @xmath24 missing elements in the upper triangle of @xmath12 supposing reciprocity ( in all there are @xmath25 unknown entries in the matrix ) . the new matrix is denoted by @xmath26 as its elements are the functions of the variables : @xmath27 ( in)complete pairwise comparison matrices can be represented by graphs @xcite .
let @xmath12 be a pairwise comparison matrix of size @xmath0 .
then @xmath28 , where @xmath29 , vertices correspond to the alternatives , and @xmath30 , thus @xmath31 represents the structure of known elements .
a recent result in this field is the extension of em and llsm to the incomplete case @xcite .
the em solution is arising from the idea that inconsistency could be measured by the maximal eigenvalue , so it is a natural approach to choose the unknown elements for minimizing the dominant eigenvalue . therefore the task is : @xmath32 , where @xmath33 .
as proved in @xcite , the solution is unique if and only if graph @xmath34 is connected , thus two alternatives could be compared via directly or indirectly through other alternatives .
an algorithm is also proposed for the @xmath35-optimal completion based on cyclic coordinates .
another opportunity is the llsm method . in this case
only the known elements of @xmath12 are examined : @xmath36 ^ 2 \ } \rightarrow \min$ ] .
the solution is unique again if and only if graph @xmath34 is connected , thus it depends strictly on the position of known elements , not on their exact values @xcite
. calculation of the optimal weights requires only the inverse of the upper - left @xmath37 submatrix in the @xmath38 laplace - matrix of graph @xmath34 and some additional matrix multiplication .
the problem can be solved quickly by any commonly used office softwares , like ms excel .
an example for an incomplete pairwise comparison matrix is the result matrix of a chess olympiad described in the introduction . in the 39th chess olympiad 2010 open tournament officially 149 teams participated .
all of them played 11 matches , except for some teams due to their late arrivals .
the matrix consists of 810 results from played matches , the ratio of known elements is small ( 7.3% ) as 11,026 elements is placed in the upper triangle of a @xmath39 matrix .
the diagonal elements of the matrix contains unities .
there was no significant tendency in number of draws during the 11 rounds and it is the least frequent result .
it implies that weaker teams played no simple draws in the final rounds , when they still had no chances to reach a good position .
this is important in chess , as draws could have arranged by the mutual agreement of players .
however , that possibility is not totally excluded . the distribution of results is presented in figure [ fig1 ] . in the tournament ,
all teams constituted by 5 players ( 4 and 1 reserve ) , a match between two of them includes 4 games with 3 different results ( white win or loss , draw ) . in both teams
2 players have white , implying symmetricity as the chances of the match are a priori equal .
the winner of the game gets 1 game point , the loser 0 , while draw means 0.5 game points for each player
. therefore the final result of the match for one team , the sum of players game points , ranges from 0 to 4 , by 0.5 .
if a team achieves in the match minimum 2.5 game points , it gets 2 match points .
if it scores 2 game points ( thus its opponent has likewise 2 game points ) , it gets 1 match point .
if it secures 1.5 or fewer game points , it counts as 0 match points for this match .
the sum of allocated match points is always 2 .
the official ranking method is a lexicographical order determined by the application of 4 tie - breaking procedures in sequence , proceeding from tb1 to tb4 to the extent required.[multiblock footnote omitted ] the primary criteria of the lexicographical order is tb1 , the number of match points .
however , as at most 22 match points could be scored in 11 matches , tb2 rules certainly should be applied , which means that teams are strongly interested in increasing their game points , players can not be satisfied with a simple 2.5:1.5 win .
consequently , the size of wins reliably reflects the difference of teams performance and it is justified to give higher weights to bigger wins .
this is not the case in a lot of sport events .
it was presented that match results correspond to the main conception of ratios used in pairwise comparison matrices , and they could be transformed into values ( ratios ) .
draws ( 2:2 ) are obviously converted to 1 , but in the other 4 sort of results 4 different rules were applied ( reciprocity ensures that it is enough to see the results from the viewpoint of the winner ) .
the variants are presented in table [ table2 ] .
.transformation of match results into pairwise comparison values ( ratios ) [ cols="<,^,^,^,^,^",options="header " , ] the ratio between the maximal and minimal weights has remarkable implications highlighted in the last 2 rows of table [ table7 ] .
average win s ratio corresponds to the mean of wins in each coding : @xmath40 for example , it is @xmath41 for b variant .
finally , power equals to @xmath42 , which reflects a kind of order of magnitude in the tournament .
for example , in a variant the factor between the first and the last team is 150-fold , while a standard victory in a match corresponds to a pairwise comparison ratio of 3.2 . as @xmath43 ,
a virtual chain of 5 teams should exist , where each team defeated the following with this average difference .
however , if one of the first and last teams played against each other , the ratio as a known element in the incomplete pairwise comparison matrix is at most 5 according to a 4:0 win in a variant .
it implies participants should be put into 4 - 5 groups .
figure [ fig6 ] presents the weights derived from llsm vectors . due to the pairing method
, the classification of teams partly happens .
the structure of known elements in the pairwise comparison matrix shows an impressive picture ; figure [ fig4 ] indicates them with a filled box covering approximately 7% of the whole area , which is the ratio of known elements relative to all possible matchings . on the two charts matrix rows ( teams )
are ordered according to final ( a , left ) and a - llsm ( b , right ) rankings , respectively . known elements representing matches knit around the diagonal ,
this effect is more stronger in the second case .
it is affirmed by some calculations : the average difference between the rank of two teams played against each other is 28.70 in final and 25.32 in a - llsm , while the median is 22.5 and 19 .
thus matches were taken place between contestants with a similar performance , the matching algorithm operates not randomly . despite the fact that the pairing method was not examined , the above numbers suggest that the proposed pairing based on pairwise comparisons is better regarding the classification of teams if the number of matches is limited , they should be played between participants whose internal ranking is difficult to decide . amid these circumstances , it is favourable that the number of rounds in the tournament remains small compared to the number of participants , as it makes possible the appearance of significant ratios not permitted by the transformation of match results into pairwise comparison ratios .
it raises the question whether all countries should participate in the same tournament which is not the case in almost all team sports like soccer or ice hockey .
rankings based on pairwise comparison matrices have revealed some interesting facts about the performance of certain teams , nonetheless , the long lists do not give an impressive picture about their relation .
the @xmath44 distance seems to be a good starting point , as it makes possible to plot @xmath15 rankings in a @xmath45 dimensional space .
in addition , it is possible to decrease number of dimensions in order to give a better overview if it goes without significant loss of information .
multidimensional scaling ( mds ) is a statistical method in information visualization for exploring similarities or dissimilarities in data ; a basic book in this field is @xcite .
a classical application of mds is to draw cities on a map from the matrix consisting of their air distance . for sufficiently small @xmath15
, it requires only 1 or 2 dimensions to visualize all points representing different rankings .
the distance matrix of the 7 rankings is still calculated in section [ rankings ] , but the applied software ( spss v18.0 ) requires at least 10 cases to be evaluated .
thus , regarding former observations , 3 further rankings ( sonneborn - berger , buchholz , mix ) were defined on the basis of different components of the original lexicographical order.[multiblock footnote omitted ] mds maps the original distances in interval , ratio or ordinal scales ; the most general interval scale were applied . here
discrepancies on the reduced - dimension map are related to the original @xmath24 distances by a linear function : @xmath47 , which means it is indifferent to multiplying the distances by a constant factor caused by choosing a different base of logarithm for @xmath44 .
the goodness of mapping ( the information loss derives from dimension reduction ) is measured by kruskal s stress and rsq .
the pairwise @xmath44 distances 10 rankings could be mapped appropriately in two dimensions , but one dimension is not enough .
the coordinates on the mds map has no further meaning , distances on the map reflects the original distances as well as possible , while the mean of @xmath48 and @xmath49 coordinates is 0 .
stress is 0.1431 , a middle - strength relation .
rsq reaches 0.9468 , approximately 94.5% of variance of the scaled data can be accounted for by the mds procedure .
the reduced map is plotted in figure [ figure3 ] .
start and final rankings are the farthest from each other , the 5 rankings derived from the incomplete pairwise comparison matrices are in a small cluster approximately at the same distance from start and final .
the proposed rankings are almost the same , only d - llsm differs from the others marginally ( it gives a great weight to all wins , but hardly rewards its size ) .
the 3 newly defined rankings ( sonneborn - berger , buchholz and mix ) are nearer to the final than to the start which is reassuring , while buchholz and mix are still nearer to the proposed rankings .
it backs the former observation that methods based on pairwise comparison matrices overweighs tb4 ( buchholz points ) to tb2 ( sonneborn - berger points ) , while the official lexicographical order prefers teams with higher tb2 ( sonneborn - berger points ) .. ] it is not by chance , as tb4 ( buchholz points ) rather reflects the performance of opponents , the key idea beyond pairwise comparison matrices .
this paper presents an alternative method to determine the final ranking of teams participated in the 39th chess olympiad 2010 open tournament based on incomplete pairwise comparison matrices . in lack of former experience
, 4 arbitrary scales were used to transform the match results into ratios .
two well - known techniques are used , the eigenvector method ( em ) and the logarithmic least squares method ( llsm ) .
the decision between them is far from trivial , while other methods deserve testing , like the one given by fedrizzi and giove in @xcite .
results show that in some cases the incompleteness of the pairwise comparison matrix has favourable effects if it has a special structure , namely , if ordering on the priority vector derived from the matrix , known elements are located near the diagonal with high probability : some ratios given by the optimal completion can significantly exceed the original scale .
the chosen variant for transforming match results into ratios affects the absolute priorities significantly , but final rankings served by weights are relatively robust .
it was confirmed by the application of a distance based on the asymmetry of committed mistakes in rankings , however , it was strongly influenced by some particular teams .
the construction of better indices is a topic of future research .
matches with the same result were represented by the same ratios .
it means some simplification as a 4:0 win in the final round against a strong opponent seems to be more valuable than a similar result in the first rounds against underdogs .
however , the official ranking does not take into account the opponent s strength in awarding victories , and a priori it is not reasonable to transform these outcomes differently .
nevertheless , further investigations are needed how this aspect could be incorporated into the pairwise comparison method .
alternative proposals revealed some oddities of the current fide olympiad tie - breaking rules .
as there is no commonly accepted ranking method in chess , other techniques are worth examining .
visualization implies that certain lexicographical orders , despite the lack of continuity , approximates the results of the proposed method well .
while it is really impossible to find a perfect final ranking for a similarly complex swiss - system tournament , it was justified that in some cases the use of pairwise comparison matrices give robust and intuitively better results than currently used ranking techniques .
this approach can be extended to other sport championships .
the author is very grateful to the anonymous reviewers for their valuable comments and suggestions .
the research was supported by otka grant k-77420 . 99 arsigny , v , fillard , p , pennec , x , ayache , n ( 2006 ) fast and simple calculus on tensors in the log - euclidean framework . magnetic resonance in medicine 56(2 ) : 411421 bozki , s , flp , j , rnyai , l ( 2010 ) on optimal completion of incomplete pairwise comparison matrices .
mathematical and computer modelling 52(1 - 2 ) : 318333 crawford , g , williams , c ( 1980 ) analysis of subjective judgment matrices .
the rand corporation , office of the secretary of defense usa , r-2572-af crawford , g , williams , c ( 1985 ) a note on the analysis of subjective judgment matrices .
journal of mathematical psychology 29 : 387405 de graan , jg ( 1980 ) extensions of the multiple criteria analysis method of t.l .
presented at euro iv conference , cambridge , uk , july 22 - 25 , 1980 fedrizzi , m , giove , s ( 2007 ) incomplete pairwise comparison and consistency optimization .
european journal of operational research 183(1 ) : 303313 gass , si ( 1998 ) tournaments , transitivity and pairwise comparison matrices .
journal of the operational research society 49 : 616624 handbook of fide .
regulations for specific competitions , chess olympiad , olympiad pairing rules .
+ approved by the 1994 and 1998 executive councils , amended by the 2006 presidential board and the 2007 and 2009 executive council .
+ http://www.fide.com/fide/handbook.html?id=95&view=article harker , pt ( 1987 ) incomplete pairwise comparisons in the analytic hierarchy process .
mathematical modelling 9(11 ) : 837848 kri , g ( 2011 ) on qualitatively consistent , transitive and contradictory judgment matrices emerging from multiattribute decision procedures .
central european journal of operations research 19(2 ) : 215224 kruskal , jb , wish , m ( 1978 ) multidimensional scaling .
sage publications , beverly hills and london norris , an ( 2005 ) the isotropic material closest to a given anisotropic material . j. mech . materials structures 1(2 ) : 231246 perron , o ( 1907 ) zur theorie der matrizen .
mathematische annalen 64 : 248 - 263 results of the 39th olympiad khanty - mansiysk 2010 open tournament .
+ http://chess-results.com/tnr36795.aspx?lan=1 saaty , tl ( 1980 ) the analytic hierarchy process .
mcgraw - hill , new - york spearman , c ( 1904 ) the proof and measurement of association between two things .
american journal of psychology 15 : 72101 tarsitano , a ( 2008 ) nonlinear rank correlations .
working paper .
agostino tarsitano , italy .
( in press ) http://eprints.bice.rm.cnr.it/586/
[ table10 ] lcccccc ukraine & 2 & 1 & 2 & 2 & 2 & 2 + russia 1 & 1 & 2 & 1 & 1 & 1 & 1 + israel & 11 & 3 & 4 & 4 & 4 & 4 + hungary & 5 & 4 & 3 & 3 & 3 & 3 + china & 3 & 5 & 6 & 5 & 5 & 7 + russia 2 & 4 & 6 & 7 & 7 & 7 & 9 + armenia & 6 & 7 & 5 & 6 & 6 & 5 + spain & 16 & 8 & 11 & 9 & 10 & 10 + united states of america & 9 & 9 & 10 & 11 & 11 & 8 + france & 10 & 10 & 8 & 10 & 9 & 6 + poland & 15 & 11 & 9 & 8 & 8 & 11 + azerbaijan & 7 & 12 & 12 & 12 & 12 & 13 + russia 3 & 14 & 13 & 13 & 14 & 13 & 12 + belarus & 35 & 14 & 17 & 17 & 17 & 17 + netherlands & 13 & 15 & 16 & 16 & 16 & 14 + slovakia & 22 & 16 & 22 & 22 & 22 & 22 + brazil & 24 & 17 & 25 & 26 & 25 & 24 + india & 19 & 18 & 20 & 21 & 20 & 20 + denmark & 44 & 19 & 46 & 47 & 47 & 42 + czech republic & 17 & 20 & 14 & 13 & 14 & 15 + italy & 30 & 21 & 18 & 19 & 18 & 18 + greece & 25 & 22 & 24 & 23 & 24 & 25 + cuba & 18 & 23 & 19 & 18 & 19 & 19 + england & 12 & 24 & 21 & 20 & 21 & 21 + argentina & 26 & 25 & 34 & 34 & 34 & 33 + estonia & 48 & 26 & 32 & 31 & 32 & 32 + kazakhstan & 41 & 27 & 35 & 36 & 36 & 35 + moldova & 31 & 28 & 37 & 38 & 37 & 36 + iran & 38 & 29 & 31 & 32 & 31 & 30 + georgia & 20 & 30 & 15 & 15 & 15 & 16 + bulgaria & 8 & 31 & 28 & 28 & 28 & 31 + croatia & 28 & 32 & 26 & 25 & 26 & 26 + serbia & 21 & 33 & 23 & 24 & 23 & 23 + sweden & 34 & 34 & 29 & 30 & 30 & 28 + lithuania & 39 & 35 & 55 & 55 & 55 & 58 + slovenia & 29 & 36 & 30 & 29 & 29 & 29 + canada & 53 & 37 & 42 & 42 & 42 & 41 + austria & 45 & 38 & 45 & 43 & 46 & 48 + russia 4 & 52 & 39 & 49 & 49 & 49 & 51 + iceland & 54 & 40 & 47 & 46 & 45 & 44 + egypt & 40 & 41 & 50 & 50 & 50 & 50 + montenegro & 56 & 42 & 43 & 44 & 43 & 43 + qatar & 55 & 43 & 77 & 77 & 76 & 79 + peru & 46 & 44 & 51 & 52 & 51 & 49 + turkey & 50 & 45 & 57 & 58 & 58 & 54 + uruguay & 74 & 46 & 65 & 66 & 65 & 65 + zambia & 121 & 47 & 91 & 89 & 91 & 92 + icsc & 72 & 48 & 75 & 76 & 75 & 73 + uzbekistan & 33 & 49 & 39 & 39 & 39 & 40 + philippines & 37 & 50 & 44 & 45 & 44 & 46 + norway & 23 & 51 & 36 & 35 & 35 & 37 + vietnam & 27 & 52 & 27 & 27 & 27 & 27 + chile & 51 & 53 & 38 & 37 & 38 & 38 + colombia & 57 & 54 & 40 & 40 & 40 & 39 + australia & 49 & 55 & 48 & 48 & 48 & 47 + former yug rep of macedonia & 43 & 56 & 58 & 57 & 57 & 61 + albania & 68 & 57 & 59 & 59 & 59 & 57 + singapore & 73 & 58 & 60 & 60 & 60 & 55 + finland & 60 & 59 & 64 & 64 & 64 & 63 + belgium & 71 & 60 & 68 & 69 & 69 & 70 + united arab emirates & 88 & 61 & 72 & 70 & 71 & 72 + pakistan & 123 & 62 & 73 & 74 & 74 & 71 + ipca & 70 & 63 & 81 & 81 & 81 & 80 + germany & 42 & 64 & 41 & 41 & 41 & 45 + switzerland & 47 & 65 & 56 & 54 & 54 & 62 + bosnia & herzegovina & 32 & 66 & 33 & 33 & 33 & 34 + indonesia & 67 & 67 & 53 & 53 & 53 & 56 + kyrgyzstan & 77 & 68 & 63 & 61 & 63 & 64 + latvia & 58 & 69 & 54 & 56 & 56 & 53 + russia 5 & 61 & 70 & 62 & 63 & 62 & 60 + mongolia & 66 & 71 & 61 & 62 & 61 & 59 + mexico & 36 & 72 & 52 & 51 & 52 & 52 + bangladesh & 82 & 73 & 92 & 92 & 92 & 89 + south africa & 81 & 74 & 80 & 80 & 80 & 78 + portugal & 59 & 75 & 66 & 65 & 66 & 66 + turkmenistan & 69 & 76 & 67 & 68 & 68 & 69 + jordan & 79 & 77 & 71 & 73 & 72 & 67 + libya & 105 & 78 & 95 & 95 & 94 & 95 + paraguay & 84 & 79 & 85 & 85 & 85 & 84 + faroe islands & 83 & 80 & 79 & 79 & 79 & 75 + venezuela & 64 & 81 & 76 & 78 & 78 & 74 + costa rica & 80 & 82 & 84 & 84 & 84 & 83 + scotland & 63 & 83 & 70 & 72 & 70 & 68 + yemen & 85 & 84 & 88 & 90 & 89 & 85 + ecuador & 65 & 85 & 69 & 67 & 67 & 76 + tajikistan & 62 & 86 & 78 & 75 & 77 & 81 + andorra & 89 & 87 & 82 & 82 & 82 & 82 + ireland & 75 & 88 & 74 & 71 & 73 & 77 + algeria & 91 & 89 & 98 & 97 & 98 & 99 + dominican republic & 87 & 90 & 89 & 87 & 87 & 93 + new zealand & 92 & 91 & 86 & 86 & 86 & 87 + malaysia & 86 & 92 & 96 & 96 & 96 & 96 + thailand & 94 & 93 & 87 & 88 & 88 & 91 + panama & 107 & 94 & 105 & 105 & 104 & 101 + barbados & 96 & 95 & 97 & 98 & 97 & 97 + japan & 101 & 96 & 102 & 102 & 102 & 103 + luxembourg & 90 & 97 & 94 & 94 & 95 & 94 + cyprus & 109 & 98 & 103 & 103 & 103 & 100 + guatemala & 103 & 99 & 99 & 100 & 99 & 98 + malta & 106 & 100 & 116 & 116 & 116 & 114 + nigeria & 139 & 101 & 112 & 114 & 113 & 109 + ibca & 78 & 102 & 90 & 91 & 90 & 88 + iraq & 76 & 103 & 83 & 83 & 83 & 86 + sri lanka & 115 & 104 & 101 & 101 & 101 & 104 + jamaica & 97 & 105 & 100 & 99 & 100 & 107 + uganda & 127 & 106 & 114 & 115 & 114 & 113 + nepal & 114 & 107 & 113 & 110 & 112 & 116 + puerto rico & 100 & 108 & 93 & 93 & 93 & 90 + lebanon & 99 & 109 & 104 & 104 & 105 & 102 + monaco & 95 & 110 & 106 & 106 & 106 & 106 + honduras & 125 & 111 & 108 & 108 & 108 & 112 + palestine & 134 & 112 & 123 & 123 & 122 & 124 + korea & 116 & 113 & 120 & 120 & 120 & 120 + bolivia & 104 & 114 & 107 & 107 & 107 & 105 + trinidad & tobago & 108 & 115 & 117 & 117 & 117 & 115 + botswana & 102 & 116 & 110 & 112 & 110 & 110 + bahrain & 112 & 117 & 115 & 113 & 115 & 117 + mauritius & 110 & 118 & 124 & 122 & 123 & 126 + chinese taipei & 131 & 119 & 132 & 132 & 132 & 132 + kenya & 133 & 120 & 127 & 126 & 126 & 127 + aruba & 120 & 121 & 128 & 128 & 128 & 130 + wales & 93 & 122 & 109 & 111 & 111 & 108 + jersey & 113 & 123 & 118 & 118 & 118 & 111 + angola & 98 & 124 & 111 & 109 & 109 & 118 + mali & 145 & 125 & 129 & 129 & 129 & 128 + namibia & 129 & 126 & 133 & 133 & 133 & 134 + malawi & 141 & 127 & 131 & 131 & 131 & 131 + ethiopia & 132 & 128 & 119 & 119 & 119 & 119 + hongkong & 124 & 129 & 126 & 127 & 127 & 123 + guernsey & 128 & 130 & 141 & 142 & 142 & 141 + mauritania & 146 & 131 & 125 & 124 & 124 & 125 + surinam & 111 & 132 & 130 & 130 & 130 & 129 + macau & 122 & 133 & 135 & 135 & 135 & 136 + mozambique & 130 & 134 & 134 & 134 & 134 & 133 + madagascar & 140 & 135 & 121 & 121 & 121 & 122 + netherlands antilles & 118 & 136 & 138 & 139 & 139 & 138 + cameroon & 144 & 137 & 139 & 137 & 137 & 139 + sao tome and principe & 138 & 138 & 143 & 143 & 143 & 142 + haiti & 135 & 139 & 136 & 136 & 136 & 137 + ghana & 137 & 140 & 140 & 140 & 140 & 140 + bermuda & 126 & 141 & 144 & 144 & 144 & 144 + sierra leone & 148 & 142 & 145 & 145 & 145 & 145 + papua new guinea & 117 & 143 & 142 & 141 & 141 & 143 + san marino & 119 & 144 & 137 & 138 & 138 & 135 + burundi & 143 & 145 & 146 & 146 & 146 & 146 + rwanda & 142 & 146 & 148 & 148 & 148 & 148 + u.s .
virgin islands & 149 & 147 & 149 & 149 & 149 & 149 + seychelles & 136 & 148 & 147 & 147 & 147 & 147 + senegal & 147 & 149 & 122 & 125 & 125 & 121 + | pairwise comparison matrices are widely used in multicriteria decision making .
this article applies incomplete pairwise comparison matrices in the area of sport tournaments , namely proposing alternative rankings for the 2010 chess olympiad open tournament .
it is shown that results are robust regarding scaling technique . in order to compare different rankings ,
a distance function is introduced with the aim of taking into account the subjective nature of human perception .
analysis of the weight vectors implies that methods based on pairwise comparisons have common roots .
visualization of the results is provided by multidimensional scaling on the basis of the defined distance .
the proposed rankings give in some cases intuitively better outcome than currently used lexicographical orders . |
pattern formations in diffusion fields have been studied as a typical nonlinear and nonequilibrium problem.@xcite the dendrite is a typical pattern observed in crystal growth . the tip velocity and the shape near the tip region
have been intensively studied .
the growth law of the dendrite is expressed as @xmath1 , where @xmath2 is the tip velocity and @xmath3 is the radius of curvature of the parabolic tip.@xcite the parabolic structure of the dendritic tip is linearly stable and the formation of sidebranches is considered to be due to some noise effects.@xcite perturbations are expected to grow as @xmath4 , where @xmath5 is the distance from the dendritic tip,@xcite and they grow into sidebranches .
well - developed sidebranches can be easily observed in experiments of succinonitrile and nh@xmath6cl .
there are some quantitative investigations about the well - developed sidebranches .
huang and glicksman investigated the development of sidebranches in an experiment of succinonitrile.@xcite they found that the sidebranching spacing , which is the interval between neighboring active sidebranches , grows in a power law of exponent 1.3 as a function of distance from the dendritic tip .
they found the angle of the envelope with respect to the main branch increases as a function of the distance from the tip until 41@xmath0 .
they expected that the angle increases up to 45@xmath0 finally , where the sidebranches grow with the same velocity as the main branch .
li and beckermann analyzed the shape of the envelope of the sidebranches for succinonitrile on the microgravity flight and found that the envelope grows with a power law of exponent 0.859 as a function of the distance from the dendritic tip.@xcite the exponent is less than 1 , and it implies that the envelope angle decreases as a function of distance from the tip , which is different from the observation by huang and glicksman .
recently , honda and honjo are investigating the shape of the sidebranches of kinetic dendrites in viscous fingering on the plate with many grooves of square symmetry.@xcite theoretically , well - developed sidebranches have not been studied in detail .
we will study well - developed sidebranches using a coupled map lattice model .
we proposed a coupled map lattice model as a simple simulation method for crystal growth and succeeded in generating various patterns such as dla , dbm and dendrites.@xcite the dendrite with anisotropic surface tension obeys the growth law @xmath1 also in our coupled map lattice , when the undercooling is sufficiently small.@xcite the model is a deterministic model similar to the phase field model.@xcite our coupled map lattice models involves the kinetic effect and we have not succeeded in removing the kinetic effect as in the phase - field model by karma and rappel.@xcite however , our coupled map model is much simpler and numerically preferable for a large scale simulation even at small undercooling such as @xmath7 .
previously , different types of coupled map lattice models were proposed by kaneko and kessler et al.@xcite their coupled map lattice models may be interpreted as direct discretization of the phase - field models , but they have not reported dendritic growth . our model is a simpler model , in that only the linear diffusion process is assumed in lattice points other than interface sites .
packard and li - godenfeld proposed cellular automaton models for the crystal growth.@xcite in their cellular automaton models , the order parameter takes 0 ( liquid ) or 1 ( solid ) . in our coupled map lattice model
, the order parameter takes a continuous value between 0 and 1 at the interface sites .
the interface grows therefore more smoothly than the cellular automaton models .
our coupled map lattice model is considered to be a preferable model compared to the previous coupled map lattice models and cellular automaton models .
we will perform numerical simulation of dendrites using the coupled map lattice model to study the growth law of the envelope . and
( b ) 0.7 .
the dendritic patterns are plotted as the tip position locates at the origin.,width=566 ] and 0.9 , which are plotted with the coordinates scaled with the diffusion length @xmath8 .
the numbers in the figure denote the values of @xmath9 ( b ) double - logarithmic plots of the envelopes for @xmath10 and 0.9 .
two linear lines denote power laws with exponents 0.5 and 1.4.,width=491 ] as a function of the scaled distance from the tip @xmath5 for @xmath11 and 0.9.,width=302 ] as a function of the scaled distance from the tip @xmath5 for @xmath12 ( solid curve ) and 0.7 ( dashed curve).,width=264 ]
we consider melt growth of crystal with four - fold rotational symmetry .
our model consists of two steps of time evolution on a square lattice .
the four - fold rotational symmetry of crystal is involved in performing the simulation on the square lattice .
if the triangular lattice is used , a pattern like a snowflake with the six - fold rotational symmetry grows as shown in the previous paper.@xcite the first step is a diffusion process and the second step is a growth process at the interface .
the diffusion process of the latent heat released at the interface is expressed as @xmath13 where @xmath14 denotes a dimensionless temperature , @xmath15 is the step number , @xmath16 denotes the lattice point and @xmath17 is a diffusion constant .
this process is interpreted as the simplest discretization of the laplacian .
the second step is a growth process at the interface .
the order parameter is expressed with the variable @xmath18 , where @xmath18 is 0 in the liquid phase and 1 in the solid phase .
the order parameter @xmath18 changes only at the interface sites between the liquid sites and the solid sites .
the growth rule is written as @xmath19 where the melting temperature is assumed to be 1 .
there are two parameters @xmath20 and @xmath9 in our model , and the ratio @xmath21 means the latent heat .
this growth process is done only at the interface sites , and @xmath22 for the other sites keep the values @xmath23 at the previous step .
when @xmath18 goes over a threshold @xmath24 , the interface site changes into a solid site . the diffusion process ( 1 ) and the growth process ( 2 ) at the interface
are performed alternately .
the initial temperature is assumed to be @xmath25 .
then , the normalized undercooling @xmath26 is given by @xmath27 .
we have set a seed of crystal at the origin as an initial condition . to save computation time
, we have calculated only the upper half of the crystal assuming the mirror symmetry .
the system size is @xmath28 in most simulations .
the parameter @xmath20 is fixed to be 0.1 and the diffusion constant is @xmath29 . the surface tension effect is neglected in this paper .
( a stable parabolic structure is obtained if the term for the surface tension is included in this model as shown in ref.11 . )
our dendrite is therefore a kinetic dendrite , since the growth direction is determined by the kinetic anisotropy.@xcite our simulation may be closely related to the experiment by honda and honjo , since they investigate kinetic dendrites in two dimensions .
figure 1 displays snapshot patterns of dendrites for ( a ) @xmath30 and ( b ) @xmath10 . to include small noise effect ,
we have assumed that the threshold @xmath31 at the tip site ( only one site ) takes a random value between 0.99 and 1.01 .
sidebranches are well developed in the region far from the tip .
the sidebranches which locate in @xmath32 are not shown .
the sidebranches are more developed for @xmath30 than for @xmath10 , since the undercooling is large .
we have calculated the envelope line for each snapshot pattern and constructed an average curve of the envelope for @xmath33 and 0.9 . in our numerical simulation , an envelope of sidebranches
is defined as the line which connects the tip positions of active sidebranches .
a sidebranch is judged as active if it is larger than the neighboring active sidebranch locating in the left of itself ( with smaller @xmath34 ) . as a result of our definition ,
the envelope is a monotonously increasing function .
a relevant length scale of our model is the diffusion length @xmath35 .
the tip velocities are numerically obtained respectively as 0.0058,0.0081,0.0124 and 0.0227 for @xmath36 and 0.3 .
the velocities do not obey the law @xmath37 like the surface tension dendrite in two dimensions , but seem to obey @xmath38 for the kinetic dendrites in our model .
figure 2(a ) displays the averaged envelope , which is plotted with the scaled length unit @xmath39 .
( we have performed numerical simulation with larger size @xmath40 only for @xmath12 , since the development of sidebranches is slow at the parameter . ) the envelope for @xmath30 can be approximated as @xmath41 for large @xmath5 .
that means that the angle of the envelope with respect to the @xmath5-direction approaches 45@xmath0 .
the envelopes overlap with each other rather well , especially , the envelope for @xmath12 overlaps with that for @xmath10 very well
. there may exist a universal scaling function for the envelope function for sufficiently small undercooling .
figure 2(b ) displays logarithmic plots of the envelope functions for @xmath10 and 0.9 . for small @xmath5 ,
the envelope is approximated as @xmath42 , where the sidebranches are not well developed and the parabolic structure of the tip region is maintained .
for relatively larger @xmath5 , the envelope , roughly speaking , obeys a power law with exponent about 1.4 .
however , the envelope in the logarithmic scale has a tendency to curve upwards as @xmath5 is increased .
the derivative @xmath43 is calculated and displayed in fig . 3 for @xmath11 and 0.9 .
the derivative is about 0.5 for small @xmath5 , which corresponds to the parabolic shape , however , it increases from 1.2 to 1.9 in the intermediate region as @xmath5 is increased .
this implies that the envelope does not obey a simple power law .
we have also calculated the spacing between neighboring active sidebranches and obtained the average curve of the spacing between the active sidebranches as a function of the distance from the tip position .
figure 4 displays the average spacing @xmath44 for @xmath10 and 0.9 as a function of the scaled distance @xmath5 .
the two curves again overlap very well .
the sidebranch spacing is almost a linear function of @xmath5 . by the fitting to a power law @xmath45 , the exponent @xmath46 is evaluated to be about 1.15 .
the sidebranch spacing increases with the distance from the tip position because of the competition among sidebranches .
smaller sidebranches stop to grow and the survived sidebranches become larger .
competitive processes between survived large sidebranches go on further , until the spacing between the sidebranches becomes sufficiently larger than the diffusion length . .
( b ) contour curves of @xmath47 and @xmath48 for the temperature field @xmath49 at the same time as ( a).,width=680 ] of the largest branch for @xmath10 .
( b ) comparison of the time evolution of the tip position of the largest branch for the unidirectional growth from a linear seed ( solid curve ) with the envelope shown in fig . 2 for @xmath10 ( dashed curve ) .
they overlap very well.(c ) comparison of the time evolution of the spacing between the neighboring large branches for the unidirectional growth ( solid curve ) with the spacing between the active sidebranches shown in fig . 4 for @xmath10 ( dashed curve).,width=491 ] for @xmath10 .
the linear line denotes the power law with exponent 0.7 .
( b ) numerically evaluated exponent @xmath50 of the power - law growth as a function of the width @xmath51 of the rectangular box .
, width=491 ]
we can see the competitive growth of dendritic sidebranches more explicitly using another type simulation , which is the crystal growth starting from a linear seed .
we have performed a numerical simulation in a square box of @xmath52 .
we have imposed periodic boundary conditions in the @xmath34 direction such as @xmath53 at @xmath54 and 0 , and fixed boundary conditions such as @xmath55 at @xmath56 and no - flux boundary conditions @xmath57 at @xmath58 . as an initial condition for a linear seed
, we assumed that @xmath59 , @xmath60 , and @xmath61 took a random number between 0 and 0.1 .
the initial value of @xmath18 and @xmath62 for @xmath63 are all 0 .
figure 5(a ) is a snapshot pattern of @xmath18 at @xmath10 .
many small dendritic branches are created initially , but the number of growing branches decreases in time by the competitive interaction between the neighboring branches .
the velocities of delayed branches decrease and finally become almost zero .
large branches seen in fig .
5(a ) are survived branches for the competitive process .
figure 5(b ) displays contour lines of @xmath47 and @xmath48 .
the large neighboring branches interact competitively through the diffusion field @xmath49 .
the temperature @xmath49 is almost 1 in the regions between the delayed branches .
figure 6(a ) displays the time evolution of the tip postition @xmath64 of the largest branch as a function of the step number @xmath15 .
the ensemble average with respect to 20 different initial conditions is taken .
the form of this time evolution is similar to the envelope of sidebranches shown in fig . 2 .
we have compared the two curves by rescaling the coordinates .
the step number @xmath15 can be converted into a length by @xmath65 where @xmath2 is the tip velocity of the main stem and the scaled coordinates @xmath66 and @xmath67 are used in fig .
6(b ) . figure 6(b ) displays the relation of @xmath68 and @xmath69 ( solid curve ) and it is compared with the envelope of sidebranches at @xmath10 ( dashed curve ) .
the two plots overlap very well , although the numerical factor 0.67 is suitably chosen .
we have also calculated the time evolution of the average interval @xmath70 between active branches for the crystal growth starting from a linear seed . here , the active branches at step @xmath15 are chosen as the branches whose tip positions are larger than @xmath71 .
the average interval between active branches increases almost in proportion to @xmath15 , which corresponds to the linear growth of the sidebranch spacing shown in fig .
4 . figure 6(c ) compares the time evolution of the rescaled quantity @xmath72 as a function of @xmath69 ( solid curve ) with the average spacing between active sidebranches shown in fig . 4 for @xmath10 ( dashed curve ) .
( the numerical factor 0.8 is again suitably chosen . )
the good agreement of two plots in fig .
6(b ) and 6(c ) implies that the competitive dynamics among dendritic branches starting from a linear seed and that for sidebranches in the region far from the dendritic tip are essentially the same .
the tip velocity of the largest branch is small at small @xmath15 and it increases with time as is seen in fig .
this is because the spacings between the survived branches increase with time and the tip velocities become larger for the larger spacing . to understand the relation between the tip velocity and the spacing between the branches , we have performed numerical simulations in a rectangular box of size @xmath73 with periodic boundary conditions in the direction of @xmath5 . as an initial condition ,
we have put a small branch of size @xmath74 at @xmath75 and @xmath76 . since we have assumed periodic boundary conditions , the growth condition is equivalent to the case involving many branches , which have completely the same size and are periodically located with the same interval @xmath51 , in an infinitely wide system .
the system size @xmath77 is fixed to be 700 , and @xmath51 is changed from 10 to 120 , and the parameter @xmath9 is 0.7 .
figure 7(a ) displays the time evolution of the tip position of the branch in a logarithmic scale .
the time evolution seems to obey a power law @xmath78 with exponent 0.7 in fairly wide region of @xmath79 . in a narrow channel , the steady growth of dendrite
is not possible when the undercooling is small .
brener et al . investigated the parameter region for the existence of steadily growing dendrites.@xcite figure 7(b ) displays numerically evaluated exponents @xmath50 as a function of @xmath51 .
the flat interface grows with a power law of exponent 1/2 , which is shown theoretically and was numerically shown also in our coupled map lattice in the previous paper,@xcite which corresponds to the limit of @xmath80 .
inversely , the exponent becomes 1 , when the undercooling and the spacing @xmath51 are larger than a certain threshold , and the steady growth becomes possible .
figure 7(b ) shows that the exponent increases with the spacing @xmath51 for our kinetic dendrites . in the crystal growth from a linear seed ,
the interface is almost flat and the growth exponent is nearly 1/2 in an initial stage .
then , many small branches are created . at this stage , the branch spacing is small , the growth is slow , and the growth exponent is still close to 1/2 .
as the competitive process proceeds , the average spacing between active branches increases as a result of competition , and the tip velocities of the survived branches increase further . in this time scale
, the tip position seems to grow like a power law with exponent larger than 1 .
the exponents of the power law may be evaluated roughly as follows . if the tip velocity @xmath81 grows in a power law with exponents @xmath50 as shown in fig .
7(b ) , the exponents depend on the average spacing @xmath82 , and the average spacing @xmath82 increases roughly as @xmath83 , the tip position will grow as @xmath84 .
if @xmath50 grows as @xmath85 for small @xmath79 with a certain constant @xmath86 , @xmath87 and the exponent for the tip position is evaluated as @xmath88 for small @xmath79 , although this is a very rough approximation .
if the spacing becomes comparable or larger than the diffusion length , the tip velocities of survived branches approach the constant tip velocity for a single branch in an infinitely wide system .
we have performed a large scale numerical simulation of kinetic dendrites with a coupled map lattice model .
well - developed sidebranches have appeared in the coupled map lattice .
many sidebranches are naturally created and the competitive dynamics among sidebranches determines the shape of the envelope .
the envelope has a parabolic shape near the tip of the dendrite , and the angle of the envelope with respect to the main stem grows up to 45@xmath0 finally . in an intermediate region
, the envelope grows , roughly speaking , in a power law , however , the exponent seems to increase gradually as a function of the distance from the tip .
similar competitive dynamics among many dendritic branches is observed in the unidirectional growth .
we could compare quantitatively the temporal evolution of the unidirectional growth with the envelope of sidebranches as a function of the distance from the tip position .
the temporal evolution of a single branch in a narrow rectangular box seems to obey a power law in a wide range of @xmath79 .
the exponent increases with the width of the rectangular box .
the width can be interpreted to correspond to the spacing betweeen the sidebranches for the dendritic growth . as a result of the competition among sidebranches ,
the spacing between the active sidebranches increases , and then the tip velocities of the sidebranches are accelerated further . when the spacing becomes sufficiently large , the tip velocities of the active branches become nearly equal to the tip velocity of a single branch in an infinitely wide system .
we have calculated numerically the spacing @xmath44 and the envelope @xmath89 of sidebranches .
the competitive dynamics among the active sidebranches via the diffusion field determines @xmath44 and @xmath89 . expecting that the large scale competitive dynamics may not depend on the surface tension effect and
we have neglected it in this paper .
the surface tension is important to determine the tip radius and the tip velocity as in the experiments by glicksman et al .
it is not difficult to involve the surface tension effect in our model .
however , the surface tension effect makes the parabolic shape of the dendrites stable , and strong noises are necessary to construct well - developed sidebranches .
it is a future problem to study the sidebranch structure for the surface tension dendrites .
99 j. s. langer : rev .
* 52 * ( 1980 ) 1 .
t. vicsek : _ fractal growth phenomena _
( world - scientific , singapore , 1991 ) 2nd ed .
s. c.huang and m. e. glicksman : acta metal .
* 29 * ( 1981 ) 701 .
h. honjo and y. sawada : j. cryst .
growth * 58 * ( 1982 ) 297 .
a. dougherty , p. d. kaplan and j. p. gollub : phys .
* 58 * ( 1987 ) 1652 .
r. pieters and j. s. langer : phys .
* 56 * ( 1986 ) 1948 .
s. c.huang and m. e. glicksman : acta metal .
* 29 * ( 1981 ) 717 .
q. li and c. beckermann : phys .
e * 57 * ( 1998 ) 3176 .
t. honda and h. honjo : talk at the 58th annual meeting of the physical society of japan ( 2003 ) .
h. sakaguchi : j. phys .
* 67 * ( 1998 ) 96 .
h. sakaguchi and m. ohtaki : physica a * 272 * ( 1999 ) 300 .
a. karma and w. j. rappel : phys . rev .
e * 53 * ( 1996 ) r3017 .
k. kaneko : in _ formation , dynamics and statistics of patterns _ ( world scientific , singapole , 1990 ) ed . by k. kawasaki , m. suzuki and a. onuki .
d. a. kessler , h. levine and w. n. reynolds : phys .
a * 42 * ( 1990 ) 6125 .
n. h. packard : in _ theory and application of cellular automata _
( world scientific , 1986 ) ed . by s.
f. liu and n. goldenfeld : physica d * 47 * ( 1991 ) 124 .
e. ben - jacob , p. garik , t. mueller and d. grier : phys .
a * 38 * ( 1988 ) 1370 .
e. a. brener , m. b. geilikman and d. e. temkin : sov .
jetp * 67 * ( 1988 ) 1002 . | dendrites with developed sidebranches are numerically studied with a coupled map lattice model .
the competitive dynamics among sidebranches determines the shape of the envelope .
the envelope has a parabolic shape near the tip of the dendrite and the envelope angle with respect to the main branch increases up to 45@xmath0 finally . in an intermediate region ,
the envelope grows roughly in a power law , however , the exponent increases gradually as a function of the distance from the tip .
the competitive dynamics among many branches is also observed in a unidirectional growth from a linear seed , and it is compared with the competitive dynamics of sidebranches . |
the neutrino telescope @xcite is used to study astrophysical sources by detecting the high - energy neutrinos that these sources may emit .
the detector is deployed on the seabed at a depth of 2.5 km , and consists of a large three - dimensional array of 900 light - sensitive photo - multiplier tubes ( pmts @xcite , @xcite ) .
neutrinos are detected indirectly , after an interaction inside or in the vicinity of the detector .
the produced charged particles emit cherenkov light , which can be detected by the pmts . from the known positions of the pmts , and the measured arrival times of the cherenkov photons
, the signal produced by these charged particles can be distinguished from the background .
the main purpose of the data acquisition ( daq ) system is to convert the analogue signals from the pmts into a format suitable for the physics analysis . to achieve this , the daq system has the task to prepare the detector for data taking , convert the analogue signals from the pmts into digital data , transport the data to shore , filter the different physics signals from the background , store the filtered data on disk , and archive the run settings .
the daq hardware and software are described in sections [ sec : daq_hardware ] and [ sec : daq_software ] respectively , and the overall performance of the daq system is summarised in section [ sec : performance ] .
the hardware of the daq system is a network consisting of hundreds of processors , as shown schematically in figure [ fig : daq_hardware ] .
( 12,18 ) ( 0,0 ) a significant part of this network is located off shore .
the off - shore processors are integrated in custom made electronics .
the off - shore part of the daq system is connected to the on - shore part by a single electro - optical cable .
the on - shore processors are standard pcs and are located in the shore station .
most of the off - shore electronics modules are used to read out the photo - multiplier tubes ( pmts ) .
these modules are indicated in figure [ fig : daq_hardware ] by ` sector module ' and ` storey module ' .
a detector string is divided into 25 storeys , each consisting of a triplet of pmts and a titanium container . this container houses the electronics needed for the digitisation of the analogue signals from the pmts , and the data transport .
the hardware components that play a key role in the daq system are shown schematically in figure [ fig : electronics ] .
these components include several custom designed analogue ring sampler ( ars ) chips @xcite , a local clock , a field programmable gate array ( fpga ) , an sdram of 64 mb , and a processor with an ethernet port .
( 12,8 ) ( 0,0 ) during data taking the analogue signals from the pmts are processed by the ars chips .
this includes the digitisation of the timing of each pmtsignal and the total charge of the pulse . the settings of each ars chip can be adjusted , including the two main parameters : threshold and integration gate .
the voltage threshold is set to eliminate small pulses due to the dark current in the pmt .
its typical value corresponds to 0.3 photo - electrons .
the gate is set to integrate most of the pmt signal , while limiting the contribution of electronic noise .
the typical width of the integration gate is 35 ns .
a local clock is used by the ars chips to timestamp each pmt signal above threshold . the combined time and
charge information of the pmt signal within the integration time is referred to as a single photo - electron hit .
after the integration time , the ars chip has a dead time of about 200 ns due to the limited transfer speed to the analogue pipeline . to compensate for this dead time
, two ars chips are connected to each pmt .
these ars chips operate in a token ring scheme . after the integration time of the first ars chip ,
the second takes over . only after the integration time of the second ars chip , the dead time of the first plays a role . for special cases , like large or double pulses
, the pulse shape can be read out by the ars chip with a sampling frequency that is tunable between 150 mhz and 1 ghz .
this feature is particularly useful for calibration and tuning of detector parameters , but can also be enabled for physics triggers .
the pulse shape information is not used in the data filter algorithms ( see section [ subsec : data_taking ] ) , but it can be stored on disk with the associated hit information .
the readout system of the ars chips is implemented in a high density fpga xilinx virtex - e xcv1000e .
the data from the ars chips are buffered by the fpga in the 64 mb sdram into separate frames covering a certain period .
this period can be set to values between 10 and 100 ms , and corresponds to a predefined number of clock cycles .
the clock system is described in more detail in section [ subsec : clock ] .
the processor in the off - shore electronics modules is the interface between the ars chips and the online data processing system .
it is the motorola mpc860p , which has a fast ethernet port ( 100 mb / s ) .
the operating system on these processors is vxworks .
tcp / ip is used for the data transport .
the module on every fifth storey on a detector string , in figure [ fig : daq_hardware ] specified by ` sector module ' , has some additional functionality .
these modules contain an ethernet switch , and the transceiver that plays a role in the data transport to and from shore .
the ethernet port on each storey is connected to a sector module using an optical bidirectional 100 mb / s link .
the links from five storeys are merged by the ethernet switch in the sector module into a single gb / s ethernet link .
the group of storeys with a common gb / s ethernet link is referred to as a sector . for a total of 25 storeys on a detector string
, each detector string has five such sectors .
each detector string has at the bottom an extra container which houses an electronics module , in figure [ fig : daq_hardware ] specified by ` string module ' .
the string module is used for the slow control of the electrical power system and the calibration systems of the detector string , and also for the distribution of the clock signal .
this module has a separate 100 mb / s link to shore . for the data transport between the off - shore and on - shore parts of the detector the dense wavelength division multiplexing ( dwdm ) technique
is used @xcite .
this is an optical technology that uses multiple wavelengths to transmit different streams of data along a single fibre .
the dwdm system can be considered as a large fibre - optic network .
the dwdm transceivers are used to extend the ethernet link to shore .
the five dwdm channels from ( to ) the sectors and the dwdm channel from ( to ) the string module are optically ( de)multiplexed in the string module . each sector and each string module has a unique pair of wavelengths that is used to transmit the data .
each string is connected with an electro - optical cable to the so - called junction box , the container where the cables from the 12 detector strings meet .
the junction box is connected to the shore station with a single 40 km long electro - optical cable . on shore
, there is a multiplexer and a demultiplexer for each detector string .
the dwdm system is also used for the transmission of the slow control data , and the distribution of initialisation and configuration data .
the data that are sent from the shore station to the detector are multiplexed on shore , and demultiplexed in the string modules .
the on - shore part of the daq system is located in the shore station .
it consists of a farm of standard pcs , a large ethernet switch , the dwdm hardware , and the master clock system ( described in section [ subsec : clock ] ) .
all pcs in the farm have software programs running that are needed for the detector operation , data transfer and communication , data processing , monitoring , and data storage .
these programs are all part of the daq software , discussed in section [ sec : daq_software ] .
the pc farm consists of some tens of pcs , all connected to the ethernet switch . with 12 detector strings , each consisting of 25 storeys and an electronics module at the bottom
, there are 312 off - shore processors , each with an ethernet port for the connection to shore .
all on - shore and off - shore processors that are part of the daq system are connected to the same on - shore ethernet switch .
together they form a large ethernet network .
each of these processors can be considered as a node in the daqnetwork , addressable by its unique ip address .
all off - shore processors can be reached by a telnet connection .
this allows system tests to be performed and new software and firmware to be downloaded .
the ethernet network makes the distribution of the data from the arschips to the on - shore processes completely transparent , and it enables the communication with all processors in the system .
like in every string module , a ( de)multiplexer and dwdm transceiver is available on shore for each detector string .
the multiplexers are used for multiplexing the data streams that are meant for the initialisation and configuration of the detector , and the demultiplexers are used for demultiplexing the data streams from the ars chips , and slow control data for monitoring purposes .
the main purpose of the clock system is to provide a common clock signal to all ars chips .
it consists of a clock signal generator on shore , a clock signal distribution system , and a clock signal transceiver in each detector storey .
the on - shore master clock generates a 20 mhz clock signal that is synchronised internally to the gps time with an absolute accuracy of 100 ns .
the clock signal is distributed to all off - shore clock transceivers using the available optical fibre network .
therefore each detector storey has one local clock that is synchronised to the master clock .
the phase offset of each local clock can be determined by measuring the return time of a calibration signal .
the local clock is used to synchronise the ars chips .
the relative time accuracy is about 50 ps .
the clock system operates in parallel and independent of the daq system . as a result
, the internal clock calibration does not induce any dead time .
the main software processes in the daq system are indicated schematically in figure [ fig : daq_software ] .
( 12,10 ) ( 0,0 ) part of the processes run on the off - shore processors , part of them run on the on - shore processors .
there are basically three types of processes : processes that are used for the data transfer and communication ( see section [ subsec : data_transfer ] ) , processes that take care of the operation of the detector ( see section [ subsec : detector_operation ] ) , and processes that are involved in the data taking and data handling ( see ) . in total , with hundreds of off - shore processors and the on - shore pc farm , there will be several hundreds of processes in the daq system . a daq model , that is implemented in each process , has been defined to synchronise all processes .
the hundreds of processes in the distributed multiprocessor system operate independently . in order to synchronise all processes in the system
, a finite state machine has been designed with a fixed set of states and transitions , represented by the state diagram shown in figure [ fig : state_m ] .
each of the processes that participates in the daq system , except for the data transfer package ( see section [ subsec : data_transfer ] ) , has this state machine implemented , for which the chsm language @xcite is used .
the state transitions are accompanied by actions that are specific to each type of process in the system .
these actions should be performed before the data taking can start and after the data taking has stopped .
the state machine ensures that these actions are executed for all processes in the required order ( as specified by the state transitions ) .
( 12,3.8 ) ( 0,0 ) the steps involved in the preparation of the daq system for data taking include the configuration of the hardware components and software processes .
the required steps for stopping the data taking ( and switching off the detector ) include the handling of the data that are still being processed , closing the data files , and storing the run information in the database .
data are only taken in state running .
each start transition corresponds to a new data taking run , identified by a unique run number .
the run number is added to the raw data by the off - shore processes , and is also used by the data storage program to archive the data .
processes in the daq system can only be in one of the predefined states .
state transitions are initiated from a central application ( see section [ subsec : detector_operation ] ) , and apply to all processes in the system . during a state transition
, all processes perform the necessary actions independently .
consequently , the actual state of the system is associated with all actions that took place during the state transitions .
normally all processes are in the same state at any time .
all data transfer and communication between processes in the daq system are done with the package @xcite .
implements the tagged data concept , which means that any program in the system can send data accompanied by a tag to a server .
the server will distribute the data to all processes that subscribed to the specific tag using tcp sockets .
most of the pcs in the shore station have a server running .
the package has been optimised for gb / s ethernet , and its overhead is typically 5% of the total bandwidth .
is used for the data transfer between processes at three different stages in the daq : data processing , data storage , and detector operation .
the data from the off - shore processes ( ) are transferred to the on - shore processes that take care of the processing of the data ( , see section [ subsec : data_taking ] ) . on each pc in the shore station that has such a data processing program running , a server is running for passing the data .
all data that passed the data processing , and that have to be stored on disk , are transferred from the data processing pcs via one server to the data storage program ( , see section [ subsec : data_storage ] ) .
the program , that is used to operate the detector ( see section [ subsec : detector_operation ] ) , communicates with all processes in the daq system .
all state transitions are initiated by the .
the commands for the state transitions , as well as the messages for the handshaking mechanism are distributed via a central server
. takes care of the distribution of a single message to multiple processes on multiple hosts . without knowing
how many processes should receive a certain message , each will receive the message if it subscribed to the corresponding tag .
the implementation of also enables the exchange of information between processes written in different programming languages ( @xmath0 and java are used in the daq system ) .
the program is written in java because of the straightforward possibility for remote operation of the detector , the user friendly interface with the oracle@xmath1 database that is used in the daq system , and the advanced toolkit for developing graphical user interfaces .
the on - shore programs used for data processing and data handling are written in @xmath0 because of the required execution speed .
the required compatibility of the off - shore software with the related firmware also implies the use of the @xmath0 programming language . the main program in the daq system that is used to operate the detector is the .
it has a graphical user interface from which the operator submits the requests for the execution of state transitions by all processes in the system ( shown in figure [ fig : state_m ] ) .
all processes that are subjected to these state transitions can be considered as client processes . with these state transitions
the detector can be prepared for data taking , and the necessary actions are performed when the data taking has ended .
all data needed by the daq system to operate the detector are stored in the central database system , for which the relational database system oracle@xmath1 is used .
these data are retrieved from the database by the only .
these data include the detector information ( the location of each component in the detector ) , the client process list , the different configurations for each client setup , the initialisation and configuration data , and the information about the data taking .
the client process list determines which processes participate in the system , and thus which parts of the detector are active during data taking .
the retrieves the necessary information from the database by using the input from the operator and the relations between the database tables .
a separate graphical user interface is used to control the power system in each string module . when the off - shore electronics modules are powered on , the off - shoreprocessors boot .
the launches each client process that is part of the client setup chosen by the operator by executing its start command on the host with its associated ip address ( both retrieved from the database ) . at startup
the client processes enter the general state machine .
once the client processes are started , the general mechanism that is used for the communication between the and its clients is done with a handshaking method based on the unique i d of each process .
this i d consists of the nickname of the process ( which corresponds to the client type ) , and the ip address of its host .
the implementation of the handshaking method is that , after the request for a state change from the via the central server , each client sends a message containing this unique i d , and the state it just entered , to the same server that indicates that the state change has succeeded for this client .
as the uses the same unique coding for each client , it is aware of the current state of each client . as a consequence ,
the also has a monitoring task , visualising the state of each client .
the package is designed in such a way that it broadcasts a message when a process connects to or disconnects from a server .
these messages are also intercepted by the .
as each client connects to the server when it is started , and disconnects when it terminates , the also visualises the running state of each client .
in addition , the provides the means to modify the client setup during operation in case of a client error to prevent interruption of the data taking .
depending on the type of data that will be taken , a predefined configuration for the chosen client setup is selected by the operator .
the specific initialisation and configuration data required by each process are retrieved from the database by the , and sent , together with the corresponding state transition request , via the server to the particular client process . along with the state transitions for starting and stopping a run , the master clock on shore sends a hardware signal to the off - shore clocks for the synchronisation of the ars chips . the scharness process that runs on each off - shore processor controls the high voltage applied to the pmts during the initialisation and configuration phase .
it is also used for the readout of various instruments .
the data from these instruments are sent to shore , and stored in the database by the program .
these data are then used to monitor the detector and its environment .
the scdatapolling program schedules the read requests of all these instruments in the detector .
the updates the database with the run information and the reference to the detector settings to be able to couple the detector settings to the data when they are eventually analysed .
a new run is automatically started after a few hours of data taking , or when the output data file , created and updated by the data storage program ( see section [ subsec : data_storage ] ) , reaches its maximum size ( typically 1 gb ) .
the then triggers the necessary state transitions automatically to stop the run , and start a new run . during data taking ,
the raw data are monitored online using dedicated software .
for this , the data are taken from and histogrammed using the root package @xcite .
a custom made presentation tool based on root displays and compares histograms , showing the basic measurements such as the time and amplitude information from the pmts . during data taking , all signals that are recorded by the pmts and digitised off shore
are transported to shore without any off - shoredata selection .
this implementation is known as the ` ' concept . as a result ,
all raw data are available on shore where the required data processing method can be applied to the data .
this minimises the overall loss of physics signal , and allows different processing methods to be applied for specific physics analyses .
the on - shore handling of all raw data is the main challenge in the daq system , because of the high background rates , mainly caused by bioluminescence and @xmath2k decay . as a result of this background
, the average photon detection rate per pmt varies between 60 khz and 90 khz during periods with low bioluminescence activity @xcite .
the data output rate of the detector is then 0.3 gb / s to 0.5 gb / s .
the on - shore handling of the data is done with the processes that run on each pc in the on - shore data processing farm .
the processing of the data involves the separation of signal from background , and as a result the necessary reduction of the data flow for storage . during data taking ,
the programs process all data online as the detector is in principle operating continuously .
the programs apply various algorithms , each aimed at the search for specific physics signals .
this includes a muon filter that looks in all directions for the signals in the detector compatible with a muon .
depending on the random background rate , the filter settings can be adjusted so that the trigger rate is dominated by muons passing the sensitive volume of the detector . for an average data size of a physics signal of 2.5 kb , and an expected atmospheric muon rate of about 10 hz
, the data reduction is then about @xmath3 .
the bandwidth required for the data transport scales linearly with the background photon detection rate . the data processing speed scales with some power of this rate , this power being the required minimum number of photons to identify a physics signal ( see section [ sec : performance ] ) .
the processes receive the raw data from the off - shore processes , running on each off - shore processor in the detector .
all data that are produced by each ars chip in a certain time window are buffered in the sdram by the fpga in a so - called frame .
the length of this time window can be set to values between 10 and 100 ms .
a separate thread that is spawned by the process takes the buffered data from the sdram , and sends each frame as a single packet to shore using tcp / ip ( see figure [ fig : timeslices ] ) .
the frames from all processes that belong to the same time window are sent to the same pc in the on - shore data processing system , identified by its ip address .
the large ethernet switch thus operates as a so - called level 2 switch .
the list of ip addresses as possible destinations for the raw data is provided to the processes by the .
the collection of frames belonging to the same time window is called a time slice . as a result
, a time slice contains all data that were registered during the same time interval by all ars chips in the detector .
the programs alternately receive the frames belonging to the same time window from the off - shore processes via the server that runs on the same pc .
each program has to be finished with processing a time slicebefore it receives the next .
this imposes an optimisation of the configuration of the programs in terms of processing speed , and it determines the number of pcs required for online data processing and the specifications of these pcs .
when a sufficient number of correlated single photo - electron ( spe ) hits is found that is consistent with a specific physics signal , the data are considered as a physics event , that is written to disk . the period covered by a time slice
is chosen such that its duration is long compared to the duration of a physics event in the detector ( approximately 1 @xmath4s ) , in order to minimise the chance of having an event crossing the boundaries of a time slice .
the algorithms implemented in the data processing software are designed to find a physics signal by looking for space - time correlations in the data . from the time of the spe hits , and the positions of the pmts ,
these algorithms calculate in real time if hits could originate from light produced by this signal . in the case of standard data processing , when the muon filter is used , this is the signal produced by a muon traversing the detector .
apart from the standard muon filter there are various other algorithms implemented that are used for the search for specific physics signals .
the configuration of the programs , that indicates which algorithm is used , is done by the , as described in section [ subsec : detector_operation ] . among the presently available filter algorithms is a magnetic monopole filter , an optical beacon filter , and a directional filter .
the monopole filter looks for space - time correlations in the data that are caused by particles that traverse the detector at lower speeds than a muon .
the optical beacon filter is used when the optical beacons , that are attached to the detector strings , are fired .
these optical beacons produce a typical light signal , that is used for time calibration .
the directional filter is used when searching for a muon coming from a specific direction .
this filter method takes into account the known direction of the neutrinos , and is used for astrophysical sources with known positions on the sky . for gamma - ray bursts , a subgroup of these sources ,
this filter method can be used offline .
the combination of the concept , the use of an on - shore pc farm , the specific features of gamma - ray bursts , and the information provided by external satellite triggers , make the detector particularly sensitive to neutrinos from gamma - ray bursts @xcite . any other possible filter algorithm for a specific physics signal can easily be implemented in the data processing system . as the whole data processing system is implemented in software ,
a new filter algorithm only involves adding the specific code and the corresponding configuration option in the database . if necessary , the computing power can be increased by adding pcs to the on - shore system .
the physics events selected by the programs are passed via one server to the , which formats the events and writes them to disk in root format @xcite .
these disks are located in the shore station . for each run , the data are stored in a separate file .
the event files are used for the physics analyses , and are copied regularly from the shore station to a computer centre elsewhere .
the performance of the daq system has been evaluated on the basis of the operation of a prototype detector @xcite , and has been quantified using a detailed simulation of the response of the complete detector to muons @xcite . during periods of low bioluminescence activity
( when the average photon detection rate is 6090 khz per pmt ) all data are transmitted from the off - shore electronics modules to the shore station .
the total data output rate of the complete detector is then 0.30.5 gb / s .
however , at higher background rates congestion of the data flow can occur at the processors in the off - shore storey modules .
although these off - shore processors have a 100 mb / s ethernet port , they can not make use of the full available bandwidth .
the maximum data throughput of a single storey is at present 20 mb / s .
this corresponds to a photon detection rate of about 150 khz per pmt . with the zero - copy feature of the vxworks operating system running on the off - shore processors
, this maximum data flow could be increased to about 40 mb / s .
this corresponds to a maximum manageable photon detection rate of about 300 khz per pmt .
the sector modules have a gb / s ethernet link to shore which can easily accommodate the maximum rate . on shore
, there is a single large ethernet switch to which all data processing pcs , each with a gb / s ethernet link , are connected . to avoid congestion of the data flow at the data processing pcs
, a barrel shifting mechanism of the data has been implemented in the off - shore processes . in this mechanism
, the data sending threads do not send the data for a given time slice simultaneously to the same on - shore pc , but desynchronise the data sending according to a predefined scheme .
the filter programs that run on the data processing pcs have to filter the data in real time . the number of required filter programs , which corresponds to the number of data processing pcs involved in the daq system , depends mainly on the background rate in the sea .
the standard muon filter is configured such that it triggers when it finds 5 or more space - time correlated level 1 hits ( two hits within 20 ns in one detector storey , or one hit with an amplitude larger than 2.5 photo - electrons ) .
therefore a minimum of 10 detected photons is required to trigger an event .
the trigger efficiency of this filter has been evaluated using a simulation of the detector response to muons @xcite , and the result is shown in figure [ fig : efficiency ] as a function of the number of detected photons for a given event .
( 10,9 ) ( 0,0 ) the efficiency rises rapidly above 10 detected photons , and reaches 1 at 40 detected photons . during periods of low bioluminescence activity ( see above ) , and using a trigger based on a majority logic , the trigger rate due to random background is a few hundred hz .
simulations have shown that with a more sophisticated trigger , this rate can be reduced to less than 1 hz , without loss of efficiency . with an expected atmospheric muon trigger rate of about 10 hz
, the standard muon filter then has a purity of 90% . with the filter conditions mentioned above
, 10 pcs with a processing speed of 3 ghz are needed .
different filter configurations are used for the various physics filters described in section [ subsec : data_taking ] . for physics filters that require more computing power , or that use less strict filter conditions ( as used for example in the directional filter ) ,
more data processing pcs are used to be able to filter the data online . in principle the data taking and online data processing take place continuously .
there is no dead time in the data transport , nor in the online filtering of the data . only a small dead time
is encountered in the ars chips , such that a third photon is lost when it is detected by one pmt within 235 ns ( i.e. integration gate and dead time of the ars chip ) .
the effective dead time is then less than 165 ns , taking into account the integration gate of both ars chips .
however , at high background photon detection rates , the maximum bandwidth of the off - shore processors is reached ( see above ) .
part of the data is then lost , and this will result in a loss of the overall detection efficiency . with the prototype detectors and the first complete detector string that have been operational , significantly higher background rates than the maximum manageable background rate have been measured , especially during spring .
the daq system has been operational since 2003 , and has been used for various systems that have been deployed in the deep - sea environment .
the concept offers a flexible data filtering system .
a simulation of the detector response to muons has shown that a high detection efficiency can be achieved .
the results obtained with one of the prototype detector strings have been published @xcite . for this detector ,
the data taking was more than 90% efficient during periods of low bioluminescence activity .
the daq system has also been used to read out the first complete detector string . by enlarging the on - shore computer farm ,
the same system will be used to read out the complete detector , that will consist of twelve detector strings .
the authors would like to thank r. gurin and a. maslennikov for providing the package .
the authors also acknowledge the financial support of the funding agencies : centre national de la recherche scientifique ( cnrs ) , commissariat lenergie atomique ( cea ) , commission europenne ( feder fund ) , rgion alsace ( contrat cper ) , rgion provence - alpes - cte dazur , dpartement du var and ville de la seyne - sur - mer , in france ; bundesministerium fr bildung und forschung ( bbf ) , in germany ; istituto nazionale di fisica nucleare ( infn ) , in italy ; stichting voor fundamenteel onderzoek der materie ( fom ) , nederlandse organisatie voor wetenschappelijk onderzoek ( nwo ) , in the netherlands ; russian foundation for basic research ( rfbr ) , in russia ; ministerio de educacin y ciencia ( mec ) , in spain . 99 e. aslanides et al . , the proposal , astro - ph/9907432 p. amram et al .
instrum . meth .
a484 ( 2002 ) 369 j.a .
aguilar et al . , nucl .
a555 ( 2005 ) 132 f. feinstein et al .
a504 ( 2003 ) 258 j.m .
senior , optical fiber communications : principles and practice , prentice hall , 1992 p.j .
lucas , an object - oriented language system for implementing concurrent , hierarchical , finite state machines , m.s .
thesis , technical report : uiucdcs - r-94 - 1868 , university of illinois at urbana - champaign , department of computer science , 1993 r. gurin and a. maslennikov , : distributed data handling package , 1995 r. brun and f. rademakers , nucl .
instrum . meth .
a389 ( 1997 ) 81 j.a .
aguilar et al . , first results of the instrumentation line for the deep - sea neutrino telescope , astro - ph/0606229 , accepted for publication in astroparticle physics m.c .
bouwhuis , detection of neutrinos from gamma - ray bursts , phd thesis , university of amsterdam , 2005 d.j.l .
bailey , monte carlo tools and analysis methods for understanding the experiment and predicting its sensitivity to dark matter , phd thesis , university of oxford , 2002 | the neutrino telescope is being constructed in the mediterranean sea .
it consists of a large three - dimensional array of photo - multiplier tubes .
the data acquisition system of the detector takes care of the digitisation of the photo - multiplier tube signals , data transport , data filtering , and data storage .
the detector is operated using a control program interfaced with all elements .
the design and the implementation of the data acquisition system are described . , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , neutrino telescope , data acquisition system |
recently pedestrian dynamics has been gaining increasing interest .
one focus of this research area is the security of people , e.g. , trying to optimize evacuation processes by minimizing evacuation time . to achieve this `` benchmark '' architects and civil engineers need to have an idea about the minimal necessary width of exit doors , their optimal placement , length of escape routes , etc .
usually legal regulations and other descriptive specifications in handbooks provide requirements , which are in general not flexible enough for complex buildings .
however , simulations of pedestrian dynamics offer the possibility to analyze and understand the dynamics of pedestrian streams in the building to be designed .
this is useful to ease decisions e.g. , dimensioning of emergency doors , and make them more realistic and adapted to different architectures . in this work ,
we address the possibility of describing quantitatively the pedestrian dynamics , by proposing a space - continuous model . in section 2 the theory of mathematical modeling of pedestrian dynamics
is briefly discussed . in section 3 our model
is introduced . then in section 4 numerical results of pedestrian flow through bottleneck and density measurements inside and in front of the entrance to a bottleneck will be presented .
section 5 gives concluding remarks .
there is a wide range of mathematical models to simulate pedestrian dynamics .
generally , these models are subdivided into macroscopic and microscopic models @xcite . in macroscopic models
the system is described by variables like crowd - density and flows of continua , whereas microscopic models consider the movement of individual persons separately .
microscopic models can be subdivided into models which are discrete or continuous in space .
widely - used discrete models for pedestrian dynamics are cellular automata models ( ca ) , which describe phenomena in space - time by assigning discrete states to a grid of space - cells .
these cells can be occupied " by a pedestrian or empty " .
thus the movement of pedestrians in space is done by passing them from cell to cell in discrete time by simple rules that reflect e.g. psychological aspects of the motion .
space - continuous force - based models determine the continuous movement of well distinguished individuals from their desires to reach certain targets and from influences of the space geometry as well as actual states and positions of the other individuals .
inspired by the field theory of lewin @xcite , force - based models assume that not only physical forces do affect the movement of pedestrians but similarly do _ social _ forces . a mathematical description of such _ social _ forces that determine the movement of pedestrians was first given by helbing and molnr in 1995 @xcite .
though force - based models pose quite some computational problems that do not occur in ca , their investigation appears to be worthwhile , since they permit higher resolution of geometry and time .
force - based models take newton s second law of dynamics as guiding principle .
the movement of each pedestrian is thus described by @xmath0 here @xmath1 denotes the repulsive force exerted by pedestrian @xmath2 on pedestrian @xmath3 .
@xmath4 is the repulsive force emerging from borders ( e.g. walls ) and @xmath5 is a driving force . @xmath6 is the mass of pedestrian @xmath3 and @xmath7 is the number of pedestrians acting on pedestrian @xmath3 .
pedestrians try to avoid collisions and contact with other pedestrians and objects by changing their direction .
the repulsive forces are introduced to model this avoidance of physical contact .
they can be social forces @xcite as well as physical forces @xcite , where physical forces come into play mainly to prevent overlapping with other pedestrians and with obstacles .
several different forms for these repulsive forces have been proposed .
the driving force @xmath5 represents the intention of a pedestrian to move towards a given destination .
the set of equations ( [ eq : maineq ] ) for all pedestrians results in a high - dimensional system of second order ordinary differential equations .
the time evolution of the positions and velocities of all pedestrians is then obtained by numerical integration .
most force - based models describe the movement of pedestrians qualitatively well : self - organisation phenomena e.g. , lane formations @xcite , oscillations at bottlenecks @xcite , `` faster - is - slower '' effect @xcite , and clogging at exit doors @xcite are reproduced .
these achievements indicate that the models could reliably describe pedestrian dynamics . however , a qualitative description is not sufficient if one has to make reliable statements about critical processes , e.g. , emergency egress .
+ moreover , rather often software implementations of the model do not rely on one sole approach . especially in high density situations
the simple numerical treatment has to be supplemented by additional techniques to obtain sensible results .
examples are restrictions on state variables and sometimes even totally different procedures which replace the above equations of motion ( [ eq : maineq ] ) to avoid overlappings among pedestrians @xcite and negative and high velocities @xcite .
necessarily a model has to describe pedestrian dynamics in a quantitative manner , e.g. , by reproducing the fundamental diagram , the flow through bottlenecks and the density inside and in front of the entrance of a bottleneck .
additionally , for further investigation , it is advantageous that it uses a single set of forces . in the next section ,
we propose such a model , which is based on a single set of formulae .
furthermore the model incorporates free parameters which allow adaption of the model to quantitative data .
pedestrians are modelled as circles moving in two - dimensional space , see figure [ fig : fuss ] .
we model the repulsive force exerted by pedestrian @xmath2 on pedestrian @xmath3 as depending on the inverse of their distance .
additionally , relative velocity between two pedestrians is taken into account and assumed to influence the strength of the repulsive force .
moreover , we assume proportionality between the repulsive forces acting on a pedestrian @xmath3 and its desired velocity . thus a pedestrian tending to move with high velocity `` feels '' intense repulsive actions .
the repulsive force , as a dimensionless relationship between the above mentioned quantities , is defined by @xmath8 where @xmath9 and @xmath10 , \label{eq : relv}\ ] ] and @xmath11 .
\label{eq : k}\ ] ] here @xmath12 denotes the desired velocity of pedestrian @xmath3 and @xmath6 its mass .
@xmath13 is the projection of the relative velocity of pedestrian @xmath2 and pedestrian @xmath3 onto the direction of @xmath14 , @xmath15 is the distance between @xmath3 and @xmath2 , @xmath16 is a coefficient that reduces the action - field of the repulsive force to the angle of vision of each pedestrian ( @xmath17 ) , and @xmath14 is the normalised direction - vector between @xmath3 and @xmath2 .
@xmath18 denotes the euclidian norm in @xmath19 and @xmath20 the absolute value in @xmath21 . by means of the parameter
@xmath22 the strength of the force can be adjusted while the diameter of pedestrians @xmath23 depends linearly on the velocity .
this incorporates the dynamic space requirement of pedestrians , modeling the fact that faster pedestrians require more space than slower pedestrians , due to increasing step sizes @xcite .
@xmath24 with free parameters @xmath25 and @xmath26 .
we use the driving force as defined in @xcite : @xmath27 with @xmath28 the velocity of pedestrian @xmath3 , @xmath29 its desired direction and @xmath30 a time constant .
the above model has been developed from the centrifugal force model ( cfm ) of yu et al .
their expression for the repulsive force reads as follows : @xmath31 with the quantities @xmath16 , @xmath13 , @xmath32 and @xmath14 as defined in equations ( [ eq : k ] ) , ( [ eq : relv ] ) , and ( [ eq : re ] ) . when pedestrians come near to one another the repulsive force is expected to grow . in the cfm
, this is in general not the case due to the relative velocity term in the nominator of the repulsive force ( [ eq : cfmfrep ] ) .
if both the distances between two pedestrians and their relative velocities are small , their quotient does not grow as expected .
consequently the repulsive force does not become high enough to prohibit overlappings .
introducing the intended speed in the nominator of the repulsive force ( [ eq : frep ] ) we eliminate this side - effect .
furthermore , there is a compensation between repulsive and driving forces at low velocities , which damps oscillations . due to these changes we can do in most cases without the extra collision detection technique ( cdt ) which takes over control in @xcite in case of formation of dense crowds .
the solution of the initial value problem ( [ eq : maineq ] ) was done in all simulations using a fourth - order runge - kutta scheme with a fixed - step size of @xmath33s .
the repulsive interaction between two pedestrians is neglected when the distance between them is larger than two meters . in order to investigate the influence of our modifications ,
different model approaches were simulated .
then the results were compared with empirical data of pedestrian flow through a bottleneck .
the flow of @xmath34 pedestrians through the bottleneck situation as described in @xcite was simulated .
the width of the bottleneck was changed from @xmath35 m to @xmath36 m in steps of @xmath37 m .
the free parameter @xmath22 in equation ( [ eq : frep ] ) is set to @xmath38 for the repulsive force among pedestrians and to @xmath39 for the repulsive force emerging from obstacles . for the free parameters in equation ( [ eq : diam ] ) we set @xmath40 several parameter values were tested .
with this parameter set the results of the simulations are in good agreement with the empirical flow measurements presented in @xcite , see figure [ fig : flowa ] .
for comparison the flow through the same bottleneck was measured by the original cfm without cdt .
the results presented in figure [ fig : flowb ] show very high values of the flow .
these are possible , because cdt does no more care for volume exclusion .
actually , control of bottleneck - flow seems to be dominated by the cdt in the cfm - approach . to back this hypothesis we measured pedestrian flow through the bottleneck with a cfm like model without repulsive forces , managing collisions with ctd , only .
the gained values are in the range of experimental data , see figure [ fig : flowc ] .
volume exclusion of pedestrians seems thus to be a favorable feature of successful models .
a second validation of our modifications of cfm comes from measurements of density inside the bottleneck as well as in front of the entrance to the bottleneck , see figure [ fig : scrshot ] .
the density in front of the entrance to the bottleneck is presented in figure [ fig : rhovor ] .
the results are in good agreement with the experimental data in @xcite
. additionally , the measured density values inside the bottleneck are in accordance with the published empirical results in @xcite , see figure [ fig : rhoin ] . though the model is already a useful tool to describe quantitatively the bottleneck - situations as defined in @xcite , it is not yet fully satisfactory .
depending on the specific situation oscillations and overlappings can still occur .
determination of uniformly usable parameters is a difficult endeavour ; cf .
the parameter estimation in simwalk @xcite .
the proposed space - continuous force - based model quantitatively describes the movement of pedestrians in 2d - space in an almost satisfactory way . besides being a remedy for numerical instabilities in cfm
the model simplifies the approach of yu et al . since we can dispense with their extra cdt without deteriorating performance .
the implementation of the model is straightforward and does not use any restrictions on the velocity as was the case e.g. , in @xcite .
simulation results show good agreement with empirical data .
the model contains natural free parameters which can further be tuned to adapt the model to certain scenarios . introducing density - dependence of the strength of the repulsive force
promises further ameliorations of the model . | this paper introduces a space - continuous force - based model for simulating pedestrian dynamics .
the main interest of this work is the quantitative description of pedestrian movement through a bottleneck .
measurements of flow and density will be presented and compared with empirical data .
the results of the proposed model show a good agreement with empirical data .
furthermore , we emphasize the importance of volume exclusion in force - based models . |
two - body breakup processes involving nuclei at high momentum and energy transfer play an important role in studies of nuclear qcd .
the uniqueness of these processes is in the effectiveness by which large values of invariant energy are produced at rather moderate values of beam energy .
the kinematics of two - body photodisintegration provide the relation @xmath8 in which the produced invariant energy grows with the energy of the probe twice as fast as compared , for example , to hard processes involving two protons , in which case @xmath9 . as it follows from eq.([s ] ) , already at photon energies of @xmath10 gev the produced invariant mass on one nucleon , @xmath11 , exceeds the threshold at which deep - inelastic processes become important , @xmath12 gev . combining the above property with a requirement that the momentum transfer in the reaction exceeds the masses of the particles
involved in the scattering ( @xmath13 ) one sets into a hard scattering kinematic regime , in which case we expect that only the minimal fock components dominate in the wave function of the particles involved in the scattering .
assuming that all the constituents of minimal fock component participate in a hard scattering one arrives at the constituent - counting rule@xcite .
according to this rule we are able to predict the energy dependence of any hard processes at fixed and large center of mass ( c.m . ) scattering angles .
for example , for the @xmath14 hard scattering process the constituent - counting rule predicts the following energy dependence for the scattering amplitude : @xmath15 with @xmath16 in eq.([qqrule_m ] ) @xmath17 is the number of constituents in the minimal fock component of the wave function of particle @xmath18 involved in the scattering .
these predictions were confirmed for a wide variety of hard processes involving leptons and hadrons .
one of the most interesting aspects of the constituent - counting rule is that its application allows us to check the onset of quark degrees of freedom in hard reactions involving nuclei @xcite .
this is essential for probing the quark - gluon structure of nuclei .
for example if quarks are involved in hard photodisintegration of the deuteron then according to eqs.([qqrule_m ] ) and ( [ crs_scale ] ) one expects that @xmath19 @xcite . during the last decade
there were several experiments in which 90@xmath20 c.m .
photodisintegration of the deuteron had been studied at high photon energies @xcite .
these experiments clearly demonstrated the onset of @xmath2 scaling for the differential cross section at 90@xmath20 c.m .
, starting at @xmath21 gev . also , the polarization measurements@xcite were generally in agreement with the prediction of the helicity conservation a precursor of the dominance of the mechanism of hard gluon exchange involving quarks . even though two - body scattering experiments demonstrate clearly an onset of quark degrees of freedom in the reaction
, they do not affirm the onset of the perturbative qcd ( pqcd ) regime .
indeed it has been argued that the validity of constituent - counting rule does not necessarily lead to the validity of pqcd(see , e.g. , refs@xcite ) . in several measurements in which the constituent quark rule works
pqcd still underestimates the observed cross sections sometimes by several orders of magnitude ( see e.g. refs.@xcite ) .
the latter may indicate a substantial contribution due to nonperturbative effects although one still may expect sizable contributions from pqcd due to generally unaccounted hidden color components in the hadronic and nuclear wave functions@xcite .
a similar situation also exists for the case of hard photodisintegration of the deuteron . even though experiments clearly indicate the onset of @xmath2 scaling for the cross section of these reactions at [email protected] .
, one still expects sizable nonperturbative effects . theoretical methods of calculation of these effects are very restricted .
they use different approaches to incorporate nonperturbative contributions in the process of hard photodisintegration of the deuteron .
the reduced nuclear amplitude ( rna ) formalism includes some of the nonperturbative effects through the nucleon form factors@xcite , while in the quark - gluon string model ( qgs)@xcite nonperturbative effects are accounted for through the reggeization of scattering amplitudes .
also recently , large c.m .
angle photodisintegration of the deuteron for photon energies up to 2 gev was calculated within point - form relativistic quantum mechanics approximation@xcite in which the strength of the reaction was determined by short range properties of the @xmath1 interaction potential . in the qcd hard rescattering model (
hrm)@xcite it is assumed that the energetic photon knocks - out a quark from one nucleon in the deuteron which subsequently experiences hard rescattering with a quark of the second nucleon .
the latter leads to the production of two nucleons with large relative momentum .
the summation of all the relevant rescattering diagrams results in a scattering amplitude in that the hard rescattering is determined by the large - momentum transfer @xmath6 scattering amplitude , which includes noncalculable nonperturbative contributions .
experimental data are used to estimate the hard @xmath6 scattering amplitude .
the hrm allows us to calculate the absolute cross section of @xmath22 c.m .
hard photodisintegration of the deuteron without using additional adjustable parameters .
also , within the qgs@xcite approximation and the hrm@xcite rather reasonable agreement has been obtained for polarization observables@xcite .
although all the above - mentioned models describe the major characteristics of hard photodisintegration of the deuteron they are based on very different approaches in the calculation of the nonperturbative parts of the photodisintegration reaction .
to investigate further the validity of these approaches it was suggested in ref.@xcite to extend the studies of high energy two - body photodisintegration to the case of large angle c.m .
breakup of two protons from the @xmath0he target . in this case
not only do the predictions of the above - described models ( rna , qgs , hrm ) for absolute cross section diverge significantly , but also the two - proton breakup reaction from @xmath0he provides additional observables such as spectator - neutron momentum distributions that can be used to check further the validity of the models .
the analysis of the first experimental data on @xmath5 photodisintegration of @xmath0he nucleus at high momentum transfer is currently in progress at jefferson lab@xcite .
it may significantly advance our understanding of the mechanism of hard breakup reactions involving nuclei .
in the present work we carry out detailed studies of reactions involving hard breakup of both @xmath5 and @xmath6 pairs from the @xmath0he target .
we demonstrate that comparative study of @xmath5 and @xmath6 breakup processes allows us to gain new insight into the nature of large c.m .
angle scattering .
one of the observations is that the relative strength of @xmath5 to @xmath6 breakup is larger than the one observed in low energy reactions .
this is related to the onset of quark degrees of freedom in hard breakup reactions in which effectively more charges are exchanged between two protons than between proton and neutron .
another signature of the hrm is that the shapes of the energy dependencies of @xmath3-scaled differential cross sections of @xmath5 and @xmath6 breakup reactions mirror the shapes of the energy dependencies of @xmath4-scaled differential cross sections of hard elastic @xmath5 and @xmath6 scatterings . within
the hrm one observes also that @xmath5 and @xmath6 hard breakup processes are sensitive to different components of the @xmath0he ground state wave function .
this results in different spectator - nucleon momentum dependencies for @xmath5 and @xmath6 hard breakup cross sections . because of the different ground state wave function components involved in @xmath5 and @xmath6 breakup reactions , the hrm also predicts significantly different magnitudes for transferred longitudinal polarizations for these two processes .
the article is organized as follows : in sec .
[ ii ] , within the hrm , we present a detailed derivation of the differential cross section of the reaction of hard breakup of two - nucleons from a @xmath0he target . in sec . [ iii ] we apply the formulas derived in the previous section to calculate the differential cross section of a proton - neutron breakup reaction , while in section [ iv ] calculations are done for a two - proton breakup reaction .
[ v ] considers the relative contribution of two- and three - body processes for hard breakup reactions involving @xmath23 nuclei . in sec .
[ vi ] we present numerical estimates for differential cross sections of @xmath6 and @xmath5 breakup reactions . in sec .
[ vii ] we discuss the polarization transfer mechanism of the hrm and estimate the asymmetry of the cross section with respect to the helicity of the outgoing proton .
results are summarized in sec .
[ viii ] . in appendix
a the details of the derivation of the hard rescattering amplitude are given .
in appendix b we discuss the quark - interchange contribution to hard @xmath1 elastic scattering .
appendix c discusses the method of calculation of quark - charge factors within quark - interchange mechanism of @xmath1 hard elastic scattering . in appendix
d we present the complete list of the hrm helicity amplitudes for high energy two - nucleon breakup of the @xmath0he nucleus .
we are considering a hard photodisintegration of two nucleons from the @xmath0he target through the reaction : @xmath24 in which two nucleons @xmath25 are produced at large angles in the `` @xmath26-@xmath1 '' center of mass reference frame with momenta comparable to the momentum of the initial photon , @xmath27 ( @xmath28@xmath29 gev/@xmath30 ) .
the third nucleon , @xmath31 , is produced with very small momentum @xmath32 .
( definitions of four - momenta involved in the reaction are given in fig.[fig.1 ] . )
we consider `` @xmath26-@xmath1 '' in a `` @xmath33 '' reference frame , where the light - cone momenta , where @xmath34 . here
the @xmath35 axis is defined in the direction opposite to the incoming photon momentum .
] of the photon and the @xmath1 pair are defined as follows : @xmath36 where @xmath37 , @xmath38 , and @xmath39 . here
the invariants , @xmath40 and @xmath41 are defined as follows : @xmath42 as it follows from eq.([q+=0 ] ) in the limit of @xmath43 the `` @xmath44 '' reference frame coincides with the center of mass frame of the @xmath26-@xmath1 system . as such it is maximally close to the reference frame used for the @xmath45 reaction in refs.@xcite and @xcite .
photodisintegration from the @xmath0he target.,width=325,height=325 ] the hard rescattering model is based on the assumption that in the hard two - nucleon photodisintegration reaction , two nucleons with large relative momenta are produced because the hard rescattering of a fast quark from one nucleon with a quark from the other nucleon . in this scenario
the fast quark is knocked out from a low - momentum nucleon in the nucleus by an incoming photon .
this approach is an alternative to the models in which it is assumed that the incoming photon breaks the preexisting two - nucleon state which has very large relative momentum in the nucleus .
the validity of the hrm is based on the observation that the ground state wave functions of light nuclei peak strongly at small momenta of bound nucleons , @xmath46 .
thus , diagrams in which an energetic photon interacts with bound nucleons of small momenta will strongly dominate the diagrams in which the photon interacts with bound nucleons that have relative momenta @xmath47 .
the resulting scenario that the hrm sketches out is as follows ( see e.g. fig.[fig.1 ] ) : first , the incoming photon will knock out a quark from one of the nucleons in the nucleus and then the struck quark that now carries almost the whole momentum of the photon will share its momentum with a quark from the other nucleon through the exchanged gluon .
the resulting two energetic quarks will recombine with the residual quark - gluon systems to produce two nucleons with large relative momentum ( @xmath48@xmath27 ) .
this recombination will contain gluon exchanges and also incalculable nonperturbative interactions .
note that for the quark - gluon picture discussed above to be relevant the intermediate masses @xmath49 produced after the photon absorption should exceed the mass scale characteristic for deep inelastic scattering , @xmath50 gev . using the relation @xmath51 , from the requirement that @xmath52 one obtains the condition @xmath53 gev .
additionally , to ensure the validity of quark degrees of freedom in the final state rescattering , one requires @xmath54
gev/@xmath30 for the relative momentum @xmath55 , of two outgoing nucleons .
all of these imposes a restriction on the incoming photon energy , @xmath56 gev , and for transferred momenta @xmath57 gev@xmath58 .
note that provided a smooth transition from hadronic to quark - gluon degrees of freedom in nuclei one expects the validity of the hrm to extend to even lower values of @xmath59 ( @xmath60@xmath29 gev ) .
this expectation was confirmed in recent measurements of angular dependencies of the @xmath61 cross section for a wide range of incoming photon energies@xcite . to calculate the differential cross section of the hard photodisintegration reaction of eq.([ghenn_n ] ) within the hrm one needs to evaluate the sum of hard rescattering diagrams similar to the one presented in fig.[fig.1 ] .
we start with analyzing the scattering amplitude corresponding to the diagrams of fig.[fig.1 ] . using feynman rules and applying the light - cone wave function reduction described in appendix a
, we obtain @xmath62 { i[{/ \hskip-5pt } p_{1i}-{/ \hskip-5pt } k_1 + { / \hskip-5pt } q + m_q]\over ( p_{1i}-k_1+q)^2-m_q^2 + i\epsilon } [ -iq_ie{\bf \epsilon^{\lambda_\gamma}_\perp\gamma^\perp}]u_{\eta_{1i}}(p_{1i}-k_1 ) \left .
{ \psi_n^{\lambda_{1i},\eta_{1i}}(p_{1i},x_1,k_{1\perp})\over ( 1-x_1)}\right\}_1\times \nonumber \\ & & \left\ { { \psi^{\dagger\lambda_{1f},\eta_{1f}}_n(p_{1f},x'_1,k_{1\perp})\over 1-x'_1 } \right .
\bar u_{\eta_{1f}}(p_{1f}-k_1)[-igt^f_c\gamma^\mu]u_{\eta_{2i}}(p_{2i}-k_2 ) \left .
{ \psi_n^{\lambda_{2i},\eta_{2i}}(p_{2i},x_2,k_{2\perp})\over ( 1-x_2 ) } \right\}_2\times \nonumber \\ & & g^{\mu,\nu}(r ) { dx_1\over x_1}{d^2k_{1\perp}\over 2 ( 2\pi)^3 } { dx_2\over x_2}{d^2k_{2\perp}\over 2 ( 2\pi)^3 } { \psi_{^3\textnormal{\scriptsize { he}}}^{\lambda_a,\lambda_{1i},\lambda_{2i},\lambda_s}(\alpha,{p_\perp},p_s ) \over ( 1-\alpha)}{d\alpha\over \alpha } { d^2p_{\perp}\over 2(2\pi)^3 } - \left(p_{1f}\longleftrightarrow p_{2f}\right ) , \label{ampl0}\end{aligned}\ ] ] where the @xmath63 part accounts for the diagram in fig.[fig.1](b ) . here the four - momenta , @xmath64 , @xmath65 , @xmath66 , @xmath67 , @xmath68 , @xmath69 , @xmath70 and @xmath71 are defined in fig.[fig.1 ] .
note that @xmath67 and @xmath68 define the four - momenta of residual quark - gluon system of the nucleons without specifying their actual composition .
we also define @xmath72 , @xmath73 , @xmath74 and @xmath75 as the light - cone momentum fractions of initial and final nucleons carried by their respective residual quark - gluon systems : @xmath76 and @xmath77 . for the @xmath0he wave function , @xmath78 is the light - cone momentum fraction of the @xmath1 pair carried by one of the nucleons in the pair , and @xmath79 is their relative transverse momentum . the scattering process in eq.([ampl0 ] ) can be described through the combination of the following blocks : ( a ) @xmath80 , is the light - cone @xmath0he - wave function that describes a transition of the @xmath0he nucleus with helicity @xmath81 into three nucleons with @xmath82 , @xmath83 , and @xmath84 helicities , respectively .
( b ) the term in @xmath85 describes the `` knocking out '' of an @xmath86-helicity quark from a @xmath82-helicity nucleon by an incoming photon with helicity @xmath87 .
subsequently , the knocked - out quark exchanges a gluon , ( @xmath88 $ ] ) , with a quark from the second nucleon producing a final @xmath89-helicity quark that combines into the nucleon `` @xmath90 '' with helicity @xmath91 .
( c ) the term in @xmath92 describes the emerging @xmath93-helicity quark from the @xmath83 -helicity nucleon then then exchanges a gluon , ( @xmath94 $ ] ) , with the knocked - out quark and produces a final @xmath95-helicity quark that combines into the nucleon `` @xmath96 '' with helicity @xmath97 .
d ) the propagator of the exchanged gluon is @xmath98 with polarization matrix , @xmath99 ( fixed by the light - cone gauge ) , and @xmath100 , with @xmath101 . in eq.([ampl0 ] ) the @xmath102 represents everywhere an @xmath103-helicity single quark wave function of a @xmath104-helicity nucleon as defined in eq.([nwf ] ) and @xmath105 is the quark spinor defined in the helicity basis .
we now consider the denominator of the struck quark s propagator which can be represented as follows : @xmath106 where @xmath107 \label{alphac}\ ] ] here @xmath108 and @xmath109 are defined in eqs.([k1onshell ] ) and ( [ tilmass ] ) , and @xmath110 represents the current quark mass of the knocked out quark .
our further discussion is based on the use of the fact that the @xmath0he wave function strongly peaks at @xmath111 , which corresponds to the kinematic situation in which two constituent nucleons have equal share of the @xmath1 pair s light - cone momentum .
thus one expects that the integral in eq.([ampl0 ] ) is dominated by the value of the integrand at @xmath112 .
this allows us to perform @xmath113-integration in eq.([ampl0 ] ) through the pole of the denominator ( [ denom ] ) at @xmath114 ( i.e. keeping only the @xmath115 part of the relation @xmath116 ) and later replacing @xmath117 by @xmath118 .
using this relation to estimate the propagator of the struck quark at its on - mass shell value ( @xmath114 ) allows us to represent , @xmath119 .
then for the scattering amplitude of eq.([ampl0 ] ) one obtains @xmath120\right .
\nonumber \\ & & { i\cdot u_\zeta(p_1-k_1+q)\bar u_\zeta(p_1-k_1+q)\over ( 1-x_1)s ' } [ -iq_ie{\bf \epsilon^{\lambda_\gamma}_\perp\gamma^\perp}]u_{\eta_{1i}}(p_{1i}-k_1 ) \left .
{ \psi_n^{\lambda_{1i},\eta_{1i}}(p_{1i},x_1,k_{1\perp})\over ( 1-x_1)}\right\}_1 \nonumber \\ & & \left\ { { \psi^{\dagger\lambda_{1f},\eta_{1f}}_n(p_{1f},x'_1,k_{1\perp})\over 1-x'_1}\right . \bar u_{\eta_{1f}}(p_{1f}-k_1)[-igt^f_c\gamma^\mu]u_{\eta_{2i}}(p_{2i}-k_2 ) \left .
{ \psi_n^{\lambda_{2i},\eta_{2i}}(p_{2i},x_2,k_{2\perp})\over ( 1-x_2 ) } \right\}_2 \nonumber \\ & & g^{\mu,\nu}(r ) { dx_1\over x_1}{d^2k_{1\perp}\over 2 ( 2\pi)^3 } { dx_2\over x_2}{d^2k_{2\perp}\over 2 ( 2\pi)^3 } { \psi_{^3\textnormal{\scriptsize { he}}}^{\lambda_a,\lambda_{1i},\lambda_{2i},\lambda_s}(\alpha_c , p_{\perp},p_s)\over ( 1-\alpha_c ) \alpha_c } { d^2p_{\perp}\over 4(2\pi)^2}- \left(p_{1f}\longleftrightarrow p_{2f}\right ) .
\label{ampl1}\end{aligned}\ ] ] next we evaluate the matrix element of the photon - quark interaction using on - mass shell spinors for the struck quark . taking into account the fact that @xmath121 , for this matrix element we obtain @xmath122u_{\eta_{1i}}(p_{1i}-k_1 ) = ieq_i2\sqrt{2 e_2e_1}(\lambda_\gamma)\delta^{\lambda_\gamma\zeta}\delta^{\lambda_\gamma\eta_{1i } } \label{melement}\end{aligned}\ ] ] where @xmath123 and @xmath124 .
further explicit calculations of eq.([ampl1 ] ) require the knowledge of quark wave functions of the nucleon .
also , one needs to sum over the multitude of the amplitudes representing different topologies of quark knock - out rescattering and recombinations into two final nucleon states .
we attempt to solve this problem by using the above observation that the @xmath0he wave function strongly peaks at @xmath125 .
we evaluate eq.([ampl1 ] ) setting everywhere @xmath126 .
such approximation significantly simplifies further derivations . as it follows from eq.([alphac ] ) the @xmath127 condition restricts the values of @xmath72 of the recoil quark - gluon system to @xmath128 , thereby ensuring that the quark - interchange happens for the valence quarks with @xmath129 .
the latter allows us to simplify eq.([melement ] ) setting @xmath130 . using these approximations and substituting eq.([melement ] ) into eq.([ampl1 ] )
one obtains @xmath131e\sum\limits_{(\eta_{1f},\eta_{2f}),(\eta_{2i}),(\lambda_{1i},\lambda_{2i } ) } \int { q_i \over \sqrt{2s ' } } \times \nonumber \\ & & \left [ \left\ { { \psi^{\dagger\lambda_{2f},\eta_{2f}}_n(p_{2f},x'_2,k_{2\perp})\over 1-x'_2}\bar u_{\eta_{2f}}(p_{2f}-k_2 ) [ -igt^f_c\gamma^\nu]\right .
u_{\lambda_\gamma}(p_1-k_1+q ) \left .
{ \psi_n^{\lambda_{1i},\lambda_\gamma}(p_{1i},x_1,k_{1\perp})\over ( 1-x_1 ) } \right\ } \nonumber \\ & & \left\ { { \psi^{\dagger\lambda_{1f},\eta_{1f}}_n(p_{1f},x'_1,k_{1\perp})\over 1-x'_1 } \bar u_{\eta_{1f}}(p_{1f}-k_1)[-igt^f_c\gamma^\mu ] \right .
u_{\eta_{2i}}(p_{2i}-k_2 ) { \psi_n^{\lambda_{2i},\eta_{2i}}(p_{2i},x_2,k_2)\over ( 1-x_2 ) } \right\ } \nonumber \\ & & \left .
g^{\mu,\nu}(r ) { dx_1\over x_1}{d^2k_{1\perp}\over 2 ( 2\pi)^3 } { dx_2\over x_2}{d^2k_{2\perp}\over 2 ( 2\pi)^3 } \right]_{\textnormal{\scriptsize { qim } } } \psi_{^3\textnormal{\scriptsize { he}}}^{\lambda_a,\lambda_{1i},\lambda_{2i}}(\alpha={1\over 2},p_{2\perp ) } { d^2p_{2\perp}\over ( 2\pi)^2 } \ \ \ \ \ - \left(p_{1f}\leftrightarrow p_{2f}\right )
. \label{ampl2}\end{aligned}\ ] ] note that due to the @xmath132 factors in eq.([melement ] ) the helicity of the knocked out quark in eq.([ampl2 ] ) is equal to the helicity of incoming photon , that is @xmath133 .
to proceed , we observe that the kernel , @xmath134_{\textnormal{\scriptsize qim}}$ ] representing the quark - interchange mechanism ( qim ) of the rescattering in eq.([ampl2 ] ) can be identified with the quark - interchange contribution in the @xmath1 scattering amplitude ( see appendix b ) .
such identification can be done by observing that in the chosen reference frame , @xmath33 , and the quark wave function of the nucleon depends on the quark s light - cone momentum fraction and transverse momentum only , which are the same in both eqs.([ampl2 ] ) and ( [ nn_qim ] ) . for our derivation
we also use the above - discussed observation that the @xmath135 condition ensures that the quark - interchange happens for the valence quarks with @xmath129 .
this justifies our next assumption , that valence quarks carry the helicity of their parent nucleon ( i.e. @xmath136 ) .
the last assumption allows us to perform the summation of eq.([ampl2 ] ) over the helicities of the exchanged quarks ( @xmath137 ) and to use eq.([nn_qim ] ) to express the qim part in eq.([ampl2 ] ) through the corresponding qim amplitude of @xmath1 scattering .
summing for all possible topologies of quark - interchange diagrams we arrive at @xmath138\times \nonumber \\ & & \left\ { \sum\limits_{i \in n_1}\sum\limits_{\lambda_{2i } } \int { q_i^{n_1}\over \sqrt{2s ' } } \langle \lambda_{2f};\lambda_{1f}\mid t_{nn , i}^{\textnormal{\scriptsize { qim}}}(s , l^2)\mid \lambda_\gamma;\lambda_{2i}\rangle \psi_{^3\textnormal{\scriptsize { he}}}^{\lambda_a}(p_1,\lambda_\gamma;p_2,\lambda_{2i},p_s,\lambda_s ) { d^2p_\perp \over ( 2\pi)^2 } \right . \nonumber \\ & + & \left .
\sum\limits_{i \in n_2}\sum\limits_{\lambda_{1i } } \int { q_i^{n_2}\over \sqrt{2s ' } } \langle \lambda_{2f};\lambda_{1f}\mid t_{nn , i}^{\textnormal{\scriptsize { qim}}}(s , l^2)\mid \lambda_{1i};\lambda_{\gamma}\rangle \psi_{^3\textnormal{\scriptsize { he}}}^{\lambda_a}(p_1,\lambda_{1i};p_2,\lambda_{\gamma},p_s,\lambda_s ) { d^2p_\perp \over ( 2\pi)^2 } \right\ } \label{ampl4}\end{aligned}\ ] ] where nucleon momenta @xmath139 and @xmath140 have half of their c.m .
momentum fractions and @xmath141 is their relative transverse momentum with respect to the direction of the photon momentum [ see eq.([momrel ] ) ] . here , for example , @xmath142 represents the quark - interchange amplitude of @xmath1 interaction weighted with the charge of those interchanging quarks @xmath143 that are struck from a nucleon @xmath144 by the incoming photon . the sum ( @xmath145 )
can be performed within the quark - interchange model of @xmath1 interaction which allows us to represent the @xmath1 scattering amplitude as follows @xcite : @xmath146 f_{i , j}(s , t)|ab\rangle \label{nnqim}\ ] ] where @xmath147 and @xmath148 are the identity and pauli matrices defined in the su(2 ) flavor ( isospin ) space of the interchanged quarks .
the kernel @xmath149 describes an interchange of @xmath18 and @xmath150 quarks .
\tilde f_{i , j}(s , t)$ ] , where @xmath147 and @xmath151 operate in the su(2 ) helicity ( @xmath152-spin ) space of exchanged ( @xmath153 ) quarks@xcite .
however for our discussion the assumption of helicity conservation is not required . ] . using eq.([nnqim ] ) one can calculate the quark - charge weighted qim amplitude , @xmath154 , as follows : @xmath155 ( q_i)f_{i , j}(s , t)|ab\rangle \nonumber\\ & = & q^n_f\cdot \langle a'b ' |t_{nn}^{\textnormal{\scriptsize { qim}}}|ab\rangle , \label{qnnqim}\end{aligned}\ ] ] where @xmath156 are the charge factor that are explicitly calculated using the method described in appendix c. these factors can be expressed through the combinations of valence quark charges @xmath157 of nucleon @xmath144 and the number of quark interchanges for each flavor of quark , @xmath158 , necessary to produce a given helicity @xmath1 amplitude , as follows , @xmath159 next we discuss the light - cone wave function of @xmath0he that enters in eq.([ampl4 ] ) .
the important result that allows us to evaluate the wave function is the observation that two nucleons that interact with the photon share equally the @xmath1 pair s c.m .
momentum ( @xmath160 ) , i.e. , @xmath125 .
if we constrain the third nucleon s light - cone momentum fraction @xmath161 and transverse momentum @xmath162 , then the momenta of all the nucleons in the nucleus are nonrelativistic . in this case one can use the calculation of triangle diagrams , which provides the normalization of nuclear wave functions based on baryonic number conservation to relate lc and nonrelativistic nuclear wave functions as follows@xcite @xmath163 where for @xmath164 we can use known nonrelativistic @xmath0he wave functions ( see e.g.,@xcite ) .
substituting eqs.([qnnqim ] ) and ( [ wfnr ] ) into eq.([ampl4 ] ) for the two - nucleon photodisintegration amplitude we obtain @xmath165e\sqrt{2}(2\pi)^3\over \sqrt{2s^\prime_{nn}}}\times \nonumber \\ & & \ \
\left\ { q_f^{n_1 } \sum\limits_{\lambda_{2i } } \int \langle \lambda_{2f};\lambda_{1f}\mid t_{nn}^{\textnormal{\scriptsize { qim}}}(s_{nn},t_{n})\mid \lambda_\gamma;\lambda_{2i}\rangle \psi_{^3\textnormal{\scriptsize { he , nr}}}^{\lambda_a } ( \vec p_1,\lambda_\gamma;\vec p_2,\lambda_{2i};\vec p_s,\lambda_s)m_n{d^2p_{\perp } \over ( 2\pi)^2 } + \right .
\nonumber \\ & & \ \ \left .
q_f^{n_2 } \sum\limits_{\lambda_{1i } } \int \langle \lambda_{2f};\lambda_{1f}\mid t_{nn}^{\textnormal{\scriptsize { qim}}}(s_{nn},t_{n})\mid \lambda_{1i};\lambda_{\lambda}\rangle \psi_{^3\textnormal{\scriptsize { he , nr}}}^{\lambda_a } ( \vec p_1,\lambda_{1i};\vec p_2,\lambda_{\gamma};\vec p_s,\lambda_s)m_n{d^2p_{\perp } \over ( 2\pi)^2}\right\ } \nonumber \\ \label{ampl5}\end{aligned}\ ] ] where in the lab frame of the @xmath0he nucleus , defining the @xmath35 direction along the direction of @xmath166 one obtains @xmath167 with all the momenta defined in the lab frame .
equation([ampl5 ] ) allows us to calculate the unpolarized differential cross section of two nucleon breakup in the form @xmath168 where @xmath169 and @xmath170 as follows from eq.([ampl5 ] ) the knowledge of quark - interchange helicity amplitudes of @xmath1 elastic scattering will allow us to calculate the differential cross section of hard @xmath1 breakup reaction without introducing any adjustable parameter . because the assumption of @xmath127 plays a major role in the above derivations we attempt now to estimate the theoretical error introduced by this approximation
this approximation by its nature is a `` peaking '' approximation that is used in loop calculations involving feynman diagrams ( one such example is the calculation of radiative effects in electroproduction processes ; see , e.g. , @xcite ) .
one way to estimate the accuracy of the approximation is to identify the main dependence of the integrand in eq.([ampl1 ] ) on @xmath117 which can be evaluated exactly and compare with its evaluation at @xmath127 . using eq.([melement ] ) as well as eq.([alphac ] ) that allows us to relate @xmath171 to @xmath172 , and assuming that the quark wave functions of nucleons at @xmath173 are less sensitive to @xmath117 , one arrives at @xmath174 this ratio depends on the kinematics of the spectator nucleon , and for the case of @xmath175 mev/@xmath30 , @xmath176 , which corresponds to @xmath48@xmath177 of uncertainty in the cross section of the reaction calculated with the @xmath127 approximation .
the uncertainty increases with an increase of the momentum of the spectator nucleon .
this can be understood qualitatively because , for large center of mass momenta of the @xmath1 pair , the @xmath111 peak of the nuclear wave function is less pronounced .
the possibility of using @xmath1 elastic scattering data to calculate the cross section in eqs.([hrm_crs ] ) and ( [ m2 ] ) is based on the assumption that the quark - interchange mechanism provides the bulk of the @xmath1 elastic scattering strength at high energies and large c.m .
this is a rather well - justified assumption .
experiments on exclusive large @xmath178 two - body reactions @xcite demonstrated clearly the dominance of the quark - interchange mechanism for the scattering of hadrons that share common quark flavors .
the analysis of these experiments indicate that contributions from competing mechanisms such as pure gluon exchange or quark - antiquark annihilation are on the level of few percent .
this justifies our next approximation , to substitute quark - interchange @xmath1 amplitudes in eq.([ampl5 ] ) with actual @xmath1 helicity amplitudes as follows : @xmath179 all other helicity combinations can be related to the above amplitudes through the parity and time - reversal symmetry . the minus sign in the last equation above is due to the jackob - wick phase factor ( see , e.g. , ref.@xcite ) , according to which one gains a phase factor of ( @xmath180@xmath29 ) if two quarks that scatter by @xmath181 angle in c.m .
have opposite helicities @xcite .
note that @xmath182 s are normalized in such a way that the cross section for @xmath1 scattering is defined as @xmath183 because in the hard breakup regime the momentum transfer @xmath184 , one can factorize the helicity @xmath1 amplitudes from eq.([ampl5 ] ) at @xmath40 and @xmath185 values defined as follows : @xmath186 using this factorization in eq.([ampl5 ] ) for the spin averaged square of the breakup amplitude one obtains @xmath187s_{34}\right\ } , \label{m2_fct}\end{aligned}\ ] ] where @xmath188 and @xmath189 , @xmath190 , and @xmath191 are partially integrated nuclear spectral functions : @xmath192 @xmath193 and @xmath194 in the above equations @xmath195 and @xmath196 are the isospin projections of nucleons in @xmath0he .
the wave function is normalized to @xmath7 for proton and @xmath197 for neutron .
the normalization constants , @xmath198 renormalize the wave function to one @xmath5 and two @xmath199 effective pairs in the wave function with @xmath200 and @xmath201 .
equations.([hrm_crs ] ) and ( [ m2_fct ] ) together with eqs.([phis]),([s12]),and ( [ s34 ] ) allow us to calculate the differential cross section of both @xmath5 and @xmath6 breakup reactions off the @xmath0he target .
notice that , on the qualitative level , as it follows from eqs.([hrm_crs ] ) and ( [ m2_fct ] ) in the limit of @xmath202 and @xmath203 , the hrm predicts an @xmath2 invariant energy dependence of the differential cross section provided that the @xmath1 cross section scales as @xmath204 .
however the numerical calculations of eq.([m2_fct ] ) require a knowledge of the @xmath1 helicity amplitudes at high energy and momentum transfers .
our strategy is to use eq.([crs_nn ] ) to express @xmath1 breakup reactions directly through the differential cross section of @xmath6 and @xmath5 elastic scatterings rather than to use helicity amplitudes explicitly .
we consider now the reaction @xmath205 in which one proton is very energetic and produced at large c.m .
angles with the neutron , while the second proton emerges with low momentum @xmath206
@xmath207 mev/@xmath30 . in this case
the hard rescattering happens in the @xmath6 channel . using the @xmath208 relation for hard @xmath6 scattering amplitude ( see , e.g. , refs @xcite ) for breakup amplitude of eq.([m2_fct ] ) one obtains @xmath209 where @xmath210 can be calculated using eqs.([qnnqim ] ) and ( [ qf ] ) .
based on su(6 ) flavor - spin symmetry of nucleon wave functions , for the helicity amplitudes of eq.([phis ] ) using the method described in appendix c one obtains @xmath211 we can further simplify eq.([m2_fct_pn ] ) noticing that for the @xmath6 pair in @xmath0he one has @xmath212 .
this is due to the fact that in the dominant @xmath213 state two protons have opposite spins and therefore the probability of finding one proton with a helicity opposite to that of the neutron is equal to the other proton having the same helicity as the neutron s . using this relation and eq.([crs_nn ] ) for
the @xmath6 breakup reaction one obtains @xmath214 inserting it in eq.([hrm_crs ] ) for the differential cross section one obtains @xmath215 where @xmath216 and @xmath217 is the differential cross section of hard @xmath6 scattering evaluated at values of @xmath40 and @xmath218 defined in eq.([stun ] ) .
the spectral function @xmath219 is defined in eq.([s0 ] ) and corresponds to : @xmath220
we now consider the reaction @xmath221 in which two protons are produced at large c.m .
angles while the neutron emerges as a spectator with small momentum ( @xmath222 mev/@xmath30 ) .
we observe now that the relation between @xmath189 and @xmath190 is very different from that of the @xmath6 case .
due to the fact that two protons can not have the same helicity in the @xmath213 state one has @xmath223 .
the estimates of the spectral functions based on the realistic @xmath0he wave function@xcite gives @xmath224 .
therefore one can neglect the @xmath189 term in eq.([m2_fct ] ) .
the next observation is that for @xmath5 scattering the helicity amplitudes @xmath225 and @xmath226 have opposite signs due to the pauli principle ( see , e.g. , refs.@xcite ) .
using the above observations and neglecting the helicity - nonconserving amplitude @xmath227 for the @xmath5-breakup amplitude we obtain @xmath228 the charge factor @xmath229 depends on the helicity amplitude it couples ; therefore one estimates it for the combination of @xmath230 . using su(6 ) symmetry for the distribution of given helicity - flavor valence quarks in the proton and through the approach described in appendix c
we obtain @xmath231 with @xmath232 it is worth noticing that due to explicit consideration of quark degrees of freedom the effective charge involved in the breakup is larger for the case of two protons than for proton and neutron [ see eq.([qfpn ] ) ] .
this is characteristic of the hrm model in which a photon couples to a quark and more charges are exchanged in the @xmath5 case than in the @xmath6 case .
this is rather opposite to the scattering picture considered based on hadronic degrees of freedom in which case the photon will couple to an exchanged meson and @xmath5 contribution will be significantly suppressed because no charged mesons can be exchanged within the @xmath5 pair . to be able to estimate the cross section of the @xmath5 breakup reaction through the elastic @xmath5 scattering cross section we introduce a parameter @xmath233 which allows to express the differential cross section of the reaction ( [ pp_hrm ] ) in the following form : @xmath234 where we also introduced a factor @xmath235 , @xmath236 which accounts for the suppression due to the cancellation between @xmath225 and @xmath226 helicity amplitudes of elastic @xmath5 scattering .
photodisintegration from @xmath0he target ( see , e.g. , @xcite ) . ] the spectral function @xmath237 in eq.([pp_hrm_crs ] ) is expressed through the @xmath0he wave function according to eq.([s34 ] ) as follows : @xmath238
for a two - body hard @xmath1 breakup mechanism to be observed it must dominate the three - body / two - step processes .
this is especially important for @xmath5 breakup processes , ( [ pp_hrm ] ) because according to eqs.([pp_hrm_crs ] and [ betafactor ] ) the two - body contribution is suppressed because of a cancellation between @xmath225 and @xmath226 helicity amplitudes . at low to intermediate range energies ( @xmath239 mev )
it is rather well established that the @xmath5 breakup reaction proceeds overwhelmingly through a two - step ( three - body ) process@xcite in which the initial breakup of the @xmath6 pair ( dominated by @xmath240 exchange ) is followed by a charge - interchange final state interaction of the neutron with the spectator proton .
other two - step processes include the excitation of intermediate @xmath241 isobars in the @xmath6 system with the subsequent rescattering off the spectator neutron , which produces two final protons .
pair is followed by a soft charge - exchange rescattering of the neutron off the spectator proton . ]
the dominance of three - body processes at low energies is related mainly to the fact that the two - body @xmath5 breakup is negligible because of the impossibility of charged - pion exchanges between two protons that absorb an incoming photon . at high energy kinematics within the hrm the interaction between protons
is carried out by exchanged quarks because of which the relative strength of @xmath5 breakup is larger . to estimate the strength of three - body processes at high energy kinematics
, one needs to calculate the contribution of diagrams similar to fig .
[ 3bfig ] .
because the charge - exchange rescattering at the final stage of the process in fig.[3bfig ] takes place at proton momenta @xmath242 , one can apply an eikonal approximation@xcite to estimate its contribution . for @xmath56 gev
assuming that the hrm is valid for the first ( @xmath6 breakup ) stage of the reaction , for the amplitude of three - body / two - step process within the eikonal approximation@xcite one obtains @xmath243 where we suppressed helicity indices for simplicity and choose the isospin projections , @xmath244 , corresponding to the initial @xmath6 pair that interacts with the photon . here
@xmath245 is the hard elastic @xmath6 scattering amplitude and @xmath246 represents the amplitude of the soft charge - exchange @xmath6 scattering . because of the pion - exchange nature of the latter it is rather well established that this amplitude is real and can be represented as @xmath48@xmath247 , where @xmath248 and @xmath249 are approximately constants@xcite .
two main observations follow from eq.([3b ] ) and the above - mentioned property of the charge - exchange amplitude : first , three - body and two - body amplitudes [ see , e.g. , eq.([ampl5 ] ) ] will not interfere , since one is real and the other is imaginary .
the fact that these two amplitudes differ by order of @xmath18 follows from the general structure of rescattering amplitudes ( see , e.g. , ref.@xcite ) .
equation ( [ ampl5 ] ) corresponds to a single rescattering amplitude , while eq.([3b ] ) to a double rescattering amplitude .
+ second , because of the energy dependence of the charge - exchange scattering amplitude at small angles , the three - body contribution will scale like @xmath250 as compared to the two - body breakup contribution . using eq.([3b ] )
one can estimate the magnitude of the contribution of three - body processes in the @xmath5 breakup cross section as follows : @xmath251 where @xmath252 is defined in eq.([s0pn ] ) and for @xmath253 based on eq.([3b ] ) one obtains @xmath254 here both spectral functions are defined at @xmath111 . using eqs.([3to2 ] ) and ( [ spnp ] ) and the parametrization of @xmath246 from ref.@xcite one can estimate the relative contribution of three - body processes numerically .
note that this contribution is maximal at @xmath255 and increases with an increase of the momentum of @xmath66 .
however because of the charge - exchange nature of the second rescattering , this contribution decreases linearly with an increase of @xmath256 .
rescattering @xmath257 and as a result the probability of the rescattering is energy independent . ]
for numerical estimates we consider the center of mass reference frame of @xmath26@xmath1 system , for which according to eq.([mans_st ] ) one obtains @xmath258 where @xmath259 and @xmath260 . using the above equation we obtain for @xmath185 [ eq.([stun ] ) ] , which defines the effective momentum transfer in the @xmath1 scattering amplitude , the following relation : @xmath261
one can also calculate the effective c.m .
angle that enters in the @xmath1 scattering amplitude as follows : @xmath262 the above equations define the kinematics of hard @xmath1 rescattering . weighted differential cross sections at @xmath263 c.m .
angle scattering in `` @xmath26-@xmath1''system . in these calculations
one integrated over the spectator nucleon momenta in the range of @xmath264-@xmath207 mev/@xmath30.,width=325,height=325 ] we are interested in energy dependences of the hard breakup reactions of eqs.([pn_hrm ] ) and ( [ pp_hrm ] ) at fixed and large angle production of two fast nucleons in the `` @xmath26-@xmath1 '' center of mass reference frame .
particularly interesting is the case of @xmath265=@xmath266 for which as it follows from eq.([theta_cmn ] ) @xmath267 .
this means that the cross sections of hard breakup reactions at these kinematics will be defined by the @xmath1 elastic scattering at @xmath268 . in fig.[fig.s_dep ] the @xmath59 and @xmath256 dependencies of the @xmath269 weighted differential cross sections are presented for the cases of the @xmath5 and @xmath6 breakup reactions . in the calculation we integrated over the spectator nucleon s momentum in the range of ( 0 - 100 ) mev/@xmath30 and over the whole range of its solid angle . also for the parameter @xmath270 in eq.([cfactor ] ) we used @xmath271 , consistent with an estimate obtained within the quark - interchange model of @xmath5 scattering ( see , e.g.,@xcite ) .
the estimation of the factor @xmath235 , which takes into account the cancellation between @xmath225 and @xmath226 helicity amplitudes in eq.([pp_hrm_crs ] ) requires the knowledge of the angular dependence for helicity amplitudes . for this we used the helicity amplitudes calculated within thequark - interchange model@xcite with phenomenological angular dependencies
estimated using @xmath272 ( see , e.g. , refs .
@xcite ) which describes reasonably well the data at hard scattering kinematics .
weighted differential cross sections of elastic @xmath5 scattering at @xmath273 and @xmath274 .
curves are the fits @xcite of the available world data @xcite.,width=325,height=325 ] several features of the hrm calculations are worth discussing in fig.[fig.s_dep ] : first , the breakup cross sections in average scale like @xmath275 .
note that the absolute ( nonscaled ) values of the cross sections drop by five orders of magnitude in the 2 - 8 gev of photon energy range .
next , the shapes of the @xmath3 weighted differential cross sections reflect the shapes of the @xmath4 weighted differential cross sections of @xmath5 and @xmath6 scattering at @xmath274 .
[ see figs.([fig.pp ] ) and ( [ fig.pn ] ) ] .
it is worth noting that as follows from figs.([fig.pp ] ) and ( [ fig.pn ] ) the fits used in the calculation of @xmath5 and @xmath6 breakup reactions contain uncertainties on the level of 10% for @xmath5 breakup ( for @xmath276 gev@xmath58 ) and up to 30% for @xmath6 breakup reactions .
based on this , one can conclude that the calculated shape of the energy dependence of the @xmath6 breakup reaction in fig.([fig.s_dep ] ) does not have much predictive power .
however , for the @xmath5 breakup the calculated shape , for up to @xmath277 gev@xmath58 , is not obscured by the uncertainty of the @xmath5 data and can be considered as a prediction of the hrm .
it is worth mentioning that considered features of the hrm are insensitive to the choice of the above - discussed parameters of @xmath270 and @xmath235 , because they only define the absolute magnitude of the @xmath5 breakup cross section .
weighted differential cross sections of elastic @xmath6 scattering at @xmath273 and @xmath274 .
curves are the fits @xcite of the available world data @xcite.,width=325,height=325 ] the next feature of the calculations in fig.[fig.s_dep ] is the magnitude of the @xmath6 and @xmath5 breakup cross sections .
the @xmath6 breakup cross section [ eq.([pn_hrm_crs ] ) ] does not contain any free parameter , and similar to the hrm prediction for the breakup of the deuteron@xcite , it is expressed through the rather well - defined quantities . for the estimate of @xmath5 breakup , however , one needs to know the relative strength of the @xmath225 and @xmath226 amplitudes as compared to @xmath278 as well as the extent of their cancellation at kinematics of @xmath40 and @xmath218 defined in eqs.([stun ] ) and ( [ tn ] ) .
our calculation , based on phenomenologically justified estimates of factors @xmath270 and @xmath235 in eq.([pp_hrm_crs ] ) results in the @xmath5 breakup cross section which is about ten times smaller than the cross section for the @xmath6 breakup .
this magnitude indicates however , an increase of @xmath5 breakup cross section relative to the @xmath6 breakup cross section as compared to the results from the low energy breakup reactions . as it was mentioned in sec.[v ] at low energies ( @xmath48 200 mev )
the cross section of @xmath5 photodisintegration from @xmath0he is significantly smaller ( by almost two orders of magnitude according to ref.@xcite ) than the @xmath6-breakup cross section .
note that the factors @xmath270 and @xmath235 introduce an additional uncertainty in the estimation of the magnitude of the @xmath5-breakup cross section . while the factor @xmath270 can be evaluated in the quark - interchange model thus staying within the framework of the considered model , the factor @xmath235 is not constrained by the theoretical framework of the model .
the latter is sensitive to the angular dependence , @xmath279 , of the helicity amplitudes . to estimate the uncertainty due to @xmath279 we varied it around the form , @xmath280 in such a way that the results were still in agreement with angular distribution of @xmath5 scattering at @xmath281 gev@xmath58 .
we found that this variation changes the hrm prediction for the magnitude of @xmath5-breakup cross section as much as @xmath282 .
breakup at @xmath283 gev on ( a ) transverse momentum of the spectator neutron @xmath284 at @xmath255 and on ( b ) @xmath285 at @xmath286.,width=325,height=325 ] because @xmath5 breakup cross section is still by a factor of 10 smaller than the @xmath6 cross section , one needs to estimate the contribution due to three - body processes in which hard @xmath6 breakup is followed by soft charge - exchange rescattering . the estimate based on eq.([3to2
] ) is given in fig .
[ fig.3b_2b ] where the ratio of three - body to two - body breakup cross sections is evaluated for different values of @xmath285 and transverse momentum of the spectator neutron , @xmath287 . because of the eikonal nature of the second rescattering in three - body processes , one expects the cross section to be maximal at @xmath255 . as fig.[fig.3b_2b](a )
shows in this case , the three - body process is a correction to the two - body breakup process , @xmath482% for @xmath288 mev/@xmath30 and @xmath289% for @xmath290 mev/@xmath30 .
then , starting at @xmath287 > 300 mev/@xmath30 the three - body process dominates the two - body contribution .
the latter can be verified by observing an onset of @xmath250 scaling at large ( @xmath291 mev/@xmath30 ) transverse momenta of the spectator neutron in the case of hard @xmath5 breakup reactions .
fig.[fig.3b_2b](b ) shows also that the three - body contribution will be always small for @xmath292 mev/@xmath30 , and for a wide range of @xmath285 , which again reflects the eikonal nature of the second order rescattering in which case the recoiling of the spectator nucleons happens predominantly at @xmath293 ( see , e.g. , @xcite ) .
note that one expects the above estimate of the three - body contribution to contain an uncertainty of 10 - 15% , representing the general level of accuracy of eikonal approximations .
based on fig.[fig.3b_2b ] one can expect that overall , for small values of @xmath294100 - 150 mev/@xmath30 in the high energy limit ( @xmath295 gev ) one expects two - body breakup mechanisms to dominate for both @xmath5 and @xmath6 production reactions .
the presence of a spectator nucleon in the hard two - nucleon breakup reaction from @xmath0he gives us an additional degree of freedom in checking the mechanism of the photodisintegration .
as follows from eqs.([pn_hrm_crs ] ) and ( [ s0pn ] ) and eqs.([pp_hrm_crs ] ) and ( [ s34pp ] ) the @xmath6 and @xmath5 breakup cross sections within the hrm are sensitive to different components of the nuclear spectral function .
this is due to the fact that the @xmath5 component with the same helicities for both protons is suppressed in the ground state wave function of the @xmath0he target .
thus one expects rather different spectator - momentum dependencies for @xmath5 and @xmath6 breakup cross sections .
weighted 90@xmath296 c.m .
breakup differential cross section on the light - cone momentum fraction of of spectator nucleon , @xmath285 , calculated at @xmath297 gev and @xmath298 .
( a ) the solid line is for @xmath5 breakup reactions , and the dashed line is for @xmath6 breakup reactions .
calculations are normalized to the cross sections at @xmath255 .
( b ) ratio of the @xmath6 to @xmath5 breakup cross sections normalized to their values at @xmath255.,width=325,height=325 ] the quantity that we consider for numerical estimates is not the momentum of the spectator but rather the momentum fraction of the target carried by the spectator nucleon , @xmath285 .
this quantity is lorentz invariant with respect to boosts in the @xmath27 direction , which allows us to specify it in the lab frame as follows : @xmath299 where @xmath300 for @xmath301 , @xmath302 , @xmath303 and @xmath35 axis in the lab frame is defined parallel to the momentum of incoming photon @xmath27 .
note that the photon does not contribute to the above equation because @xmath304 . in definition of @xmath285
we use a normalization such that for a stationary spectator @xmath255 .
the @xmath285 dependencies of the differential cross sections for @xmath5 and @xmath6 breakup reactions normalized to their values at @xmath255 are given in fig.[fig.al_dep](a ) .
one feature of @xmath285 dependence is the asymmetry of the cross section around @xmath255 with cross sections dominating at @xmath305 .
this property can be understood from the fact that the momentum fraction of the @xmath1 pair that breaks up is defined through @xmath285 as follows : @xmath306 the latter quantity defines the invariant energy of the @xmath1 pair as follows @xmath307 because the cross section within the hrm is proportional to @xmath308 , factor in the differential cross section of the reaction [ see eqs.([pn_hrm_crs],[pp_hrm_crs ] ) ] . ]
it will be enhanced at small values of @xmath40 that will correspond to smaller values of @xmath309 or larger values of @xmath285 .
the difference of the cross sections due to the different composition of the nuclear spectral functions entering the @xmath5 and @xmath6 breakup reactions can be seen in fig.[fig.al_dep](b ) in which case one calculates the ratio of @xmath6 to @xmath5 breakup cross sections normalized to their values at @xmath255 .
the drop of the ratio in fig.[fig.al_dep].(b ) at values close to @xmath255 is due to the suppression of the same - helicity two - proton component in the ground state nuclear wave function at small momenta . in this case
the spectral function is sensitive to the higher angular momentum components of the ground state nuclear wave function .
this yields a wider momentum distribution for the @xmath5 spectral function as compared to that for @xmath6 because no same - helicity state suppression exits for the latter .
the estimates indicate that differences in @xmath285 dependencies of @xmath5 and @xmath6 breakup cross sections are rather large and can play an additional role in checking the validity of the hrm .
one of the unique properties of the hard rescattering mechanism of two - nucleon breakup is that the helicity of the nucleon from which a quark is struck is predominantly defined by the helicity of the incoming photon @xmath310 [ see eq.([ampl5 ] ) ] .
this is based on the fact that in the massless quark limit the helicity of the struck quark equals the helicity of the photon , @xmath133 , and assuming that at large @xmath311 the quark carries almost all the helicity of the parent nucleon one obtains @xmath312 .
because within the hrm , the energetic struck quark shares its momentum with a quark of the other nucleon through a hard gluon exchange , it will retain its initial helicity when it merges into the final outgoing nucleon .
it will also have @xmath313 , which allows us to conclude that the final outgoing nucleon will acquire the large part of struck quark s ( as well as the photon s ) helicity .
this mechanism will result in a large ( photon ) polarization transfer for the hard two - nucleon breakup reactions .
an observable that is sensitive to polarization transfer processes is the quantity @xmath314 , which for a circularly polarized photon measures the asymmetry of the hard breakup reaction with respect to the helicity of the outgoing proton .
a large value of @xmath314 was predicted within the hrm for the hard breakup of the deuteron in ref.@xcite that was observed in the recent experiment of ref.@xcite .
for the case of the @xmath0he target an additional experimental observation will be a comparison of @xmath314 asymmetries for @xmath5 and @xmath6 breakup channels .
for the @xmath0he target we define @xmath314 as follows : @xmath315 using eq.([ampl5 ] ) and the definitions of eq.([phis ] ) for @xmath314 one obtains @xmath316 where @xmath317 and @xmath318 . as follows from eqs.([cz2 ] ) and
( [ spm ] ) one predicts significantly different magnitudes for @xmath314 for @xmath5 and @xmath6 breakup cases . for the @xmath5 breakup , @xmath319 due to the smallness of the nuclear wave function component containing two protons in the same helicity state .
as a result one expects @xmath320 while for the @xmath6 break up case @xmath321 then one obtains @xmath322 where in the last part of the equation we assumed that @xmath323 .
the hard rescattering mechanism of a two - nucleon breakup from the @xmath0he nucleus at large c.m .
angles is based on the assumption of the dominance of quark - gluon degrees of freedom in the hard scattering process involving two nucleons .
the model explicitly assumes that the photodisintegration process proceeds through the knock - out of a quark from one nucleon with a subsequent rescattering of that quark with a quark from the second nucleon .
while photon - quark scattering is calculated explicitly , the sum of all possible quark rescatterings is related to the hard elastic @xmath1 scattering amplitude .
such a relation is found assuming that quark - interchange amplitudes give the dominant contribution in the hard elastic @xmath1 scattering .
the model allows one to calculate the cross sections of the hard breakup of @xmath6 and @xmath5 pairs from @xmath0he expressing them through the amplitudes of elastic @xmath6 and @xmath5 scatterings , respectively .
several results of the hrm are worth mentioning : first , the hrm predicts an approximate @xmath2 scaling consistent with the predictions of the quark - counting rule .
however , the model by itself is nonperturbative because the bulk of the incalculable part of the scattering amplitude is hidden in the amplitude of the @xmath1 scattering that is taken from the experiment .
second , because the hard @xmath1 scattering amplitude enters into the final amplitude of the photodisintegration reaction , the shape of the energy dependence of the @xmath3 weighted breakup cross section reflects the shape of the @xmath4 weighted @xmath1 elastic scattering cross section . because of a better accuracy of @xmath5 elastic scattering data for @xmath277 gev@xmath58 , we are able to predict a specific shape for the energy dependence of the hard @xmath5 breakup cross section at photon energies up to @xmath324 gev .
another observations is that , when @xmath2 scaling is established , the hrm predicts an increase of the strength of the @xmath5 breakup cross section relative to the @xmath6 breakup as compared to the low energy results .
this is due to the feature that within the quark - interchange mechanism of @xmath1 scattering one has more charges flowing between nucleons in the @xmath5 pair than in the @xmath6 pair .
this situation is opposite in the low energy regime when no charged meson exchanges exist for the @xmath5 pair .
even though the large charge factor is involved in the @xmath5 breakup its cross section is still by a factor of ten smaller than the cross section of the @xmath6 breakup . within the hrm ,
this is due to cancellation between the helicity conserving amplitudes @xmath225 and @xmath226 , which have opposite signs for the @xmath5 scattering .
because of the smallness of the @xmath5 breakup cross section , within the eikonal approximation , we estimated the possible contribution of three - body / two - step processes in which the initial two - body hard @xmath6 breakup is followed by the charge - exchange rescattering of an energetic neutron off the spectator proton .
we found that this contribution has @xmath250 energy dependence and is a small correction for spectator nucleon momenta @xmath325150 mev/@xmath30 .
however , the three - body / two - step process will dominate the hard @xmath5 breakup contribution at large transverse momenta of the spectator nucleon starting at @xmath326 mev/@xmath30 .
the next result of the hrm is the prediction of different spectator - momentum dependencies of breakup cross section for the @xmath5 and @xmath6 pairs .
this result follows from the fact that the ground state wave function of @xmath0he containing two protons with the same helicity is significantly suppressed as compared to the same component in the @xmath6 pair .
because of this , the @xmath5 spectral function is sensitive to the higher angular momentum components of the nuclear ground state wave function .
these components generate wider momentum distribution as compared to say the @xmath213 component of the wave function . as a result
the cross section of the @xmath5 breakup reaction exhibits wider momentum distribution as compared to the @xmath6 cross section .
additionally because of the strong @xmath256 dependence of the reaction , the cross section exhibits an asymmetry in the light - cone momentum distribution of the spectator nucleon , favoring larger values of @xmath285 .
the final result of the hrm is the strong difference in prediction of the polarization transfer asymmetry for @xmath5 and @xmath6 breakup reactions for circularly polarized photons . because of the suppression of the same helicity @xmath5 components in the @xmath0he ground state wave function
, the dominant helicity conserving @xmath278 component will not contribute to the polarization transfer process involving two protons .
because of this effect , the hrm predicts longitudinal polarization transfer @xmath327 , for the @xmath5 breakup to be close to zero .
because no such suppression exists for the @xmath6 breakup , the hrm predicts a rather large magnitude for @xmath328 . even though the hrm model does not contain free parameters , for numerical estimates we use the magnitude of elastic @xmath1 cross sections as well as some properties of the @xmath1 helicity amplitudes
this introduces certain error in our prediction of the magnitudes of the breakup cross sections . for the @xmath6 breakup
this error is mainly related to the uncertainty in the magnitude of the absolute cross section of hard elastic @xmath6 scattering which is on the level 30% . for the @xmath5 breakup
the main source of the uncertainty is the magnitude of the cancellation between @xmath225 and @xmath226 , which is sensitive to the angular distribution of helicity amplitudes .
the uncertainty due to the angular function is on the level of 40 % .
these uncertainties should be considered on top of the theoretical uncertainties that the hrm contains due to approximations such as estimating the scattering amplitude at maximal value of the nuclear wave function at @xmath125 .
the latter may introduce an uncertainty as much as @xmath177 in the breakup cross section .
in conclusion we expect that experimental verification of all above - mentioned predictions may help us to verify the validity of the hard rescattering model
. also , the progress in extracting the helicity amplitudes of the hard @xmath1 scattering will allow us to improve the accuracy of the hrm predictions .
we are grateful to drs .
ronald gilman , leonid frankfurt , eli piasetzky and mark strikman for many useful discussions and comments .
special thanks to drs .
ronald gilman , eli piasetzky and ishay pomerantz for the discussion of their experimental results that allowed us to find the cancellation factor in the @xmath5-breakup cross section , which was overlooked in our calculations earlier .
this work is supported by u.s .
department of energy grant under contract de - fg02 - 01er41172 .
applying feynman diagram rules for the scattering amplitude corresponding to the diagram of fig.[fig.1](a ) one obtains @xmath330\over ( p_{1f}-k_1)^2-m_q^2+i\epsilon}i s(k_1 ) \cdots [ -igt^f_c\gamma_\mu]\cdots { i[{/ \hskip-5pt } p_{1i } - { / \hskip-5pt } k_1 + m_q](-i)\gamma_{n_{1i } } \over ( p_{1i}-k_{1})^2-m^2+i\epsilon } { d^4k_{1}\over ( 2\pi)^4 } \nonumber \\ & & ( \gamma q ) : { i[{/ \hskip-5pt } p_{1i}-{/ \hskip-5pt } k_1 + q + m_q]\over ( p_{1i}-k_1+q)^2-m_q^2+i\epsilon } [ -iq_ie\epsilon^\perp\gamma^\perp ] \nonumber \\ & & ( n2 ) : \int { -i\gamma_{n_{2f}}i[{/ \hskip-5pt } p_{2f}-{/ \hskip-5pt } k_2+m_{q}]\over ( p_{2f}-k_2)^2-m_q^2+i\epsilon}is(k_2)\cdots [ -igt^f_c\gamma_\nu ] { i[{/ \hskip-5pt } p_{2i } - { / \hskip-5pt } k_2 + m_q](-i)\gamma_{n_{2i } } \over ( p_{2i}-k_{2})^2-m^2+\epsilon } { d^4k_{2}\over ( 2\pi)^4 } \nonumber \\ & & ( ^3he ) : \int { -i\gamma_{^3\textnormal{\scriptsize { he}}}\cdot\bar{u}_{\lambda_s}(p_s ) i[{/ \hskip-5pt } p_{nn } - { / \hskip-5pt } p_{2i } + m_n]\over ( p_{nn}-p_{2i})^2 - m_n^2+i\epsilon } { i[{/ \hskip-5pt } p_{2i}+m_n]\over p_{2i}^2-m_n^2+i\epsilon } { d^4p_{2i}\over ( 2\pi)^4}\nonumber \\ & & ( g ) : { i d^{\mu,\nu}\delta_{ab}\over [ ( p_{2i}-k_2)-(p_{1i}-k_1)-(q - l)]^2 + i\epsilon } , \nonumber \\ \label{amplitude}\end{aligned}\ ] ] where the momenta involved above are defined in fig.[fig.1 ] . note that the terms above are grouped according to their momenta . as such they do not represent the correct sequence of the scattering presented in fig.[fig.1 ] . to indicate this we separated the disconnected terms by `` @xmath331 '' . the covariant vertex function
, @xmath332 describes the transition of the @xmath0he nucleus to a three - nucleon system .
the vertex function @xmath333 describes a transition of a nucleon to one - quark and a residual spectator quark - gluon system with total momentum @xmath334 , @xmath335 .
the function @xmath336 describes the propagation of the off - mass shell quark - gluon spectator system of the nucleon .
as is shown below , this nonperturbative function can be included in the definition of a nonperturbative single quark wave function of the nucleon . using the reference frame and
the kinematic conditions described in sec.[ii ] we now elaborate each labeled term of eq.([amplitude ] ) separately . *
( @xmath0he)-term . * using the light - cone representation of four - momenta and introducing the light - cone momentum fraction of the @xmath1 pair carried by the nucleon @xmath337 as @xmath338 , one represents the nucleon propagators as well as the momentum integration @xmath339 in the following form : @xmath340 using these relations in eq.([amplitude ] ) we can integrate over @xmath341 taking the residue at the pole of the @xmath342-nucleon propagator , i.e. , @xmath343d p_{2i-}\over p_{2i-}-{m_n^2+p_{2i\perp}^2\over \alpha p_{nn+}}+i\epsilon } = -2\pi i[ ... ]\mid_{p_{2i-}={m_n^2+p_{2i\perp}^2\over\alpha p_{nn+}}}. \label{pole}\ ] ] after this integration one can use the following relations in eq.([amplitude ] ) : @xmath344 furthermore we use the condition @xmath345 to neglect the second term of the right - hand part of the third equation in eq.([afterpole ] ) .
this relation is justified for the high energy kinematics described in sec.[ii ] as well as from the fact that in the discussed model the scattering amplitude is defined at @xmath346 . introducing the light - cone wave function of @xmath0he @xcite @xmath347 and collecting all the terms of eq.([afterpole ] ) in the * ( @xmath0he : ) * part of eq.([amplitude ] ) one obtains @xmath348 * ( n1 : ) . * to evaluate this term in eq.([amplitude ] )
we first introduce @xmath349 where @xmath350 .
furthermore we perform the @xmath351 integration such that it puts the spectator system of the @xmath352 nucleon at its on - mass shell .
this will results in @xmath353 where @xmath354 represents the nucleon s spectator wave function with mass @xmath355 , and spin @xmath256 .
note that in the definition of @xmath356 one assumes an integration over all the internal momenta of the spectator system . using eq.([k1onshell ] ) for the ( n1 ) term one
obtains @xmath357\over ( p_{1f}-k_1)^2-m_q^2+i\epsilon}\psi_s(k_1)\cdots \nonumber \\ & & \cdots [ -igt^f_c\gamma_\mu]\psi_s^\dagger(k_1 ) { i[{/ \hskip-5pt } p_{1i } - { / \hskip-5pt } k_1 + m_q](-i)\gamma_{n_{1i } } \over ( p_{1i}-k_{1})^2-m^2+i\epsilon } \times { dx_1\over x_1 } { d^2k_{1\perp}\over 2(2\pi)^3}. \label{n1}\end{aligned}\ ] ] now we evaluate the propagator of the off - shell quark with the momentum , @xmath358 .
this yields : @xmath359 where the effective off - shell mass of the nucleon is defined as @xmath360 as it follows from eq.([q1on ] ) at the high energy limit , @xmath361 , one can neglect the second term of the rhs ( off - shell ) part of the equation if @xmath362@xmath363 @xmath364 . as is shown in sec.[ii ]
[ see discussion before eq.([ampl2 ] ) ] , the essential values that contribute in the scattering amplitude correspond to @xmath346 and @xmath365 .
therefore the second term in the right - hand side part of eq.([q1on ] ) can be neglected . using the closure relation for the on - shell spinors for eq.([q1on ] ) one obtains @xmath366 similar considerations yield the following expression for the propagator of the quark entering the wave function of the final nucleon `` @xmath302 '' : @xmath367 where @xmath368 is defined in eq.([xes ] ) . by inserting eqs.([q1on2 ] ) and ( [ q1onfin ] ) into eq.([n1 ] ) and defining quark wave function of the nucleon as @xmath369 for the @xmath370 term we obtain @xmath371u_{\eta_{1i}}(p_{1i}-k_1 ) { \psi^{\lambda_{1i},\eta_{1i } } ( p_{1i},x_1,k_{1\perp } ) \over ( 1-x_1 ) } { dx_1\over x_1}{d^2 k_{1\perp}\over 2 ( 2\pi)^3}. \label{n1fin}\end{aligned}\ ] ] * ( n2 : ) .
* this term can be evaluated following similar considerations used above in the evaluation of the ( n1 : ) term . introducing light - cone momentum fraction of the spectator system of the second nucleon as @xmath372 for the ( n2 : ) term we obtain @xmath373 collecting the expressions of eqs.([^3he : ] ) , ( [ n1fin ] ) and ( [ n2fin ] ) in eq.([amplitude ] ) and rearranging terms to express the sequence of the scattering , we obtain the expression of the scattering amplitude presented in eq.([ampl0 ] ) .
in this section we consider a hard @xmath1 elastic scattering model in which two nucleons interact through the(qim ) . the typical diagram for such scattering
is presented in fig .
[ fig.b1 ] . applying feynman diagram rules for these diagrams
one obtains @xmath374\over ( p_{1f}-k_1)^2-m_q^2+i\epsilon}i s(k_1 ) \cdots [ -igt^f_c\gamma_\mu ] { i[{/ \hskip-5pt } p^{*}_{1i } - { / \hskip-5pt } k_1 + m_q](-i)\gamma_{n_{1i } } \over ( p^{*}_{1i}-k_{1})^2-m^2+i\epsilon } { d^4k_{1}\over ( 2\pi)^4 } \nonumber \\ & & ( n2 ) : \int { -i\gamma_{n_{2f}}i[{/ \hskip-5pt } p_{2f}-{/ \hskip-5pt } k_2+m_{q}]\over ( p_{2f}-k_2)^2-m_q^2+i\epsilon}s(k_2 ) \cdots [ -igt^f_c\gamma_\nu ] { i[{/ \hskip-5pt } p_{2i } - { / \hskip-5pt } k_2 + m_q](-i)\gamma_{n_{2i } } \over ( p_{2i}-k_{2})^2-m^2+\epsilon } { d^4k_{2}\over ( 2\pi)^4 } \nonumber \\ & & ( g ) : { i d^{\mu,\nu}\delta_{ab}\over r^2 + i\epsilon}\nonumber \\ & & \hspace{2 in } - ( p_{1f}\leftrightarrow p_{2f } ) , \label{nnamp}\end{aligned}\ ] ] where definitions of the momenta are given in fig.([fig.b1 ] ) . the procedure of reducing the above amplitude is similar to the one used in the previous section .
first we estimate the propagators of each nucleon s spectator system at their pole values , @xmath375 ( @xmath376 ) by performing the @xmath377 integration , which yields @xmath378s(k_i ) d k_{i- } = -{2\pi i [ ... ] \over x_ip_{+ } } \sum_{s } \psi_s(k_i)\psi^\dagger_s(k_i ) \mid_{k_{i,- } = { m_s^2+k_{i,\perp}^2\over x_i p_{+}}}. \label{kpolint}\end{aligned}\ ] ] furthermore , because @xmath379 one can apply similar to eqs.([q1on2 ] ) and ( [ q1onfin ] ) , approximations for propagators of interchanging quarks leaving and entering the corresponding nucleons . then using the definition of single quark wave function according to eq.([nwf ] ) for the @xmath380 and @xmath381 terms ,
one obtains similar expressions that can be presented in the following form : @xmath382u_{\eta_{1i}}(p^*_{1i}-k_1 ) { \psi^{\lambda_{1i},\eta_{i1 } } ( p_{1i},x_1,k_{1\perp } ) \over ( 1-x_1 ) } { dx_1\over x_1}{d^2 k_{1\perp}\over 2 ( 2\pi)^3}. \label{nn_n1}\end{aligned}\ ] ] the @xmath383 term is obtained from the above equation by replacing @xmath384 .
regrouping ( n1 ) and ( n2 ) terms given by eq.([nn_n1 ] ) into eq.([nnamp ] ) , for the amplitude of nucleon - nucleon scattering in qim we obtain @xmath385\right .
u_{\eta_{1i}}(p^{*}_{1i}-k_1 ) { \psi_n^{\lambda_{1i},\eta_{1i}}(p^{*}_{1i},x_1,k_{1\perp})\over ( 1-x_1 ) } \right\ } \times \nonumber \\ & & \left\ { { \psi^{\dagger\lambda_{1f},\eta_{1f}}_n(p_{1f},x'_1,k_{1\perp})\over 1-x'_1 } \bar u_{\eta_{1f}}(p_{1f}-k_1)[-igt^f_c\gamma^\mu ] u_{\eta_{2i}}(p_{2i}-k_2 ) { \psi_n^{\lambda_{2i},\eta_{2i}}(p_{2i},x_2,k_{2\perp})\over ( 1-x_2 ) } \right\ } \nonumber \\ & & \left .
g^{\mu,\nu}(r ) { dx_1\over x_1}{d^2k_{1\perp}\over 2 ( 2\pi)^3 } { dx_2\over x_2}{d^2k_{2\perp}\over 2 ( 2\pi)^3 } \right ] .
\label{nn_qim}\end{aligned}\ ] ] note that in the above expression we redefined the initial momentum of `` @xmath352 '' nucleon to @xmath386 to emphasize its difference from @xmath64 which enters in the photodisintegration amplitude . in the latter case @xmath64 is not independent and it is defined by the momenta of the two remaining nucleons in the @xmath0he nucleus .
we consider now the hard elastic @xmath1 scattering ( described through @xmath387 reaction ) within quark - interchange approximation following the approach used in ref.@xcite . in this case one can represent the scattering amplitude as follows : @xmath388 where ( @xmath389 ) , ( @xmath390 ) and ( @xmath391 ) describe the spin - flavor quark states before and after the hard scattering , @xmath152 , and @xmath392 describes the probability amplitude of finding an @xmath393 helicity - flavor combination of three valence quarks in the nucleon @xmath150 .
the @xmath394 terms can be calculated by representing the nucleon wave functions on the helicity - flavor basis . for su(6 ) symmetric wave function such representation reads @xmath395 where we separated the wave function into two parts with two quarks ( e.g. 2nd and 3rd ) being in helicity zero - isosinglet and helicity one - isotriplet states . here
@xmath396 is the helicity wave function with total spin @xmath150 and helicity @xmath397 and @xmath398 is the isospin wave function with total isospin @xmath399 and three - dimensional component @xmath400 .
also , @xmath401 and @xmath402 are the isospin projection and helicity of the nucleon , respectively .
assuming helicity conservation in the hard kernel of quark interactions , @xmath403 one obtains a rather simple relation for @xmath1 scattering amplitude in the form@xcite @xmath404 with @xmath405 where one sums over all the possible values of @xmath235 and @xmath26 . using the above equations for the helicity amplitudes defined in eq.([phis ] ) and normalized according to eq.([crs_nn ] ) @xcite for @xmath406
one obtains @xmath407 and for @xmath408 one obtains @xmath409 with @xmath410 due to helicity conservation .
the above - described formalism gives an explicit form for calculation of the charge weighted @xmath1 amplitudes described in sec.[sec.2b ] .
for example eq.([qnnqim ] ) can be calculated by modifying eq.([ampl2su ] ) in the following form : @xmath411 where @xmath412 is the charge of the quark @xmath413 in @xmath414 units .
replacing qim amplitudes in eq.([ampl5 ] ) by @xmath1 helicity amplitudes of eq.([phis ] ) and using the antisymmetry of the ground state wave function with respect to the exchange of quantum numbers of any two nucleons one obtains the following expressions for the helicity amplitudes of two nucleon breakup reactions off the @xmath0he nucleus , @xmath415 : + for a positive helicity photon , @xmath416m_n { d^{2}p_{\perp}\over ( 2\pi)^2}\nonumber \\ \langle+,-,\lambda_s\mid a \mid + , \lambda_a\rangle & = & b \int \left[(q_f^{n_1}\phi_3 + q_f^{n_2}\phi_4 ) \psi_{^3\textnormal{\scriptsize { he}}}^{\lambda_a}(+,-,\lambda_s ) - q_f\phi_5\psi_{^3\textnormal{\scriptsize { he}}}^{\lambda_a}(+,+,\lambda_s)\right]\nonumber \\ & & \hspace{3.6 in } \times m_n { d^{2}p_{\perp}\over ( 2\pi)^2}\nonumber \\ \langle-,+,\lambda_s\mid a \mid + , \lambda_a\rangle & = & b \int \left[-(q_f^{n_1}\phi_4 + q_f^{n_2}\phi_3 ) \psi_{^3\textnormal{\scriptsize { he}}}^{\lambda_a}(+,-,\lambda_s ) + q_f\phi_5\psi_{^3\textnormal{\scriptsize { he}}}^{\lambda_a}(+,+,\lambda_s)\right]\nonumber \\ & & \hspace{3.6 in } \times m_n { d^{2}p_{\perp}\over ( 2\pi)^2}\nonumber \\ \langle-,-,\lambda_s\mid a \mid + , \lambda_a\rangle & = & b \int\left[q_f\phi_5 \psi_{^3\textnormal{\scriptsize { he}}}^{\lambda_a}(+,-,\lambda_s ) + q_f\phi_2\psi_{^3\textnormal{\scriptsize { he}}}^{\lambda_a}(+,+,\lambda_s)\right]m_n { d^{2}p_{\perp}\over ( 2\pi)^2}\nonumber \\ \ ] ] and for a negative helicity photon , @xmath417m_n { d^{2}p_{\perp}\over ( 2\pi)^2}\nonumber \\ \langle+,-,\lambda_s\mid a \mid - , \lambda_a\rangle & = & -b \int \left[-(q_f^{n_1}\phi_4 + q_f^{n_2}\phi_3 ) \psi_{^3\textnormal{\scriptsize { he}}}^{\lambda_a}(-,+,\lambda_s ) - q_f\phi_5\psi_{^3\textnormal{\scriptsize { he}}}^{\lambda_a}(-,-,\lambda_s)\right ] \nonumber \\ & & \hspace{3.8 in } \times m_n { d^{2}p_{\perp}\over ( 2\pi)^2}\nonumber \\ \langle-,+,\lambda_s\mid a \mid - , \lambda_a\rangle & = & -b \int \left[(q_f^{n_1}\phi_3 + q_f^{n_2}\phi_4 ) \psi_{^3\textnormal{\scriptsize { he}}}^{\lambda_a}(-,+,\lambda_s ) + q_f\phi_5\psi_{^3\textnormal{\scriptsize { he}}}^{\lambda_a}(-,-,\lambda_s)\right]\nonumber \\ & & \hspace{3.8 in } \times m_n { d^{2}p_{\perp}\over ( 2\pi)^2}\nonumber \\ \langle-,-,\lambda_s\mid a \mid - , \lambda_a\rangle & = & -b \int\left[-q_f\phi_5 \psi_{^3\textnormal{\scriptsize { he}}}^{\lambda_a}(-,+,\lambda_s ) + q_f\phi_1\psi_{^3\textnormal{\scriptsize { he}}}^{\lambda_a}(-,-,\lambda_s)\right]m_n { d^{2}p_{\perp}\over ( 2\pi)^2},\nonumber \\\end{aligned}\ ] ] where @xmath418 . because the scattering process is considered in the `` @xmath26-@xmath1 '' center of mass reference frame where @xmath35 direction is chosen opposite to the momentum of incoming photon , the bound nucleon helicity states correspond to the nucleon spin projections @xmath118 for positive and @xmath419 for negative helicities . 08 s.j . brodsky and g.r .
farrar , phys .
. lett . * 31 * , 1153 ( 1973 ) ; phys .
rev . * d11 * , 1309 ( 1975 ) v. matveev , r.m .
muradyan and a.n .
tavkhelidze , lett .
nuovo cimento * 7 * , 719 ( 1973 ) .
s.j . brodsky and b.t .
chertok , phys .
lett . * 37 * , 269 ( 1976 ) .
r. j. holt , phys .
c * 41 * , 2400 ( 1990 ) .
j. napolitano _ et al .
* 61 * , 2530 ( 1988 ) ; s.j .
freedman _ et al .
_ , phys . rev . *
c48 * , 1864 ( 1993 ) .
belz _ et al .
74 * , 646 ( 1995 ) . c. bochna
_ et al . _ ,
lett . * 81 * , 4576 ( 1998 ) .
schulte _ et al .
* 87 * , 102302 ( 2001 ) .
k. wijesooriya _ et al . _
lab hall a collaboration .
] , phys .
lett . * 86 * , 2975 ( 2001 ) . n. isgur and c.h .
llewellyn smith , phys .
lett . * 52 * , ( 1984 ) 1080 .
a. radyushkin , acta phys .
b15 * , 403 ( 1984 ) . g. r. farrar , k. huleihel and h. y. zhang , phys .
lett . * 74 * , 650 ( 1995 ) .
t. c. brooks and l. j. dixon , phys . rev .
d * 62 * , 114021 ( 2000 ) .
s. j. brodsky , aip conf .
* 792 * , 279 ( 2005 ) .
s.j . brodsky and j.r .
hiller , phys .
c * 28 * , 475 ( 1983 ) .
s.j . brodsky and j.r .
hiller , phys .
c * 30 * , 412 ( 1984 ) .
v. y. grishina , l. a. kondratyuk , w. cassing , a. b. kaidalov , e. de sanctis and p. rossi , eur .
j. a * 10 * , 355 ( 2001 ) n. a. khokhlov , v. a. knyr and v. g. neudatchin , phys .
c * 75 * , 064001 ( 2007 ) l. l. frankfurt , g. a. miller , m. m. sargsian and m. i. strikman , phys .
lett . * 84 * , 3045 ( 2000 ) [ arxiv : hep - ph/9904222 ]
. m. m. sargsian , in proceedings of the workshop on `` exclusive processes at high momentum transfer , '' newport news , virginia , may 15 - 18 , 2002 . published in
* newport news 2002 , exclusive processes at high momentum transfer , edited anatoly radyushkin and paul stoler , pp.369 - 376 .
arxiv : nucl - th/0208027 .
r. gilman and e. piasetzky ( spokespersons ) , `` hard photodisintegration of a proton pair , '' jefferson lab proposal , e-03 - 101 , 2003 .
v. y. grishina , l. a. kondratyuk , w. cassing , e. de sanctis , m. mirazita , f. ronchetti and p. rossi , eur .
j. a * 19 * , 117 ( 2004 ) . m. m. sargsian , phys .
b * 587 * , 41 ( 2004 ) [ arxiv : nucl - th/0309054 ] .
g. r. farrar , s. gottlieb , d. w. sivers and g. h. thomas , phys .
d * 20 * , 202 ( 1979 ) .
s. j. brodsky , c. e. carlson and h. j. lipkin , phys .
d * 20 * , 2278 ( 1979 ) .
frankfurt and m.i .
strikman , phys . rep . * 76 * , 214 ( 1981 ) ; l.l .
frankfurt and m.i .
strikman , phys .
rep.*160 * , 235 ( 1988 ) .
l. w. mo and y. s. tsai , rev .
phys . * 41 * , 205 ( 1969 ) .
a. nogga a. kievsky , h. kamada , w. gloeckle , l. e. marcucci , s. rosati and m. viviani , phys .
c * 67 * , 034004 ( 2003 ) .
m. sargsian and m. strikman , phys .
b * 639 * , 223 ( 2006 ) c. g. white _ et al .
_ , phys .
d * 49 * , 58 ( 1994 ) .
m. jacob and g.c .
wick , annals phys .
* 7 * , 404 ( 1959 ) [ annals phys .
* 281 * , 774 ( 2000 ) ] .
ramsey and d. w. sivers , phys .
d * 45 * , 79 ( 1992 ) .
laget , nucl .
a * 497 * , 391c ( 1989 ) . j. m. laget , phys .
b * 151 * , 325 ( 1985 ) .
n. dhose _ et al .
_ , phys .
* 63 * , 856 ( 1989 ) . l. machenil , m. vanderhaeghen , j. ryckebusch and m. waroquier , phys .
b * 316 * , 17 ( 1993 ) j. ryckebusch , m. vanderhaeghen , l. machenil and m. waroquier , nucl .
a * 568 * , 828 ( 1994 ) [ arxiv : nucl - th/9307004 ] . c. giusti and f. d. pacati , nucl .
a * 535 * , 573 ( 1991 ) .
w. r. gibbs and b. loiseau , phys .
c * 50 * , 2742 ( 1994 ) . c. granados ,
( to be published ) .
hepdata , the durham hep database , http://durpdg.dur.ac.uk/hepdata , 2008 .
l. l. frankfurt , m. m. sargsian and m. i. strikman , phys .
c * 56 * , 1124 ( 1997 ) m. m. sargsian , int .
j. mod .
e * 10 * , 405 ( 2001)[arxiv : nucl - th/0110053 ] .
r. j. glauber , phys .
* 100 * ( 1955 ) 242 ; in . lectures in theoretical physics , edited by w. brittain and l. g. dunham ( interscience publ .
, n.y . 1959 ) . | we investigate a large angle photodisintegration of two nucleons from the @xmath0he nucleus within the framework of the hard rescattering model ( hrm ) . in the hrm a quark of one nucleon knocked out by an incoming photon rescatters with a quark of the other nucleon leading to the production of two nucleons with large relative momentum . assuming the dominance of the quark - interchange mechanism in a hard nucleon - nucleon scattering , the hrm allows the expression of the amplitude of a two - nucleon break - up reaction through the convolution of photon - quark scattering , @xmath1 hard scattering amplitude and nuclear spectral function which can be calculated using a nonrelativistic @xmath0he wave function .
the photon - quark scattering amplitude can be explicitly calculated in the high energy regime , whereas for @xmath1 scattering one uses the fit of the available experimental data .
the hrm predicts several specific features for the hard breakup reaction .
first , the cross section will approximately scale as @xmath2 .
secondly , the @xmath3 weighted cross section will have the shape of energy dependence similar to that of @xmath4 weighted @xmath1 elastic scattering cross section .
also one predicts an enhancement of the @xmath5 breakup relative to the @xmath6 breakup cross section as compared to the results from low energy kinematics .
another result is the prediction of different spectator momentum dependencies of @xmath5 and @xmath6 breakup cross sections .
this is due to the fact that same - helicity @xmath5-component is strongly suppressed in the ground state wave function of @xmath0he . because of this suppression the hrm predicts significantly different asymmetries for the cross section of polarization transfer @xmath1 breakup reactions for circularly polarized photons . for the @xmath5 breakup
this asymmetry is predicted to be zero while for the @xmath6 it is close to @xmath7 . |
the bilayer heisenberg antiferromagnet , @xmath4 where the intralayer couplings run over nearest neighbors on square lattices and the @xmath5 are @xmath3 quantum spins , has turned out to be an excellent testing ground for notions of quantum criticality .
a variety of controlled calculations , such as strong - coupling expansions at zero temperature ( starting with the work of hida@xcite and extended by two of the present authors@xcite ) , finite - temperature quantum monte carlo calculations@xcite and high - temperature expansions@xcite fit extremely well into the general field - theoretic framework for zero - temperature order - disorder critical points in unfrustrated two - dimensional quantum antiferromagnets.@xcite the critical point in the @xmath3 bilayer heisenberg model is between a nel ordered phase when @xmath6 is sufficiently large ( greater than the critical value @xmath7 ) and a magnetically disordered `` dimer '' phase which is the adiabatic continuation of the @xmath8 ground state ( which is a product of interlayer singlet pair wavefunctions ) .
the universality class of the phase transition is the same as for finite - temperature transitions in @xmath9 , @xmath10 symmetric models .
it seemed most natural to us that for larger values of the quantum spin @xmath0 the same scenario would play out : the value of @xmath11 would change , but qualitatively nothing would be different .
it therefore came as a surprise when ng , zhang , and ma@xcite described the results of schwinger boson mean - field ( sbmf ) calculations , which indicate that for @xmath12 the transition between magnetically ordered and disordered phases is _ first - order_. the calculation leads to @xmath13 , and hence indicates that for all physical values of @xmath0 the transition should be first - order ; but it is reasonable to accept the suggestion of ng _ et al . _
that sbmf theory leads to phase diagram of @xmath0 versus @xmath14 which is qualitatively valid even if @xmath15 is underestimated .
two more salient points regarding the sbmf results should be noted . at large @xmath0 , the value of @xmath14 at the dimer - nel phase boundary scales as @xmath16 ; and at sufficiently large @xmath0 the ground state on the dimer side of the boundary is the pure interlayer singlet pair state ( _ i.e. , _ identical to the ground state at @xmath8 ) . here
we will argue that the results of sbmf theory for phase transitions in the bilayer heisenberg model are qualitatively incorrect , even in the limit of large @xmath0 where one might expect such an approximation to be reliable .
ng _ et al .
_ offer an elementary argument in support of their proposed scenario , which it is informative to reconsider .
they note that the ideal nel state and the interlayer singlet state become degenerate for @xmath17 ( with the interlayer singlet state energetically favored for @xmath18 ) , while linear spin - wave theory indicates that the magnetization should vanish at @xmath19 ( with nel order existing for @xmath20 ) . at large @xmath0
, it is clear that @xmath21 .
_ suggest that , on decreasing @xmath14 from infinity , the transition that one might suspect should take place at @xmath22 ( presumably a continuous transition from the nel to dimer phase ) is pre - empted by a first - order transition at @xmath23 .
this is certainly consistent with their sbmf calculations .
however , we suggest an alternative interpretation .
we believe that the argument in favor of a first - order transition at @xmath23 is too simplistic : using simple trial states to represent the ground states in the respective phases is dangerous , particularly since it is a comparison of subleading ( in powers of @xmath0 ) terms in the variational energies for each phase which leads to the estimate for @xmath23 .
furthermore , there is an elementary calculation which strongly suggests that the dimer phase is unstable with respect to the development of nel order for @xmath24 .
consider a single dimer , subject to a staggered magnetic field which is produced self - consistently by the staggered magnetization of the system .
let @xmath25 denote the staggered susceptibility of a single dimer .
then , within this `` dimer mean - field approximation '' one has @xmath26 .
one can calculate @xmath27 directly from second - order perturbation theory , based on explicit formulae for 3-@xmath28 symbols,@xcite to find that @xmath29 . as promised , the domain of stability of the dimer phase is @xmath30 , in agreement with the spin - wave estimate for the stability of the nel phase .
it should be clear that the dimer mean - field approximation underestimates the stability of the dimer phase , just as the standard weiss theory overestimates curie temperatures .
one could reasonably object to our argument , then , by suggesting that the corrections to this mean - field theory could be large : in particular , @xmath31 might be pushed up to a value of order @xmath16 . in the following section ,
we describe high - order perturbation expansions about @xmath8 for bilayers with @xmath32 .
combined with the existing results for @xmath3 , they strongly suggest that the corrections to dimer mean - field theory are order unity rather than order @xmath0 .
in addition , for @xmath2 bilayers we present a wide variety of results based on expansions about @xmath8 and about ising models .
they are all qualitatively similar to what is found for @xmath3 bilayers , providing additional confirmation that the phase diagram is unchanged with increasing @xmath0 ( in particular , there is no evidence for any phase besides dimer or nel ) and offering accurate results which could prove useful in interpreting data from experimental realizations of @xmath2 bilayers .
perturbation expansions@xcite for the ground - state energy @xmath33 , the triplet excitation spectrum , and the antiferromagnetic susceptibility @xmath34 have been carried out for bilayers with all integer and half - integer values of @xmath0 from 1 to 4 .
the series coefficients for integer @xmath0 bilayers for the minimum gap @xmath35 , corresponding to wave vector @xmath36 , and @xmath34 are presented in table [ table : gaps ] . for @xmath3 bilayers such expansions
have already been presented by zheng@xcite to @xmath37 .
[ gapapprox ] is a set of `` scatter plots '' of estimated critical @xmath14 values versus estimated critical exponents derived from unbiased @xmath38log - pad approximants to the gap series for @xmath2 , 2 , 3 and 4 bilayers .
an analogous plot for @xmath3 is presented in ref . .
in every case they have very similar character , with nearly all exponent estimates lying slightly above the anticipated value of 0.71 , and with clear correlations between critical point and exponent values .
( scatter plots for the antiferromagnetic susceptibility have very similar character ; likewise for the gap and susceptibility of half - integer @xmath0 bilayers . )
as @xmath0 increases the series actually become better behaved , in that the approximants derived from a fixed number of terms are more tightly clustered ( neglecting the outliers ) and estimated exponents are closer to the expected value . biasing the exponent
one can obtain more precise estimates of the critical points @xmath11 , as listed in table [ table : xc ] .
the trend is clear : with increasing @xmath0 , the critical value of @xmath14 is approaching a constant multiple of the dimer mean - field value @xmath39 . if we plot @xmath40 as a function of @xmath41 ( shown in fig .
[ lambda_c ] ) , we can see that the results can be remarkably well fitted by a straight line @xmath42 with an intercept at @xmath43 slightly above @xmath44 .
the constants @xmath45 and @xmath46 determined by a linear least squares fit are : @xmath47 for general @xmath0 we have found empirically , based on the ground state energy @xmath33 , the minimum triplet gap @xmath35 , and the antiferromagnetic susceptibility @xmath34 series for the eight values of @xmath0 that have been explicitly calculated , that the coefficients of @xmath48 can be expressed as polynomials of order @xmath49 in the variable @xmath50 .
for example , to third order @xmath51\lambda^2 \label{delta}\\ & & + [ ( 56/135)r-(116/135)r^2-(3824/1215)r^3]\lambda^3\ .\nonumber\end{aligned}\ ] ] further coefficients are listed in the appendix .
one can then define @xmath52 and consider the @xmath53 , @xmath54 limit .
the resulting series in @xmath55 corresponds to the terms in the double - series ( in @xmath50 and @xmath14 ) of the form @xmath56 . its @xmath38log - pad approximants are as well behaved as the series for the larger values of @xmath0 shown above , with unbiased approximants clustering and exponent estimates just above the anticipated value .
biasing the critical point estimates leads to a critical @xmath55 of 0.2691(7 ) ( this is consistent with the value of @xmath45 in eq .
( [ r_c ] ) ) , so the ratio of the exact critical point value in the large-@xmath0 limit to that coming from the dimer mean - field theory is 1.435(4 ) .
we can also try to use the double - expansions in @xmath50 and @xmath14 to analytically continue to unphysically small values of @xmath0 .
an interesting feature of the general-@xmath0 double - series ( both for the @xmath35 and @xmath34 ) is that all terms of the form @xmath57 appear to have vanishing coefficients .
consequently , for @xmath58 there is no phase transition to a nel - ordered state .
presumably there is a critical spin @xmath15 , less than @xmath59 , at and below which the dimer state is stable for all values of @xmath14 . it has not been possible to obtain a reliable estimate of @xmath15 by constructing @xmath38log - pad approximants at fixed values of @xmath0 : for @xmath60 the approximants depend strongly on the number of terms used .
however , the extrapolation from larger @xmath0 , summarized in eqs .
( [ fit_lambdac ] ) and ( [ r_c ] ) , suggests that @xmath61 .
although we have presented evidence above that the dimer phase is unstable for @xmath14 larger than a critical value of @xmath30 , we did not firmly establish that the structure of the phase diagram is simply dimer phase critical point
nel phase .
the fact that the unbiased exponent values are consistent with a lorentz - invariant , @xmath9 , @xmath10 universality class is certainly suggestive . for the @xmath2 bilayers we have done much more .
in addition to expansions about @xmath8 we have constructed expansions about ising models , generating a set of results analogous to those presented by zheng@xcite for @xmath3 bilayers . taken together , these leave little room for doubt about the phase diagram . because the calculations follow those presented in ref .
so closely we refer the reader to sec .
iii of that paper for a description of the calculation ( note that the parameter @xmath63 in that paper corresponds to @xmath64 ) .
expansions were obtained to @xmath65 ( with @xmath66 denoting the ratio of transverse to longitudinal exchange strength ) for the sublattice magnetization @xmath67 , uniform transverse susceptibility @xmath68 , and triplet excitation spectrum . from these
we could derive the gap to the optical branch of the spin - wave spectrum at wave vectors @xmath69 ( @xmath70 , the minimum gap ) and @xmath71 ( @xmath72 ) , spin - wave velocity @xmath73 , and spin - wave stiffness @xmath74 .
the expansion coefficients will not be presented here but are available from the authors on request .
results of series extrapolations are shown in figs .
[ optical_gap ] , [ m_chi ] , and [ c_rho ] .
they demonstrate that expansions about the nel state lead to estimates of the domain of stability of the nel phase which are completely consistent with the critical point found by expanding about @xmath8 .
this is nicely confirmed by fig .
[ e0 ] , which displays the ground - state energy per site .
the values from the dimer expansion match very smoothly on to those from the ising expansion around the critical point , whereas for a first - order transition there would be a discontinuity in slope at the transition .
all of the results presented in this subsection represent the best available theoretical values for experimentally accessible properties of nel - ordered @xmath2 bilayers , which should prove useful in the interpretation of experimental data if any such systems are studied .
( in real compounds single - ion anisotropy is always present to some degree .
the present calculations could readily be extended to include such terms in the hamiltonian . ) finally , it is amusing to note that in the dimer phase for @xmath75 bilayers , sufficiently close to @xmath8 , there exist spin-2 elementary excitations .
let us consider @xmath2 bilayers specifically .
triplet spectra for various values of @xmath14 in the dimer phase are shown in fig .
[ mk_triplet ] while _ quintuplet _ spectra are shown in fig .
[ mk_quintuplet ] .
( these spectra are obtained by direct sums of the terms in the series to the maximum order available . )
the latter spectra lose physical significance when the quintuplet excitations can decay into multiple triplet excitations .
one might imagine this happens for arbitrarily small values of @xmath14 , since at @xmath8 the quintuplet gap is three times the triplet gap .
however , the parity , with respect to interchange of layers in the bilayer system , of the quintuplet excitations is opposite to that of the triplet excitations , and symmetry forbids decay of the quintuplet excitations into an odd number of triplet excitations .
hence stable spin-2 excitations lie between the 2-triplet and 4-triplet continua for sufficiently small @xmath14 .
the order - disorder transition in bilayer heisenberg antiferromagnets appears to be a problem for which schwinger boson mean - field theory leads to qualitatively incorrect results even in the limit of large @xmath0 .
we have shown by high order perturbation expansions about the limit of uncoupled interlayer singlets that there is strong evidence for continuous transitions between dimer and nel phases with critical values of @xmath76 scaling as @xmath77 .
there seems to be no evidence in favor of the scenario described by ng _
et al._,@xcite in which the transition would become first - order for sufficiently large @xmath0 . of course this does not imply that the phase diagram for bilayer heisenberg antiferromagnets is completely universal .
if further - neighbor interactions or higher - order exchange interactions ( for example , terms in the hamiltonian of the form @xmath78 ) were allowed then a wide variety of new phases could be stabilized .
however , for the simplest bilayer models it appears that simple arguments based on the instability of the dimer phase to nel order ( using dimer mean - field theory ) and the instability of the nel phase to a spin - disordered phase ( using linear spin - wave theory to determine where the staggered magnetization vanishes ) , which both suggest a critical point at @xmath79 , are correct .
the double expansions in @xmath14 and @xmath41 for the ground - state energy per site @xmath80 , triplet minimum energy gap @xmath35 , and antiferromagnetic susceptibility @xmath34 for spin-@xmath0 bilayers are presented in full here .
@xmath81\lambda^2\\ % & & + [ ( 56/135)r-(116/135)r^2-(3824/1215)r^3]\lambda^3+\cdots\ .\nonumber \chi & & = { 4r\over 3 } { \big [ } 1 + { { 16\,\lambda \,r}\over 3 } + { { { { \lambda } ^2}\,r\,\left ( -1 + 140\,r \right ) } \over 6 } \nonumber \\ & & + { { \lambda } ^3}\,r\,\left ( -{1\over 8 } - { { 143\,r}\over { 36 } } + { { 2677\,{r^2}}\over { 27 } } \right ) \nonumber \\ & & + { { \lambda } ^4}\,r\ , { \big ( } -0.10671296 - 1.45847442\,r \nonumber \\ & & - 33.2428954\,{r^2 } + 403.712863\,{r^3 } { \big ) } \nonumber \\ & & + { { \lambda } ^5}\,r\ , { \big ( } -0.090168701 - 0.58931360\,r \nonumber \\ & & - 3.49146301\,{r^2 } - 221.1724119\,{r^3 } + 1622.218477\,{r^4 } { \big ) } \nonumber \\ & & + \ ! { { \lambda } ^6}r { \big ( } \ ! -0.076292486 + 0.12455955 r + 2.92215451 { r^2 } \nonumber \\ & & + 8.15178055\,{r^3 } - 1231.683801\,{r^4 } + 6416.371180\,{r^5 } { \big ) } \nonumber \\ & & + o(\lambda^7 ) { \big ] } \end{aligned}\ ] ] k .- k .
ng , f. c. zhang , and m. ma , phys .
b * 53 * , 12196 ( 1996 ) .
m. rotenberg , r. bivins , n. metropolis , and john k. wooten , jr . , _ the 3-j and 6-j symbols _
( cambridge , massachusetts : the technology press , 1959 ) .
see especially pp . 4 , 12 . | spin-@xmath0 bilayer heisenberg models ( nearest - neighbor square lattice antiferromagnets in each layer , with antiferromagnetic interlayer couplings ) are treated using dimer mean - field theory for general @xmath0 and high - order expansions about the dimer limit for @xmath1 .
we suggest that the transition between the dimer phase at weak intraplane coupling and the nel phase at strong intraplane coupling is continuous for all @xmath0 , contrary to a recent suggestion based on schwinger boson mean - field theory .
we also present results for @xmath2 layers based on expansions about the ising limit : in every respect the @xmath2 bilayers appear to behave like @xmath3 bilayers , further supporting our picture for the nature of the order - disorder phase transition . |
we thank aveek bid , dwarkanath k.s .
, subroto mukerjee , robert myers , rajaram nityananda , barry sanders , rafael sorkin , radhika vatsan and gregor weihs for useful discussions .
we thank raymond laflamme and anthony leggett for reading through the draft and helpful comments and discussions . as
acknowledges partial support from a ramanujan fellowship , govt . of india .
caldwell , r. & kamionkowski , m. cosmology : dark matter and dark energy .
_ nature _ * 458 * , 587 ( 2009 ) .
kirshner , r.p .
throwing light on dark energy .
_ science _ * 300 * , 1914 ( 2003 ) .
amelino - camelia , g. gravity - wave interferometers as quantum - gravity detectors .
_ nature _ * 398 * , 216 ( 1999 ) .
visser , m. quantum gravity : progress at a price .
_ nature phys . _ * 5 * , 385 ( 2009 ) .
feynman , r. p. leighton , r. & sands , m. _ the feynman lectures on physics vol
. 3 _ ( addison - wesley , reading mass , 1963 ) .
cohen - tannoudji , c. diu , b. & laloe , f. _ quantum mechanics i _ , ( wiley - vch , 2nd edition , 2005 ) shankar , r. _ principles of quantum mechanics _ , ( springer , 2nd edition , 1994 ) .
yabuki , h. feynman path integrals in the young double - slit experiment .
. _ * 25 * , no.2 , 159 - 174 ( 1986 ) .
raedt , h.d .
michielsen , k. hess , k. analysis of multipath interference in three - slit experiments .
a. _ * 85 * , 012101 ( 2012 ) .
feynman , r. p. & hibbs , a. r. _ quantum mechanics and path integrals _ ( mcgraw - hill , new york , 3rd .
experimentally realizable .
aharonov , y. & bohm , d. significance of electromagnetic potentials in the quantum theory .
rev . _ * 115 * , 485 - 491 ( 1959 ) .
berry , m.v .
asymptotics of the many - whirls representation for aharnonov - bohm scattering .
a : math.theor._ * 43 * , 354002 ( 2010 ) .
bachtold , a. _ et al_. aharonov - bohm oscillations in carbon nanotubes .
_ nature _ * 397 * , 673 - 675 ( 1999 ) .
yacoby , a. schuster , r. & heiblum , m. phase rigidity and h/2e oscillations in a single - ring aharonov - bohm experiment .
b _ * 53 * , 9583 - 9586 ( 1996 ) .
sinha , u. couteau , c. jennewein , t. laflamme , r. weihs , g. ruling out multi - order interference in quantum mechanics .
_ science _ * 329 * , 418 - 421 ( 2010 ) .
richter , k. and seiber , m .
semiclassical theory of chaotic quantum transport .
_ phys.rev.lett . _
* 89 * , 206801 ( 2002 ) .
sorkin , r. d. quantum mechanics as quantum measure theory .
* 9 * , 3119 ( 1994 ) .
sllner , i. _ et al_. testing born s rule in quantum mechanics for three mutually exclusive events .
_ found phys_. * 42 * , 742 - 751 ( 2012 ) .
park , d. k. moussa , o. laflamme , r. three path interference using nuclear magnetic resonance : a test of the consistency of born s rule .
_ new j. phys_. * 14 * , 113025 ( 2012 ) .
niestegge , g. three - slit experiments and quantum nonlocality .
_ found phys_. * 43 * , 805 - 812 ( 2013 ) ududec , c. barnum , h. & emerson , j. three slit experiments and the structure of quantum theory .
_ found . phys_. * 41 * , 396 - 405 ( 2011 ) .
lapkiewicz , r. _ et al_. experimental non - classicality of an indivisible quantum system .
_ nature _ * 474 * , 490 - 493 ( 2011 ) .
kolenderski , p. _ et al_. aharon - vaidman quantum game with a young - type photonic qutrit .
a _ * 86 * , 012321 ( 2012 ) .
franson , j. d. pairs rule quantum interference .
_ science _ * 329 * , 396 - 397 ( 2010 ) .
hickmann , j .m .
fonseca , e. j. s & jesus - silva , a. j. born s rule and the interference of photons with orbital angular momentum by a triangular slit . _
europhysics lett . _
* 96 * , no.6 , 64006 ( 2011 ) .
zurek , w. h. environment - assisted invariance , entanglement , and probabilities in quantum physics .
rev . lett_. * 90 * , 120404 ( 2003 ) .
supporting material quantum field theory effects lead to a well known nonlinear correction to maxwell equations arising from the euler - heisenberg lagrangian @xcite .
the contribution to @xmath9 from this correction is found to be of the order @xmath74 where @xmath75 is the intensity measured in si units .
for instance , for a typical titanium - sapphire laser with power output 10 watts and beam diameter 20@xmath67 m , the correction to @xmath9 is around 10 orders of magnitude smaller than what is reported here .
bleinstein , n. handelsman , r. a. _ asymptotic expansions of integrals .
_ ( dover publications , incorporated , new york , 1975 ) .
bach , r. pope , d. liou , s. batelaan , h. controlled double - slit electron diffraction . _ new j. phys _ , * 15 * , 033018 ( 2013 ) . we have verified that in the absence of non - classical terms , the parameter @xmath9 evaluates to zero within numerical accuracy .
heisenberg w. & euler , h. folgerungen aus der diracschen theorie des positrons
_ z. phys .
_ * 98 * , 714 ( 1936 ) .
see supplemental material [ url ] , which includes refs . @xcite-@xcite .
landau , l. d. & lifshitz , e. m. _ classical theory of fields _ ( pergammon press , oxford , fourth revised english edition , 1975 ) . born , m. and wolf , e. _ principles of optics _ ( cambridge university press , seventh expanded edition , 1999 ) .
sunil kumar p. b. and ranganath g. s. , pramana-j.phys .
* 3 * 6 ( 1991 ) .
feynman , r.p .
space - time approach to non - relativistic quantum mechanics .
_ _ rev . of mod .
phys.__**20 * * , no.2 , 367 ( 1948 ) .
@#1@@@(s#1@italiccorr ) * supplementary material *
we suppose that we have a monochromatic source of light ( monoenergetic source of electrons ) and that the detectors integrate over the duration @xmath76 of the experiment . assuming that @xmath76 is much longer than any other time scale in the problem , like the travel time across the apparatus
, we can use a steady state description . both electrons and light
are then described ( in a scalar approximation ) by the helmholtz equation @xmath77 which is satisfied away from the sources and detectors .
@xmath78 is a scalar field representing the wave function of the electron or a component of the electromagnetic vector potential . for light @xmath79 and for electrons @xmath80 ( setting @xmath81 ) .
both electron and photon diffraction can be treated on the same footing in the time independent case .
below we will drop the superscript @xmath23 on physical quantities to avoid cluttering the formulae with it .
we will also suppose throughout that @xmath82 is much smaller than any other length scale in the problem , the sizes and separations of the slits and the distance to the source and the detector .
to describe free propagation , we use the kernel ( see section 59 of @xcite ) @xmath83 which a. satisfies the helmholtz equation@xcite away from @xmath84 and b. satisfies the fresnel - huygens principle : @xmath85 where @xmath86 is integrated over any plane between @xmath87 and @xmath88 @xcite and perpendicular to @xmath89 . by repeated application of equation s[fold ] , we can express the propagator for free space in the form @xmath90 \exp{[ik\int ds ] } , \label{kpath}\ ] ] where @xmath91 is the contour length along the path @xmath92 and the sum is over all paths connecting @xmath93 with @xmath94 . in the classical limit of @xmath95 ,
paths near the straight line path joining @xmath93 to @xmath94 contribute by stationary phase .
we refer to these as `` classical paths '' in the text .
all of these would contribute `` in phase '' .
paths away from the classical path are expected to contribute with rapidly oscillating phase . in describing diffraction by a system of slits
, we would have to sum over all paths connecting the source to the detector .
this would include paths of the kind shown in purple in fig .
2 , which are far from classical . we would expect ( in the limit of small @xmath82 ) that the contributions of such paths are negligible because of rapid oscillations of the phase .
we would like to know just how small " these contributions are .
in fresnel s theory of diffraction by a slit @xcite we use equation s[fold ] to insert a single intermediate state on the slit plane and find the amplitude @xmath96 to reach @xmath8 from @xmath97 : @xmath98 where @xmath99 is an intermediate point on the slit plane ( taken to be the @xmath100 plane ) .
the range of integration in equation s[ksd ] is over the _ two dimensional _
region @xmath101 in the slit plane , where @xmath101 is the union of the open slits .
this gives a remarkably accurate account of the phenomenon of diffraction .
in fresnel s theory we find because there is a single integration over @xmath101 , that the outcomes of the three possible experiments with two slits ( a , b and ab open ) are related by @xcite @xmath102 going beyond fresnel s theory , we insert two intermediate points on the slit plane and integrate twice@xcite over @xmath101 , the open parts of the slit plane . @xmath103 a typical path in this integration has two `` kinks '' ( at @xmath87 and @xmath104 ) and thus the integral is highly oscillatory . these integrals in @xmath105 can be computed numerically and seen to be much smaller than @xmath106 . in this order of approximation the amplitude for detection at @xmath8
is given by @xmath107 where @xmath108 is the leading order contribution to the amplitude due to non - classical paths @xmath30 .
as mentioned before @xmath105 is expected to be much smaller than @xmath106 .
most importantly , because of the two integrations over @xmath101 in equation s[ktwo ] , @xmath105 results in violations of equation [ k]sum .
sorkin suggested that the born rule for probabilities in quantum mechanics can be tested by performing a three slit experiment@xcite . by keeping each slit either open or closed
, we can perform seven distinct non trivial experiments .
theoretical predictions for the outcomes of these seven experiments are given by choosing @xmath101 to be one of the seven domains @xmath109 .
following sorkin , we define the quantity @xmath110 a straightforward calculation shows that after cancellations and to linear order in @xmath108 , @xmath111 this final expression shows clearly that it is the @xmath108 terms that violate @xmath112 and make @xmath11 non zero . in this equation @xmath96 would also receive contributions from non - classical paths , for example one which crosses the slit plane exactly once , but has a kink in it
. however , it is evident that these would contribute at a sub - leading order to a non - zero @xmath11 .
@xmath113 has been computed numerically and the resulting graphs are shown in fig.3 .
there are several subtleties associated with huygens principle , which do not however affect the order of magnitude we get for @xmath9 .
huygens initially gave a construction for evolving the wavefront using secondary wavelets .
it was fresnel @xcite who applied huygen s construction to understand diffraction effects . however , fresnel had to introduce _ ad hoc _ some `` inclination factors '' .
subsequently kirchoff derived fresnel s theory from maxwell s equations using an integral equation derived from helmholtz s equation ( see equation 17 on page 422 of @xcite ) .
he was also able to derive ( and correct ) fresnel s `` inclination factors '' . in equation s[ktwo ] these factors result in an additional factor of @xmath114 ( because of two right angle kinks ) and this leads to a factor of @xmath114 multiplying @xmath11 .
classically the path taken by a particle is a path of least action .
but quantum mechanics tells us that each physically possible path has a probability amplitude associated to it .
the final probability amplitude is summation of probability amplitudes from all paths @xcite . +
in solving the problem of scattering due to the presence of slits using feynman path integral ( equation s[kpath ] ) we simply excise all those paths which go through the solid portion of the slits . in fresnel
s theory we suppose that the amplitude at slit @xmath34 is the same as it would be in free space . in the next order , we allow for the fact that the amplitude at @xmath34 could also be influenced by waves arriving through slit @xmath52 ( if @xmath52 is open ) .
this is why the non - classical effects violate the simple minded application of the superposition principle .
the classical amplitude is given by , @xmath115}\,.\nonumber\\\end{aligned}\ ] ] here the integrals for @xmath56 and @xmath116 variables are independent .
the @xmath56 integral evaluates to a complex number , @xmath117}\,.\nonumber\\\end{aligned}\ ] ] this is same for all the slits @xmath118 and @xmath53 .
the non - classical amplitude within the fraunhofer approximation is , @xmath119^{-1/2 } e^{ik\left[\frac{y_1 ^ 2+z_1 ^ 2}{2l}+\sqrt{(y_2-y_1)^2+(z_2-z_1)^2 } + \frac{(y_d - y_2)^2+z_2 ^ 2}{2d}\right ] } \ \ \ \ \ \ \ \ \ \\end{aligned}\ ] ] for the @xmath120 integral using the stationary phase approximation ( s[stationary ] ) with @xmath121 as stationary point and @xmath122 the non - classical amplitude reduces to , @xmath123}\,.\end{aligned}\ ] ] again the integrals for @xmath56 and @xmath116 variables are independent and the @xmath124 integral is equal to @xmath125 . as the numerator and denominator in @xmath9
are both linear combination of multiplications of type @xmath126 , the factor @xmath127 cancels out .
we assume that the penetration of light through the opaque metal is zero .
the transmission amplitude can be found heuristically .
if @xmath128 is the attenuation constant and if @xmath28 is the incident wave amplitude the transmission amplitude is given by , @xmath129 .
@xmath130 is the thickness of the layer in units of the wavelength .
this quantifies the approximation that there is no path passing through the solid metal .
here @xmath131 , therefore the transition amplitude is , @xmath132 . @xmath133 as refractive index of steel
is @xmath134 .
@xmath135 , therefore the error is @xmath136 we assumed that the propagator from the source to the slit and the slit to the detector is a free particle propagator . to quantify this approximation we integrate over an intermediate plane ( @xmath137 ) similar to equation s[fold ] .
the integral is of the form , @xmath138 here , @xmath139 is the total distance from the source to the slit .
@xmath116 is a variable .the stationary point for the above integral is a point lying on the straight line joining the source and the detector . in the fraunhofer limit @xmath153 ,
the errors due to this approximation are of the order @xmath154 .
+ these errors result in error in calculating @xmath9 .
the final leading order error in @xmath9 is @xmath155 .
landau , l. d. & lifshitz , e. m. _ classical theory of fields _ ( pergammon press , oxford , fourth revised english edition , 1975 ) .
born , m. and wolf , e. _ principles of optics _ ( cambridge university press , seventh expanded edition , 1999 ) .
sunil kumar p. b. and ranganath g. s. , pramana-j.phys .
* 3 * 6 ( 1991 ) .
feynman , r. p. leighton , r. & sands , m. _ the feynman lectures on physics vol .
3 _ ( addison - wesley , reading mass , 1963 ) .
cohen - tannoudji , c. diu , b. & laloe , f. _ quantum mechanics i _ , ( wiley - vch , 2nd edition , 2005 ) shankar , r. _ principles of quantum mechanics _ , ( springer , 2nd edition , 1994 ) . strictly speaking , in equation s[ktwo ]
one should excise the region where @xmath156 ( the diagonal of the @xmath157 integration ) .
this integral is already accounted for in @xmath158 .
however this does not affect any of the terms in the @xmath11 expression ( s[final ] ) .
feynman , r.p .
space - time approach to non - relativistic quantum mechanics .
_ _ rev . of mod .
phys.__**20 * * , no.2 , 367 ( 1948 ) .
sorkin , r. d. quantum mechanics as quantum measure theory .
lett . a_. * 9 * , 3119 ( 1994 ) . | in a double slit interference experiment , the wave function at the screen with both slits open is not exactly equal to the sum of the wave functions with the slits individually open one at a time .
the three scenarios represent three different boundary conditions and as such , the superposition principle should not be applicable
. however , most well known text books in quantum mechanics implicitly and/or explicitly use this assumption which is only approximately true . in our present study , we have used the feynman path integral formalism to quantify contributions from non - classical paths in quantum interference experiments which provide a measurable deviation from a naive application of the superposition principle .
a direct experimental demonstration for the existence of these non - classical paths is hard .
we find that contributions from such paths can be significant and we propose simple three - slit interference experiments to directly confirm their existence .
quantum mechanics has been one of the most successful theories of the twentieth century , both in describing fundamental aspects of modern science as well as in pivotal applications .
however , inspite of these obvious triumphs , there is universal agreement that there are aspects of the theory which are counter - intuitive and perhaps even paradoxical .
furthermore , understanding fundamental problems involving dark matter and dark energy @xcite in cosmology may need a consistent quantum theory of gravity .
unification of quantum mechanics and general relativity towards a unified theory of quantum gravity @xcite is the holy grail of modern theoretical physics .
such unification attempts involve modifications of either or both theories .
however , all such attempts would rely very strongly on precise knowledge and understanding of the current versions of both theories .
this makes precision tests of fundamental aspects of both quantum mechanics and general relativity very important to provide guiding beacons for theoretical development .
the double slit experiment ( figure 1 ) is one of the most beautiful experiments in physics .
in addition to its pivotal role in optics , it is frequently used in classic textbooks on quantum mechanics @xcite to illustrate basic principles .
consider a double slit experiment with incident particles ( eg . photons , electrons ) .
the wave function at the detector with slit a open is @xmath0 .
the wavefunction with the slit b open is @xmath1 .
what is the wavefunction with both slits open ?
it is usually assumed to be @xmath2 @xcite .
this is illustrated in figure 1 . from the mathematical perspective of solving the schrdinger equation ,
this assumption is definitely not true .
the three cases described above correspond to three different boundary conditions @xcite and as such the application of the superposition principle can at best be approximate .
recent numerical simulations of maxwell s equations using finite difference time domain analysis have shown this to be true in the classical domain @xcite .
how do we quantify this effect in quantum mechanics ? .
] an intuitive and simple way of understanding this problem is to appeal to feynman s path integral formalism @xcite .
the path integral formalism involves an integration over all possible paths that can be taken by the particle through the two slits .
this not only includes the nearly straight paths from the source to the detector through either slit ( the classical paths ) like the green paths in figure 2 but also includes paths of the type shown in purple in figure 2 ( non - classical paths ) .
these looped paths are expected to make a much smaller contribution to the total intensity at the detector screen as opposed to the contribution from the straight line paths . however , their contribution is finite .
formally , a classical path is one that extremizes the classical action .
any other path is a non - classical path .
this leads to a modification of the wave function at the screen which now becomes : @xmath3 where @xmath4 is the contribution due to the looped _
i.e. , _ non - classical paths . that @xmath4 is non - zero was first pointed out in @xcite in the non - relativistic domain where certain unphysical approximations were made in computing @xmath4 and hence the results or the methods can not be used in an experimental situation
recently , @xcite has reiterated the point that @xmath4 can be non - zero without attempting to quantify it in quantum mechanics .
= the interslit distance , @xmath5 = the slit width , @xmath6 = the slit height , @xmath7 = the distance from the source to the slit plane and @xmath8 = the distance from the slit plane to the detector plane . ] in this paper , we will quantify the effect of such non - classical paths in interference experiments , thus quantifying the deviation from the common but incorrect application of the superposition principle in different possible experimental conditions .
a well - known example of a direct experimental demonstration of such non - classical paths involves the measurement of the aharonov - bohm phase @xcite .
berry s many - whirls " representation @xcite provides insight into simple explanations of the aharonov bohm effect in terms of interference between whirling waves passing around the flux tube .
however , in most experimental attempts to measure the aharonov bohm phase , the detection relies on rather complicated experimental architecture and the results are also open to interpretational issues and further discussion @xcite .
in this work , we propose simple triple slit based interference experiments @xcite which can be used as table top demonstrations of non - classical paths in the path integral formalism .
non - classical paths have been used to compute the semi - classical off - diagonal contributions to the two - point correlation function of a quantum system whose classical limit is chaotic @xcite .
the paths in this case are real . in the feynman path integral approach , all possible paths going from the initial to final state need to be considered with an appropriate weight . in this sense
all paths are real although in a physical quantity the contribution from certain paths may be suppressed .
the triple slit experiment provides a simple way to quantify the effects from non classical paths in terms of directly measurable quantities . the triple slit ( path )
set - up has been used as a test - bed for testing fundamental aspects of quantum mechanics over the last few years @xcite .
three - state systems are also fast becoming a popular choice for fundamental quantum mechanical tests @xcite . in order to analyse the effect of non - classical paths in interference experiments ,
we have considered the effect of such paths on an experimentally measurable quantity @xmath9 .
@xmath9 ( defined below ) has been measured in many experiments over the last few years in order to arrive at an experimental bound on possible higher order interference terms in quantum mechanics @xcite and in effect the born rule for probabilities @xcite .
investigations of this quantity may also be relevant to theoretical attempts to derive the born rule @xcite . if born s postulate for a square law for probabilities is true and if @xmath10 , then the quantity @xmath11 defined by @xmath12 is identically zero in quantum mechanics . here
@xmath13 is the probability at the detector when all three slits are open , @xmath14 is the probability when slits a and b are open and so on . in the experiments reported in the literature
, the normalization factor has been chosen to be the sum of the three double slit interference terms called @xmath15 given by : @xmath16 where @xmath17 and so on .
this choice of normalization can sometimes lead to false peaks in the @xmath9 as a function of detector position due to the denominator becoming very small at certain positions .
we use a somewhat different normalization , @xmath18 , where @xmath19 is the intensity at the central maximum of the triple slit interference pattern to avoid this problem
. then the normalized quantity @xmath9 is given by : @xmath20 in discussions which invoke the zeroness " of @xmath9 , it is implicitly assumed that only classical paths contribute to the interference . in his seminal work @xcite
, sorkin had also assumed that the contribution from non - classical paths was negligible .
now , what is the effect of non - classical paths on @xmath9 ? if one can derive a non - zero contribution to @xmath9 by taking into account all possible paths in the feynman path integral formalism , that would mean @xmath21 is not strictly true and experimentalists should not be led to conclude that a measurement of non - zero @xmath9 would immediately indicate a falsification of the born rule for probabilities in quantum mechanics .
a measured non - zero @xmath9 could also be explained by taking into account the non - classical paths in the path integral .
there is thus a theoretical estimate for a non - zero @xmath9 .
of course , the immediate expectation would be a clear domination of the classical contribution and perhaps a very negligible contribution from the non - classical paths which would in turn imply that @xmath21 is true in all experimentally observable conditions .
" however , what we go on to discover is that this expectation is not always true .
it is possible to have experimental parameter regimes in which @xmath9 is measurably large .
this in turn leads to a paradigm shift in such precision experiments .
observation of a non - zero @xmath9 which is expected from the proposed correction to @xmath21 would in fact also serve as an experimental validation of the full scope of the feynman path integral formalism .
as mentioned before , in calculating @xmath9 , one inherently assumes contributions only from the classical straight line paths as shown in green in figure 2 . in this paper , we have estimated the contribution to @xmath9 from non - classical paths , thus providing the first theoretical estimate for @xmath9 . for simplicity
, we will use the free particle propagator in our calculations . for a particle in free space and away from the slits , this is a reasonable approximation .
we account for the slits by simply removing from the integral all paths that pass through the opaque metal .
an estimate for the error due to this assumption has been worked out in @xcite .
the normalized energy space propagator @xmath22 @xcite for a free particle with wave number @xmath23 from a position @xmath24 to @xmath25 is given by @xmath26 although in this paper , we will be mainly focusing on analyzing optics based experiments using photons , this propagator equation can be used both for the electron and the photon as argued in @xcite .
we should point out that there are corrections to the propagator due to closed loops in momentum space from quantum field theory considerations .
we have explicitly estimated that the effects of such corrections will be negligibly small @xcite .
consider the triple slit configuration shown in figure 2 . according to the path integral prescription ,
all paths that go from the source to the detector should contribute in the analysis . in the quantity of interest , @xmath9
, some important simplifications occur .
only those non - classical paths that involve propagation between at least two slits would contribute to the leading non - zero value .
this is because any non - classical path that goes through _ only _ the @xmath27th slit can be taken into account in the wave - function @xmath28 at the detector and hence would cancel out in @xmath9 as can be easily checked . in light of the above , the entire set of paths from the source to the detector through the slits
can be divided into two classes : 1 .
paths which cross the slit plane exactly once pertaining to a probability amplitude @xmath29 ; a representative path is shown by the green line and 2 .
paths which cross the slit plane more than once at two or more slits pertaining to a probability amplitude @xmath30 @xcite as for instance , represented by the purple line .
@xmath31 we wish to estimate @xmath30 relative to @xmath29 .
an example of a representative @xmath29 in our problem is the probability amplitude to go from the source @xmath32 to the detector @xmath33 through slit @xmath34 which we call @xmath35 .
this uses the general scheme that a path in feyman s path integral formalism can be broken into many sub - paths and the propagator is the product of the individual propagators @xcite .
for instance , @xmath36 where @xmath37 is the interslit distance , @xmath5 is the slit width , @xmath6 is the slit height , @xmath38 and @xmath39 as shown in figure 2 . for the source and the detector far apart from one another , _
i.e. , _ in the fraunhofer regime , @xmath40 @xmath41 @xmath42 in the region of integration , therefore , @xmath43 .
similarly @xmath44 giving @xmath45 .
thus we have @xmath46}\,.\nonumber\\\end{aligned}\ ] ] here @xmath47 .
these are fresnel integrals and have been evaluated using mathematica .
let us now proceed to the probability amplitude for multiple slit crossings _
i.e. , _ @xmath30 .
an example of a representative @xmath30 in our problem is the probability amplitude to go from the source @xmath32 to the detector @xmath33 following the kind of path shown in figure 2 . in this case
, the particle goes from the source to the first slit and then loops around the second and third slits before proceeding to the detector .
we represent this by @xmath48 .
this is approximated by @xcite : @xmath49 here the @xmath50 integral runs over slit @xmath34 and @xmath51 integral runs over slits @xmath52 and @xmath53 and where @xmath54 making approximations appropriate to the fraunhofer regime , using stationary phase approximation @xcite for the oscillatory integrals the integral becomes : @xmath55}\,.\end{aligned}\ ] ] an important simplification occurs at this stage : the @xmath56 integral in @xmath57 is same as in the integral for @xmath58 .
since we are just concerned with ratios , the contributions from the @xmath56 integrals cancel out . in terms of @xmath29 and @xmath30 ,
the propagator to go from the source to the detector when all three slits are open is given by : @xmath59 where @xmath60 include non - classical terms arising when all slits are open .
similarly : @xmath61 @xmath62 are non - classical terms involving only @xmath34 and @xmath52 .
similarly for @xmath63 and @xmath64 .
thus , in terms of propagators , @xmath65 as a function of detector position . here
@xmath66 .
a ) this is for incident photons , slit width = 30@xmath67 m , interslit distance = 100@xmath67 m , distance between source and slits and slits and detector = 18 cm and incident wavelength = 810 nm @xcite .
b ) this is for incident electrons , slit width = 62 nm , interslit distance = 272 nm , distance between source and slits = 30.5 cm and slits and detector = 24 cm and debroglie wavelength = 50pm @xcite . ]
and the normalization @xmath15 is given by @xmath68 , where @xmath69 is the value of @xmath70 at the central maximum . by numerical integration
, we find @xmath9 at the central maximum of the triple slit interference pattern to be of the order of @xmath71 for the parameters used in the triple slit experiment reported in reference @xcite .
what would have been expected to be zero considering only straight line paths now turns out to be measurably non - zero having taken the non - classical ones into account @xcite . in figure 3 , we show @xmath9 as a function of detector position .
we also show a plot of the triple slit interference pattern as a function of detector position which gives a clearer understanding of the modulation in the plot for @xmath9 .
the experiment reported in reference @xcite was not sensitive to a theoretically expected non - zeroness in @xmath9 due to systematic errors .
however , in the absence of such systematic errors , it is definitely possible to use a similar set - up to measure a non - zero @xmath9 .
simulation results indicate that the set - up could have measured a much lower value of @xmath9 but the presence of the systematic error due to one misaligned opening in the blocking mask set the limitation of the experiment making it possible to only measure a value of @xmath9 upto @xmath72 .
there is no reason why this systematic error can not be removed in a future version of the experiment thus making it a perfect table - top experiment to test for the presence of non - classical paths in interference experiments .
however , experiments of the kind reported in @xcite are not as ideally suited for this purpose .
this is because , in our analysis , we have worked in the thin - slit approximation .
the effective `` slit - thickness '' in a diffraction grating based interferometer set - up would be quite big and hence the resulting @xmath9 would certainly be smaller .
what we go on to also find in our current analysis is that @xmath9 is very strongly dependent on certain experimental parameters and one can definitely find a parameter regime where @xmath9 would be even bigger , hence easier to observe .
we find that keeping all other experimental parameters fixed , @xmath9 increases with an increase in wavelength .
thus , for instance , for an incident beam of wavelength 4 cm ( microwave regime ) and slit width of 120 cm and interslit distance of 400 cm , a theoretical estimate for @xmath9 would be @xmath73 .
this is an experiment which can be performed for instance in a radio astronomy lab .
experiments of this kind where the value of @xmath9 due to non - classical paths can be estimated would definitely be of great interest as they would serve as a simple experimental demonstration of how the basic assumption that a composite wavefunction is just the sum of component wavefunctions is not always true . in a sense
they would also serve as a direct table - top demonstration of the complete scope of the feynman path integral formalism where not only the straight line paths are important but also the looped paths can make a sizeable contribution depending on one s choice of experiment .
the effects due to such non - classical paths may also be used to model possible decoherence mechanisms in interferometer based quantum computing applications . |
it is well known that different types of topological objects may have been formed in the early universe after planck time by the vacuum phase transition @xcite . depending on the topology of the vacuum manifold these are domain walls , strings , monopoles and textures . among them
the cosmic strings are of special interest . although the recent observational data on the cosmic microwave background radiation have ruled out cosmic strings as the primary source for primordial density perturbations , they are still candidates for the generation of a number of interesting physical effects such as the generation of gravitational waves , gamma ray bursts and high - energy cosmic rays ( see , for instance , @xcite ) . recently , cosmic strings attract a renewed interest partly because a variant of their formation mechanism is proposed in the framework of brane inflation sara02 . in the simplest theoretical model describing the infinite straight cosmic string
the spacetime is locally flat except on the string where it has a delta shaped riemann curvature tensor . in quantum field theory the corresponding non - trivial topology induces non - zero vacuum expectation values for physical observables .
explicit calculations for the geometry of a single cosmic string have been done for different fields @xcite - spin08 .
vacuum polarization effects by higher - dimensional composite topological defects constituted by a cosmic string and global monopole are investigated in refs .
@xcite for scalar and fermionic fields .
another type of vacuum polarization arises when boundaries are present .
the imposed boundary conditions on quantum fields alter the zero - point fluctuations spectrum and result in additional shifts in the vacuum expectation values of physical quantities .
this is the well - known casimir effect ( for a review see @xcite ) . in ref .
@xcite , we have studied both types of sources for the polarization of the scalar vacuum , namely , a cylindrical boundary and a cosmic string , assuming that the boundary is coaxial with the string and that on this surface the scalar field obeys robin boundary condition . for a massive scalar field with an arbitrary curvature coupling parameter we evaluated the wightman function and the vacuum expectation values of the field squared and the energy - momentum tensor .
the polarization of the electromagnetic vacuum by a conducting cylindrical shell in the cosmic string spacetime is investigated in @xcite ( for a combination of topological and boundary induced quantum effects in the gravitational field of a global monopole see refs .
@xcite . )
continuing in this line of investigation , in the present paper we analyze the polarization of the fermionic vacuum by a cylindrical shell coaxial with the cosmic string on which the field obeys mit bag boundary condition .
we evaluate the fermionic condensate and vacuum expectation values of the energy - momentum tensor in both interior and exterior regions of the shell .
the renormalized vacuum expectation value of the energy - momentum tensor for a fermionic field in the geometry of a cosmic string without boundaries is investigated in @xcite .
in addition to describing the physical structure of the quantum field at a given point , the energy - momentum tensor acts as a source of gravity in the einstein equations and plays an important role in modelling a self - consistent dynamics involving the gravitational field . in the problem under consideration
all calculations can be performed in a closed form and it constitutes an example in which the topological and boundary - induced polarizations of the vacuum can be separated in different contributions . from the point of view of the physics in the region outside the string , the geometry considered in the present paper can be viewed as a simplified model for the non - trivial core .
this model presents a framework in which the influence of the finite core effects on physical processes in the vicinity of the cosmic string can be investigated .
in particular , it enables to specify conditions under which the idealized model with the core of zero thickness can be used .
the corresponding results may shed light upon features of finite core effects in more realistic models , including those used for string - like defects in crystals and superfluid helium .
in addition , the problem considered here is of interest as an example with combined topological and boundary induced quantum effects in which the vacuum characteristics such as energy density and stresses can be found in closed analytic form . from the results of the present paper , as a special case , we obtain the fermionic casimir densities for a cylindrical shell with mit boundary conditions in the minkowski background ( for the combined effects of a magnetic fluxon and mit boundary conditions on the vacuum energy of a dirac field see refs .
@xcite ) . note that , in addition to traditional problems of quantum field theory under the presence of material boundaries , the casimir effect for cylindrical geometries can also be important to the flux tube models of confinement @xcite and for determining the structure of the vacuum state in interacting field theories @xcite .
we have organized the paper as follows . in the next section ,
the eigenspinors for the region inside the cylindrical boundary are constructed and the eigenvalues of the corresponding quantum numbers are specified .
these eigenspinors are the basis for the anlysis of the casimir densities in the following sections . in section [ sec : fermc ] , by using a variant of the generalized abel - plana formula , we extract from the mode - sum of the fermionic condensate the part corresponding to the geometry of a cosmic string without the shell . the part induced by the shell
is investigated in various asymptotic regions for the parameters .
section [ sec : emt ] is devoted to the investigation of the boundary induced parts in the vacuum expectation value of the energy - momentum tensor inside the cylindrical shell .
the vacuum densities in the region outside the cylindrical shell are discussed in section [ sec : exterior ] .
the main results of the paper are summarized in section [ sec : conc ] . in appendix
[ sec : appendix ] we give an alternative representation for the fermionic condensate and the expectation value of the energy - momentum tensor for a massive fermionic field in the geometry of a cosmic string without boundaries .
we consider the background spacetime corresponding to an infinitely long straight cosmic string with the conical line element@xmath0where @xmath1 and the spatial points @xmath2 and @xmath3 are to be identified .
the planar angle deficit is related to the mass per unit length of the string , @xmath4 , by @xmath5 , where @xmath6 is the newton gravitational constant . in the discussion below , in addition to the parameter @xmath7
, we will use the combination@xmath8 the dynamics of a massive fermionic field is governed by the dirac equation @xmath9with the covariant derivative operator defined as @xmath10here @xmath11 are the dirac matrices in curved spacetime and @xmath12 is the spin connection given in terms of the flat space dirac matrices @xmath13 by the relation @xmath14note that in this formula @xmath15 means the standard covariant derivative for vector fields . in the relations above @xmath16 is the tetrad basis satisfying @xmath17 , where @xmath18 is the minkowski spacetime metric tensor .
in this paper we are interested in the change of the vacuum expectation values ( vevs ) of the fermionic condensate and the energy - momentum tensor for a fermionic field , induced by a cylindrical shell coaxial with the string on which the field obeys mit bag boundary condition : @xmath19where @xmath20 is the cylinder radius and @xmath21 is the outward - pointing normal to the boundary . for the evaluation of the vevs we will use the direct mode summation procedure . in this approach
we need to have the complete set of the eigenfunctions satisfying boundary condition ( [ bagcond ] ) . in order to find these eigenfunctions ,
we will use the standard representation of the flat space dirac matrices:@xmath22with @xmath23 being the pauli matrices .
we take the tetrad fields in the form used before in @xcite ( see also beze06,skar94):@xmath24where the index @xmath20 identifies the rows of the matrix . with this choice
, the gamma matrices are given by@xmath25where we have introduced the @xmath26 matrices@xmath27for the spin connection and the combination appearing in the dirac equation we find@xmath28and the dirac equation takes the form@xmath29 for positive frequency solutions , assuming the time - dependence of the eigenfunctions in the form @xmath30 and decomposing the bispinor @xmath31 into the upper and lower components , denoted by @xmath32 and @xmath33 , respectively , we find the equations@xmath34substituting the function @xmath33 from the first equation into the second one , we obtain the second order differential equation for the function @xmath35:@xmath36\varphi = 0 , \label{phieq}\]]where @xmath37 .
the same equation is obtained for the function @xmath33 . because the above equation is in diagonal matrix form
, we decompose the spinor @xmath32 into the upper , @xmath38 , and lower , @xmath39 , components . taking the eigenfunctions corresponding to these components in the form @xmath40 , @xmath41
, we can see that the solutions of the equations for the radial functions , regular on the string , are expressed in terms of the bessel function of the first kind : @xmath42 , where the order is defined by the relations@xmath43with @xmath44 , and@xmath45hence , the components of the upper spinor are given by the formula@xmath46 , \;l=1,2 .
\label{phil}\]]having the upper spinor , we can find the components @xmath47 of the lower one by using the first equation in ( [ phixieq ] ) . from this equation
we find the following relations@xmath48and@xmath49 , \;l=1,2 , \label{xil}\]]with the coefficients@xmath50we can see that the bispinor with the components defined by relations ( phil ) and ( [ xil ] ) is an eigenfunction of the projection of the total momentum along the cosmic string:@xmath51where@xmath52 for the further specification of the eigenfunctions we can impose an additional condition relating the constants @xmath53 and @xmath54 . as such a condition , following @xcite , we will require the following relations between the upper and lower components:@xmath55from the expressions for the spinor components we find the eigenvalues of the parameter @xmath56,@xmath57and the relation@xmath58for the coefficients in ( [ phil ] )
. hence , the positive frequency solutions to the dirac equation , specified by the set of quantum numbers @xmath59 , has the form@xmath60 , \label{psi+}\]]where the orders of the bessel functions are defined in terms of @xmath61 as@xmath62 in ( [ psi+ ] ) and in the consideration below we use the notations@xmath63note that one has the relation @xmath64 .
the eigenvalues of the radial quantum number @xmath65 are determined from the boundary condition ( [ bagcond ] ) imposed on the eigenspinor ( [ psi+ ] ) . for fixed values of @xmath61 and @xmath66
, this leads to the single equation@xmath67using the recurrence relations for the bessel functions , this equation may also be written in the form@xmath68here and in what follows we use the notation@xmath69 , \label{ftilde}\end{aligned}\]]with @xmath70 .
we will denote the solutions of the equation ( lambvalue1 ) by @xmath71 , @xmath72 , assuming that they are arranged in ascending order .
now the set of quantum numbers is specified by @xmath73 .
the coefficient @xmath74 in ( [ psi+ ] ) is determined from the normalization condition @xmath75where @xmath76 is the determinant of the spatial metric and the integration goes over the region inside the cylindrical shell .
the delta symbol on the right - hand side of eq .
( [ normcond ] ) is understood as the dirac delta function for continuous quantum numbers ( @xmath77 ) and the kronecker delta for discrete ones ( @xmath78 ) .
substituting the eigenspinors ( [ psi+ ] ) into eq .
( [ normcond ] ) and using the value of the standard integral involving the square of the bessel function @xcite , we find@xmath79 , \label{csig+}\]]with the notation @xmath80 .
the negative frequency eigenspinors are found in a way similar to that used above for the positive frequency ones and have the form@xmath81 , \label{psi-}\]]with the same notations as in ( [ psi+ ] ) .
the boundary condition imposed on this eigenspinor leads to the same equation ( [ lambvalue ] ) for the eigenvalues of @xmath65 .
the coefficient @xmath82 is found from the orthonormalization condition which is similar to ( [ normcond ] ) and has the form @xmath83 .
fermionic condensate is evaluated by using the mode - sum formula @xmath84where @xmath85 is the dirac adjoint and@xmath86substituting the eigenspinor ( [ psi- ] ) into ( [ fcmodesum ] ) , we find @xmath87_{\lambda = \lambda _ { \beta , l}/a } , \label{fc1}\]]where @xmath88 is defined by expression ( [ kappas ] ) and @xmath89the fermionic condensate given by formula ( [ fc1 ] ) is divergent and some regularization procedure is necessary .
we will assume that a cutoff function is introduced in formula ( [ fc1 ] ) without explicitly writing it .
as the explicit form for @xmath90 is not known , formula ( fc1 ) is not convenient for the direct evaluation of the condensate .
in addition , the separate terms in the mode - sum are highly oscillatory for large values of the quantum numbers .
a convenient form can be obtained by applying to the series over @xmath91 the summation formula , previously derived in ref .
@xcite ( see , also , @xcite ) . in @xcite by using the generalized abel - plana formula , it has been shown that for a function @xmath92 analytic in the half - plane @xmath93 and satisfying the condition @xmath94with @xmath95 and @xmath96 for @xmath97 , the following formula takes place @xmath98 \notag \\ & & { } -\frac{1}{\pi } \int_{0}^{\infty } dx\,\bigg[e^{-\beta \pi i}h(xe^{\pi i/2})\frac{k_{\beta } ^{(+)}(x)}{i_{\beta } ^{(+)}(x)}+e^{\beta \pi i}h(xe^{-\pi i/2})\frac{k_{\beta } ^{(-)}(x)}{i_{\beta } ^{(-)}(x)}\bigg]\ , \label{sumlamb}\end{aligned}\]]where @xmath99 , @xmath100 are the modified bessel functions , and @xmath101 is defined by the relation @xmath102\
. \label{r1}\]]in formula ( [ sumlamb ] ) we used the notations @xmath103for a given function @xmath104 . by taking into account
the relation@xmath105we can write the mode - sum for the fermionic condensate in the form @xmath106with the notation@xmath107note that in ( [ fc2])@xmath108and we have the property@xmath109for @xmath110 . now we apply to the sum over @xmath91 in ( [ fc2 ] ) formula ( sumlamb ) taking @xmath111 . for this function the residue term in ( [ sumlamb ] )
the part in the fermionic condensate with the last integral on the right - hand side of ( [ sumlamb ] ) vanishes in the limit @xmath112 , whereas the part with the first integral does not depend on @xmath20 . from here
it follows that the latter presents the fermionic condensate in the geometry of a cosmic string without boundaries .
this can also be seen by direct evaluation .
indeed , when the cylindrical boundary is absent , the positive and negative frequency eigenspinors are still given by formulae ( [ psi+ ] ) and ( psi- ) , where now the spectrum for @xmath65 is continuous , @xmath113 .
the corresponding normalization coefficients are found from the condition ( [ normcond ] ) and have the form@xmath114substituting the eigenspinors into the mode - sum formula ( [ fcmodesum ] ) , for the fermionic condensate in the geometry of a cosmic string without boundaries we find@xmath115 .
\label{fc0}\end{aligned}\]]this coincides with the result obtained from the first term on the right of formula ( [ sumlamb ] ) applied to mode - sum ( [ fc2 ] ) .
formula ( [ fc0 ] ) is further simplified by taking into account the expression for @xmath116 and after the summation over @xmath66:@xmath117 .
\label{fc0b}\]]here and in what follows@xmath118as it is seen from formula ( [ fc0b ] ) , for a massless field the fermionic condensate vanishes in the boundary - free cosmic string spacetime . since the geometry outside of the string is flat , the renormalization of the fermionic condensate given by ( [ fc0b ] ) is done by subtracting the corresponding quantity for the boundary - free minkowski spacetime .
the latter is obtained from ( [ fc0b ] ) taking in this formula @xmath119 .
note that for the minkowski case the summation over @xmath61 is explicitly done by using the formula @xmath120=2% \sideset{}{'}{\sum}_{n=0}^{\infty } j_{n}^{2}(\lambda r)=1 , \label{sumj2}\]]where the prime on the sign of summation means that the @xmath121 term should be halved .
the renormalized fermionic condensate in the geometry of a cosmic string without boundaries is further investigated in appendix sec : appendix . from the discussion above it follows that , after the application of the summation formula ( [ sumlamb ] ) , the part in the fermionic condensate with the second integral on the right - hand side of this formula corresponds to the vev induced by the presence of the cylindrical shell . by using property ( [ fprop ] ) and noting that under the change @xmath122 we have @xmath123 , @xmath124 , the fermionic condensate in the geometry with the cylindrical shell
is presented in the decomposed form @xmath125here the second term on the right - hand side,@xmath126 , \label{fc3}\end{aligned}\]]is the part in the fermionic condensate induced by the cylindrical boundary .
the number of the integrations in this formula is reduced by using the relation@xmath127further , redefining @xmath122 in the part of the summation over the negative values @xmath61 , it can be seen that the negative and positive values of @xmath61 give the same contributions to the fermionic condensate .
finally , we arrive to the following formula @xmath128 , \label{fccyl}\]]where @xmath129and the function @xmath130 is given by the expression @xmath131 in ( [ fccyl ] ) , for the modified bessel functions we use the barred notations@xmath132with @xmath133 , and @xmath134 , @xmath135 .
the boundary induced part ( [ fccyl ] ) is finite for points away from the cylindrical shell and the renormalization is necessary for the boundary - free part , @xmath136 , only .
note that the ratio in the integrand of eq .
( [ fccyl ] ) may also be presented in the form@xmath137 + 2\mu xi_{\beta } ( x)i_{\beta + 1}(x ) } , \label{kbib}\]]where@xmath138 + 2\mu xi_{\beta } ( x)k_{\beta + 1}(x ) .
\label{wbeta}\]]for a massless fermionic field , from ( [ fccyl ] ) and ( [ kbib ] ) we find@xmath139 in this case the boundary - free part vanishes and the fermionic condensate is always negative .
now we turn to the investigation of the fermionic condensate given by eq .
( [ fccyl ] ) in the asymptotic regions of the parameters . for large values of the cylinder radius
we use the asymptotic formulae for the modified bessel functions when the argument is large . by taking into account that the main contribution into the integral in ( [ fccyl ] ) comes from the lower limit of the integral , to the leading order we find@xmath140for @xmath141 . for a massless field , expanding the integrand in ( [ fcm0 ] ) we see that the main contribution is due to the term with @xmath142 and one has@xmath143for @xmath144 . for points near the string , @xmath145 , the main contribution into the boundary - induced vev ( [ fccyl ] ) comes from the lowest mode @xmath142 and to the leading order we find@xmath146 .
\label{fconaxis}\]]this quantity is non - zero in the case of the cylindrical boundary in the minkowski bulk and vanishes for the cosmic string geometry with @xmath147 . for a massless field formula ( [ fconaxis ] )
is reduced to eq .
( [ fclargeam0 ] ) . for points near the boundary
the main contribution comes from large values of @xmath61 and in this case we can use the uniform asymptotic expansions for the modified bessel functions for large values of the order ( see , for instance , @xcite ) . in this way , to the leading order we have@xmath148for @xmath149 .
as we see , the leading term does not depend on the planar angle deficit and corresponds to the same one obtained for a cylindrical boundary in the minkowski bulk .
now we consider the limit when the parameter @xmath150 is large which corresponds to small values of @xmath7 and , hence , to a large planar angle deficit . in this limit
the order of the modified bessel functions in ( [ fccyl ] ) is large and we replace these functions by their uniform asymptotic expansions . assuming fixed value of the ratio @xmath151 , the integral
is estimated by the laplace method and to the leading order we have@xmath152}{2\pi ^{2}r^{3}[1-(r / a)^{2}]}\left ( \frac{r}{a}% \right ) ^{q},\;q\gg 1 .
\label{fclargeq}\]]hence , for large values of the angle deficit the fermionic condensate is exponentially suppressed .
in this section we consider the vev of the energy - momentum tensor of the fermionic field inside a cylindrical shell on which the field satisfies the boundary condition ( [ bagcond ] ) .
this vev can be evaluated by making use of the mode - sum formula @xmath153\ , \label{modesum}\]]with the negative frequency eigenspinors given by ( [ psi- ] ) .
we can see that the vacuum energy - momentum tensor is diagonal and the separate components are given by the formulae ( no summation over @xmath154 ) @xmath155where the following notations were introduced@xmath156 , \notag \\
f_{\beta } ^{(1)}(x , y ) & = & j_{\beta } ^{2}(y)+j_{\beta + \epsilon _ { j}}^{2}(y)-% \frac{2\beta + \epsilon _ { j}}{y}j_{\beta } ( y)j_{\beta + \epsilon _ { j}}(y ) , \label{fbetamu } \\ f_{\beta } ^{(2)}(x , y ) & = & \frac{2\beta + \epsilon _ { j}}{y}j_{\beta } ( y)j_{\beta + \epsilon _ { j}}(y),\;\,f_{\beta } ^{(3)}(x , y)=(k^{2}/\omega ^{2})f_{\beta } ^{(0)}(x , y ) .
\notag\end{aligned}\]]the other notations in ( [ emt ] ) are the same as in ( [ fc2 ] ) .
the vev given by ( [ emt ] ) is divergent and , as in the case of the fermionic condensate , it is assumed that a cutoff function is present .
now , we can explicitly verify that the vevs ( [ emt ] ) satisfy the trace relation @xmath157 in order to extract from the vev of the energy - momentum tensor the part corresponding to the geometry of a string without boundaries , we apply to the series over @xmath91 in ( [ emt ] ) the summation formula ( [ sumlamb ] ) with@xmath158 in a way similar to that used in the case of the fermionic condensate , the vev can be written in the decomposed form:@xmath159where @xmath160 is the fermionic energy - momentum tensor in the geometry of a cosmic string when the cylindrical shell is absent .
the second term on the right - hand side of formula ( [ emt1 ] ) is induced by the cylindrical shell and is given by the formula ( no summation over @xmath154)@xmath161 , \label{emtcyl}\]]where @xmath162 is defined by eq .
( [ betaj ] ) . in this formula
we have introduced the notations@xmath163and @xmath164 .
note that the function for the radial component is also presented in the form@xmath165as we see , the vacuum stress along the axis of the string is equal to the energy density . of course , this property is a direct consequence of the bust invariance along this axis . by using eq .
( [ kbib ] ) , we may write the expressions for the components of the energy - momentum tensor in more explicit form given by @xmath166 - 2\mu i_{\beta _ { j}}^{2}(xr / a)}{i_{\beta _ { j}+1}^{2}(x)+i_{\beta _ { j}}^{2}(x)+2(\mu /x)i_{\beta _ { j}}(x)i_{\beta _ { j}+1}(x ) } , \label{t00int}\end{aligned}\]]for the energy density and by ( no summation over @xmath154)@xmath167f_{\beta _ { j}}^{(\nu ) } ( x , xr / a)}{i_{\beta _ { j}+1}^{2}(x)+i_{\beta _ { j}}^{2}(x)+2(\mu /x)i_{\beta _ { j}}(x)i_{\beta _ { j}+1}(x ) } , \label{tnuint}\]]for the radial and azimuthal stresses , @xmath168 .
the vev of the fermionic energy - momentum tensor in the geometry of a cosmic string when the cylindrical shell is absent , corresponds to the first term on the right - hand side of summation formula ( [ sumlamb ] ) . for this vev
we have the mode - sum ( no summation over @xmath154)@xmath169the summation over @xmath66 in this formula is done explicitly and noting that the negative and positive values of @xmath61 give the same contributions , we find@xmath170where now@xmath171 , \notag \\ g_{\beta _ { j}}^{(1)}(\lambda , y ) & = & j_{\beta _
{ j}}^{2}(y)+j_{\beta _ { j}+1}^{2}(y)-\frac{2qj}{y}j_{\beta _ { j}}(y)j_{\beta _ { j}+1}(y ) , \label{gmu } \\ g_{\beta _ { j}}^{(2)}(\lambda , y ) & = & \frac{2qj}{y}j_{\beta _ { j}}(y)j_{\beta _ { j}+1}(y),\;g_{\beta _ { j}}^{(3)}(\lambda , y)=(k^{2}/\omega ^{2})g_{\beta _ { j}}^{(0)}(\lambda , y ) .
\notag\end{aligned}\]]as in the case of the fermionic condensate , the vev ( [ emts1 ] ) is renormalized by subtracting the corresponding vev in the minkowski spacetime without boundaries .
the boundary - free renormalized vev of the energy - momentum tensor for a massive fermionic field is further investigated in appendix [ sec : appendix ] .
it can be explicitly verified that the boundary - induced parts satisfy the trace relation @xmath172 and the covariant conservation equation @xmath173 .
for the geometry under consideration the latter is reduced to the single equation@xmath174=\left\langle t_{2}^{2}\right\rangle _ { \mathrm{cyl}}. \label{conteq}\]]this equation is easily checked by using the relation @xmath175 between the radial and azimuthal functions in ( [ fbetamu ] ) . in the case of a massless fermionic field , for the vev induced by the cylindrical boundary from ( [ t00int ] ) and ( [ tnuint ] ) we have ( no summation over @xmath154 ) @xmath176where @xmath177 for @xmath168 , and the corresponding function for the energy density is defined as@xmath178 .
\label{fbeta0m0}\]]in the special case with @xmath119 , from the formulae given above we obtain the fermionic casimir densities for a cylindrical boundary in the minkowski spacetime . on the left panel of figutre [ fig1 ]
we have plotted the corresponding vevs for the energy density and radial stress as functions of the radial coordinate for a massless fermionic field . on the right panel
the boundary induced parts are presented in the geometry of a cosmic string with the parameter @xmath179 . [ cols="^,^ " , ]
in this section we consider the fermionic condensate and the vev of the energy - momentum tensor in the region outside the cylindrical shell . as in the interior case
this can be done by the direct mode summation .
the corresponding eigenspinors have the form ( [ psi+ ] ) and ( [ psi- ] ) with the difference that now , instead of the bessel functions @xmath180 , the linear combinations of the functions @xmath180 and @xmath181 should be taken , with @xmath182 being the neumann function . the ratio of the coefficients in this combination is determined from the boundary condition ( [ bagcond ] ) imposed on the cylindrical surface . in this way for the positive and negative frequency eigenspinors we have the expressions @xmath183 , \label{psiext}\]]where the function @xmath184 is defined by the formula@xmath185and@xmath186with @xmath187 given by relations ( [ bet12a ] ) . here the notation @xmath188 is defined by ( [ ftilde ] ) with the replacement @xmath189 .
the eigenspinors are orthonormalized by condition ( [ normcond ] ) , where now the radial integration goes over the exterior region .
the eigenvalues for @xmath65 are continuous and on the right - hand side of the normalization condition we have @xmath190 . since the radial integral diverges for @xmath191 , the main contribution to this integral comes from large values @xmath192 and we cane replace the bessel and neumann functions with the arguments @xmath193 , by the corresponding asymptotic expressions . in this way , for the normalization coefficients in ( [ psiext ] ) we find:@xmath194 , \label{csigext}\]]where @xmath88 and @xmath195 are defined by eqs .
( [ kappas ] ) and ( [ bsplmin ] ) . substituting the eigenspinors ( [ psiext ] ) into the mode - sum formula ( fcmodesum ) , for the fermionic condensate in the region outside the cylindrical shell
we obtain the formula@xmath196 } \notag \\ & & \times \lbrack ( -m+s\sqrt{\lambda ^{2}+m^{2}})z_{\beta } ^{2}(\lambda a,\lambda r)-(m+s\sqrt{\lambda ^{2}+m^{2}})z_{\beta + \sigma _ { j}}^{2}(\lambda a,\lambda r ) ] , \label{fcext}\end{aligned}\]]where@xmath197 in order to extract from the vev ( [ fcext ] ) the part induced by the cylindrical shell , we subtract the fermionic condensate for the geometry of a string without the shell .
as it has been shown before , the latter is given by formula ( [ fc0 ] ) .
for further evaluation of the difference , we use the identities@xmath198where @xmath199 or @xmath200 , and @xmath201 are the hankel functions . as a result for the fermionic condensate we obtain @xmath202 .
\label{fcext1}\end{aligned}\ ] ] now , in the complex plane @xmath65 we rotate the integration contour in the integral over @xmath65 on the right - hand side of formula ( [ fcext1 ] ) by the angle @xmath203 for @xmath204 term and by the angle @xmath205 for @xmath206 term . by using the symmetry properties of the integrands , it can be seen that the parts of the integrals over @xmath207 and @xmath208 are cancelled .
the number of the remained integrations is reduced by using formula ( [ intformula ] ) . in this way , introducing the modified bessel functions , we present the fermionic condensate in the decomposed form ( [ fc4 ] ) , where the part induced by the cylindrical boundary in the region @xmath209 is given by the expression@xmath210 , \label{fccylext}\]]where @xmath162 is defined by eq .
( [ betaj ] ) and we are using notation@xmath211note that the ratio of the combinations of the modified bessel functions in ( [ fccylext ] ) can be written in the form@xmath212 - 2\mu xk_{\beta } ( x)k_{\beta + 1}(x ) } , \label{ibkb}\]]where the function @xmath213 is defined by eq .
( [ wbeta ] ) . for a massless fermionic field one has the formula@xmath214and the fermionic condensate is positive .
let us consider the behavior of the fermionic condensate in asymptotic regions of the parameter . in the limit @xmath215 with fixed values @xmath192
, we introduce in ( [ fccylext ] ) a new integration variable @xmath216 and expand the integrand in powers of @xmath20 .
the main contribution comes from the mode @xmath142 and we have the leading term given below@xmath217 . \label{fcsmalla}\]]for a massless fermionic field , by using the result from @xcite for the integral involving the square of the macdonald function , from here we find@xmath218 at large distances from the cylindrical boundary , for a massive field under the condition @xmath219 the main contribution into the integral in ( fccylext ) comes from the lower limit of the integration and to the leading order we find@xmath220 .
\label{fclarger}\]]here the imaginary part is easily taken by using eq .
( [ ibkb ] ) . as we could expect , in this limit the vev is exponentially suppressed . at large distances and for a massless field
the behavior of the fermionic condensate is described by eq .
( [ fclargerm0 ] ) .
note that the decreasing of the fermionic condensate at large distances is stronger than in the case when the string is absent . for points near the boundary , by using the uniform asymptotic expansions for the modified bessel functions
, we can see that the leading term in the asymptotic expansion of the fermionic condensate over the distance from the boundary is given by the same expression ( [ fcnear ] ) as in the interior region .
the vev for the energy - momentum tensor in the exterior region is found in the way similar to that for the fermionic condensate .
here we omit the details of the calculations and give the final result .
the vev is decomposed into the sum of boundary - free and boundary - induced parts in the form given by ( [ emt1 ] ) . in the region outside the cylindrical shell
the boundary - induced part is ( no summation over @xmath154)@xmath221 .
\label{emtext}\]]in this formula we introduced the notations@xmath222and @xmath223 .
as an additional check we can see that these vevs satisfy the trace relation and the covariant conservation equation . by using formula ( [ ibkb ] ) , we can write the vacuum densities in the form@xmath224 + 2\mu k_{\beta _ { j}+1}^{2}(xr / a)}{k_{\beta _ { j}+1}^{2}(x)+k_{\beta _ { j}}^{2}(x)-2(\mu /x)k_{\beta _ { j}}(x)k_{\beta _ { j}+1}(x ) } , \label{t00ext}\end{aligned}\]]for the energy density and in the form ( no summation over @xmath154 ) @xmath167f_{\beta _ { j}}^{\mathrm{(ex)}(\nu ) } ( x , xr / a)}{k_{\beta _ { j}+1}^{2}(x)+k_{\beta _ { j}}^{2}(x)-2(\mu /x)k_{\beta _ { j}}(x)k_{\beta _ { j}+1}(x ) } , \label{tnuext}\ ] ] for the radial and azimuthal stresses , @xmath168 .
we recall that the summation over @xmath61 in these formulae goes in accordance with eq .
( sumjnot ) . in the case of a massless fermionic field from ( [ emtext ] )
we find the following expressions ( no summation over @xmath154)@xmath225where@xmath226 , \label{f0ext}\]]and for the corresponding functions for the radial and azimuthal stresses we have @xmath227 , @xmath168 .
now we turn to the investigation of the vev in the energy - momentum tensor induced by the cylindrical shell in the exterior region in limiting cases .
first let us consider the limit @xmath215 for fixed values @xmath192 .
expanding the integrands in powers of @xmath20 , we can see that the main contribution comes from the terms with @xmath142 , and the leading terms are given by the expressions ( no summation over @xmath154)@xmath228with @xmath168 .
for a massless field these terms vanish . in this case from eq .
( [ emtextm0 ] ) we find the following leading behavior ( no summation over @xmath154):@xmath229with the coefficients@xmath230for @xmath119 the vevs behave like @xmath231 . at large distances from the cylinder and for a massive field
the main contribution comes from the lower limit of the integral in ( [ emtext ] ) . by using the asymptotic formulae for the macdonald function for large values of the argument
, we find@xmath232 , \label{emtextlargr}\end{aligned}\]]for @xmath219 . as we see , in this limit @xmath233 .
at large distances from the cylinder and for a massless field the asymptotic behavior of the boundary induced parts is given by formula ( [ emtextlargm0 ] ) .
the asymptotic behavior of the vev for the energy - momentum tensor near the cylindrical shell is found in a way similar to that for the interior region and the leading terms are given by the formulae ( [ t22nearcyl ] ) , ( t00nearcyl ) .
hence , near the boundary the energy density and the azimuthal stress in the interior and exterior regions have opposite signs , whereas the radial stresses have the same sign .
the boundary induced parts in the vevs of the energy density and the radial stress for the exterior region are plotted in figures [ fig1]-[fig3 ] as functions of the radial coordinate and the parameters @xmath150 and @xmath234 .
in this paper the vacuum polarization effects are investigated for a fermionic field in the geometry of a cosmic string with a coaxial cylindrical shell .
we have assumed that on the shell the field obeys the mit bag boundary condition . in order to evaluate the fermionic condensate and the vev of the energy - momentum tensor one needs the complete set of normalized eigenspinors satisfying the boundary condition .
this set for the region inside the cylindrical shell is considered in section sec : eigspinor .
the corresponding mode - sums for both fermionic condensate and the energy - momentum tensor contain series over the zeros of the combination ( [ ftilde ] ) of the bessel function of the first kind and its derivative .
for the summation of these series we used a variant of the generalized abel - plana formula previously derived in ref .
this formula allows us to extract from the respective vevs the parts corresponding to the cosmic string geometry without a cylindrical shell and to present the part induced by the shell in terms of exponentially convergent integrals for points away from the boundary . in this way the renormalization procedure for the fermionic condensate and the energy - momentum tensor is reduced to the renormalization of the corresponding quantities in the geometry of the boundary - free cosmic string .
the renormalized vev of the energy - momentum tensor for a fermionic field in the boundary - free geometry is well investigated in literature . in appendix
[ sec : appendix ] , by using the abel - plana summation formula , we give alternative integral representations for both fermionic condensate and the energy - momentum tensor in the case of a massive field . in the region inside the shell ,
the parts in the vevs induced by the presence of the cylindrical boundary are given by formula ( [ fccyl ] ) for the fermionic condensate and by eq .
( [ t00int ] ) , ( [ tnuint ] ) for the vacuum energy densities and stresses .
these formulae are further simplified for a massless fermionic field with the vacuum densities given by eqs .
( fcm0 ) and ( [ emtm0 ] ) . for points near the cylindrical shell
the boundary induced parts in the vevs dominate over the boundary - free parts and diverge on the cylindrical shell .
these type of divergences are well known in quantum field theory with boundaries and are investigated for various bulk and boundary geometries . in the problem under consideration , the leading terms in the asymptotic expansions in powers of the distance from the boundary are given by eq .
( [ fcnear ] ) for the fermionic condensate and by eqs .
( [ t22nearcyl ] ) and ( [ t00nearcyl ] ) for the components of the energy - momentum tensor .
these leading terms do not depend on the planar angle deficit and are the same as for a cylindrical boundary in the minkowski bulk .
the boundary induced parts in the vevs vanish on the string axis for @xmath147 and are non - zero in the case of a cylindrical boundary in the minkowski bulk . since the boundary - free part diverges on the string axis , for points near the string it dominates . for large values of the parameter @xmath235 , which corresponds to a large planar angle deficit ,
the boundary - induced vevs are suppressed by the factor @xmath236 .
the boundary induced vevs have non - trivial dependence on the mass of the field and , as it is illustrated by figure [ fig3 ] , for a massive field the polarization effects can be stronger than for a massless one .
fermionic vacuum densities in the region outside a cylindrical shell with mit bag boundary condition are investigated in section [ sec : exterior ] .
subtracting from the mode - sums the parts corresponding to the geometry of a string without boundaries and by making use of a complex rotation , we have derived explicit expressions for the boundary induced vevs .
the corresponding parts in the fermionic condensate and the energy - momentum tensor are given by eqs .
( [ fccylext ] ) , ( [ t00ext ] ) and ( [ tnuext ] ) .
when the cylinder radius goes to zero , for a fixed value of the radial distance , the boundary induced part in the vev of the energy - momentum tensor vanishes as @xmath237 for a massive field and as @xmath238 for a massless one . at large distances from the cylindrical shell
this part is exponentially suppressed for a massive field and decay as @xmath239 in the case of a massless field .
note that in the latter case the boundary - free part behaves as @xmath240 and it dominates at large distances . for points near the cylindrical shell the leading terms in the asymptotic expansions in powers of the distance from the boundary
are given by the same formulae as for the interior region . in this limit
the total vev is dominated by the boundary induced part . in dependence of the mass , the vacuum stresses can be either positive or negative , whereas the energy density is positive . in the special case @xmath119 , from the formulae derived in the present paper we obtain the fermionic casimir densities for a cylindrical boundary in the minkowski spacetime .
we have considered the idealized geometry of a cosmic string with zero thickness .
a realistic model for cosmic string has a non - trivial structure on a length scale defined by the phase transition at which it is formed . as it has been shown in refs .
@xcite , the internal structure of the string may have non - negligible effects even at large distances . here
we note that when the cylindrical boundary is present , the vevs of the physical quantities in the exterior region are uniquely defined by the boundary conditions and the bulk geometry .
this means that if we consider a non - trivial core model with finite thickness @xmath241 and with the line element ( [ ds21 ] ) in the region @xmath242 , the results in the region outside the cylindrical shell will not be changed .
as regards to the interior region , the formulae given in this paper are the first stage of the evaluation of the vevs and other effects could be present in a realistic cosmic string .
note that from the point of view of the physics in the exterior region the cylindrical surface with mit bag boundary condition can be considered as a simple model of non - trivial string core .
e.r.b.m . and v.b.b .
thank conselho nacional de desenvolvimento cientfico e tecnolgico ( cnpq ) , fapesq - pb / cnpq ( pronex ) and fapes - es / cnpq ( pronex ) for partial financial support .
a.a.s . was supported by the armenian ministry of education and science grant no .
119 and by conselho nacional de desenvolvimento cientfico e tecnolgico ( cnpq ) .
a.s.t . was supported in part by grant nfsat - crdf uc-06/07 .
in this appendix we consider the renormalized fermionic condensate and the vev of the energy - momentum tensor in the cosmic string geometry when the cylindrical shell is absent . for a massless field
the vacuum energy - momentum tensor was found in @xcite .
fermionic propagators for a massive field are considered in refs .
@xcite . in the case of a massive field ,
a representation of the vevs for the energy - momentum tensor in terms of contour integrals is given in @xcite .
here an alternative integral formulae are given by applying to the corresponding mode - sums the abel - plana formula .
we will do these calculations by using the mode - sum formulae ( [ fc0b ] ) and ( [ emts1 ] ) for the boundary - free vevs .
first let us consider the fermionic condensate .
the renormalization is done by subtracting the corresponding quantity for the minkowski background .
the latter is obtained from ( [ fc0b ] ) putting @xmath119 . substituting in the corresponding formulae@xmath243integrating over @xmath77 and @xmath65 , and introducing a new integration variable @xmath244 , we find the following representation of the renormalized fermionic condensate:@xmath245 .
\label{fc01}\end{aligned}\]]next , we apply to the series over @xmath61 the abel - plana summation formula in the form ( see , for example , @xcite ) @xmath246it is easily seen that for the summand in ( [ fc01 ] ) the first integral on the right - hand side of ( [ abel ] ) vanishes and , hence , the divergent parts are explicitly cancelled . introducing the macdonald function
, we arrive to the following expression@xmath247 . \label{fcren}\]]where the notation@xmath248
, \label{gqx}\]]is introduced . in a similar way we can find the formula for the renormalized vev of the energy - momentum tensor . for the renormalized energy density and the azimuthal stress we have the representations given below:@xmath249 , \notag \\
\left\langle t_{2}^{2}\right\rangle _ { \mathrm{s , ren } } & = & \frac{r^{-4}}{\pi ^{2}}\int_{0}^{\infty } dy\,ye^{-m^{2}r^{2}/2y - y}\sum_{\delta = \pm 1}\delta \,\sum_{j}j\left [ q^{2}i_{qj-\delta /2}(y)-i_{j-\delta /2}(y)\right ] .
\label{t00t22free}\end{aligned}\]]by making use of summation formula ( [ abel ] ) to the series over @xmath61 , we find the following formulae for the renormalized vevs:@xmath250 , \notag \\
\left\langle t_{2}^{2}\right\rangle _ { \mathrm{s , ren } } & = & \frac{4}{\pi ^{3}r^{4}}\int_{0}^{\infty } dy\,ye^{-m^{2}r^{2}/2y - y}\int_{0}^{\infty } dxxg(q , x){\mathrm{re}}\left [ k_{ix+1/2}(y)\right ] .
\label{t00sren}\end{aligned}\]]the radial stress is found from ( [ fcren ] ) and ( [ t00sren ] ) by using the trace relation .
formulae ( [ t00sren ] ) are further simplified for a massless fermionic field .
the integration over @xmath251 is done by using the formula@xmath252substituting this into eq .
( [ t00sren ] ) and integrating over @xmath253 , we find @xmath254 in the massless case the radial stress is equal to the energy density .
smith , in _ the formation and evolution of cosmic strings _ , proceedings of the cambridge workshop , cambridge , england , 1989 , edited by g.w .
gibbons , s.w . hawking , and t. vachaspati ( cambridge university press , cambridge , england , 1990 ) .
mostepanenko and n.n .
trunov , _ the casimir effect and its applications _ ( oxford university press , oxford , 1997 ) ; m. bordag , u. mohidden , and v.m .
mostepanenko , phys . rep . * 353 , * 1 ( 2001 ) ; k.a .
the casimir effect : physical manifestation of zero - point energy _
( world scientific , singapore , 2002 ) .
s. leseduarte and a. romeo , commun .
phys . * 193 * , 317 ( 1998 ) ; c.g .
beneventano , m. de francia , k. kirsten , and e.m .
santangelo , phys .
d * 61 * , 085019 ( 2000 ) ; m. de francia and k. kirsten , phys .
d * 64 * , 065021 ( 2001 ) . | the vacuum expectation values of the energy momentum tensor and the fermionic condensate are analyzed for a massive spinor field obeying the mit bag boundary condition on a cylindrical shell in the cosmic string spacetime .
both regions inside and outside the shell are considered . by applying to the corresponding mode - sums a variant of the generalized abel
plana formula , we explicitly extract the parts in the expectation values corresponding to the cosmic string geometry without boundaries . in this way
the renormalization procedure is reduced to that for the boundary - free cosmic string spacetime .
the parts induced by the cylindrical shell are presented in terms of integrals rapidly convergent for points away from the boundary .
the behavior of the vacuum densities is investigated in various asymptotic regions of the parameters . in the limit of large values of the planar angle deficit ,
the boundary - induced expectation values are exponentially suppressed . as a special case
, we discuss the fermionic vacuum densities for the cylindrical shell on the background of the minkowski spacetime .
pacs numbers : 03.70.+k , 98.80.cq , 11.27.+d |
intracellular sodium triggers a cell to progress to cell division @xcite . in many biological activities like creation of daughter cells ,
salt enacts a pivotal role .
it is also a known fact that the two strands of dna carry negative charge due to their phosphate groups ( @xmath0 ) @xcite .
the two strands are connected with each other through the hydrogen bonds between the bases on the opposite strands .
however , the stability of the dna double helix also depends on the charge neutrality of the system . to counter the negative charges on the strands , the cellular environment requires some positive charges in order to nullify the repulsion between negative charges .
such positive counter ions derive from salts present in the body .
the primary salts are @xmath1 and @xmath2 which on reaction with water release @xmath3 and @xmath4 , while the chlorine rearranges itself in the form of hydrogen chloride ( @xmath5 ) .
these cations act as electrostatic screening agents by creating an attractive force between the positive ions @xcite and the negative charges on the strands .
the stability of the dna helix is then governed by the balance between the attractive and repulsive forces exist in the system .
these forces exist in the form of hydrogen bonding between the bases on the opposite strands , the stacking interaction along the strands , the repulsion between same kind of charges and attraction between opposite kind of charges .
various experiments followed by theoretical investigations on the role of salt in screening the repulsive forces in dna revealed many interesting results .
not only the stability of dna molecule @xcite but effects of salt on the @xmath6 transition , and on the condensation of the dna molecule have been studied in detail by several researchers @xcite . in all these studies
the stability of dna has been studied under low or moderate ( 0.1 - 1.0 m ) concentrations of salt .
several experiments , conducted by the groups of owczarzy @xcite and santalucia @xcite revealed that short as well long dna molecules become more stable as the content of cations in the solution increases , with the stability of the structure being determined by calculating the melting temperature ( @xmath7 ) of the molecule . in all these experiments ,
the melting temperature was found to have a logarithmic dependence on the amount of cations present in the solution .
the theoretical work following these experiments attempted to explain this logarithmic dependence , based either on a semi - empirical formula @xcite or on statistical models @xcite .
the counterion theory proposed by manning explains the stability of the dna molecule due to the presence of salt in the solution @xcite .
the stretching and unzipping behaviour of the dna molecule in the presence of cations have also been discussed by several groups @xcite .
these results show that the mechanical stability of the dna molecule also increases with the concentration of cations . in another set of experiments @xcite that were executed at relatively high salt concentration ,
some strikingly different behaviour in the dna molecule was observed .
these experiments found that in this range of salt concentrations the stability of dna gets shattered .
interestingly , in a similar set of experiments , several research groups tested the condensation process and melting profile in dna under different ethanol concentrations @xcite .
one of the findings of these experiments revealed that at a relatively high concentration of ethanol , the melting temperature of the system increases .
the role of cations _ in vivo _ is manifold . in a recent review , bose _
et al _ @xcite showed that abundant cations inside the cell as well as outside the cell play important roles in sustaining cancer cells and at the same time in the decay of immune cells .
cancer cells possess a significant electrical character@xcite compared to normal somatic cells .
cancer cells become more electrostatic @xcite during oncogenesis . keeping in view the important aspect of salt in the cell
, we propose a theoretical description of the stability of dna molecules at higher salt concentration .
we consider one of the dna sequences studied by khimji _
@xcite , who studied the dna duplex stability in crowded polyanion solutions . for the current investigations
we adopt the standard peyrard bishop dauxois ( pbd ) model @xcite and modify the potentials appearing in the model . in our earlier work
, we showed that the pbd model has enough detail to explain the thermal as well as mechanical stability of dna molecules at lower strengths of salt and with varying salt concentration @xcite .
our presentation in the current manuscript is organized as follows . in section [ model ] , we discuss the pbd model and the suitable modifications we did to incorporate the high concentration of cations .
how the stability of dna in thermal as well as in force ensembles varies in a wide range of salt concentrations in solution is discussed in section [ result ] .
we finally conclude our results in section [ concl ] .
we adopt the pbd model to study the response of dna in different ensembles .
we use the linear version of the pbd model .
although this version simplifies the helicoidal geometry of the molecule , it has enough details to delineate the stability of the molecule in thermal as well as in force ensembles @xcite .
the model represents the hydrogen bonding between the bases in a pair through the morse potential while the stacking interaction is represented by an anharmonic potential .
the hamiltonian of the system is , @xmath8 + \sum_{i=1}^{n-1 } v_s(y_i , y_{i+1})\ ] ] where @xmath9_y@xmath10 _ ) represents the momentum of a base pair where @xmath11 is the mass of a base pair and @xmath12 represents the stretching of the @xmath13 base pair .
the stacking interaction between the nearest neighbours is represented by an anharmonic term , @xmath14 , which is , @xmath15.\ ] ] here @xmath16 represents the single strand elasticity .
the anharmonicity in the strand elasticity is taken care of by the term @xmath17 while @xmath18 represents the range of anharmonicity . for our studies we tuned the model parameters and found the values of @xmath19 , @xmath20 as suitable values
the potential @xmath21 describes the interaction between the two bases in a pair @xcite @xmath22 where @xmath23 is the equivalence of the dissociation energy of a pair .
the constant , @xmath24 , represents the inverse of the width of the potential well .
these two parameters have a crucial role in dna denaturation .
the dissociation energy is a representation of the hydrogen bond energy that binds the @xmath25 and @xmath26 .
it is a known fact that the bond strength of these two pairs are not the same but are in an approximate ratio of 1.25 - 1.5 as the @xmath27 pairs have three while @xmath28 pairs have two hydrogen bonds @xcite .
the last term in the hamiltonian is the solvent term @xmath29 .
this is an additional term which simulates the formation of h - bonds with the solvent , once the hydrogen bonds are stretched by more than their equilibrium values .
this is expressed as@xcite @xmath30\ ] ] where @xmath31 is the solvent interaction factor and it reduces the height of the barrier appearing in the potential @xcite . the concentration of salt that stabilizes the dna molecule can be incorporated through the on - site potential term @xcite in the hamiltonian .
the presence of cations around the dna molecule are shown , schematically , in fig .
[ fig01 ] . the schematic representation of cations and dna .
the figure shows the probable distribution of cations around the two negative strands which are standing around 20 apart and the cations trying to nullify the repulsion between the two strands and possible distribution of excess cations that interact among themselves . , width=288,height=192 ] it is a known fact that the cations are required to reduce the repulsion between and along the dna strands so that the dna is stabilised in its double stranded configuration .
however , there is a critical concentration of cations in the solution below which these cations act as shielding agents . in case
the number of cations in the solution crosses this critical number , the repulsive forces in the system dominate over the attractive forces which further shatter the stability of the dsdna .
the overall activities in the surroundings of dna molecule at higher concentration might be complicated to analyse .
however , as an intuitive guess , one can anticipate the behaviour of cations as coloumbic particles .
we are helped by the basics of electrostatics here . from elementary knowledge we know that for a monopole the potential scales as @xmath32 , where @xmath33 is the distance between the charges .
similarly for dipole it scales as @xmath34 , for quadrupole @xmath35 and for octupole as @xmath36 etc @xcite .
we extend this concept and introduce a scaling term that represents the repulsive contribution in the complete potential .
the cation s concentration , @xmath37 around the dna molecule can be mapped with the inter particle distance @xmath33 .
the logic behind this proposal lies in the fact that when @xmath37 increases , @xmath33 will decrease and vice versa . in the pbd model ,
the prime contributor to the stability of the molecule is the potential depth of the morse potential ( @xmath23 ) .
thus , we adopt this parameter ( @xmath23 ) as a function of the salt concentration of the solution . considering all of the above forces that are responsible for the stability of dna molecules , we express the parameter , @xmath23 , as ( in terms of salt concentration ) , @xmath38\ ] ] where the first three terms are taken from our previous work @xcite which show the logarithmic dependence of @xmath7 on salt concentration . here
we make @xmath37 as a dimensionless quantity by choosing the reference value @xmath39 @xcite .
the fourth term is added which takes care of the behaviour of cations in the surrounding of dna molecules at high salt concentrations .
this is a pure coloumbic term .
a point to note is that @xmath40 is an important parameter .
since the nature of cations and their exact distribution is not known , how the coloumbic interactions modify the interaction between the two strands is not precisely known .
at the same time , it is also a fact that cations shield the repulsion between the negative ions .
we are not sure about the nature of the _ pole _ that might be existing due to presence of these cations . the best we can think of is that they obey some power law .
we propose that the potential depth will scale with the concentration of salt as @xmath41 .
this term will have a proportionality factor @xmath42 which can be found by comparing the theoretical results with the experimental findings . at lower values of @xmath37 ,
the logarithmic term will dominate , so the molecule will be more stable with increasing values of @xmath37 . for higher concentration the interaction between the cations will dominate . for our calculations we found the best match for @xmath43 and @xmath44 .
the other potential parameters are @xmath45 ev , @xmath46 , @xmath47 , @xmath48 , @xmath49 .
we take @xmath50 in this problem as a suitable choice for our calculations .
the partition function can be written with the complete hamiltonian of the system like this , @xmath51 dy_i dp_i\ ] ] where @xmath52 . the momentum part can be easily integrated and is equal to @xmath53 .
the configurational partition function is defined as , @xmath54 for the homogeneous chain , one can evaluate the partition function by transfer integral ( ti ) method by applying the periodic boundary condition @xcite . in the case of a heterogeneous chain , with an open boundary ,
the onfigurational part of the partition function can be integrated numerically with the help of the matrix multiplication method @xcite .
the important part of this integration is the selection of proper cut - offs for the integral appearing in eq.6 to avoid the divergence of the partition function .
the method to identify the proper cut - off has been discussed by several groups @xcite . in the calculations by t.s .
et al _ it was shown that the upper cut - off should be @xmath55 50 with our model parameters at @xmath56 k while the lower cut - off is -0.3 . in the earlier work by dauxois and peyrard
it was shown that the @xmath7 converges rapidly with the upper limit of integration @xcite . in this work
the equation for the partition function for an infinite homogeneous chain was solved using the ti method . for a short chain , we calculated @xmath7 for different values of upper cut - offs and found that an upper limit of 200 is sufficient to avoid the divergence of the partition function .
thus the configurational space for our calculations extends from -5 to 200 .
once the limit of integration has been chosen , the task is reduced to discretize the space to evaluate the integral numerically . the space is discretized using the gaussian quadrature formula . in our previous studies
@xcite , we observed that in order to get precise value for melting temperature ( @xmath7 ) one has to choose large grid points .
we found that 900 is a sufficient number for this purpose . as all matrices in eq .
[ eq8 ] are identical in nature the multiplication is done very efficiently .
the thermodynamic quantities of interest can be calculated by evaluating the helmholtz free energy of the system .
the free energy per base pair is , @xmath57 the thermodynamic quantities like specific heat ( @xmath58 ) as a function of temperature or the applied force can be evaluated by taking the second derivative of the free energy .
the peak in the specific heat corresponds to the melting temperature or the critical force of the system .
other quantities like the average fraction @xmath59 of bonded ( or open ) base pairs can be calculated by introducing the dsdna ensemble(dsdna ) @xcite or using the phenomenological approach @xcite .
in general , the @xmath60 is defined as , @xmath61 @xmath62 is the average fraction of strands forming duplexes , while @xmath63 is the average fraction of unbroken bonds in the duplexes .
the opening of long and short chains are completely different . for long chains ,
when the fraction of open base pairs , @xmath64 , goes practically from 0 to 1 at the melting transition , the two strands are not yet completely separated . at this point , a majority of the bonds are in a broken state and the dsdna is denatured .
however , few bonds are still in an intact state , preventing the two strands separating from each other . only at high temperatures
there is a real separation .
therefore for very long chains the double strand is always a single macromolecule through the transition , thus one can calculate the fraction of intact or broken base pairs only . while for short chains , the process of single bond disruption and strand dissociation tends to happen in the same temperature range .
thus , the computation of both @xmath63 and @xmath62 is essential @xcite .
first we discuss the thermal denaturation of the dna molecule . to find the thermal stability of the molecule we calculated the melting temperature with the help of eq .
[ eq9]&[eq10 ] . here
we discus the stability of the short dna sequences that are used by khimji _
et al _ @xcite .
we consider the sequence @xmath65 for our studies .
the peak in the specific heat and in the differential melting curve indicate the melting temperature of the molecule ( see fig .
[ fig02 ] ) .
we calculate the melting temperature of dna for values of salt ranging from 0.05 to 5.0 .
the resultant phase diagram is shown in fig.[fig03 ] . the specific heat as a function of temperature for the 12 base pair chain for the model parameters @xmath66 . the figure is plotted for @xmath67 m.
the peak corresponds to the melting temperature.,width=249,height=220 ] from the figure we find that the first three terms in eq.(4 ) are responsible for the behaviour of the molecule in the low salt region , between 0.01 - 1.0 .
once the concentration is above 1.0 the coloumbic repulsion among the cations increases .
this leads to the instability in the molecule .
as discussed earlier , the parameter @xmath40 plays an important role here .
we calculate the phase diagram for different values of @xmath40 and find the best match for @xmath43 for the experimental results of khimji @xcite .
the maximum deviation is at @xmath68 which is about 0.8 k. thus , we can say that our results are in close match with the experimental findings of @xcite .
the temperature - salt phase diagrams showing the variation of @xmath7 with salt concentration .
the comparison of our results with the experimental results of @xcite , width=268,height=220 ] in order to validate our results for long chains , we increased the length of the molecule by repeating the same sequence . from fig .
[ fig04 ] we find the behaviour of phase diagram is the same with increasing lengths of dna molecule .
however , as the chain length increases the phase diagram for different lengths comes closer .
this indicates the limit of infinite chain length .
we do not have the experimental data for infinite chains at this moment to compare our results .
the phase diagram for different chain lengths in order to validate the results for infinite chain .
the lengths are 36 , 60 , 156 , 252 , 276 bps.,width=268,height=220 ] it is important to note here the difference in the nature of the slope at lower and higher concentrations .
experimental as well as theoretical findings clearly indicate that the stability rate is much sharper than the instability rate . the prime reason behind this difference might be the effect of excluded volume at higher concentrations . as the concentration of the salt increases the number of cations increase correspondingly and therefore the volume available for the movement will lessen .
the cations have charge as well as size and so occupy the flexible volume ( they crowd the dna ) . at higher concentrations although there is a instability due to repulsive forces , there is a stability due to excluded volume . while the movement of cations at lower concentration is faster , at high concentration their movement might be hampered . after evaluating the stability of dna in a thermal ensemble we calculated the stability of the molecule in a force ensemble for
the above mentioned range of salt concentration in the solution .
it is a known fact that the replication process is initiated by the force exerted by dna polymerase on a segment of dna chain .
replication starts at the replication origin @xcite and the replication fork propagates bidirectionally .
mathematically one can model the replication as the force applied on an end of the dna chain @xcite .
the physics of opening the chain in these ensembles is completely different @xcite.whereas in thermal denaturation , the opening is due to an increase in the entropy of the system , for a mechanically stretched dna chain the opening is enthalpic .
the modified hamiltonian for the dna that is pulled mechanically from an end is , @xmath69 + \sum_{i=1}^{n-1}\left[w_s(y_i , y_{i+1})\right ] -f\cdot y_1\ ] ] where the force @xmath70 is applied on the @xmath71 base pair @xcite .
the thermodynamic quantities of interest , from the modified hamiltonian , can be calculated using eq .
[ eq8 ] & [ eq9 ] .
here we consider an infinite chain of 300 base pairs which is generated by repetition of the 12 base pair chain @xcite . _ in vitro , _ experiments on dna unzipping are executed either at constant displacement or at varying loading rates .
although in both the set - ups , the microscopic and dynamic behavior of unzipping are different , the critical force comes out to be the same for an infinite chain ( or in thermodynamic limit ) , _ i.e. _ for @xmath72-phase dna . for theoretical investigation on force
induced unzipping of dna , the chain length of the molecule should be large enough to consider the chain as infinite .
we calculated the melting temperature , @xmath7 for different chain lengths ( in increasing order ) and found that a length of about 300 base pairs is sufficient to consider the chain as an infinite chain @xcite .
all the base pairs of dsdna kept in a thermal bath share equal amount of energy while in the case when the molecule is pulled from an end there is a differential distribution of the applied force ( from the pulling point to the other end of the chain ) .
the critical force - salt phase diagram showing the variation of @xmath73 with salt concentration .
as there is no experimental data is available to compare , we show only the calculated points on the graph.,width=268,height=220 ] the phase diagram for the mechanically pulled dna shows that the force required to open the chain decreases at higher salt concentrations in the solution , as shown , fig.[fig05 ] .
since the molecule itself is unstable at higher salt concentrations , the force required is less .
unfortunately , we do not have any experimental data for force induced unzipping at higher salt values .
we hope that some experiments may be performed to validate these results .
in the current work , we propose a theoretical description of the behaviour of dna molecules at relatively high concentrations of salt .
there are several biological events where the concentration of cations increases in the cell .
for example , in somatic cells the concentrations of sodium ions substantially increase during activation to become cancer cells @xcite . from most of the previous studies
it was concluded that these cations contribute to the stability of the dna in the cell @xcite .
these cations act as shielding agents for the negatively charged strands of dna .
however , there is a range up to which they contribute to the stability of the molecule .
when the concentration exceeds this range , their role in stability may be shattered @xcite .
our work is an attempt to explain the possible mechanism that might be taking place at higher concentrations of salt in the solution .
the volume available in the cell is fixed and there is a limit up to which these cations contribute to the overall charge neutrality of the system .
a greater number of cations in a fixed volume may disrupt the balance between the positive and negative charge forces . through the schematic diagram and by calculating the free energy as a function of temperature and force , we explain the cause of the instability of the dna molecule at higher concentrations .
our results are in very good agreement with the experimental findings @xcite . however ,
due to unavailability of results for infinite chains we can not compare our results for force induced unzipping of dna molecules at high salt concentration .
it would be interesting to calculate the kinetics of duplex unzipping at higher salt concentration .
the movement of cations at higher salt concentrations might also be interesting to analyse .
we are thankful to s. kumar and y. singh , department of physics , banaras hindu university , india , for useful discussions .
we acknowledge the financial support provided by department of science and technology , new delhi under the project grant ( sb / s2/cmp-064/2013 ) . a.s . would like to thank ugc , india for the financial support through bsr fellowship . | intracellular positive ions neutralise negative charges on the phosphates of a dna strand conferring greater strength on the hydrogen bonds that connect complementary strands into a double helix and so confer enhanced stability . beyond a certain value of salt concentration
, the dna molecule displays a unstable nature _ in vivo _ as well as _ in vitro_. we consider a wide range of salt concentrations and study the stability of the dna double helix using a statistical model . through numerical calculations we attempt to explain the different behaviour exhibited by dna molecules in this range .
we compare our results with experimental data and find a close agreement . |
from asteroids to nuclei , fragmentation phenomena can be seen everywhere .
many scientists and engineers have been fascinated by the fragmentation process , and much effort has been made to understand fragmentation @xcite .
in particular , impact fragmentation of brittle solids has been investigated by simulation and experimentally @xcite .
the results have shown that cumulative fragment mass ( or size ) distributions exhibit a power - law dependence , in which the exponent depends on the dimensionality of the fractured objects .
these results do not depend on the details of fragmentation , such as the material the fractured object is made from or the fragmentation method .
consequently , oddershede et al . concluded that brittle fragmentation is a self - organized critical phenomenon @xcite .
cumulative distributions obey power - law dependence when the imparted energy is sufficiently large , but brittle solids do not break when the imparted energy is very small .
therefore , one might assume that there is a transition in the fragmentation phenomenon .
recently , kun and herrmann examined critical behavior in fragmentation transition and concluded that fragmentation transition belongs to the percolation transition universality class @xcite .
in addition , campi has shown that nuclei fragmentation has the same statistical properties as percolation @xcite .
by contrast , strm et al . performed other fragmentation model simulations and obtained a scaling law whose critical exponents differ from those of percolation @xcite .
while there are some simulation results , there have been no experiments on critical fragmentation . for a proper analysis ,
the existing simulation results must be compared to experimental data . here , we experimentally investigate brittle fragmentation as a kind of critical phenomenon , and we report experimental results for glass tube fragmentation near the critical point .
figure [ fig : apparatus ] is a schematic illustration of the experimental apparatus .
we used glass tubes ( @xmath1 mm outside diameter , @xmath2 mm thick , and @xmath3 mm , @xmath4 mm , or @xmath1 mm long ) as the fractured objects .
they have 2-d geometry , and it is easy to impart the impact vertically because they can stand by themselves .
kun and herrmann @xcite and strm et al .
@xcite performed numerical simulation of 2-d samples .
a glass tube was put in a plastic bag , which was in turn placed between a hard stainless steel stage ( dimensions @xmath5 mm @xmath6 @xmath7 mm @xmath6 @xmath1 mm ) and a stainless steel plate ( @xmath8 mm thick ) .
the stage was fixed by placing heavy weights on it .
a cylindrical brass weight with a flat bottom ( @xmath9 kg ) was dropped on the stainless steel plate vertically , and a planar failure wave propagated to the glass tube from the impact circle ( cross section of the glass tube ) .
consequently , the glass tube was cleaved by the impact .
this is a type of 2-d fragmentation .
four poles ( @xmath8 mm diameter ) were used as guide poles for the falling brass weight .
stoppers for the stainless steel plate were placed at each guide pole to prevent a secondary impact .
while the samples were not annealed after the cutting processing , they had almost the flat end .
we used a level in order to check the verticality between the weight and the stainless steel plate at each fragmentation .
the variable in this experiment was the height that the weight was dropped from .
[ 0.4 ] is the falling height of the weight.,title="fig : " ] after fragmentation , we collected the fragments in the plastic bag and measured the mass of each fragment with an electronic balance .
the data were analyzed , and cumulative fragment mass distributions were obtained .
we fractured a total of @xmath10 samples ( @xmath11 mm , @xmath12 mm , and @xmath13 mm length ) .
we defined the falling height @xmath14 as a length from the top of stainless steel plate to the bottom of the falling weight . in our experiments , it was set in the range @xmath15 mm @xmath16 @xmath17 mm .
this range corresponds to @xmath18 ( nm / g ) in terms of the imparted energy ( released potential energy of the dropped weight ) per unit sample mass , @xmath19 .
when the height was too low to break the glass tube , no visible cracks were seen .
once a visible crack was produced , fragmentation occurred suddenly .
we defined the point at which a sample began to cleave as the fragmentation critical point .
we could not obtain samples that had only visible macro - cracks , but did not fragment .
we show some photos of fragments in fig .
[ fig : samples ] .
we observed that with a flat impact , low impact energy fragmentation events produced vertical main cracks . therefore , vertical cleaving produces a few , large fragments ( fig .
[ fig : samples](a ) ) . in the mediate range ,
some large fragments remained ( fig .
[ fig : samples](b ) ) . for large impact energy events ,
the largest fragment was smaller , and it was impossible to distinguish the main cracks from the many small fragments by observation ( fig .
[ fig : samples](c ) ) .
[ 0.65 ] for ( a ) , ( b ) , and ( c ) are @xmath20 , @xmath21 , and @xmath22 ( nm / g ) , respectively.,title="fig : " ] first , we plotted the cumulative mass distribution of each fragmentation .
figure [ fig : cd ] shows some typical cumulative fragment mass distributions , @xmath23 , for @xmath3-mm - long samples .
@xmath23 is defined as follows : @xmath24 where @xmath25 and @xmath26 are the fragment mass and the number of fragments of mass @xmath25 , respectively .
the distribution curves for samples of different sizes were similar to the example in fig .
[ fig : cd ] .
one can see the clear power - law dependence for fully fragmented samples ( far from the fragmentation transition point ) .
the distribution curves near the critical point do not show clear power - law dependence , and they have a rather flat part .
this tendency is quite different from that of percolation universality . in percolation universality
, the clearest power - law dependence should appear beneath the critical point .
in the fully fragmented data curves , we can observe a region that satisfies @xmath27 , i.e. , @xmath28 .
the obtained value of the characteristic exponent @xmath29 is nearly @xmath30 ( @xmath31 ) .
this result conflicts with the scaling ansatz of percolation @xcite .
therefore , we believe fragmentation criticality to differ from that of percolation .
conversely , our result @xmath32 concurs with strm s result @xcite .
in addition , the value of @xmath29 obtained in this experiment is consistent with hayakawa s scaling for 2-d fragmentation with a planar failure wave @xcite . [ 1.0]-mm - long samples .
power - law behavior is confirmed in the sufficiently high impact energy curves , while the distribution curves for low impact energy do not show a clear trend .
the corresponding values of @xmath33 for plus , diamond , triangle , square , and circle marks are @xmath34 , @xmath35 , @xmath36 , @xmath37 , and @xmath38 , respectively.,title="fig : " ] since fragmentation events near the fragmentation critical point have less fragments , they can not be analyzed statistically using only @xmath23 .
therefore , we used the @xmath39-th order moment of the fragment mass distribution , @xmath40 , defined as @xmath41 since we are interested in fragmentation criticality , the height of the falling weight was controlled near the critical point . despite our control of the height ( imparted energy ) ,
the @xmath40 data fluctuate too much due to individual differences in the glass tube samples , e.g. , the initial density of micro - cracks and residual stress distributions .
when we considered the imparted energy as a control parameter , we could not find clear critical behavior quantitatively .
for example , we show a log - log plot of @xmath42 vs. @xmath19 in fig . [
fig : am](a ) .
where , @xmath42 is the weighted mean fragment mass .
it is hard to say the data in fig .
[ fig : am](a ) are particularly convincing about showing linear behavior .
we could not consider that power - law fitting was appropriate with confidence .
alternatively , we used multiplicity @xmath33 analysis , which campi introduced to the analysis of nuclei fragmentation @xcite .
multiplicity is defined as @xmath43 where @xmath44 , @xmath45 , and @xmath46 correspond to the number of fragments , the total mass of the fragments , and the smallest cut - off fragment mass , respectively .
we fix @xmath47 for all the analyses .
this value usually corresponds to the smallest 2-dimensionality for the samples .
according to this definition , the fragmentation critical point corresponds to @xmath48 .
while the multiplicity is a resultant parameter , rather than a control parameter , it can be considered the control parameter that incorporates individual differences in the samples . furthermore , because the multiplicity is easy to measure and calculate , it has been applied to other fragmentation systems @xcite .
we use @xmath33 as a pseudo - control - parameter in this paper .
certainly , @xmath33 and the imparted energy are positively correlated , roughly .
the larger the impact energy imparted , the larger @xmath33 becomes .
[ 1.0 ] : ( a ) @xmath42 vs. @xmath19 .
( b ) @xmath42 vs. @xmath33 .
the solid line indicates the form of the power - law @xmath49 ( @xmath50).,title="fig : " ] [ 1.0 ] : ( a ) @xmath42 vs. @xmath19 .
( b ) @xmath42 vs. @xmath33 .
the solid line indicates the form of the power - law @xmath49 ( @xmath50).,title="fig : " ] in fig .
[ fig : am](b ) , we plot @xmath42 vs. @xmath33 of all the fragmentation event data .
it indicates the tendency toward divergence near the critical point .
the criticality appears independent of sample size .
all the data match the same power - law line .
hence , we assume that the relationship between the weighted mean fragment mass @xmath42 and the multiplicity , @xmath33 , holds as , @xmath51 the obtained value of @xmath52 is @xmath53 .
this value is nontrivial because @xmath54 .
we discuss the possibility of scaling @xmath23 curves using @xmath52 . for large @xmath33 data ,
@xmath23 curves have the same power - law , in which the power is @xmath32 .
this value is independent of the size of the samples .
this similarity allows us to collapse the @xmath23 curves of different - sized samples into a unified scaling function .
we expect the following scaling for distribution curves : @xmath55 where @xmath56 is the probability distribution of fragments whose masses are larger than @xmath25 .
we show the scaling result in fig .
[ fig : sf ] with the approximation @xmath57 .
the curves in fig .
[ fig : sf ] include the results for samples of three different sizes ( 11 curves of large @xmath33 data ) .
figure [ fig : sf ] shows good collapse of @xmath23 curves into the scaling function @xmath58 and @xmath59 .
the function @xmath58 has a clear power - law regime , in which the power is @xmath0 , and a unified cut - off scale .
the former corresponds to the scaling region and the latter corresponds to exponential decay due to the finite size effect .
accordingly , the scaling function @xmath59 has a constant part and a rapidly decaying part .
the flat part in @xmath59 exists only for the fully fragmented data , because clear power - law form @xmath60 can be satisfied with well fractured samples .
the form of scaling function might change with @xmath33 , however , we consider they will not depend on the sample size
. we could not confirm the clear collapse for small @xmath33 .
more detailed experiments are necessary to verify the scaling functions in wide @xmath33 range .
[ 1.0 ] and @xmath59 approximated using ( a ) @xmath61 and ( b ) @xmath62 .
the values of @xmath29 and @xmath52 are obtained from scaling figs .
[ fig : cd ] and [ fig : am](b ) as @xmath30 and @xmath63 , respectively.,title="fig : " ] [ 1.0 ] and @xmath59 approximated using ( a ) @xmath61 and ( b ) @xmath62 .
the values of @xmath29 and @xmath52 are obtained from scaling figs .
[ fig : cd ] and [ fig : am](b ) as @xmath30 and @xmath63 , respectively.,title="fig : " ] bak et al . reported a similar scaling function for earthquake statistics @xcite . while they discussed the time correlation of earthquakes ,
their scaling function is analogous to ours .
they concluded that the unified scaling function for earthquakes depends only on the number of earthquakes occurring within the area and period considered . in eq .
[ eq : sf ] , our scaling function depends on the variable @xmath64 , which is the normalized mass of the fragments . moreover , the scaling eq .
[ eq : sf ] resembles the scaling function strm et al . obtained @xcite .
next , we discuss the behavior of the higher - order moments of the fragment mass distributions .
we can easily expand eq .
[ eq : sigmadef ] and introduce the higher - order weighted mean fragment mass scaling as , @xmath65 the open circles in fig .
[ fig : sigma ] show the @xmath66 values obtained from fitting the all experimental data like fig .
[ fig : am](b ) . while the value @xmath67 is the trivial value @xmath68 , @xmath69 approaches a nontrivial value of approximately @xmath70 .
[ 1.0 ] ( @xmath71).,title="fig : " ] in order to understand this behavior , we consider a simple biased cascade model .
let us set a unit mass fragment initially .
first , this unit mass is divided into two pieces of mass @xmath72 and @xmath73 @xmath74 .
the same @xmath72-biased partition occurs for each fragment in the next step .
this cascade continues until the imparted energy dissipates . from the above definition and eq .
[ eq : sigmakdef ] , the following equation holds at each step : @xmath75 substituting @xmath76 into eq .
[ eq : biasedef ] , the trivial value @xmath77 is obtained for all @xmath39 .
this case corresponds to mono - scaling critical fragmentation . in fig .
[ fig : sigma ] , however , @xmath66 varies with @xmath39 .
this indicates a multi - scaling property of critical impact fragmentation . the solid curve in fig .
[ fig : sigma ] is the value calculated using eq .
[ eq : biasedef ] with @xmath71 .
the curve and experimental data roughly agree within the error bars .
although this model is very simple , it explains the trend of the higher - order behavior .
we conclude that fragmentation near the critical point has a multi - scaling nature .
however , this only alters the problem .
the origin of the symmetry breaking given with @xmath72 and the relation between @xmath72 ( or @xmath66 ) and @xmath29 remain unsolved
. the cascade model does not seem to yield the appropriate value of @xmath29 ( @xmath78 ) directly .
since this model is too simple , it needs to be improved using more detailed analyses and experiments .
similar multiplicative model was proposed for the energy cascade of eddy by meneveau and sreenivasan @xcite . according to their result
, the cascade of division into @xmath79 and @xmath80 can explain the energy cascade of turbulent flow .
however , the origin of the value @xmath79 has remained unsolved .
the value @xmath79 slightly agrees the value @xmath81 we obtained .
kun and herrmann have analyzed the largest fragment mass as an order parameter @xcite . in our experiments ,
the largest fragment mass had a very wide distribution , as seen in fig .
[ fig : sf ] , while the weighted mean fragment mass behaved more calmly .
figure [ fig : cd ] suggests that the power - law exponent might change with @xmath33 .
ching et al .
reported the impact energy dependence of the power - law exponent @xcite .
their interest focused on the very large imparted energy region and not on the neighborhood of the fragmentation critical point .
in addition , if we fit the power - law form to all @xmath23 curves , the results involve large uncertainty , and do not show a clear trend . ishii and matsuhita reported that the distribution function for the low impact energy region has a log - normal form @xcite .
we have not investigated the distribution of the concrete function form .
the relationship between the two energy regions and the study of the concrete function form remain open to study .
our experiments were focused on the 2-d fragmentation with the _ flat _ impact .
kun and herrmann investigated the _
impact 2-d fragmentation @xcite .
this might be a reason of the discrepancy between our results and their ones .
hayakawa suggested that the universality of the fragmentation depends not only on the dimensionality of fractured object , but also on the propagation dimensionality of the failure wave @xcite .
strm et al .
performed numerical simulations of 2-d fragmentation with the _ flat _ expansion @xcite .
their results resemble ours in many points as described above .
we believe that our results are valid for 2-d fragmentation by the _ flat _
impact to one side .
other dimensional objects or other failure wave propagation manner may produce other values of scaling exponents such as @xmath29 and @xmath66 .
moreover , scaling - law among these exponents may exist . as written in sec .
[ sec : introduction ] , the exponent @xmath29 is independent of the material of fractured objects in fully fragmented state . we guess the other critical exponents are also independent of that , however , it should be examined .
these problems are open questions . in general , the low impact energy region is difficult to study .
small differences of initial condition are enlarged by nonlinearity of fragmentation dynamics .
figure [ fig : am](a ) presents this difficulty well . even in fig .
[ fig : am](b ) , the fluctuating data can be seen .
some unrealized experimental parameters might not be controled .
we think that the coincidence between the experimental result and the model shown in fig .
[ fig : sigma ] is rough due to this reason . in conclusion
, we investigated the 2-d impact fragmentation of brittle solids experimentally .
the measured characteristic exponents imply that the impact fragmentation of brittle solids does not belong to the percolation transition universality class .
instead , we found that the weighted mean fragment mass @xmath42 and the multiplicity parameter @xmath33 are related to the nontrivial scaling exponent @xmath52 .
the cumulative fragment mass distributions of different - sized samples could be collapsed into the unified scaling function in the large @xmath33 regime .
we calculated the generalized weighted mean fragment mass scaling @xmath82 .
the behavior of the multi - scaling exponent @xmath66 was modeled using a simple biased cascade partition model .
the critical behavior of other fragmentation systems should be studied in order to understand the universality of fragmentation phenomena in detail .
experiments with other materials , such as ceramics , or with samples of different dimensions may prove interesting .
c. meneveau and k. r. sreenivasan , phys .
lett . * 59 * , 1424 ( 1987 ) ; c. meneveau and k. r. sreenivasan , in _ physics of chaos and systems far from equilibrium _ , edited by minh - duong van and b. nichols ( north - holland , amsterdam , 1987 ) . | we investigated two - dimensional brittle fragmentation with a flat impact experimentally , focusing on the low impact energy region near the fragmentation - critical point .
we found that the universality class of fragmentation transition disagreed with that of percolation .
however , the weighted mean mass of the fragments could be scaled using the pseudo - control - parameter _
multiplicity_. the data for highly fragmented samples included a cumulative fragment mass distribution that clearly obeyed a power - law .
the exponent of this power - law was @xmath0 and it was independent of sample size .
the fragment mass distributions in this regime seemed to collapse into a unified scaling function using weighted mean fragment mass scaling .
we also examined the behavior of higher order moments of the fragment mass distributions , and obtained multi - scaling exponents that agreed with those of the simple biased cascade model . |
after more than 40 years since the seminal ideas coined by a handful of theoretical pioneers@xcite , our present understanding of the electroweak symmetry breaking ( ewsb ) phenomenon through the higgs ( englert - brout and guralnik - hagen - kibble ) mechanism is still rather incomplete and experimentally inconclusive . on the one hand
, there is no compelling alternative to consistently embed the ewsb mechanism into the quantum field theoretical description of particle physics offered by the in so many regards successful standard model ( sm ) of the strong and electroweak interactions . on the other hand
, we do not have a single phenomenological hint of the existence of elementary scalar fields , not to mention the fact that we do not understand how to make compatible the ewsb mechanism and its associated vacuum energy with fundamental problems of different scope , for example in the domain of cosmology .
still , the possibility to describe the inner theoretical structure of the sm through the ewsb mechanism is so successful within the restricted particle physics domain that it would not be wise , not even advisable , to cease our pursue of the phenomenological implications of the higgs mechanism paradigm till the limits of the current experimental possibilities .
it goes without saying that the quest for experimental evidences of the higgs boson is a most preeminent milestone of the upcoming generation of collider facilities .
nonetheless , the future data might well reveal that the purportedly found higgs boson actually belongs to a richer model structure , which might be grounded somewhere beyond the minimal conception of the sm , namely of a single , spinless , fundamental constituent of matter .
if so , a few more fundamental spinless constituents could appear .
a particularly well - motivated extension is the two - higgs - doublet model structure encompassed by the minimal supersymmetric standard model ( mssm)@xcite . here the physical higgs boson spectrum contains a couple of @xmath17-even ( @xmath18 , @xmath19 ) , one @xmath17-odd ( @xmath20 ) and two charged higgs bosons ( @xmath21 ) , and the corresponding higgs potential is highly constrained by the underlying supersymmetry ( susy ) .
one particular consequence of the latter is that susy invariance restricts all higgs boson self - interactions to be gauge - like . while this represents an economy from the point of view of the number of couplings in the potential , it is strongly imbalanced by the exceeding number of parameters ( mixings and masses ) populating the other sectors of the mssm .
in addition , the pure gauge nature of the higgs boson self - couplings make them highly inconspicuous from the practical point of view , in the sense that they are unable to trigger any outstanding phenomenological signature .
the core of the enhancement capabilities of the mssm lagrangian resides , instead , in the multifarious pattern of yukawa couplings between the higgs bosons and the quarks , as well as between quarks , squarks and charginos / neutralinos .
the rich interplay of opportunities that they give rise to has been extensively analyzed in the past within a plethora of processes , see e.g. @xcite , and also @xcite for reviews on the subject .
the two - higgs @xmath22-doublet structure of the higgs sector in the mssm constitutes a genuine prediction of the susy dynamics .
nonetheless , a more general architecture can appear in the form of a non - susy framework through the so - called general ( unconstrained ) two - higgs - doublet model ( 2hdm)-conserving higgs potential .
as usually done in practice@xcite , we impose an additional ( softly - broken ) @xmath23 discrete symmetry @xmath24 , where @xmath25 denote each of the @xmath22 higgs doublets , as a sufficient condition to banish the tree - level flavor changing neutral currents . ] .
although they share a common higgs boson spectrum , the potentially most distinctive phenomenological features of both models are located in very different sectors .
the most relevant observation here is that , in the absence of an underlying susy , the triple ( 3h ) and quartic ( 4h ) higgs - boson self - interactions are no longer restrained to be purely gauge .
this can have a tremendous impact e.g. in the physics of the top quark in hadron colliders , as it was shown long ago in @xcite , and it can also trigger significant neutral flavor - changing interactions@xcite that can perfectly compete with the susy ones@xcite .
much attention has been devoted to higgs boson production and decay in hadron colliders , extending from the ongoing tevatron facility at fermilab to the brand new lhc collider recently operating at cern @xcite .
however , precision higgs boson physics will greatly benefit from the interplay@xcite of the forthcoming generation of linear colliders ( linac ) , such as the ilc and clic projects@xcite . here
a variety of processes can provide new clues to higgs boson physics , e.g. the production of triple higgs - boson final states , both in the mssm @xcite and in the general 2hdm @xcite ; the double higgs - strahlung channels @xmath26 @xcite ; and the inclusive higgs - pair production via gauge - boson fusion@xcite . in the same vein , also
the @xmath27 mode of a linac has been explored@xcite , in particular the loop - induced production of a single neutral higgs boson @xcite and of a higgs boson pair @xcite .
in all the above mentioned cases , promising signatures were singled out and illustrate that , if effectively realized in nature , such hints of a generic 2hdm structure could hardly be missed in the superbly clean environment of a linac . and , what is more , they could not be confused as having a susy origin , because of the intrinsically different nature of the higgs self - interaction sector .
besides , outstanding fingerprints of a generic 2hdm could also be stamped in the pattern of radiative corrections to higgs production processes ; for instance , quantum effects on the cross - sections for two - body higgs boson final states : @xmath28 have been attentively investigated in the mssm @xcite . as for the general 2hdm
, the efforts were first concentrated on the production of charged higgs pairs @xcite .
this program has been recently brought to completion from a full - fledged study of the quantum effects on the production cross - sections in the neutral higgs sector @xcite . alongside these processes
we find the more traditional higgs - strahlung events , in which a higgs boson is produced in association with the @xmath0 : @xmath29 ( notice that the @xmath30 final state is forbidden by @xmath31-conservation . )
processes ( [ eq : hz ] ) are complementary to the ( [ 2h ] ) ones cf .
@xcite , and also @xcite for a phenomenological analysis in the mssm context .
let us also recall that the last available limit on the sm higgs boson mass , placed by lep searches , comes precisely from investigating the `` bjorken process''@xcite , i.e. @xmath32 followed by @xmath33 , with the result : @xmath34 gev@xcite . in this article , we aim at discussing the corresponding one - loop corrections to the generalized bjorken processes within the 2hdm .
however , in contrast to the original sm case , where the radiative corrections are small , the quantum effects on the processes ( [ eq : hz ] ) can be large and are mainly driven by the 3h self - couplings .
in fact , our main aim here is to identify the regions of the parameter space where this is so , and then quantify the impact of the potentially enhanced 3h self - couplings on the final cross - sections . in this way , joining this study with that of ref .
@xcite on the pairwise higgs boson production channels ( [ 2h ] ) in the 2hdm , a rather complete panorama of the genuine quantum effects associated to the basic neutral higgs boson production channels becomes available .
a good deal of attention has been devoted in the literature to multiple properties of the higgs - strahlung process @xmath35 in the sm , see e.g. @xcite and references therein .
not surprisingly it was one of the gold - plated channels for higgs boson search at lep .
concerning the mssm , the higgs - strahlung channels have also been discussed extensively cf .
@xcite and part two of the review @xcite and are currently under investigation also in the mssm with @xmath17-violating phases@xcite ( the so - called complex mssm@xcite ) .
the upshot of these analyses spotlights the following features : i ) the dominant source of corrections at one - loop originate from the higgs boson propagators , which can be reabsorbed into an _ effective _ ( loop - corrected ) mixing angle @xmath36 ( equivalently , a more generic mixing matrix in the complex mssm case ) ; ii ) the corrections to the @xmath37 vertex are in general small ,
although they can reach the level of @xmath38 for very low ( or high ) values of @xmath39 , precisely in the regions where the higgs yukawa couplings to heavy quarks become enhanced ; iii ) the electromagnetic corrections to the initial state with virtual photonic corrections and initial - state radiation do not differ from the sm case , being in general large and positive ( except near the production threshold ) . in the present article ,
our endeavor is to seek for the genuine phenomenological imprints associated to the generic ( unconstrained ) 2hdm dynamics , most particularly to the potentially enhanced 3h self - couplings .
let us recall that the general 2hdm@xcite is obtained by canonically extending the sm higgs sector with a second @xmath22 doublet with weak hypercharge @xmath40 , so that it contains @xmath41 complex scalar fields .
the free parameters @xmath42 in the general , @xmath17-conserving , 2hdm potential can be finally expressed in terms of the masses of the physical higgs particles ( @xmath43 , @xmath44 , @xmath45 , @xmath46 ) , @xmath47 ( the ratio of the two vev s @xmath48 giving masses to the up- and down - like quarks ) , the mixing angle @xmath49 between the two @xmath31-even states and , last but not least , the self - coupling @xmath9 , which can not be absorbed in the previous quantities . therefore we end up with a @xmath50-free parameter set , to wit : @xmath51 , @xmath44 , @xmath45 , @xmath52 , @xmath53 , @xmath39 , @xmath54 . furthermore , to ensure the absence of tree - level flavor changing neutral currents ( fcnc ) , two main 2hdm scenarios arise : 1 ) type - i 2hdm , in which one higgs doublet couples to all quarks , whereas the other doublet does not couple to them at all ; 2 ) type - ii 2hdm , where one doublet couples only to down - like quarks and the other doublet to up - like quarks .
the mssm higgs sector is actually a type - ii one , but of a very restricted sort ( enforced by susy invariance)@xcite .
we refer the reader to ref .
@xcite for a comprehensive account on the structure of the model and for notational details . in particular ,
table ii of that reference includes the full list of trilinear couplings within the general 2hdm that are relevant for the present calculation .
further constraints must be imposed to assess that the sm behavior is sufficiently well reproduced up to the energies explored so far .
most particularly , we take into account i ) the approximate @xmath55 custodial symmetry , which can be reshuffled into the condition @xmath56@xcite ; and @xmath57 the agreement with the low - energy radiative @xmath58-meson decay data ( which demand @xmath59 gev for @xmath60 @xcite in the case of type - ii 2hdm ) .
additional requirements ensue from the theoretical consistency of the model , to wit i ) perturbativity @xcite ; ii ) unitarity @xcite ; and iii ) stability of the 2hdm vacuum @xcite .
let us expand a bit more on the latter conditions , which turn out to play a crucial role in our analysis .
as for the unitarity bounds , and following ref .
@xcite , the basic underlying strategy is to compute the s - matrix elements @xmath61 for the possible @xmath62 processes involving higgs and goldstone bosons in the 2hdm , and to subsequently restrain the corresponding eigenvalues @xmath63 by the generic condition @xmath64 .
the latter requirement translates into a set of upper bounds on the quartic scalar couplings @xmath65 and hence on combinations of higgs masses , trigonometric couplings and , most significantly , the parameter @xmath9 .
as far as vacuum stability restrictions are concerned @xcite , they may be written as follows : @xmath66 > 0 \label{eq : vacuum_conditions}.\end{aligned}\ ] ] more refined versions of these equations may be obtained from the renormalization group ( rg ) running of the quartic couplings @xmath67 at high energies .
nonetheless , for our current purposes we need not assume any particular uv completion of the 2hdm .
this would be an unnecessary additional assumption at this stage of the phenomenological analysis of our processes .
therefore , we are not tied to any specific uv - cutoff ( which could be , in principle , as low as @xmath68 ) . in this sense
, we may just apply eq . with all @xmath42 taken at the ew scale .
this is after all the scale at which we fix all our input parameters and perform the phenomenological analysis ( including the renormalization ) of the processes under study . by the same token
, the ( cutoff dependent ) triviality bounds are also circumvented .
the latter kind of restrictions only impose tight upper limits on the higgs boson masses when a very large uv - cutoff ( e.g. the planck scale or the gut scale ) is considered .
our choices of higgs boson mass spectra ( cf .
table [ tab : masses ] ) lie , in any case , in a sufficiently low mass range so as to conform with the typical mass requirements allowed by triviality @xcite even for large cutoff scales .
more details on the constraints setup are provided in ref .
@xcite .
c + + + + + + the dynamics of the higgs - strahlung processes under consideration is driven at leading order by the tree - level interaction lagrangians : @xmath69 where @xmath70 and @xmath71 for the electroweak mixing angle @xmath72 . since @xmath73 and @xmath74 , it is clear that the processes ( [ eq : hz ] ) are complementary to the higgs pair production ones ( [ 2h ] ) . furthermore , as all these couplings ( [ eq : lag0 ] ) are generated by the gauged kinetic terms of the higgs doublets ( cf .
( 30 ) of ref .
@xcite ) , they are fully determined by the gauge symmetry and hence show no intrinsic difference in the 2hdm as compared to the mssm . at the end of the day , this is the reason why a tree - level analysis of these events is most likely insufficient to disclose their true nature .
similarly , in the limit @xmath75 the @xmath76 coupling coincides with the analogue coupling in the sm , @xmath77 .
it is , therefore , the pattern of radiative corrections the characteristic signature associated to each one of the possible models ; most particularly , it should help to disentangle susy versus non - susy extended higgs physics scenarios .
the leading - order @xmath78 scattering amplitude follows from the @xmath79-channel @xmath0-boson exchange and renders @xmath80 here @xmath81 and @xmath82 refer to the 4-momenta and helicities of the electron and positron , and @xmath83 is the polarization 4-vector of the @xmath84 gauge boson with 4-momentum @xmath85 and helicity @xmath86 .
we have introduced also the left and right - handed weak couplings of the @xmath0 boson to the electron , @xmath87 , @xmath88 , and the left and right - handed projectors @xmath89 .
let us notice that we do not include the finite @xmath0-width corrections , since they are completely negligible for the center - of - mass energies that we consider here .
finally , the total cross section @xmath90 at the tree - level is obtained after squaring the matrix element , performing an averaged sum over the polarizations of the colliding @xmath91 beams and the outflowing @xmath0 boson , and integrating over the scattering angle .
the calculation of @xmath90 at one loop is certainly much more cumbersome . to start with , it is uv - divergent and it thus requires of a careful renormalization procedure in order to render finite results .
we adopt here the conventional on - shell scheme in the feynman gauge @xcite . in the mssm
, these cross - section calculations can be mostly carried out in an automatized fashion through standard algebraic packages which allow a rapid and efficient analysis of the electroweak precision observables@xcite .
several public codes are available , see e.g.@xcite . however , in our case the calculation is non - supersymmetric and we must deal with the renormalization of the higgs sector in the class of generic 2hdm models . a detailed description of the renormalization procedure for the 2hdm higgs sector in the on - shell scheme has been presented in @xcite and we refer the reader to this reference for all the necessary details .
we will make constant use of the framework described in this reference and sometimes we will refer to particular formulae of it .
[ cols="^ " , ] a rather comprehensive survey of the predicted cross - sections for different values of the tree - level coupling and different higgs mass setups is presented in table [ tab : xs500 ] .
we display the results for the one - loop corrected cross - sections ( in fb ) , together with the relative radiative corrections ( cf .
eq . ) , for both neutral higgs - strahlung channels @xmath1 and @xmath2 .
once more , we set @xmath92 ; and we work at @xmath93 and maximum allowed @xmath94 , since we are mostly interested in spotlighting the imprints of the 3h self - couplings in the quantum effects associated to the higgs - strahlung mechanism .
let us also recall in passing that @xmath95 maximizes the tree - level @xmath76 coupling , whilst @xmath96 is optimal in the complementary regime , @xmath97 . from the table we may read out that radiative corrections in these regimes are certainly large , regardless of the details of the chosen mass spectrum , the tree - level coupling and the actual channel under consideration . in a nutshell : the characteristic signature of the enhanced 3h self - couplings in the pattern of quantum effects on the higgs - strahlung processes is rather universal and manifests in the form of a substantial depletion of the tree - level signal ( typically in the range of @xmath98 ) .
fortunately , even under such a dramatic suppression the final loop - corrected cross sections stay at the level of a few tens of fb thereby amounting to a non - negligible yield of @xmath99 events per @xmath100 .
a note of caution should be given at this point in regard to some particular configurations of the @xmath2 channel .
it turns out that certain choices of higgs boson masses ( e.g. for mssm - like mass splittings ) may lie close , or simply beyond , the kinematical threshold for the decay process @xmath101 ; that is to say , @xmath102 . if we then consider a regime wherein @xmath97 and sizable @xmath94 , then the @xmath19 wf - correction undergoes a remarkable boost , which results from the combination of two independent sources of enhancement , namely i ) the actual strength of the @xmath103 coupling which is maximally enlarged in this regime ; and ii ) the kinematical enhancement due to the vicinity of the @xmath104 threshold , which is reflected as a sharp peak in @xmath105
however , these @xmath106 wf - corrections become so large that the perturbative formula is no longer valid and we have to keep the original ( unexpanded ) expression @xmath107^{-1/2}$ ] . out of such corner in the parameter space ,
the higher - order effects involved in the above resummed form of the wf - corrections become harmless and totally inconspicuous and can be safely discarded . if , alternatively , we were considering a mass setup such that @xmath108 , then the decay @xmath101 would be open and , in the scenario of maximum @xmath103 coupling , it would furnish a very large width for @xmath106 . in this particular setup
, the process would effectively boil down to the double higgs - strahlung channels @xmath109 previously explored in the literature @xcite .
the following comment is in order to clarify the role played by the remaining ( potential ) sources of enhancement .
we have seen that the various constraints restrict very significantly the range of allowed values of @xmath39 , and enforce it to stay around @xmath110 when @xmath94 is maximum in the region @xmath111 . as a result
, the contributions from the top quark and bottom quark yukawa couplings can not be augmented in the domain where the trilinear couplings are maximal .
remarkably enough , let us also emphasize that they can not be enhanced even in the region where @xmath94 is small or zero . to see why , notice that although in such region the parameter @xmath39 can be very large or small and still be compatible with the constraints ( cf .
fig.[fig : iso ] ) , then all the terms in the trilinear couplings ( cf .
table ii of ref.@xcite ) which are not proportional to @xmath9 become of order one ( i.e. can not be promoted to high values ) in the regime where the tree - level production cross - section for @xmath112 or @xmath106 is optimal .
not only so , in the very same regime the yukawa couplings of both the top and bottom quarks can not be enhanced either .
one can easily check all these features explicitly by observing that , in the regions where the corresponding tree - level processes are maximal , the @xmath39-enhancements are canceled in all the couplings .
we have checked numerically that the residual corrections ( positive and negative ) are merely of a few percent for any value of @xmath39 . at the end of the day , we conclude that the only sizeable and eventually measurable quantum effects on the processes under study are those stemming potentially from the enhancement of the @xmath9 coupling , not from the yukawa couplings .
let us finally emphasize that the complementarity between the neutral final states @xmath1/@xmath15 and @xmath2/@xmath16 , i.e. processes ( [ 2h ] ) and ( [ eq : hz ] ) , is a unique chance for analyzing potentially big correlations between quantum effects in the 2hdm production cross sections , as there is no similar opportunity in the charged sector . indeed , the final state charged counterparts @xmath113 are suppressed owing to the fact that the @xmath114 vertices are forbidden at the tree - level in any 2hdm extension of the sm@xcite and therefore they can only be studied in more general extensions of the higgs sector@xcite , or through loop - induced @xmath113 vertices in the 2hdm@xcite and in the mssm@xcite . in these loop - induced mechanisms , the cross section for the associated @xmath113 production in a linear collider is rather meager generally below @xmath110 fb at the startup ilc energy ( for charged higgs masses comparable to the ones we have considered for the @xmath1/@xmath15 production ) and within the region allowed by the current constraints .
we therefore deem quite difficult to use the @xmath113 channel to extract additional information . in our opinion
, the main task should be focused on performing precision tests of the @xmath1/@xmath15 final states , together with the @xmath2/@xmath16 ones ( if the mass of the heavy @xmath31-even higgs is not too large ) .
in this article , we have concentrated on the analysis of the production of neutral higgs bosons in association with the @xmath0 gauge boson at the future linac facilities within the framework of the general ( non - supersymmetric ) two - higgs - doublet - model ( 2hdm ) .
our basic endeavor has been twofold : i ) on the one hand to study the impact of radiative corrections to the final predicted rates ; and ii ) on the other hand to correlate such quantum effects with the enhancement potential of the 3h self - interactions , which are a genuine dynamical feature of the 2hdm in the sense that it is unmatched to its supersymmetric counterpart .
the upshot of our analysis singles out sizable ( although negative ) quantum effects which are correlated to the enhancement properties of the 3h self - interactions in the general 2hdm . such large effects are identified fundamentally in the region of parameter space with @xmath115 and where the coupling @xmath94 is at its maximum possible value compatible with the various constraints .
the quantum effects reach typically @xmath116 , for @xmath117 ( @xmath10 ) , this being the crucial parameter that tunes the actual size of the 3h self - couplings .
let us stress that the most stringent limits on @xmath9 are placed by the conditions of perturbative unitarity and vacuum stability - meson physics constraints over the 2hdm parameter space suggests that values of @xmath8 could be disfavored for charged higgs masses @xmath118 too near to the lowest mass limit of @xmath119 gev ( and certainly below ) @xcite .
however , the level of significance is not high and , in addition , our leading quantum corrections are basically insensitive to the charged higgs mass .
therefore , a shift of @xmath118 slightly upwards restores the possibility of @xmath8 at , say , @xmath120 without altering significantly our results ( as we have explicitly checked ) . ] .
although the vertex corrections to the @xmath121 couplings are sensitive to such 3h self - interactions through higgs - mediated one - loop diagrams , the most significant quantum effects are induced by the wave - function renormalization corrections to the @xmath122 higgs boson external lines .
these ( negative ) effects triggered by the trilinear couplings are of order @xmath123 , and are therefore very responsive to changes in the value of the parameter @xmath9 in the higgs potential of the 2hdm . in turn
, the gauge - boson and fermion one - loop contributions remain subleading ( including the effects from the yukawa couplings ) as they show no remarkable departures from their sm analogues in the relevant regions of parameter space which yield @xmath124 at the level of a few percent .
most important is also the fact that the combined analysis of such higgs - strahlung events with the previously considered higgs - pair production processes ( @xmath15 , @xmath16 )
see @xcite for details could be instrumental as a highly sensitive probe of the underlying architecture of the higgs sector . at the fiducial center - of - mass energy value of @xmath92 , and
when the genuine enhancement mechanisms of the 2hdm are active , the @xmath125 events are remarkably strengthened at the one - loop level while the @xmath126 channels are simultaneously depleted .
we can assert that the described pattern of leading quantum effects emerges as a kind of `` universal '' feature of the 2hdm as far as the predictions for the higgs - strahlung processes are concerned , meaning that this pattern is virtually independent of the details of the higgs mass spectrum and of the actual production channel ( @xmath1 , @xmath2 ) .
the same is true for the pairwise production channels@xcite .
therefore they all can be physically significant provided the produced higgs boson is not too heavy for the actual center - of - mass energy .
focusing on the higgs - strahlung processes , the rise of large ( and negative ) radiative corrections to their cross - sections within the 2hdm can be regarded as a characteristic imprint of a general two - higgs - doublet model structure .
furthermore , the presence of a @xmath0-boson in the final state ( and hence of its clear - cut leptonic signature @xmath127 ) should enable a rather comfortable tagging of these higgs events following a method entirely similar to the original bjorken mechanism@xcite in the sm .
although the corresponding cross - sections at the higher operating energies planned for the future linear colliders are smaller ( as compared to lep ) , they are nevertheless sufficiently sizeable ( typically in the range @xmath128 fb ) for a relatively comfortable practical measurement at the higher luminosities scheduled for these machines .
for example , the associated decay of the accompanying neutral @xmath17-even higgs - boson would manifest basically in the form of either i ) collimated and highly - energetic @xmath129-quark or @xmath130-lepton jets , for @xmath131 gev ; and ii ) @xmath132 + missing energy ; or @xmath133 , for @xmath134 ( in the optimal regime for the tree - level process ) .
some discussion on how the corresponding mssm higgs boson production cross - sections compare to the 2hdm ones seems also appropriate . in the mssm ,
the maximum quantum effects on @xmath126 production are typically milder . here
, the dominance of the wf - corrections is also the main source of one - loop effects@xcite .
they are usually reabsorbed into the tree - level @xmath135 couplings ( [ eq : lag0 ] ) , specifically in the @xmath31-even mixing angle @xmath49 , which then becomes an effective mixing parameter .
consider , for example , the behavior of the characteristic trigonometric coupling @xmath136 in the @xmath137 interaction vertex , which in the mssm dies away with growing values of @xmath138 ( both at the tree - level and at one - loop ) .
it follows that in a situation where @xmath138 is sufficiently heavy , say above @xmath139 gev , the mssm would predict measurable rates for the channel @xmath1 ( and also for @xmath16 , if @xmath7 is sufficiently high ) , whereas the complementary channels ( @xmath140 ) would be virtually below the observability threshold . in contrast , in this particular sm - like regime ( in which @xmath141 ) the @xmath1 cross - section in the 2hdm could well be exhibiting the trademark suppression induced at one - loop by large 3h self - couplings .
furthermore , in the mssm , the dynamical origin of the leading wf - effects does not reside in the higgs boson self - couplings , but on the yukawa couplings with fermions and also in the large yukawa - like couplings of higgs bosons with sfermions , particularly with the stop .
thus , in both models ( 2hdm or mssm ) the main source of quantum effects emanates from the renormalization of the higgs boson external lines , but the kind of interactions involved in each case is radically different . on the quantitative side ,
the impact of the quantum effects in the mssm case turns out to be considerably milder and as a consequence the cross - sections for higgs - strahlung production remain similar to the sm cross - section ( i.e. for the bjorken process ) .
the corresponding 2hdm cross - sections , instead , can be significantly smaller if taking the same higgs boson masses ( e.g. sets a and b of table [ tab : masses ] , which mimic the mssm higgs boson mass spectrum ) owing to the aforementioned large suppression effect from the higgs boson wf renormalization . at the same time , as we have already mentioned , the cross - sections for neutral higgs boson pair production in the general 2hdm cf .
eq.([2h ] ) become substantially larger than the mssm counterparts ( for the same higgs boson mass spectrum ) , carrying very sizeable and positive quantum effects@xcite . therefore , an interesting combined picture emerges in the 2hdm context , in which a large ( @xmath142 ) enhancement of the pairwise neutral higgs boson production is simultaneously accompanied by a drastic drop ( of similar size ) of the higgs - strahlung events . as this situation is completely impossible to realize in the mssm , this feature could be used as a strong characteristic signature to discriminate between these processes in the general 2hdm and in the mssm . in other words
, it could provide an essential quantum handle enabling us to perform a proper identification of the kind of higgs bosons produced in a linear collider . in summary ,
we have analyzed the classical higgs - boson strahlung processes at linear colliders in the light of the general 2hdm , and we have elucidated very significant quantum imprints which spotlight , once more , the stupendous phenomenological possibilities that triple higgs boson self - interactions could encapsulate in non - supersymmetric extended higgs sectors . to be sure , regardless of whether the lhc is finally capable to discover the higgs boson(s ) , a paramount effort will still be mandatory in order to completely settle its experimental basis and to uncover the nature of the spinless constituents behind the electroweak symmetry - breaking mechanism .
as we have shown , experiments at the future linear collider facilities could play a crucial role in this momentous endeavor . *
acknowledgments * nb thanks an er position of the eu project mrtn - ct-2006 - 035505 heptools and the hospitality at the dept .
ecm of the univ .
de barcelona ; dlv acknowledges the mec fpu grant ap2006 - 00357 .
the work of js has been supported in part by mec and feder under project fpa2007 - 66665 and by diue / cur generalitat de catalunya under project 2009sgr502 and by the spanish consolider - ingenio 2010 program cpan csd2007 - 00042 .
dlv wishes to thank the hospitality of the theory group at the physikalisches institut of the university of bonn and the bethe center for theoretical physics .
discussions with karina e. williams are gratefully acknowledged .
higgs , _ phys .
_ 12 ( 1964 ) 32 ; _ phys .
* 13 * ( 1964 ) 508 ; f. englert and r. brout , _ phys .
_ * 13 * ( 1964 ) 321 .
g.s guralnik , c.r .
hagen and t. w. b. kibble , _ phys .
_ * 13 * ( 1964 ) 585 .
r. a. jimnez , j. sol , _ phys .
_ * b389 * ( 1996 ) 53 ; j.a .
coarasa , r. a. jimnez , j. sol , _ phys .
_ * b389 * ( 1996 ) 312 ; j. guasch , r. a. jimnez , j. sol , _ phys .
* b360 * ( 1995 ) 47 ; d. garca , r. a. jimnez , j. sol , w. hollik , _ nucl . phys . _
* b427 * ( 1994 ) 53 .
s. bjar , j. guasch , d. lpez - val , j. sol , _ phys .
lett . _ * b668 * ( 2008 ) 364 , arxiv:0805.0973 [ hep - ph ] ; d. lpez - val , j. guasch , j. sol , jhep 0712 ( 2007 ) 054 , arxiv:0710.0587 [ hep - ph ] ; j. guasch , w. hollik , s. pearanda , j. sol , _ nucl . phys
_ * 157 * ( 2006 ) 152 , arxiv : hep - ph/0601218 ; g. eilam , m. frank , i. turan , _ phys
. rev . _ * d74 * ( 2006 ) 035012 , arxiv : hep - ph/0601253 .
f. cornet and w. hollik , _ phys .
_ * b669 * ( 2008 ) 58 , arxiv:0808.0719 [ hep - ph ] ; e. asakawa , d. harada , s. kanemura , y. okada and k. tsumura , _ phys . lett . _ * b672 * ( 2009 ) 354,arxiv:0809.0094 [ hep - ph ] ; a. arhrib , r. benbrik , c .- h .
chen , and r. santos , _ phys .
_ * d80 * ( 2009 ) 015010 arxiv:0902.2458 [ hep - ph ] .
v. driesen and w. hollik , _ zeitsch .
f. physik _ * c68 * ( 1995 ) 485 , arxiv : hep - ph/9504335 ; v. driesen , w. hollik and j. rosiek , _ zeitsch . f. physik _ * c71 * ( 1996 ) 259 , arxiv : hep - ph/9512441 ; v. driesen , w. hollik and j. rosiek , arxiv : hep - ph/9605437 .
a. arhrib and g. moultaka , _ nucl .
phys . _ * b558 * ( 1999 ) 3 , arxiv : hep - ph/9808317 ; a. kraft , phd thesis , universitt karlsruhe , 1999 ; j. guasch , w. hollik and a. kraft , _ nucl .
phys . _ * b558 * ( 1999 ) 3 ; arxiv : hep - ph/9911452 .
a. djouadi , j. kalinowski and p. zerwas , _ zeitsch .
f. physik _ * c57 * ( 1993 ) 569 , arxiv : hep - ph/9603368 ; p.h .
chankowski , s. pokorski and j. rosiek , _ nucl .
_ * b423 * ( 1994 ) 497 ; v. driesen and w. hollik , _ zeitsch .
f. physik _ * c68 * ( 1995 ) 485 ; v. driesen , w. hollik and j. rosiek , _ zeitsch .
f. physik _ * c71 * ( 1996 ) 259 ; a. djouadi , j. kalinowski , p. ohmann and p. zerwas , _ zeitsch . f. physik _ * c74 * ( 1997 ) 93 , arxiv : hep - ph/9605339 .
s. kanemura , t. kubota and e. takasugi , _ phys .
_ * b313 * ( 1993 ) 155 , arxiv : hep - ph/9303263 ; a.arhrib , arxiv : hep - ph/0012353 ; a. g. akeroyd , a. arhrib and e .-
naimi , _ phys .
lett . _ * b490 * ( 2000 ) 119 , arxiv : hep - ph/0006035 .
k. aoki , z. hioki , m. konuma , r. kawabe and t. muta , _ prog .
suppl . _ * g73 * ( 1982 ) 1 ; m. bohm , h. spiesberger and w. hollik , _ fortsch . phys .
_ * g34 * ( 1986 ) 87 ; a. sirlin , _ phys .
rev . _ * d22 * ( 1980 ) 971 ; w. hollik , in : _ precision tests of the standard electroweak model _ , ed .
p. langacker , advanced series on directions in high energy physics - vol .
14 ( world scientific , 1995 ) ; a. denner , _ fortsch .
phys . _ * g41 * ( 1993 ) 307 . | the associated production of neutral higgs bosons with the @xmath0 gauge boson ( @xmath1 , @xmath2 ) is investigated in the context of the future linear colliders , such as the ilc and clic , within the general two - higgs - doublet model ( 2hdm ) .
we compute the corresponding cross - section for the processes @xmath3 at one - loop , including the full set of corrections at @xmath4 together with the leading @xmath5 terms , in full consistency with the available theoretical and phenomenological constraints .
we find that the wave - function corrections to the external higgs fields are the dominant source of the quantum effects , which turn out to be large and negative ( e.g. @xmath6 ) in all the @xmath7 range , and located predominantly in the region around @xmath8 and moderate values of the parameter @xmath9 ( being @xmath10 ) .
this behavior can be ultimately traced back to the enhancement potential of the triple higgs boson self - couplings , a trademark feature of the 2hdm with no counterpart in the higgs sector of the minimal supersymmetric standard model . even under this substantial depletion of the one - loop corrected signal ( which is also highly distinctive with respect to the sm expectation for @xmath11 ) ,
the predicted higgs - strahlung rates comfortably reach a few tens of femtobarn , which means barely @xmath12 events per @xmath13 @xmath14 of integrated luminosity . due to their great complementarity , we argue that the combined analysis of the higgs - strahlung events ( @xmath1 , @xmath2 ) and the previously computed one - loop higgs - pair production processes ( @xmath15 , @xmath16 ) could be instrumental to probe the structure of the higgs sector at future linac facilities . |
poromechanics couples the mechanical deformation of a porous solid with fluid flow through its internal structure @xcite . in biophysics ,
poromechanics plays an important role in the growth and deformation of cells and tissues ( _ e.g. _ , @xcite ) , and it is the dominant mechanism underlying plant motion ( _ e.g. _ , @xcite ) . in both pure and applied geophysics , poromechanics has been studied intensely in the context of subsurface pressurization during fluid injection , such as in geothermal energy extraction or carbon dioxide sequestration ( _ e.g. _ , @xcite ) , and particularly in the context of hydraulic fracture for enhanced oil or gas recovery ( _ e.g. _ , @xcite ) .
poromechanical deformations are poro__elastic _ _ when they are controlled by the reversible storage and release of elastic energy .
the classical theory of poroelasticity couples linear elasticity with darcy s law for fluid flow through a porous medium , and the hallmark of these systems is diffusive propagation and dissipation of fluid pressure with characteristic time scale @xmath0 , where @xmath1 is the viscosity of the fluid , @xmath2 is a characteristic length scale , and @xmath3 and @xmath4 are the elastic modulus and permeability of the solid skeleton .
this approach is valid for small deformations , but many real systems feature large deformations , small - scale microstructure , or physical mechanisms such as damage , growth , or swelling , that lead to a strongly nonlinear coupling between the pore structure and the fluid flow @xcite .
many poroelastic deformations of practical interest are driven by fluid injection .
injection - driven deformations involve radial dilation ( outward expansion ) , which is particularly interesting and challenging because it leads to a nontrivial state of stress and strain ( _ e.g. _ , @xcite ) .
indeed , fluid injection into granular materials can lead to spectacular damage patterns when the injection pressure exceeds the inter - particle friction or the external confining stress ( _ e.g. _ , @xcite ) .
however , the deformation in these examples is almost completely irreversible because the solid skeleton is stiff and the fluid pressure is low , so the majority of the input energy is dissipated through frictional sliding and rearrangement , and almost none is stored elastically @xcite .
fluid injection can only drive significant storage of elastic energy when the fluid pressure becomes comparable to the stiffness of the solid skeleton .
true poroelastic deformation requires either much larger pressures or much softer materials . as a result
, it has proven difficult to study in a laboratory setting .
experiments with rocks and sands have been limited to postmortem inspection after high - pressure injection ( _ e.g. _ , @xcite ) , providing useful insight into the failure of realistic geomaterials , but at a very coarse level in time and space .
this limitation has been avoided in a one - dimensional geometry with soft , open - cell polymer foams ( kitchen sponges ; _
e.g. _ , @xcite ) .
however , these materials have proven to be experimentally challenging for a variety of reasons , with unsatisfying comparison between experiment and theory . here , we study the poromechanical deformation of a system with granular microstructure by injecting fluid into a confined monolayer of spherical particles . by using particles that are extremely soft
, we construct a system that exhibits striking poroelastic phenomena at relatively low working pressures .
high - resolution imaging and particle - tracking provide experimental access to the full , multi - scale deformation field .
we show that the smooth , quasi - reversible macroscopic deformation can be captured in part by a minimal continuum model , despite the presence of complex shear banding and structural rearrangement .
we pack a single layer of about 25,000 spherical , polyacrylamide hydrogel particles between two glass discs and we saturate the packing with the working fluid , a mixture of water and glycerol ( figure 1a ; see appendix a ) .
the discs are separated by a permeable spacer that confines the particles , but allows fluid to leave freely around the edge .
the particles are soft ( having a young modulus of @xmath520 kpa ) , nearly incompressible ( having a poisson ratio of @xmath5@xmath6 ) , hertzian ( exhibiting hertz - like contact mechanics ) , elastic ( allowing order - one elastic strains ) , non - cohesive , and very slippery ( having low friction at particle - particle and particle - wall contacts ) @xcite .
the particles have mean diameter @xmath71.2 mm with about 10% polydispersity ( figure 5 ) .
the packing has an apparent area void fraction of @xmath50.14 when viewed from above , and an actual volume void fraction of @xmath50.51 .
the former is denser than random close packing in 2d ( @xmath50.18 ) due in part to the softness of the particles , as often occurs in fluid emulsions , and also in part to their polydispersity . to perform an experiment
, we inject more of the same working fluid into the cell at a constant volume rate @xmath8 .
this fluid enters the cell via an injection port in the center of the lower disc , flows radially outward through the packing , and exits through the spacer at the outer edge .
the resulting fluid pressure gradient within the porous layer ( large pressure at the center dropping to atmospheric at the edge ) deforms the packing , driving the particles radially outward and opening a cavity in the center ( figure 1b , c ; video s1 ) .
this coupling of fluid pressure and solid deformation is the core idea behind poromechanics .
the deformation eventually reaches a steady state ( here , after @xmath5100 s ) in which the gradient of elastic stress in the solid skeleton balances the gradient of pressure in the fluid .
we then stop injecting , at which point the elastic stress relaxes as the pressure gradient dissipates and the cavity closes .
the relaxation of the packing highlights the macroscopically _
elastic _ nature of the system , demonstrating that the packing stores elastic energy during the injection phase and releases it during the relaxation phase .
we repeat this injection - relaxation cycle several times in the same packing .
we image the deformation and subsequent relaxation of the packing with a digital still camera , detecting the particle positions in each image to within about @xmath9 and then tracking the particles from image to image ( figure 1b ; see appendix a ) .
one striking aspect of the deformation is the cavity that opens and then closes in the center of the packing . despite the irregular shape of the cavity ( figure 2b )
, we find that the macroscopic dynamics of its expansion and collapse are smooth and relatively reproducible across repeated injected - relaxation cycles ( figures 2a and 7 ) .
in contrast , the shape of the cavity varies from cycle to cycle ( figure 2b ; video s2 ) .
the size of the cavity increases with injection rate roughly in accordance with the prediction of a minimal continuum model ( described in more detail below ) , but repeating an experiment at a given injection rate after `` resetting '' the packing by completely rearranging the particles leads to large variability ( figure 2c ) .
this implies that the macroscopic properties of the packing are a strong function of particle arrangement .
we measure the internal deformation of the packing via particle tracking , which provides a direct measure of the displacement field .
for this purpose , we define a rectangular coordinate system centered at the injection port , where @xmath10 is the position of particle @xmath11 at time @xmath12 and @xmath13 is its initial position .
the displacement of particle @xmath11 is then @xmath14 , with magnitude @xmath15 and radial component @xmath16 .
the deformation is primarily radial because of the axisymmetric boundary conditions , so we focus on @xmath17 .
the difference @xmath18 is a measure of the non - radial component of the displacement , which we find to be a few percent or less of the radial component ( see appendix a ) .
we find that the radial displacement is large near the cavity and fades to zero at the rigid edge , with a petal - like mesoscale structure ( figure 3a ; video s3 ) .
similar petal - like features have been observed in simulations of fluid injection into an initially dry packing of frictional particles @xcite , indicating that these structures are not an artifact of our low - friction system .
additionally , similar but much more regular features have been observed in quasi - static `` grain injection '' experiments @xcite , where they were identified with preferential directions in the far - field crystal structure . however , our packings are isotropic due to the polydispersity of the particles .
each petal represents a group of particles that move radially outward further than their neighbors , implying that the edges of each petal are bands of localized shear failure .
we confirm this by calculating the local strain field ( @xcite ; see appendix a ) , revealing a network of spiral shear bands that span the entire system ( figure 3b ) .
shear bands following logarithmic spirals are a well - known feature of failure in radial dilation and hydraulic fracture @xcite .
we see spirals with a pitch of roughly of @xmath19 , which implies that the packing has a very low shear strength ( _ i.e. _ , a very low friction angle ) .
correlations between shear strain and positive volumetric strain are evidence of shear dilation , a well - known feature of deformation in granular materials ( figure 3b , c ) .
macroscopically , elastic deformations involve the reversible storage and release of strain energy , whereas plastic deformations are dissipative and irreversible . in crystalline solids , there is a clear distinction between elasticity and plasticity at the particle scale : elastic deformations involve stretching or compressing bonds between atoms or molecules , whereas plastic deformations involve breaking and/or rearranging bonds .
this distinction is less clear in amorphous or granular materials , where deformations often involve a combination of reversible and irreversible rearrangement events @xcite . here
, the fact that the cavity closes completely upon relaxation implies that the deformation is macroscopically reversible .
however , the hysteresis in cavity shape is evidence of particle - scale irreversibility , and the shear bands are evidence of plastic failure . to investigate the apparent contradiction of strongly irreversible micromechanics coexisting with smooth , quasi - reversible macroscopic mechanics , we decompose the deformation field into reversible and irreversible components .
we calculate these by considering the transformation between three configurations : the initial ( _ before _ the deformation ) , the deformed ( at steady - state ) , and the final ( relaxed , _ after _ the deformation ) .
the total strain @xmath20 is that which transforms the initial configuration into the deformed one .
the irreversible strain @xmath21 is that which remains after the deformation relaxes ( _ i.e. _ , the residual strain ) ; this transforms the initial configuration to the final relaxed one .
the reversible strain @xmath22 is that which dissipates as the deformation relaxes ; this would transform the final relaxed configuration back to the deformed one . for infinitesimal deformations ,
these three strains are related by superposition , @xmath23 ; we calculate them independently since the deformation is large .
since we calculate strain as the locally affine best fit to the actual deformation field , we also calculate the root - mean - square difference between the affine field and the actual deformation field , @xmath24 ( @xcite ; see appendix a ) .
this is a measure of the typical non - affine displacement , which is indicative of the amount of particle - level rearrangement . comparing the reversible and irreversible components of the strain field ( figure 3f , g vs. j , k ) , we find that the inner region is dominated by a combination of reversible and irreversible volumetric expansion ( positive volumetric strain ) and shear .
volumetric expansion indicates that particles have traveled away from their neighbors , which is expected near the cavity since the particles move radially outward by several diameters .
since the packing can not support tension , this leads to local collapse or `` unjamming '' of the packing structure , which leads to large amounts of both reversible and irreversible rearrangement ( figure 3d , h , l ) .
in contrast , the outer region is dominated by relatively smooth , axisymmetric , reversible volumetric compression ( negative volumetric strain ) .
the displacement and shear strain are much smaller , and there is much less rearrangement .
the spiral shear bands span the entire system and , surprisingly , are primarily reversible ( figure 3b , f , g ) .
all rearrangements play a strong role in the dynamics since reconfiguration of the packing takes time and dissipates energy .
macroscopically , dissipative deformations that are reversible are known as `` viscoelastic '' , whereas those that are irreversible are known as `` viscoplastic '' . unlike elasticity , which is quasi - static , these viscous processes are rate dependent .
the steady - state deformation is set by the balance between the gradient in fluid pressure within the packing and the roughly axisymmetric elastic compression of the outer region of the packing .
motivated by this , we next derive a minimal axisymmetric model for this system based on the theory of poroelasticity @xcite .
our model is intended to capture four main features of this system : ( 1 ) conservation of volume , ( 2 ) poromechanical coupling between pressure gradients in the fluid and stress gradients in the solid skeleton , ( 3 ) elastic energy storage in the solid skeleton , and ( 4 ) viscous dissipation due to reversible and irreversible rearrangements .
we do not attempt to capture the evolution of effective material properties due to irreversible rearrangements ( _ c.f .
_ , figure [ fig : cavity]c ) .
we emphasize that we are not attempting to develop a general model for deforming granular materials , but rather a minimal one that captures the leading order behavior of our poroelastic system .
we outline the core assumptions of the model here and present a detailed derivation in appendix b. we assume that the packing is homogeneous , and that the flow and deformation fields are axisymmetric .
we also assume that the fluid and the solid are individually incompressible which , for the solid , implies that the beads can rearrange and deform without changing volume .
this is justified here because the working pressure is low relative to the bulk moduli of the fluid and the particles ( @xmath55 kpa _ vs. _
@xmath52 gpa ) .
this allows for a simple but exact kinematic relationship between the volumetric strain and the local porosity ( fluid or void fraction ) @xmath25 ( equation [ eq : phi_to_u ] ) .
we assume that the porosity is initially uniform and equal to @xmath26 .
we assume that the elastic stress in the solid skeleton is isotropic , meaning that the skeleton stores elastic energy due to volumetric compression but not due to shear .
this is justified by the fact that similar packings are known to have an anomalously low ratio of shear to bulk modulus due to the extreme softness and slipperiness of the contacts @xcite .
this allows us to link the stress directly to the volumetric strain , and therefore to @xmath27 .
this is useful because although the displacements are large , the changes in porosity are small .
we therefore take the elastic component of the stress to be a power law in the change in porosity with exponent @xmath28 , as appropriate for a granular material consisting of hertzian particles @xcite , with an effective ( drained ) bulk modulus @xmath3 .
the shear strain field indicates that the skeleton experiences shear failure near the cavity and along the spiral shear bands .
the shear failure of granular materials is typically modelled with a frictional ( mohr - coulomb ) failure criterion , which states that the material will yield ( fail plastically ) when the shear stress anywhere exceeds some fixed fraction of the local normal stress . after yielding , the structure of the material rearranges according to a suitable ( visco)plastic flow law .
this transition to plastic flow ( unjamming ) has been studied extensively from a variety of perspectives ( _ e.g. _ , @xcite ) .
although the dynamics of unjamming can be extremely important in many systems , we do not treat this behavior here because our system is _ compaction - controlled_. that is , the rate of shear strain near the cavity is fundamentally limited by the rate of volumetric strain in the outer region since the cavity can only expand as fast the outer region compresses .
this is consistent with the fact that we expect the effective viscosity of rearrangement in the compacting outer region to be much larger than that in the unjamming inner region @xcite . in an expansion - controlled system where the outer region is not yet jammed
, we would expect the cavity shape to exhibit regular , sharp , triangular cracks @xcite .
in contrast , we see irregular and relatively smooth cavity shapes .
we model viscous dissipation due to rearrangement in the outer region in a very simplistic way by assuming that this contributes a transient component to the volumetric stress that is linear in the rate of change of porosity with an effective viscosity @xmath29 .
this linear , kelvin - voigt - like representation is often used in viscoelasticity , but here it is intended to capture both reversible and irreversible rearrangements . the linear viscous term embodies the rate - dependence of these processes , introducing a characteristic time scale for viscous rearrangement , @xmath30 . although the viscosity should itself be a function of the local volumetric stress @xcite , we ignore this for simplicity .
finally , we assume that fluid flows relative to the solid skeleton according to darcy s law with a constant permeability @xmath4 .
the assumption of constant permeability is justified by the fact that the changes in porosity are relatively small , at most a few percent .
these assumptions lead to a nonlinear conservation law for the local porosity as a function of time : @xmath31=0,\ ] ] where @xmath32 is the normalized change in porosity , which is a measure of the eulerian volumetric strain ; @xmath33 and @xmath34 are time and radial position scaled by the poroelastic time scale @xmath35 and the outer radius @xmath36 , respectively ; and @xmath37 is the stress in the solid skeleton ( the effective stress ) scaled by the bulk modulus @xmath3 .
the model has two dimensionless parameters , @xmath38 and @xmath39 .
the former compares the pressure gradient in the fluid with the stiffness of the solid skeleton , while the latter compares the viscous time scale to the poroelastic one .
because the mechanical properties of the packing are difficult to measure and strongly dependent on the particle arrangement , we use @xmath40 and @xmath41 as fitting parameters .
the former influences the rate of deformation and sets the steady - state ; the latter influences only the rate
. we solve eq .
subject to two boundary conditions , which are that the cavity wall is a free surface where the stress in the solid skeleton vanishes , and that the outer edge is a rigid boundary where the solid is stationary .
we also solve simultaneously an evolution equation for the position of the cavity wall .
this model is a radial version of those that have been developed for the rectilinear deformation of kitchen sponges @xcite .
a similar radial model was developed in the context of blood - vessel pressurization @xcite , but our model is kinematically exact for finite axisymmetric deformations ( see appendix b ) and we incorporate a nonlinear elastic stiffness as well as viscous dissipation . in steady state , the model can be simplified to an integral coupled with a nonlinear algebraic equation for the cavity radius ; this is straightforward to evaluate numerically . despite the simplicity of the model and the large grain - scale variability in the experiments , we find that , after choosing @xmath40 to match the cavity area , the steady - state displacement field predicted by the model agrees surprisingly well with the azimuthal average from the experiments ( figure 4a ) .
we also calculate an azimuthally averaged porosity field from the displacement field via equation .
the model does not agree very well with this ( figure 4b ) , although it does capture certain aspects : the initial porosity in the experiment is @xmath42 , and both the model and the experiment show a concave - up trend from a value near 0.51 near the cavity down to a value near 0.50 at the wall .
the measured porosity field is non - monotonic with a minimum value at @xmath43 , whereas the model is monotonic with its minimum at @xmath44 , but the minimum values are similar .
the model is able to capture the displacement field relatively well because the displacement field is an integral measure of the deformation , dominated by the mean features and relatively insensitive to the local details .
in contrast , the porosity field is a direct measure of the local strain , strongly reflecting the local details of the deformation .
in addition , use of equation assumes smoothness , axisymmetry , and an initially uniform porosity field , but the experiment does not necessarily observe any these except in an averaged sense .
a more elaborate model might take advantage of plastic - failure theory from soil mechanics @xcite , the theory of shear - transformation zones in amorphous solids @xcite , or constitutive laws from the rheology of suspensions @xcite , although any of these approaches would lead to a substantial increase in complexity and several additional constitutive parameters . solving the time - dependent model numerically
, we find that it captures the dynamics of cavity expansion after choosing @xmath41 accordingly ( figure 2a ) .
the fit ( @xmath45 ) yields a poroelastic time scale of @xmath51.5 s and a viscous one of @xmath55 s , implying that viscous rearrangement controls the overall rate of cavity expansion .
given the complex nature of this process , it is surprising that our linear kelvin - voigt model captures the dynamics of cavity expansion as well as it does
. the model does not capture relaxation very well , relaxing much more slowly than the experiment .
this is not surprising since the amount of viscous dissipation due to rearrangement is likely different during relaxation than during cavity expansion .
hysteresis in effective material properties is common in plasticity , soil mechanics , and granular materials , but our minimal model does not include this .
fluid injection into a soft granular material drives deformation that is macroscopically poroelastic , despite rich micromechanical complexity .
we found that the deformation in the inner region ( near the injection port ) is dominated by irreversible structural plasticity that leads to strong variations in the cavity shape , whereas the outer region deforms smoothly and reversibly .
the latter ultimately supports the radially outward loading and controls the macroscopic mechanics of the steady state , such that the leading order features of the deformation can be captured relatively well with an axisymmetric continuum model .
we expect this coexistence of microscopic irreversibility with macroscopic reversibility to be a strong feature of elastic dilational deformation in any system with a low ratio of shear strength to bulk stiffness .
many intriguing problems remain , such as connecting changes in grain - scale structure with the evolution of macroscopic properties , examining more sophisticated material models based on plasticity theory , and exploring the dynamics of fluid - driven shear failure . more broadly , this system is a promising platform for high - resolution measurement of the dynamics of poromechanical deformation , and it has several additional features that we did not take advantage of here .
for example , polyacrylamide hydrogel is closely index - matched with water , making it extremely well - suited to visualization in three - dimensional systems @xcite .
polyacrylamide hydrogel is also sensitive to both temperature and dissolved salt concentration .
this allows for precise system - wide tuning of the size and stiffness of the particles ; it also enables studies of the dynamics of swelling or shrinking in response to local or global stimuli , which has particular relevance to biophysical systems @xcite . in much the same way that granular monolayers and rafts of bubbles have served as indispensable model systems for developing fundamental concepts in the mechanics of both crystalline and amorphous materials @xcite , so too can packings of soft particles provide unique insight into the deformation and failure of materials under nontrivial poromechanical loading , from the propagation of poroelastic waves to the coupling of deformation with flow and transport .
in addition to serving as a tool for benchmarking numerical simulations @xcite , this system offers an avenue into the experimental exploration of other fundamental problems of poroelasticity that have previously existed only as theoretical predictions or inferences from field observations .
the authors gratefully acknowledge support from the yale climate & energy institute .
cwm also thanks t. bertrand , r. f. katz , r. w. style , and l. a. wilen for helpful discussions .
erd acknowledges support from the national science foundation ( cbet-1236086 ) .
jsw acknowledges support from yale university , the swedish research council ( vetenskapsrdet ) , and a royal society wolfson research merit award .
[ [ apparatus . ] ] apparatus .
+ + + + + + + + + + we pack a single layer of about 25,000 spherical particles ( polyacrylamide hydrogel , jrm chemical ) between two borosilicate glass discs .
the particles have mean diameter @[email protected] mm with standard deviation ( sd ) 0.12 mm ( 10% polydispersity ; fig . 5 ) .
the discs are 19 mm thick and 212.7 mm in diameter , and they are separated by a plastic spacer that defines a working area of diameter @[email protected] mm and thickness @[email protected] mm .
this is two sds greater than @xmath46 , confining the particles to a two - dimensional monolayer without restricting their in - plane motion .
the spacer is permeable , allowing fluid to exit while confining the particles .
we inject fluid at a fixed volume flow rate using a syringe pump ( new era ne-4000 ) .
the fluid is a mixture of water and glycerol ( 61% glycerol by mass ) with a viscosity of @[email protected] pa@xmath49s at 20 @xmath50c .
we acquire images with a digital camera ( canon eos rebel t2i ) . ,
width=325 ] [ [ particle - detection - and - tracking . ] ] particle detection and tracking .
+ + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + we process the experimental images in ` matlab ` , detecting particle positions in each image via centroid - finding after applying an intensity threshold .
we track the particles from image to image using a standard particle - tracking algorithm @xcite .
the images have a resolution of about 7.9 pixels per 1 mm ( about 9.4 pixels per particle diameter ) .
frame - to - frame detection noise for particle centers is about 0.1 pixels ( 13 @xmath1 m or about 1% of one diameter ; figure 6 ) .
we estimate this detection noise by calculating ( a ) total particle displacement between two frames taken before injection has started , when the particles are at rest .
the noise has a mean of about 0.006@xmath46 particle diameters with a standard deviation of about 0.005@xmath46 , indicating a detection threshold of @xmath50.01@xmath46 ( vertical blue line ) . at steady - state deformation ,
( b ) the average radial displacement is about @xmath51.4@xmath46 with a standard deviation of about 1.6@xmath46 , whereas ( c ) the average non - radial displacement is about 0.03@xmath46 with a standard deviation of 0.06@xmath46 , indicating that particle motion is almost entirely radial .
[ fig : ur_hist ] , width=325 ] [ [ deformation - field . ] ] deformation field .
+ + + + + + + + + + + + + + + + + + we use the particle positions to calculate a best - fit local strain field following @xcite .
however , the quantity described in @xcite as a local strain tensor @xmath51 is more correctly identified as a local displacement gradient tensor @xmath52 , where @xmath53 is the undeformed configuration and @xmath54 is the displacement field .
the distinction is important here , where the displacements are large .
we calculate the displacement gradient tensor and then use it to calculate the green - lagrange strain tensor @xmath55 , where @xmath56 is the deformation gradient tensor and @xmath57 is the identity tensor .
the strains we report above are green - lagrange strains .
we also calculate the root - mean - square ( rms ) non - affine displacement @xmath24 from their quantity @xmath58 by dividing each local value by the number of neighbors to form a mean ( their quantity is a sum ) and then taking the square root .
the result has dimensions of length .
we derive a continuum model for this process based on the theory of poroelasticity @xcite .
we assume that the fluid and the solid are incompressible which , for the solid , implies that the beads can rearrange and deform without changing volume .
this is a common assumption , justified here by the low working pressure ( @xmath55 kpa ) .
macroscopic deformation occurs through rearrangement of the solid skeleton , which leads to variations in the porosity ( void or fluid fraction ) , @xmath27 .
assuming axisymmetry and working strictly in terms of eulerian quantities , conservation of mass dictates that where @xmath60 and @xmath61 are the velocity of the fluid and of the solid , @xmath8 is the volume rate of fluid injection , and @xmath48 is the thickness of the gap .
we assume that fluid flows relative to the solid skeleton according to darcy s law , @xmath62 where @xmath4 and @xmath1 are the permeability of the solid skeleton and the viscosity of the fluid , respectively , and @xmath63 is the fluid pressure .
poroelastic theory dictates that the internal gradient in fluid pressure acts as a body force on the solid skeleton , and mechanical equilibrium requires that this must be supported by the divergence of the stress in the solid skeleton .
we expect that the packing has an extremely low ratio of shear to bulk modulus , so for simplicity we assume that the solid can not support shear or tensile stresses .
this implies that the stress tensor is isotropic , and we write mechanical equilibrium as @xmath64 where @xmath65 is the effective stress ( _ i.e. _ , the stress in the solid skeleton ) . combining all of the above
, we obtain a conservation law for the evolution of the porosity , @xmath66 where @xmath32 is the eulerian volumetric strain , or the normalized change in porosity , and @xmath67 is the initial ( relaxed ) porosity .
we must now specify a constitutive relationship between effective stress and strain , but note that equation is valid for any stress - strain relationship that yields an isotropic stress tensor . here , we take the stress to be hertzian elastic and linearly viscous in the volumetric strain .
this can be written as @xmath68 where @xmath3 and @xmath29 are the effective bulk modulus and viscosity of the solid skeleton , respectively .
note that the stress vanishes at @xmath69 .
the linear viscous term introduces a simple `` rearrangement '' timescale , without which the mechanics would be quasi - static .
we assume that fluid is injected into a cavity in the solid of radius @xmath70 and initial radius @xmath71 .
the behavior is independent of @xmath72 for @xmath73 .
the fluid and the solid are initially at rest , @xmath74 , and the initial porosity field is @xmath75 .
injection begins at @xmath76 . at the inner boundary , @xmath77 , the normal component of the effective stress must vanish since
the cavity is a free surface of the solid skeleton .
for an isotropic stress field , this implies that @xmath78 and therefore that @xmath79 for @xmath80 . at the rigid outer boundary , @xmath81 ,
we have that @xmath82 .
this can be written @xmath83 we also require an evolution equation for the position of the moving inner boundary , @xmath84 equations constitute a one - dimensional , time - dependent , moving - boundary problem .
the solid displacement field does not appear explicitly , but we can calculate it at any time through a kinematic relationship , @xmath85 where @xmath86 is the jacobian determinant , or the jacobian of the deformation , and @xmath87 is the deformation gradient tensor ( see appendix a ) . for an axisymmetric , plane - strain deformation , eq . becomes @xmath88 where @xmath89 is the @xmath90-component of the ( eulerian ) solid displacement field .
in contrast with linear poroelasticity , this problem is nonlinear for four reasons : we account rigorously for ( 1 ) the moving boundary , ( 2 ) the solid velocity , and ( 3 ) the exact relationship between porosity and displacement , and we use ( 4 ) a nonlinear elastic law .
a similar axisymmetric model was developed by @xcite for small elastic deformations .
our model incorporates a nonlinear elastic law and linear viscous effects , and is kinematically exact
it is valid for arbitrarily large deformations as long as the constitutive laws remain valid .
we present equations and in dimensionless form in the main text ( equations [ eq : model_nd ] and [ eq : sigmaprime_nd ] , respectively ) , where @xmath91 is the radial coordinate scaled by the outer radius , @xmath92 is time scaled by the poroelastic time scale , @xmath93 is the effective stress scaled by the bulk modulus , and the two dimensionless parameters are @xmath38 and @xmath39 .
the dimensionless boundary conditions are @xmath94 @xmath95 and @xmath96 where @xmath97 .
we solve the model numerically using a finite - volume method with explicit time integration , accommodating the moving boundary with an adaptive grid . at steady - state ,
the model has implicit solution @xmath98^{2/3}\ ] ] and @xmath99 where the integral @xmath100 is @xmath101^{2/3}\,\mathrm{d}r \label{eq : i_of_r}\ ] ] and the subscript `` @xmath102 '' refers to the value of a quantity at steady state .
the steady - state cavity radius is determined implicitly by the definition @xmath103 , which leads to @xmath104 we solve equation numerically using a standard root - finding technique , evaluating the integral @xmath105 from equation using the trapezoidal rule .
the mechanical properties of the packing are difficult to measure and strongly dependent on the particle arrangement ( figure 2c ) , so we use the mechanical properties as fitting parameters .
to compare the model with the experiment at steady - state , as in figure 4 , we choose a value of the product @xmath106 to match the steady - state cavity radius .
we use a single value of @xmath106 for all of the experiments in figure 2c . to compare the dynamics , as in figures 2a and 7 , we choose @xmath106 to match the steady - state cavity radius and then @xmath107 to match the rate of deformation and relaxation ( @xmath107 plays no role in the steady state ) .
57ifxundefined [ 1 ] ifx#1 ifnum [ 1 ] # 1firstoftwo secondoftwo ifx [ 1 ] # 1firstoftwo secondoftwo `` `` # 1''''@noop [ 0]secondoftwosanitize@url [ 0 ]
+ 12$12 & 12#1212_12%12@startlink[1]@endlink[0]@bib@innerbibempty in @noop _ _ , vol .
( ) pp . @noop * * , ( ) link:\doibase 10.1093/imamat/17.2.153 [ * * , ( ) ] link:\doibase 10.1007/bf00248468 [ * * , ( ) ] @noop _ _ ( , ) link:\doibase 10.1016/0021 - 9290(91)90291-t [ * * , ( ) ] link:\doibase 10.1115/1.2894880 [ * * , ( ) ] link:\doibase 10.1016/s0021 - 9290(98)00161 - 4 [ * * , ( ) ] link:\doibase 10.1038/nature03550 [ * * , ( ) ] link:\doibase 10.1038/nmat3517 [ * * , ( ) ] link:\doibase 10.1146/annurev - fluid-120710 - 101200 [ * * , ( ) ] link:\doibase 10.1073/pnas.1115347109 [ * * , ( ) ] @noop _ _ ( , , ) link:\doibase 10.1073/pnas.1302156110 [ * * , ( ) ] link:\doibase 10.1002/2013wr015175 [ * * , ( ) ] @noop * * , ( ) , link:\doibase 10.1016/0148 - 9062(88)92299 - 1 [ * * , ( ) ] link:\doibase 10.1093/gji / ggt199 [ * * , ( ) ] @noop _ _ ( , , ) link:\doibase 10.1680/geot.1991.41.2.173 [ * * , ( ) ] link:\doibase 10.1680/geot.1992.42.3.395 [ * * , ( ) ] link:\doibase 10.1039/c3sm52689e [ * * , ( ) ] link:\doibase 10.1103/physrevlett.113.198001 [ * * , ( ) ] link:\doibase 10.1103/physreve.74.011301 [ * * , ( ) ] link:\doibase 10.1038/nphys834 [ * * , ( ) ] link:\doibase 10.1038/ncomms1289 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.108.258001 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.108.264504 [ * * , ( ) ] link:\doibase 10.1016/j.petrol.2006.01.009 [ * * , ( ) ] in link:\doibase 10.2118/151827-ms [ _ _ ] ( ) link:\doibase 10.1115/1.3423648 [ * * , ( ) ] link:\doibase 10.1115/1.3173119 [ * * , ( ) ] link:\doibase 10.1051/meca/2011115 [ * * , ( ) ] \doibase http://link.aps.org/doi/10.1103/physreve.84.011302 [ * * , ( ) ] @noop `` , '' in \doibase https://www.onepetro.org/conference-paper/arma-11-228[__ ] ( ) link:\doibase 10.1103/physrevlett.99.068001 [ * * , ( ) ] link:\doibase 10.1103/physreve.57.7192 [ * * , ( ) ] link:\doibase 10.1103/physreve.77.041505 [ * * , ( ) ] link:\doibase 10.1039/c3sm51014j [ * * , ( ) ] link:\doibase 10.1103/physrevlett.112.028302 [ * * , ( ) ] link:\doibase 10.1103/physreve.84.060402 [ * * , ( ) ] link:\doibase 10.1103/physreve.68.011306 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.105.175701 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.107.188301 [ * * , ( ) ] link:\doibase 10.1209/0295 - 5075/96/36008 [ * * , ( ) ] link:\doibase 10.1017/s0334270000008936 [ * * , ( ) ] link:\doibase 10.1146/annurev - conmatphys-062910 - 140452 [ * * , ( ) ] link:\doibase 10.1063/1.3674173 [ * * , ( ) ] link:\doibase 10.1007/s00348 - 013 - 1456-z [ * * , ( ) ] link:\doibase 10.1098/rspa.1947.0089 [ * * , ( ) ] link:\doibase 10.1016/0025 - 5416(79)90174 - 5 [ * * , ( ) ] link:\doibase 10.1038/335424a0 [ * * , ( ) ] link:\doibase 10.1126/science.252.5003.226 [ * * , ( ) ] link:\doibase 10.1103/physreve.82.046305 [ * * , ( ) ] link:\doibase 10.1002/jgrb.50204 [ * * , ( ) ] @noop `` , '' | compressing a porous , fluid - filled material will drive the interstitial fluid out of the pore space , as when squeezing water out of a kitchen sponge .
inversely , injecting fluid into a porous material can deform the solid structure , as when fracturing a shale for natural gas recovery .
these poromechanical interactions play an important role in geological and biological systems across a wide range of scales , from the propagation of magma through the earth s mantle to the transport of fluid through living cells and tissues .
the theory of poroelasticity has been largely successful in modeling poromechanical behavior in relatively simple systems , but this continuum theory is fundamentally limited by our understanding of the pore - scale interactions between the fluid and the solid , and these problems are notoriously difficult to study in a laboratory setting . here , we present a high - resolution measurement of injection - driven poromechanical deformation in a system with granular microsctructure : we inject fluid into a dense , confined monolayer of soft particles and use particle tracking to reveal the dynamics of the multi - scale deformation field .
we find that a continuum model based on poroelasticity theory captures certain macroscopic features of the deformation , but the particle - scale deformation field exhibits dramatic departures from smooth , continuum behavior .
we observe particle - scale rearrangement and hysteresis , as well as petal - like mesoscale structures that are connected to material failure through spiral shear banding . |
one of the most striking impacts of a contact with a superconductor ( s ) onto a small piece of normal metal ( n ) is the modification of the local density of states ( ldos ) .
this effect , known as the superconducting proximity effect , is related to the induction of superconducting correlations resulting in a finite value of the pair amplitude @xmath4 on the normal side @xcite . in the absence of phonon
mediated attraction between electrons on the normal side , decoherence between electronlike and holelike amplitudes leads to an exponential decay of the pair amplitude with distance from the contact , with a characteristic length scale exceeding the superconducting coherence length .
modification of the ldos on both sides of the contact strongly depends on the scattering properties of the contacts ( described in terms of transmission eigenvalues ) and the properties of the normal region ( geometry , size , and impurity concentration ) . in the case of diffusive systems , it was predicted theoretically that the ldos can even be fully suppressed in a specific energy range around the fermi energy which is known as the minigap @xcite .
the minigap width is of the order of the inverse dwell time in the normal structure , which is given by the thouless energy @xmath5 , where @xmath6 denotes the mean level spacing of the normal region and @xmath7 is the total conductance of the structure which is assumed large compared to the conductance quantum @xmath8 . in the decades after its discovery
it has in detail been studied theoretically @xcite .
the development of more elaborate experimental techniques with high spatial resolution made variations of the ldos in this energy range accessible to experiments @xcite , which was found to be in agreement with theoretical calculations to a high degree @xcite .
much interest was concentrated on systems built up of a finite normal region sandwiched between two superconductors : a josephson junction @xcite . in such systems , another parameter ,
i.e. , the phase difference between the superconducting order parameters comes into play and leads to a phase - dependent minigap @xcite .
classical ballistic systems @xcite were investigated as well as diffusive systems @xcite and the crossover between both @xcite .
it turns out that not only diffusive systems exhibit a minigap , but also ballistic systems with a chaotic classical motion @xcite . at this point
, one might think that such structures are sufficiently explored and all relevant properties are understood . however , recently levchenko reported the finding of a dip in the ldos close to the gap edge @xmath2 for short diffusive josephson junctions with ideal contacts @xcite .
actually , this dip was already seen in former publications @xcite , however , no special attention was paid to it . in a previous work @xcite
we found the peculiar result that the suppression of the ldos at @xmath2 is not limited to a dip , but a secondary gap of finite width appears for a diffusive system or chaotic cavity with the normal region connected through ballistic contacts to the superconductors .
this secondary gap has a finite width as a function of the superconducting phase difference @xmath9 symmetrically around zero and closes with the characteristic shape of a `` smile '' .
it is situated directly below the superconducting gap edge @xmath2 for large @xmath10 . for decreasing @xmath11
the upper edge of the secondary gap detaches from @xmath2 and the gap vanishes completely below a critical value of @xmath11 .
we have furthermore shown that the secondary gap is robust against asymmetries in the setup , comprising a difference in ballistic couplings or a weak spatial dependence . in this work ,
we investigate a wide range of possible nonballistic contacts and show that the secondary `` smile''-gap is not only an exotic feature which appears for ballistic contacts , but is a more general property of short diffusive or chaotic josephson systems . using quasiclassical green s functions in the form of the quantum circuit theory ,
we begin by generalizing our ballistic calculations to contacts with constant transmission eigenvalues @xmath12 , for which we calculate the density of states as a function of @xmath11 and @xmath9 .
we find a secondary gap which scales for large @xmath11 like in the ballistic case . in a specific example of contacts described by different constant transmission eigenvalues
we show that one is not limited to a single secondary gap , but this gap can be split up into multiple subgaps .
numerical considerations of continuous transmission distributions ( diffusive , dirty , double ballistic contacts ) suggest that the secondary gap below the superconducting gap edge vanishes if the contacts include channels with close - to - zero transmission coefficients .
we prove this conjecture by an analytical calculation . by considering asymmetric setups with a tunnel contact on one and a ballistic contact on the other side we show however , that in this case a secondary gap can exist at slightly smaller energies . by considering a 3-node system
we show that although the ldos varies at different nodes the secondary gap appears either in all nodes or in none of them .
it should as well be observable in the integrated dos of the normal part .
finally , we provide a simple one - dimensional ( 1d ) model in order to describe transmission through the normal region , which is able to explain the `` smile '' shape of the secondary gap .
1a.pdf ( 3,43)(0,3)@xmath13 1b.pdf ( 3,75)(0,3)@xmath14 in order to calculate the ldos in the normal region , we make use of the retarded green s function in the quasiclassical approximation . in the diffusive or dirty limit , the angle - averaged green s functions are described by the nonlinear diffusive usadel equation which has the form of a continuity equation for coherence functions including the leakage of coherence due to the finite energy difference between electrons and holes . since the spatial dependence of the green s function is not important for our needs ( for more details see sec .
f ) , we can solve the problem by applying the so - called quantum circuit theory @xcite .
we can discretize the system and reduce the equations to an algebraic problem .
a sketch of the investigated system is shown in fig .
[ fig:1 ] .
the superconductors have equal energy gaps @xmath2 , however in general the phases of the order parameters can be different . since the global phase is of no significance , only the phase difference @xmath9 enters our calculation and we can assign the phase @xmath15 to the left and right superconductors , respectively .
the green s function in the normal node @xmath16 is determined by the constraint of matrix current conservation , including the currents to the two superconductors @xmath17 ( @xmath18 being the index denoting left and right lead ) as well as the leakage current related to the volume of the normal region through @xmath11 : @xmath19=0\,.\ ] ] the scattering properties of the contacts are contained in the expressions for the matrix currents @xcite @xmath20 in terms of a set of transmission eigenvalues @xmath21 . in general
the transmission eigenvalues can be different on both sides . for continuous transmission distributions @xmath22 the sums must be replaced by integrals over the particular distributions .
the green s functions in the leads are those of a bulk superconductor , given by @xmath23 $ ] with the spectral functions @xmath24 and @xmath25 being given by @xmath26 for @xmath27 and by @xmath28 for @xmath29 , @xmath30 being the pauli matrices in nambu space of electrons and holes . in the normal node the green s function can be parametrized as @xmath31 $ ] .
@xmath32 and @xmath33 are related via the normalization condition for quasi - classical green s functions @xmath34 , which is equivalent to @xmath35 . in the general case with different contacts on both sides ,
this corresponds to the solution of two equations for two complex variables .
expanding ( [ eq : curr_cons ] ) in pauli matrices and comparing the coefficients provides two independent equations @xmath36 \left [ \frac{x_1^{-1}}{1+g_2/g_1}+\frac{x_2^{-1}}{1+g_1/g_2 } \right ] & = & 0\\ \label{eq:2 } -2i\frac{e}{e_{\textrm{th } } } f \sin(\phi ) + \left [ -g s \sin(\varphi/2)+c f \sin(\phi ) \right ] \frac{x_1^{-1}}{1+g_2/g_1 } + \left [ g s \sin(\varphi/2)+c f \sin(\phi ) \right ] \frac{x_2^{-1}}{1+g_1/g_2 } & = & 0.\end{aligned}\ ] ] note that @xmath33 and @xmath32 as well as @xmath37 are complex valued in general .
all information on the contacts is contained in the characteristic functions @xmath38 given by @xmath39 with @xmath40 and @xmath41 being the conductance of the particular side .
again , for a continuous transmission distribution , the sums must be replaced by integrals over the particular distributions @xmath42 . for a symmetric setup @xmath43 and @xmath44 .
equation ( [ eq:2 ] ) becomes trivial and only one equation in one complex variable remains . from eq .
( [ eq:1 ] ) we find @xmath45 the density of states @xmath46 is finally obtained from @xmath47 through @xmath48 , @xmath49 being the density of states at the fermi energy of the normal state .
in previous analysis @xcite this setup was investigated for ballistic contacts with all @xmath50 .
it turned out that the secondary `` smile''-gap which appears in the symmetric case is stable under asymmetries @xmath51 . for an asymmetric setup , two further gaps , complementary to the usual minigap and the `` smile''-gap ,
appear symmetrically around @xmath52 . in this work , we want to extend these calculations and consider a wider range of contact types , corresponding to a wider range of characteristic functions @xmath38 , either described by discrete transmission eigenvalues or by continuous distributions @xmath42 .
the idea of this work is to investigate the stability of the secondary gap under deviation from the ballistic limit .
especially , we want to determine which contact properties define the existence of the secondary gap , since it is known that for tunnel contacts no secondary gap is found .
for this reason , we consider symmetric as well as asymmetric setups in the intermediate regime between the tunnel and ballistic limits .
2a.pdf ( 3,66)(0,3)@xmath13 2b.pdf ( 3,66)(0,3)@xmath14 a natural generalization of the ballistic contact is to stick to constant transmission eigenvalues , however , to allow for @xmath54 .
as @xmath55 approaches 0 , the secondary gap is expected to disappear and the tunnel result for the ldos should be reproduced .
we begin by considering symmetric contacts and thus solve eq .
( [ eq:3 ] ) with the characteristic function @xmath56 we find a secondary gap in the ldos similar to the ballistic result , which survives even for small but finite @xmath55 .
the numerical results for the critical phase @xmath57 , for which the gap closes , as well as for the upper and lower gap edges @xmath58 and @xmath59 at @xmath60 , are shown in fig .
[ fig:2 ] .
the colored regions denote the gap . above a special value @xmath61 , which scales linearly with @xmath55 and
is given by @xmath62 , the upper gap edge is fixed to @xmath2 and the lower edge approaches @xmath2 for increasing @xmath11 following a power law .
the linear scaling of @xmath61 with @xmath55 follows from eq .
( [ eq:3 ] ) for @xmath63 and @xmath60 .
the dependence of the lower gap edge @xmath59 on @xmath11 for @xmath64 is derived in the following .
below @xmath61 , the upper edge is detached from @xmath2 and approaches the lower edge until the gap disappears at some critical value of @xmath11 which as well seems to scale linearly with @xmath55 . the maximum of the critical phase at which the secondary gap closes [ fig
. [ fig:2 ] ( b ) ] does not depend on @xmath55 , however , it is shifted to smaller thouless energies with decreasing @xmath55 .
the dependence of the critical phase on @xmath11 seems to scale linearly with @xmath55 . in the limit @xmath65
, the gap disappears and the tunnel result without secondary gap is reproduced .
however , for each finite value of @xmath55 the secondary gap exists , if the thouless energy is made large enough . to get further insight to the analytic properties we linearize eq .
( [ eq:3 ] ) in the energy range below @xmath63 and around @xmath60 in the limit @xmath64 .
we find @xmath66\right\ } , \end{split}\ ] ] with @xmath67 being the dimensionless energy relative to @xmath2 .
3a.pdf ( 3,66)(0,3)@xmath13 3b.pdf ( 3,66)(0,3)@xmath14 this equation can be solved analytically , however , the expression for the general solution is quite long and will not be given here . in fig .
[ fig:3 ] , it is plotted for various values of @xmath55 . figure [ fig:3 ] ( a ) shows the width of the secondary gap approaching @xmath68 with decreasing @xmath55 ; figure [ fig:3 ] ( b ) shows the structure of the density of states above the upper edge of the minigap .
note the different scales of the energy axes in the two plots .
for decreasing @xmath55 , the ldos approaches the tunnel limit without secondary gap but with the usual singularities @xcite at @xmath63 and above the minigap .
considering @xmath69 provides an analytical expression for the critical phase @xmath70 similarly for @xmath60 we find an analytical expression for the critical energy @xmath71 in the limit of large @xmath11 describing the width of the gap @xmath72 @xmath73 being a lengthy expression related to the solution of a quartic equation .
for @xmath74 it has the form @xmath75 .
the position of the minigap edge ( fig .
[ fig:3 ] ( b ) ) can as well be calculated analytically . for @xmath74
it is given by @xmath76 a generalization of the calculations from the previous section can be achieved by considering not only one constant transmission eigenvalue , but a whole set of different transmission eigenvalues , each weighted with a specific weight @xmath77 .
we stick to a symmetric system with only one set of transmission eigenvalues and weights @xmath78 describing both sides . from the huge variety of possible sets , which could be analyzed ,
we pick only one in order to demonstrate that the secondary gaps structure in principle is not limited to only a single gap : an even finer subdivision of the ldos below @xmath2 can be observed for certain contact types .
we calculate the ldos for one representative set @xmath78 given by [ cols="<,<,<,<,<,<,<,<,<,<,<,<,<",options="header " , ] a plot of the numerical results in the energy range below @xmath2 is shown in fig .
[ fig:4 ] .
we find that for certain sets of transmission eigenvalues not only one secondary gap appears , but the dos acquires an even finer structure with multiple subgaps .
the number of subgaps depends strongly on the set @xmath78 under consideration and on @xmath11 . in the presented case
we find three gaps at @xmath79 .
similar to the previously found secondary gaps they are symmetric around @xmath60 and vanish at some critical phase , which is not the same for different subgaps .
dos of a symmetric system described by a discrete set of transmission eigenvalues @xmath80 and weights @xmath77 . in the energy range below @xmath2
we find a multiply gapped density of states consisting of three secondary gaps with a finite dos between them . ]
so far , we investigated systems where scattering in the contacts is described by constant transmission eigenvalues .
however , in systems experimentally accessible scattering is rather described by a continuous transmission distribution @xmath81 than by discrete transmission eigenvalues @xmath55 .
it is thus of crucial interest whether the secondary gap appears as well if continuous transmission distributions are considered . here ,
we investigate three generic contact types , each characterized by a distribution of transmission eigenvalues of the form @xmath82 with @xmath83 .
the normalization constant is determined by the condition @xmath84 .
the distributions for @xmath85 and @xmath86 correspond to a dirty and a diffusive connector @xcite , respectively .
the distribution for @xmath87 is equivalent to two ballistic connectors with equal conductances in series @xcite .
again we stick to symmetric setups with equal contacts on both sides .
the characteristic functions corresponding to the considered distributions have a relatively compact form @xcite . for a dirty connector ( @xmath85 ) it is given by @xmath88 . in case of a diffusive contact ( @xmath86 )
the characteristic function is @xmath89 . and for the double ballistic contact @xcite we have @xmath90 .
5a.pdf ( 3,66)(0,3)@xmath13 5b.pdf ( 3,66)(0,3)@xmath14 5c.pdf ( 3,66)(0,3)@xmath91 figures [ fig:5](a)(c )
show the numerical results for the ldos calculated from eq .
( [ eq:3 ] ) for the three cases .
since in previous calculations the suppression of the ldos around @xmath2 was strongest at @xmath60 , only this case is presented .
figure [ fig:5 ] ( a ) contains the numerical results for dirty contacts ( @xmath85 ) for different values of @xmath11 .
no signature for a suppression of the ldos at the superconducting gap edge @xmath2 is found .
the plots in fig .
[ fig:5 ] ( b ) are the results for diffusive contacts ( @xmath86 ) .
for @xmath92 , a weak suppression at @xmath63 can be seen .
the inset with a higher resolution of the energy range of interest , however , shows , that this suppression is no gap . at first glance
, this seems to disagree with ref .
@xcite but is possibly due to differences in the considered geometries .
the lowest fig .
[ fig:5 ] ( c ) shows the results for the third contact type ( @xmath87 ) with the highest weight at transmission eigenvalues around @xmath93 of all three distributions .
we find a strong suppression at @xmath63 for all considered values of @xmath11 .
the inset of this plot confirms that the ldos is suppressed to 0 at @xmath63 .
in summary we find that with decreasing weight of @xmath81 at @xmath94 the suppression of energy levels at @xmath63 is reinforced .
this observation supports the idea that the existence of andreev bound states with energies directly below @xmath63 is related to the tunnel character of the boundaries , i.e. , the transport channels with transmissions at @xmath95 .
ldos at @xmath60 for a relation of conductances @xmath96 for different values of @xmath11 .
a @xmath11-dependent secondary gap is found below @xmath63 . ] in order to check this , a fourth type of transmission distribution is considered .
each contact in this system is built up of two ballistic contacts in series ( @xmath97 and @xmath98 ) having different conductances .
note that the total setup is still symmetric .
the corresponding transmission distribution fundamentally differs from the previously considered distributions in the sense that it has no contribution at small transmissions , i.e. , below a critical value @xmath99 given by @xmath100 .
for @xmath101 it follows @xmath102 which corresponds to the previously investigated distribution with @xmath87 . since we again stick to symmetric setups we drop the lead index @xmath18 in the following .
the characteristic function is given by @xcite @xmath103 the results are plotted in fig .
[ fig:6 ] . compared to the previous distributions , a gap of finite width appears directly below @xmath63 ( inset of fig .
[ fig:6 ] ) .
as expected , the gap appears when there is no contribution of the transmission distribution around @xmath94 .
this agrees with the results of sec .
a. in the following we derive a criterion that relates the existence of the secondary gap directly below @xmath63 to the weight of the transmission distribution around @xmath94 . in this section ,
we show that the existence of the secondary gap below @xmath2 is indeed directly related to the distribution @xmath81 of the transmission eigenvalues in the vicinity of @xmath94 .
we show that the existence of a minimal transmission @xmath104 results in a secondary gap below @xmath2 . to reach this conclusion , we consider eq .
( [ eq:3 ] ) for a symmetric setup containing one general function @xmath105 and @xmath60 .
it reads as @xmath106 we linearize this equation in @xmath67 in the limit @xmath107 .
the left side is expanded in @xmath108 which must be small in order to be valid .
this has to be verified for the solution . using the limiting form of @xmath109 and @xmath110 at small @xmath111 we get @xmath112 in the leading order in @xmath111 , the general expression for @xmath113 given after eq .
( 5 ) yields : @xmath114 . introducing rescaled variables
we get an equation without explicit dependence on @xmath11 . with the definitions @xmath115 and @xmath116 we have @xmath117 the ldos
is related to the real part of @xmath32 , so a purely imaginary solution for @xmath118 at small @xmath119 means having a gap at energies e close to @xmath2 . to show this we consider the sum @xmath120 in eq .
( [ eq : characteristic_funct ] ) , which is an integral in the continuous case . for an arbitrary distribution @xmath81 , which can be normalized to satisfy @xmath121
, it reads @xmath122 where the minimal transmission in @xmath81 was used to replace the lower boundary in the integral . considering small @xmath119 and assuming @xmath123 with @xmath124 ( later verified by the solution ) we can neglect all other terms in the denominator compared to @xmath125 .
we find @xmath126 @xmath127 being a constant factor defined as @xmath128 .
it is constrained to the interval @xmath129 , @xmath130 corresponding to a ballistic setup .
its value depends on the exact form of @xmath81 in the whole interval , particularly on @xmath99 . for a diffusive connector
which is cut at @xmath99 we have @xmath131 $ ] , which becomes @xmath132 for @xmath133 and approaches @xmath68 as @xmath99 approaches @xmath68 . with this eq .
( [ eq : analytic1 ] ) reduces to a quadratic equation @xmath134 with solutions @xmath135 . both solutions
are purely imaginary for @xmath136 signifying a gap in the ldos .
however only the solution @xmath137 for @xmath138 is consistent with the previously made assumption @xmath139 with @xmath124 for small @xmath119 . since
this solution is finite for small @xmath119 the second assumption , which assumed @xmath32 to be large , can always be fulfilled for @xmath140 by choosing @xmath11 sufficiently large .
this is in agreement with our numerical results which predict no secondary gap below some critical value of @xmath11 .
this critical @xmath11 depends on the value of @xmath127 and thus on the whole transmission distribution @xmath81 .
the condition @xmath140 is related to the existence of a @xmath141 , since only @xmath142 with @xmath143 leads to @xmath144 . to conclude this section ,
we have shown that for an arbitrary transmission distribution @xmath81 without contribution in a finite interval above @xmath94 a secondary gap appears directly below @xmath63 , if @xmath11 is sufficiently large . in this section , we demonstrate that having the distribution function of transmission coefficients @xmath145 at @xmath55 below some @xmath99 is a sufficient but not necessary condition for having a secondary gap .
we consider a device with two nonidentical junctions and show that the secondary gap may exist even if the latter condition on @xmath81 is violated .
the secondary gap , however , does not appear directly below @xmath2 but is shifted to slightly smaller energies , thus it appears in a different regime than considered in sec .
iii d. 7a.pdf ( 3,66)(0,3)@xmath13 7b.pdf ( 3,66)(0,3)@xmath14 the most interesting case to which we confine ourselves here is the one which combines the two extremal contact types : a tunnel contact with @xmath146 for all transport channels on one side and a ballistic contacts with @xmath93 for all channels on the other side .
the conductances of both sides enter our calculation via the relation @xmath147 , where @xmath148 denotes the conductance of the tunnel contact and @xmath149 corresponds to the ballistic contact .
for @xmath150 the role of the tunnel contact is negligible and the result of a symmetric ballistic system at @xmath60 showing a secondary gap below @xmath2 @xcite should be reproduced .
it is of particular interest how the transition from the ballistic limit @xmath151 to the tunnel limit @xmath152 occurs .
results of the previous section indicate that there should be no secondary gap just below @xmath2 : the condition on @xmath81 for the appearance of the secondary gap is violated by the presence of a tunnel junction even for @xmath151 .
figure [ fig:7 ] shows the results of our numerical calculation . in fig .
[ fig:7 ] ( a ) , the phase @xmath9 is fixed at @xmath60 and the vanishing of the secondary gap is shown for different values of @xmath11 as @xmath147 increases . at small @xmath147 , the upper gap edge is close to its ballistic value and decreases as @xmath147 increases .
similarly , the lower gap edge is close but slightly below its value of the symmetric ballistic case and decreases with increasing @xmath147 . at some critical value of @xmath147 which depends on @xmath11
the secondary gap disappears . this critical value is smaller for smaller @xmath11 . in fig .
[ fig:7](b ) , the thouless energy is fixed at @xmath79 and the phase dependence of the secondary gap is plotted for different values of @xmath147 . we find that for the considered system the secondary gap has its maximum width not at @xmath60 as one might expect , but at @xmath52 . with decreasing @xmath147 , the @xmath60 result of the symmetric ballistic system is approached at all phases .
compared to previous findings for asymmetric ballistic contacts @xcite , where a similar behavior with decreasing @xmath147 was found , no band of finite dos separates the gap at @xmath60 from the gap at @xmath52 . in order to achieve a spatial resolution of the local density of states ( ldos ) we consider a symmetric model of three normal islands connected to two superconductors at @xmath60 . due to symmetry
the green s functions in the left and right normal nodes are equal .
both nodes are thus called @xmath153 in the following , the central node is called @xmath154 .
the nodes @xmath153 are connected via a ballistic conductance @xmath148 to the superconductor and via @xmath149 to @xmath154 . in each
normal node , electron - hole - decoherence is described through thouless energies @xmath155 , respectively .
the system setup is sketched in fig .
[ fig:8 ] .
matrix current conservation in one of the nodes @xmath153 and in the node @xmath154 determines the green s function and the ldos in the particular node @xcite : @xmath156&=0,\\ -2
\hat i_{12 } + i(2 g_2)(e / e_{th}^2)[\hat\tau_3,\hat g_2(e)]&=0.\end{aligned}\ ] ] the matrix currents @xmath157 and @xmath158 are defined according to the definition ( [ eq : curr ] ) . in general , the system is described by three parameters @xmath159 , @xmath160 , and @xmath147 .
we fix one parameter by considering only systems with equal mean level spacings in all three normal nodes @xmath161 .
furthermore , we fix @xmath162 . in this case for @xmath151 the ldos in @xmath153 and @xmath154 are equal and correspond to the result of figure 1(a ) in @xcite , and for @xmath152 we have the bcs - dos in @xmath153 and again the result of figure 1(a ) of @xcite in @xmath154 .
figure [ fig:9 ] shows the ldos in the two nodes for @xmath163 [ fig .
[ fig:9 ] ( a ) ] and for @xmath164 [ fig .
[ fig:9 ] ( b ) ] . taking this into account ,
@xmath159 and @xmath160 can be expressed in terms of @xmath147 .
we find @xmath165 9a.pdf ( 3,66)(0,3)@xmath13 9b.pdf ( 3,66)(0,3)@xmath14 in fig .
[ fig:10 ] , we show the numerical results for the ldos in both normal nodes in dependence of @xmath147 and energy in the secondary gap region below the superconducting gap edge @xmath2 . figure [ fig:10 ] ( a ) shows the result in the outer nodes @xmath153 , and fig .
[ fig:10 ] ( b ) shows the result for the inner node @xmath154 .
the white region in fig .
[ fig:10 ] ( a ) denotes @xmath166 .
the main finding of our calculations is that the ldos in the two nodes differs only where @xmath167 in both nodes .
whenever it is zero in the central node it is also zero in the outer nodes and vice versa .
we thus find a behavior of the secondary gap similar to what is already known from the usual minigap @xcite .
the width of this gap is not position dependent , only the ldos above / below the particular gap edge varies with position .
10a.pdf ( 3,80)(0,3)@xmath13 10b.pdf ( 3,80)(0,3)@xmath14 we thus expect the secondary gap not only in the ldos of a singular point , but as well in the integrated dos of a finite region .
depending on the parameters , not for every system does a secondary gap appear .
however , if it appears in one point , it exists also in every other point of the normal part .
the previously used model with only a single normal node between the superconductors is thus sufficient if the main interest concerns the existence of the secondary gap and its properties . however , with this method we can not calculate a position - resolved ldos and thus can not make statements about the integrated dos in the energy interval between minigap and secondary gap .
upper plot : sketch of the system with a dirty normal metal between two superconductors @xmath168 and @xmath169 .
the upper plot shows one possible path connecting left and right superconductors with a single scattering event . in the lower plot ,
the description of excitations following such paths is shown in terms of an impurity scattering matrix ( @xmath170 ) in the normal region and andreev reflection at the superconductors . ]
the secondary gap we found for diffusive josephson systems was calculated using green s function techniques in the quasiclassical approximation . whereas this method is very powerful in calculating expectation values of physical observables
, it does not provide a simple intuitive explanation for the absence of andreev levels in the secondary gap region and the dependence of these levels on the phase difference @xmath9 between the superconductors . in this section ,
we investigate a simple 1d scattering model which is able to explain qualitatively the secondary `` smile''-gap .
however , since we deal with diffusive or chaotic scattering systems with large conductance , we should not expect to reproduce the details of 3d solutions .
we consider a semiclassical path between the left ( @xmath168 ) and the right ( @xmath169 ) superconductor [ fig .
[ fig:11 ] ] and first recall the characteristics of andreev bound states between superconductors on a ballistic trajectory .
the bound state energies follow from the semiclassical quantization condition : @xmath171 here , the first term is the phase difference acquired between electron and hole upon traversing the normal region .
@xmath11 is essentially the inverse traversal time , which could also be due to ballistic motion @xmath172 for a trajectory of length @xmath173 .
the second factor is twice the energy - dependent andreev reflection phase and the third term the phase difference @xmath9 between the superconducting order parameters .
all terms together have to add to an integer multiple of @xmath174 .
this equation reproduces two limiting cases . in long junction limit @xmath175
, we replace @xmath176 and find the usual spectrum of andreev levels @xmath177 . in this case , levels move up and down in energy linearly with the phase difference @xmath9 .
the lowest positive energy states have the energies @xmath178 .
the levels split with @xmath9 and cross 0 at @xmath179 , corresponding to the closing of the minigap . in the opposite , short junction limit @xmath180 , we neglect the first term in eq .
( [ eq : bohr_sommerfeld ] ) and find the andreev levels @xmath181 .
the most interesting case is the `` not - so - short '' junction limit @xmath182 . assuming the energy is close to @xmath2
, we can replace @xmath183 by @xmath2 in the first term of eq .
( [ eq : bohr_sommerfeld ] ) , and taking @xmath184 we obtain @xmath185\,.\end{aligned}\ ] ] thus , we obtain two states shifted in phase by the ( small ) parameter @xmath186 .
they touch the gap at the critical phase @xmath187 and the maximal distance to @xmath2 ( at @xmath60 ) is @xmath188 .
this is in quantitative agreement with the characteristics of the secondary gap found previously within the quasiclassical green s function theory .
note that in the present approximation , the two levels cross at @xmath60 .
we can expect that finite backscattering will lead to an anticrossing and the phase dependence of the level resembles the `` smile '' shape of the secondary gap .
we investigate a simple model for the anticrossing and calculate the andreev bound - state energies for a 1d - model with impurity scattering modeled by a scattering matrix .
although this model takes only backward scattering into the same trajectory into account and neglects the complex interference effects of three - dimensional impurity scattering which are covered by our original green s function calculations , the results provide an understanding of the phase - dependent andreev level density of states .
the bound - state energies are obtained from the scattering matrices in the normal region @xcite .
we consider the geometry shown in fig .
[ fig:11 ] .
the normal scattering matrix encompasses the back scattering at the impurity as well as the dynamical phases along the trajectory to the superconductor and is given by @xmath189 where @xmath190 $ ] accounts for the position of the impurity along the path and @xmath191 is the transmission probability . the normal region scattering matrix for holes is related through @xmath192 .
12b.pdf ( 3,60)(0,3)@xmath13 12a.pdf ( 3,60)(0,3)@xmath14 the scattering matrices for electron - hole conversion at the interface to the superconductors are given by @xmath193 $ ] and @xmath194 , respectively .
note that the @xmath195-space is not nambu space .
an electron arriving at either superconductor is reflected as a hole traveling towards the normal region from the same side , thus andreev reflection is described by a diagonal matrix .
the condition for a bound state reads as @xmath196=0.\ ] ] the bound - state energies in dependence of @xmath9 are plotted in fig .
[ fig:12 ] for different values of @xmath197 . without backscattering in the normal region , at @xmath60
the two andreev levels are degenerate [ fig .
[ fig:12 ] ( a ) ] . taking into account impurity scattering in the normal part [ fig .
[ fig:12 ] ( b ) ] this degeneracy is lifted ( the exact curve depends on the position where scattering occurs , i. e. , on the parameter @xmath119 ) .
this results in the characteristic shape of the minigap and the secondary `` smile''-gap below @xmath63 .
figure [ fig:12](b ) shows the @xmath119-averaged results for andreev bound states with one scattering event with @xmath198 ( weak scattering ) .
it is worth mentioning that only channels without scattering contribute to the zero - energy andreev states at @xmath179 ( not shown ) . for paths with one or more scattering events ( more scattering matrices in the normal part ) , these levels
are shifted to higher energies .
thus , we have shown that the secondary gap can be understood from the phase - dependence of the andreev level when the junction length exceeds a length of the order of the superconducting coherence length , given by @xmath182 .
the `` smile '' shape can be traced back to the effect of backscattering .
to summarize , we have calculated the local density of states for diffusive josephson systems for a wide range of contact types with attention to the energy range below @xmath2 , in which a secondary gap can appear .
we have generalized previous calculations for ballistic contacts @xcite and shown that the secondary `` smile''-gap is a robust feature in the proximity density of states for large thouless energies .
we thus suggest that this feature should be accessible to an experimental detection by means of high - resolution scanning tunneling spectroscopy and want to encourage research in this direction . _
acknowledgments_. j. r. and w. b. were supported by the dfg through sfb 767 and be 3803/5 and by the carl zeiss foundation .
y. n. and l. g. thank the aspen center for physics , supported in part by nsf grant no .
phys-1066293 , for hospitality .
work at yale is supported by nsf dmr grant no . | the superconducting proximity effect leads to strong modifications of the local density of states in diffusive or chaotic cavity josephson junctions , which displays a phase - dependent energy gap around the fermi energy .
the so - called minigap of the order of the thouless energy @xmath0 is related to the inverse dwell time in the diffusive region in the limit @xmath1 , where @xmath2 is the superconducting energy gap . in the opposite limit of a large thouless energy @xmath3 ,
a small new feature has recently attracted attention , namely , the appearance of a further secondary gap , which is around two orders of magnitude smaller compared to the usual superconducting gap .
it appears in a chaotic cavity just below the superconducting gap edge @xmath2 and vanishes for some value of the phase difference between the superconductors .
we extend previous theory restricted to a normal cavity connected to two superconductors through ballistic contacts to a wider range of contact types .
we show that the existence of the secondary gap is not limited to ballistic contacts , but is a more general property of such systems .
furthermore , we derive a criterion which directly relates the existence of a secondary gap to the presence of small transmission eigenvalues of the contacts . for generic continuous distributions of transmission eigenvalues of the contacts ,
no secondary gap exists , although we observe a singular behavior of the density of states at @xmath2 .
finally , we provide a simple one - dimensional scattering model which is able to explain the characteristic `` smile '' shape of the secondary gap . |
the study of physical phenomena that arise in strongly correlated systems far from equilibrium has become a field of highly active research recently . @xcite for the theoretical description of such systems , greens - function - based approaches starting from the keldysh formalism @xcite have proven to be very useful . a number of different approximation schemes rely on this concept .
@xcite central to these approaches is the self - energy which is related to the one - particle green s function through dyson s equation .
however , while the numerical solution of dyson s equation is rather straightforward in the equilibrium case , the computational effort is considerably increased for systems out of equilibrium since operations with matrices depending on two independent contour time variables typically scale cubically in the number of time steps .
apart from other challenges characteristic for the respective approach , already this scaling poses a severe limit on the maximal reachable propagation time in a numerical calculation .
applying additional concepts or approximations , such as the generalized kadanoff - baym ansatz@xcite or exploiting a rapid decay of the memory,@xcite are necessary to overcome this limitation .
it was proposed recently@xcite that it can be advantageous to avoid the direct inversion of dyson s equation by applying a mapping onto a markovian propagation scheme . to this end it is necessary to _ assume _ the existence of a certain functional form for the nonequilibrium self - energy , namely the existence of a lehmann representation . in the present paper
we explicitly construct this lehmann representation . with this at hand
, we pick up the proposed idea to solve dyson s equation by means of a markovian propagation and exploit the fact that the lehmann representation of the exact self - energy of a small reference system has a finite number of terms only .
this allows us to solve dyson s equation with an effort that scales linearly in the maximum propagation time @xmath1 . for equilibrium green s functions ,
the lehmann representation is a well established concept.@xcite it uncovers the analytical properties of the green s function and can be used to show that the related spectral function is positive definite .
it is further essential for the evaluation of diagrams through contour integrations in the complex frequency plane , for the derivation of sum rules , etc .
the generalization of the lehmann representation to nonequilibrium green s function is straightforward.@xcite applications include nonequilibrium dynamical mean - field theory ( dmft ) where it allows for a hamiltonian - based formulation of the impurity problem.@xcite the explicit construction of a lehmann representation for the _ self - energy _
, on the other hand , turns out to be more tedious , already for the equilibrium case : in a recent work such a construction was worked out @xcite from a diagrammatic perspective and used to cure the problem of possibly negative spectral functions arising from a summation of a subclass of diagrams . here
, we address the nonequilibrium self - energy of a general , interacting lattice - fermion model : ( i ) we rigorously show the existence of the lehmann representation by presenting an explicit construction scheme that is based on the lehmann representation of the nonequilibrium green s function .
( ii ) using a simple example , namely the cluster - perturbation theory @xcite ( cpt ) , we furthermore demonstrate that the lehmann representation of the self - energy can in fact be implemented numerically and used to study the time evolution of a locally perturbed hubbard model on a large square lattice ( @xmath0 sites ) .
propagation times of several orders of magnitude in units of the inverse hopping amplitude can be reached with modest computational resources .
( iii ) while the cpt approximation for the self - energy is rather crude and shown to violate a number of conservation laws , it is possible with this approximation to study the weak - coupling limit of the hubbard model in a reasonable way . in particular
we demonstrate that prethermalization physics is already captured on this level .
the paper is organized as follows : in section [ sec : repg ] we briefly discuss the generalization of the lehmann representation to nonequilibrium green s functions .
the main idea of ref . about the markovian propagation scheme
is recalled in sec .
[ sec : prevres ] . the explicit construction scheme for the nonequilibrium self - energy
is outlined in section [ sec : expl_const ] .
section [ sec : appl_cpt ] is devoted to the application of our formalism to the cluster - perturbation theory .
[ sec : numerics ] presents numerical results for the time evolution of a local perturbation in the fermionic hubbard model .
we conclude the paper with a summary and an outlook in sec .
[ sec : summary ] .
we consider an arbitrary , fermionic model hamiltonian @xmath2 where the indices @xmath3 run over the possible one - particle orbitals ( lattice sites , local orbitals , spin projection , ... ) .
fermions in such states are created ( annihilated ) by the operators @xmath4 ( @xmath5 ) . at time @xmath6 , the system with hamiltonian @xmath7 is assumed to be in thermal equilibrium with inverse temperature @xmath8 and chemical potential @xmath9 .
non - equilibrium real - time dynamics for @xmath10 is initiated by the time dependence of the one - particle or the interaction parameters .
this covers challenging experimental setups such as time - resolved photoemission spectroscopy@xcite or experiments with ultracold gases in optical lattices.@xcite the one - particle green s function is given by @xmath11}\right)}},\end{aligned}\ ] ] where `` @xmath12 '' traces over the fock space , i.e. , we take averages using the grand - canonical ensemble . defines the grand - canonical partition function and @xmath13 the time - ordering operator on the l - shaped keldysh - matsubara contour @xmath14 ( ) .
the time variables @xmath15 and @xmath16 are understood as contour times that can lie on the upper , lower or matsubara branch of @xmath14 .
we further introduce the convention that operators with a hat carry a time dependence according to the heisenberg picture , i.e. , @xmath17 , where @xmath18 is the system s time - evolution operator and @xmath19 the time - ordering operator .
an in - depth introduction to the keldysh formalism @xcite can be found in refs . , .
as has been shown in ref . , the one - particle green s function can be cast into the form @xmath20 which we will call its lehmann representation in the following .
@xmath21 is the non - interacting green s function of an isolated one - particle mode ( @xmath22 ) with excitation energy @xmath23 : @xmath24 e^{-i\epsilon(t - t')}.\ ] ] here , @xmath25 denotes the fermi - function while @xmath26 refers to the contour variant of the heaviside step function ( @xmath27 for @xmath28 , @xmath29 otherwise ) .
@xmath30 is defined to be equal on the upper and lower branch of the contour and furthermore constant on the matsubara branch with @xmath31 and @xmath32 $ ] . if the eigenstates @xmath33 of the initial hamiltonian ( i.e. , @xmath34 ) are used as a basis for tracing over the fock space in eq . , one has @xmath35 where @xmath36 and where the superindex @xmath37 labels the possible one - particle excitations with corresponding excitation energies @xmath38 .
note that this definition of @xmath30 indeed satisfies @xmath31 .
we emphasize that @xmath39 as a matrix is not quadratic .
our expression can be seen as a direct generalization of the time - independent @xmath39-matrix discussed in ref . .
we further note that the rows of the @xmath39-matrix fulfill the orthonormality condition @xmath40_{ij}=\sum_\alpha q_{i\alpha}(t)q_{j\alpha}^*(t ) = { \ensuremath{\langle { \left\{{\ensuremath{\hat{c}_{i}}}(t),{\ensuremath{\hat{c}^\dagger_{j}}}(t)\right\ } } \rangle}}_{h}=\delta_{ij},\ ] ] where @xmath41 denotes the anticommutator . .
@xmath42 denotes the upper branch , @xmath43 the lower branch and @xmath44 the matsubara branch . in the shown example
@xmath15 is later than @xmath16 in sense of the contour , denoted as @xmath45 in the text.,scaledwidth=40.0% ]
in several greens - function - based methods , an approximate self - energy @xmath46 is obtained from a small reference system using exact diagonalization . the desired one - particle green s function @xmath47 of a much larger system
is then obtained through dyson s equation @xmath48_{ij}(t , t ' ) \\ & \,\,\,+ \int_c\int_c { \ensuremath{\mathrm{d}t}}_1{\ensuremath{\mathrm{d}t}}_2 \sum_{k_1k_2}[g_0]_{ik_1}(t , t_1)\sigma'_{k_1k_2}(t_1,t_2 ) g_{k_2j}(t_1,t'),\nonumber\end{aligned}\ ] ] where @xmath49 denotes the non - interacting green s function ( i.e. , @xmath50 ) of the model given by eq . .
typical examples include dynamical mean - field theory ( dmft ) , @xcite where @xmath46 is obtained from a single - impurity anderson model,@xcite or cluster - perturbation @xcite and self - energy functional theory , @xcite where @xmath46 stems from a small reference system . to solve eq .
numerically , a discretization of the continuous time - contour @xmath14 is necessary .
the number of time steps required to reach a given maximal time is dependent on the lowest relevant timescale that is set by a given hamiltonian .
based on this discretization , the effort required to solve eq . for @xmath47 scales cubically in the number of time steps and also the system size . despite this challenge also the memory consumption , which scales quadratically in these quantities ,
poses a problem .
progress was made recently @xcite by introducing a mapping of eq . onto a markovian propagation - scheme .
the idea proposed by the authors of ref . relies on the _ assumption _ that the self - energy can be written in the following form : @xmath51 here , @xmath52 denotes the time - local hartree - fock term .
this decomposition is very similar to the expression eq . for the green s function .
we will refer to this as the lehmann representation of the self - energy .
the immediate and important advantage of the lehmann representation is that the self - energy can be interpreted as a hybridization function.@xcite this property allows to write down an effective non - interacting model with hamiltonian @xmath53 the @xmath54-degrees of freedom represent `` virtual '' orbitals in addition to the physical degrees of freedom labeled by @xmath55 .
they form an `` effective medium '' with on - site energies @xmath56 and hybridization strengths @xmath57 such that the _ interacting _ green s function of the original model is the same as the green s function of the effective non - interacting model on the physical orbitals : @xmath58 with this simple construction , the inversion of the dyson equation can be avoided in favor of a markovian time propagation within a non - interacting model . as a successful benchmark ,
an interaction quench in an inhomogeneous hubbard model was treated with nonequilibrium dmft in ref . using self - consistent second - order perturbation theory as impurity solver . on the theoretical side , however , it remained an open question if the existence of a lehmann representation must be postulated or if this is a general property of the nonequilibrium self - energy . in the following we explicitly derive eq .
for the exact self - energy corresponding to the general , interacting hamiltonian defined in eq . ,
i.e. , we show that the exact self - energy can always be written in the form of a lehmann representation .
the proposed construction scheme is not only useful as an analytical tool but also well suited for numerical applications where an approximate self - energy is obtained from a small reference system using exact diagonalization . in this case
the number of virtual orbitals is constant and the effort for solving eq
. scales linearly in @xmath1 .
this is a great advantage if one is interested in long - time dynamics .
we start our construction from the lehmann representation of @xmath47 as stated in eq . .
for our model hamiltonian the associated one - particle excitation energies @xmath59 and the @xmath39-matrix are given by eq . .
the self - energy is related to this representation through dyson s equation @xmath60 .
however , the inverse @xmath61 can not directly be calculated with eq . since @xmath30 is not quadratic . as a first step we block up the matrix @xmath30 to a quadratic form .
this is achieved by interpreting its orthonormal rows ( cf .
eq . ) as an incomplete set of basis vectors .
@xmath30 itself is an incomplete unitary transform from this viewpoint .
we now pick an arbitrary , pairwise orthonormal completion of this basis to find an unitary transform @xmath62 that contains @xmath30 in its upper block ( cf .
[ fig : qmat ] ) .
the next steps of our discussion will be independent of the particular completion that is chosen .
the only mathematical requirement is that it is as smooth ( and thus differentiable ) in the time variable @xmath15 as @xmath30 ; see appendix [ ap : numerics ] for numerical details on the construction of @xmath62 .
the completed unitary transform @xmath62 describes additional virtual orbitals ( labeled by the index @xmath54 , see fig .
[ fig : qmat ] and eq . ) . for convenience ,
we also absorb in the definition of @xmath62 the extra factor @xmath63 that stems from the non - interacting green s function @xmath64 ( cf .
eqs . and ) . for clarity in the notations
we use the following index convention throughout this paper @xmath65 like every time - dependent unitary transform , @xmath30 is generated by an associated hermitian matrix .
we define .
the matrix @xmath66 contains a completing set of orthonormal basis vectors in its rows . for convenience ,
the phase factor @xmath67 is also absorbed into @xmath62 . the generating , hermitian matrix @xmath68 ( cf .
can be assumed to be diagonal in the virtual sector.,scaledwidth=45.0% ] @xmath69 o_{\alpha y}^\dagger(t).\ ] ] indeed , by integration we have @xmath70 and furthermore @xmath68 is hermitian : @xmath71o^\dagger(t ) = i\partial_t [ o(t)o^\dagger(t)]-o(t)i\partial_to^\dagger(t)\nonumber\\ & = \left ( [ i\partial_t o(t)]o^\dagger(t ) \right)^\dagger = h^\dagger(t).\end{aligned}\ ] ] we now require the virtual part @xmath72 to be diagonal and time - independent , i.e. , @xmath73 . to this end
we use our freedom in choosing the completing basis vectors @xmath74 which allows us to perform the associated unitary transform in the virtual sector ( see fig . [ fig : qmat ] ) . with the resulting @xmath75
we define the single - particle hamiltonian @xmath76 @xmath77 which has precisely the form of the effective hamiltonian stated in eq . .
the requirement of a diagonal virtual sector defines the effective hamiltonian uniquely up to rotations in invariant subspaces . at time @xmath6 , the effective medium can be stated in a diagonal form which is useful for the evaluation of the corresponding one - particle green s function .
we recall that we required @xmath62 to be as smooth as @xmath30 and take a look at @xmath78_{t=0}=h(0)o(0)=o(0 ) m,\ ] ] where @xmath79 .
. implies in particular that @xmath80_{t=0}=q(0 ) m$ ] ( cf .
fig.[fig : qmat ] ) . however , from eq .
one easily evaluates @xmath81_{i\alpha}]_{t=0 } = q(0)_{i\alpha } \epsilon_\alpha $ ] and we can thus identify @xmath82 .
putting everything together we find @xmath83 we require that the effective medium is initially in thermal equilibrium with the same inverse temperature @xmath8 and the same chemical potential @xmath9 as the physical system .
the associated one - particle green s function of the medium is defined as @xmath84 recalling the diagonal form of the effective medium at @xmath6 ( cf .
eq.[eq : heff_eq ] ) and using that the effective hamiltonian is non - interacting , we can easily rewrite this expression into @xmath85 o^*_{y \alpha}(t).\ ] ] the physical sector of @xmath86 is by construction identical with the lehmann representation of @xmath47 : @xmath87 @xmath86 encodes the full information on the one - particle excitations of the system defined by the hamiltonian .
further stresses the fact that in principle any ( sufficiently smooth ) completion of @xmath30 to a unitary transform @xmath62 leads to a valid effective hamiltonian .
the physical sectors of @xmath62 and @xmath68 remain independent of its choice .
the virtual sectors , on the other hand , are affected and only the special choice of @xmath62 ( cf .
the discussion above and below eq . ) guarantees a diagonal form of the effective medium .
having found an effective , non - interacting model that reproduces the correct green s function , it remains to link this back to the self - energy .
the time - non - local ( correlated ) part @xmath88 follows by tracing out the virtual orbitals .
this procedure is straightforward as they are all non - interacting and we can use , e.g. , a cavity - like ansatz @xcite or an equation of motion based approach .
@xcite this results in a hybridization - like function @xmath89 that encodes the influence of the virtual sites on the physical sector .
the green s function at the physical orbitals is then obtained from a dyson - like equation @xmath90_{ij}(t , t'),\end{aligned}\ ] ] where @xmath91_{ij}(t , t')=[i\partial_t - h_{ij}(t)]\delta_c(t , t'),\end{aligned}\ ] ] with @xmath92 as the contour delta function . to make the final connection to the self - energy we evaluate the physical sector of @xmath93 .
with @xmath94{\ensuremath{\vert n \rangle}}\nonumber\\ & = \sum_{j}\left(t_{ij}(t)-\mu\delta_{ij}\right)q_{j(m , n)}(t)\nonumber\\ & \phantom{=}~~+\sum_{ji'j'}u_{ii'jj'}(t)z_{(m , n)}{\ensuremath{\langle m \vert } } { \ensuremath{\hat{c}^\dagger_{i'}}}(t){\ensuremath{\hat{c}_{j'}}}(t){\ensuremath{\hat{c}_{j}}}(t){\ensuremath{\vert n \rangle}}\end{aligned}\ ] ] we obtain @xmath95 at the physical orbitals the effective hamiltonian is thus determined by the hartree - fock hamiltonian . by comparison of eq . with the dyson equation @xmath96_{ij}(t , t'),\ ] ]
where @xmath97_{ij}(t , t ' ) = \left [ i\partial_t - ( t_{ij}(t)-\mu\delta_{ij } ) \right]\delta_c(t , t'),\ ] ] we finally identify @xmath98 concluding our construction of the self - energy .
let us stress that with eqs . , and we now have an _ explicit _ recipe to construct the lehmann representation of the self - energy .
this representation is further unique as follows from the uniqueness of the corresponding effective hamiltonian ( cf . the discussion above and below eq . ) . with the hamiltonian of the effective medium , eq . ,
at hand , a number of useful properties follow immediately : by taking a look at the matsubara branch only , one can link the lehmann representation of the self - energy to the positive definiteness of its equilibrium spectral function . with @xmath99
we can perform the usual fourier transform from imaginary time to matsubara frequencies and then find the analytical continuation @xmath100 to the complex - frequency plane ( see for example ref . ) .
the spectral function is defined as @xmath101\end{aligned}\ ] ] for real @xmath102 .
this can explicitly be calculated from the parameters of the effective hamiltonian .
one finds : @xmath103 where @xmath104 is the dirac delta function .
the positive definiteness for every @xmath102 is immediately evident .
the self - energy and its time derivatives can be used to calculate certain expectation values of higher order .
prominent examples include the interaction energy or the local double occupation .
their calculation is based on the evaluation of contour integrals of the form @xmath105 . by comparing the equations of motion for @xmath106 and @xmath107
one readily finds the identity @xmath108
t')\nonumber\\ & + \sum_s h_{is}(t ) f_{si'}(t , t').\end{aligned}\ ] ] this is a remarkable relation as the contour integration can be avoided in favor of a simple matrix multiplication . a convenient tool to drive quantum systems out of equilibrium is given by the so - called quantum quenches . here
, one ( or more ) parameters of the system are changed suddenly .
this sudden change reflects itself as a discontinuous time dependence of the effective hamiltonian : assume that the system is subjected to a quench at time @xmath6 , so that @xmath109 .
initially the system is in thermal equilibrium and the effective hamiltonian is given by eq . , where @xmath59 are the excitations energies of @xmath110 .
the @xmath111-matrix is continuous at @xmath6 despite the quantum quench ( it only depends on @xmath112 , cf .
its time derivative , however , is not and thus @xmath68 jumps from @xmath113 to @xmath114_{t=0^+ } o^*_{j\alpha}(0).\ ] ] after this jump , the effective hamiltonian will in general not be constant for times @xmath10 , i.e. , @xmath115 .
the simplest numerical application of our formalism is given by cluster - perturbation theory@xcite ( cpt ) .
the idea of cpt is to split the system into small clusters which can be treated by means of exact - diagonalization techniques .
the cluster self - energies are then used as approximate input for the dyson equation to obtain the cpt green s function .
the same concept is part of more powerful approaches like dmft @xcite or self - energy functional theory @xcite where the cpt green s function is self - consistently or variationally linked to the self - energy of a reference system .
the following construction of an effective hamiltonian for cpt applies to such techniques as well .
the sites @xmath3 lie within the same cluster @xmath116 , @xmath117 belongs to a different cluster @xmath118 .
the cluster diagonal part of the hopping matrix @xmath119 describes the intra - cluster , the cluster off - diagonal part @xmath120 ( @xmath121 ) the inter - cluster - hopping.,scaledwidth=20.0% ] from now on we restrict ourselves to the fermionic hubbard model .
the locality of its interaction term allows us to cast its hamiltonian into the following form : @xmath122 { \ensuremath{c^\dagger_{ii\sigma}}}{\ensuremath{c_{ij\sigma } } } + u(t)\sum_i n_{ii\uparrow}n_{ii\downarrow } \right]}_{\text{cluster system~}h_i } \nonumber\\ & \,\ , + \underbrace { \sum_{i\neq j}\sum_{ij\sigma } t^{ij}_{ij\sigma}(t){\ensuremath{c^\dagger_{ii\sigma}}}{\ensuremath{c_{jj\sigma } } } } _
\text{inter - cluster hopping}.\end{aligned}\ ] ] here , the indices @xmath123 label the cluster systems , while the indices @xmath3 run over the sites within a cluster only ( see fig . [
fig : cpt ] ) . of course
, this is fully equivalent with the usual form of the hubbard model which is re - obtained by combining @xmath124 to a superindex , i.e. , @xmath125 . the operator @xmath126 measures the particle density with spin projection @xmath127 .
the green s function of the isolated cluster @xmath116 with intra - cluster hamiltonian @xmath128 is @xmath129 so that @xmath130^{-1}_{ij\sigma}(t , t')= [ i\partial_t- ( t^{ii}_{ij\sigma}(t)-\mu\delta_{ij } ) ] \delta_c(t , t ' ) - \sigma^i_{ij\sigma}(t , t'),\ ] ] where @xmath131 denotes the corresponding self - energy .
we further define @xmath132 with the inter - cluster ( @xmath133 ) hopping @xmath134^{ij}_{ij\sigma } = ( 1-\delta_{ij } ) t^{ij}_{ij\sigma}$ ] the cpt green s function is defined as @xmath135 where @xmath136^{ij}_{ij\sigma}(t , t')=[i\partial_t -(t^{ij}_{ij\sigma}(t ) -
\mu \delta_{ij}\delta_{ij})]\delta_c(t , t')$ ] .
the definition of @xmath137 reveals that cpt becomes exact in the limit of vanishing interaction .
we then have @xmath138 and thus @xmath139^{-1}=g_0 $ ] . solving eq . in case of non - vanishing @xmath46 , on the other hand , requires the solution of a dyson equation .
this brings us back to our original problem . using our results from sec .
[ sec : lsig ] we can avoid the solution of the dyson equation and rather decompose the self - energies of the isolated clusters into their lehmann representations : @xmath140^i_{ij\sigma}(t)\nonumber\\ & + \sum_{s}h^{i}_{is\sigma}(t)g(h^i_{ss\sigma};t , t')[h^{i}]^*_{js\sigma}(t').\end{aligned}\ ] ] here , @xmath141 are the parameters of the effective medium corresponding to the @xmath116-th cluster .
we define @xmath142 it is now straightforward to realize that the inclusion of the inter - cluster hopping by means of eq .
is completely trivial in this language .
namely , @xmath143 with @xmath144^{ij}_{xy\sigma } { \ensuremath{c^\dagger_{ix\sigma}}}{\ensuremath{c_{jy\sigma}}},\ ] ] we then have @xmath145^{ij}_{ij\sigma}(t , t')=-i{\ensuremath{\langle { { \ensuremath{\mathcal{t}_c}}{\ensuremath{\hat{c}_{ii\sigma}}}(t){\ensuremath{\hat{c}^\dagger_{jj\sigma}}}(t ' ) } \rangle}}_{h^\mathrm{cpt}}.\ ] ] while this is an easy and intuitive description , we remark that @xmath146 includes virtual orbitals .
the inter - cluster hopping @xmath147 , on the other hand , is defined solely in the physical sector and has to be blocked up accordingly ( @xmath148^{ij}_{rs\sigma } = [ t^\mathrm{ic}(t)]^{ij}_{is\sigma } = [ t^\mathrm{ic}(t)]^{ij}_{rs\sigma}=0 $ ] ) . as an important observable
we briefly discuss the calculation of the total energy within cpt .
while the kinetic energy follows straightforwardly from the one - particle density matrix as @xmath149 , the interaction energy can only be accessed indirectly through the self - energy .
it is given by @xmath150 the evaluation of this contour - integral in eq .
is straightforward within our formalism by using eq . .
the study of real - time dynamics initiated by an interaction quench in the hubbard model has attracted much attention recently .
@xcite here , the system is prepared in a thermal ( usually non - interacting ) initial state and then , after a sudden change of the interaction parameter @xmath151 , evolves in time as prescribed by the interacting hamiltonian .
while the setup is apparently simple , the search for universal properties of the time evolution remains notoriously difficult due to the non - integrability of the hubbard model in two and higher dimensions .
apart from the general assumption that non - integrable models feature thermalization and thus lose memory of the initial state in the long - time limit,@xcite only the time evolution after quenches to a weak , finite hubbard @xmath151 seems to be well understood so far . here , it could be shown by means of weak - coupling perturbation theory @xcite that observables initially relax to non - thermal , quasistationary values ( the system prethermalizes ) before the significantly slower relaxation towards the thermal values sets in .
it was later worked out@xcite that the mechanism which traps the system in a quasi - stationary prethermal state is quite similar to the mechanism that hinders non - interacting systems from thermalizing . in the latter case
the integrability of the hamiltonian leads to a large number of constants of motion that highly constrain the dynamics of the system . in case of weakly interacting systems
it is the proximity to the integrable point that introduces approximate constants of motion and hinders relaxation beyond the prethermalization plateau on short timescales @xmath152 ( here , @xmath153 is the nearest - neighbor hopping ) .
relaxation towards the thermal average is delayed until later times ( @xmath154 ) . as a proof of concept of our formalism we use nonequilibrium cpt to investigate the short- and long - time dynamics of an inhomogeneous initial state after an interaction
quench in the hubbard model .
in particular we will study if and to what extent the cpt is able to describe prethermalization and the subsequent relaxation to a thermal state .
we consider the hubbard model at zero temperature ( @xmath155 ) and half - filling ( @xmath156 ) on a square lattice of @xmath157 sites with periodic boundary conditions .
cluster indices run over @xmath158 and @xmath159 , so that the system is cut into @xmath160 clusters of size @xmath161 .
the hopping is restricted to nearest neighbors and we set @xmath162 to fix energy and time units .
translational invariance of the initial state is broken by applying a local magnetic field of strength @xmath163 to an arbitrarily chosen `` impurity site '' ( here , site @xmath164 in ):
@xmath165 where @xmath166 is non - zero and unity for nearest neighbors only and where @xmath167 and @xmath168 .
initially , the magnetic field is switched on with strength @xmath169 to induce a ( nearly ) fully polarized magnetic moment on the impurity site and then switched off for times @xmath10 : @xmath170 here , @xmath171 is the heaviside step function .
furthermore , the interaction @xmath172 is switched off initially and then switched on to a non - zero value @xmath173 @xmath174 hence , in the quantum quench considered here , two parameters are changed simultaneously .
the initial hamiltonian @xmath110 features no interactions but is inhomogeneous due to the local magnetic field , the final hamiltonian @xmath175 is translationally invariant due to the absence of the magnetic field but has a finite interaction @xmath176 . to apply nonequilibrium cpt , we use exact diagonalization to solve the @xmath160 independent cluster problems and to construct the hamiltonian of the effective medium ( for details on the numerical implementation
see appendix [ ap : numerics ] ) .
finally , eq . is used to account for the inter - cluster hopping .
the number of non - zero elements of a cluster s @xmath39-matrix and therefore the computational effort of our approach increases quadratically with the number of active states in the density matrix @xmath177 ( @xmath178 ) , i.e. , states that contribute with a significant weight @xmath179 to thermal averages . for convenience
we have therefore chosen a zero - temperature initial state and consider a weak interaction @xmath180 to lift the ground - state degeneracy present in the non - interacting system ( denoted as @xmath181 in the following ) .
the effective hamiltonian @xmath182 for each cluster is then of size @xmath183 and the final cpt hamiltonian of size @xmath184 . exploiting its sparse form we are able to perform @xmath185 time steps with @xmath186 to reach a maximal time @xmath187 with modest computational effort . for comparison
we note that prior studies based on the nonequilibrium cpt , e.g. refs . ,
have been limited to @xmath188@xmath189 inverse hoppings . , blue line ) and its nearest neighbors ( @xmath190 , green line ) .
the dark - blue ( dark - green ) arrow , which is pointing from right to left , indicates the long time average of the blue ( green ) curve .
the light - blue ( light - green ) arrow , which is pointing from left to right , indicates the analytical average .
the long time average was taken over @xmath191 data points in the interval @xmath192 $ ] . ]
the partitioning of the lattice into @xmath161 clusters by cpt breaks rotational and reflection symmetries of the original problem .
these are restored by averaging the resulting one - particle density matrix over the @xmath193 possible ways to cut the lattice into @xmath161 clusters . in the following
we will show results for the time evolution of the local magnetic moment @xmath194 at the impurity ( @xmath195 ) and at its nearest neighbors ( @xmath190 ) . only the latter
are affected by the averaging .
it restores the equivalence of nearest neighbors that lie in the same and nearest neighbors that lie in a neighboring cluster of the impurity .
the extensive quantities total energy @xmath196 ( cf .
eq . and preceding discussion ) and total
magnetization @xmath197 are both unaffected by the averaging .
the initial state is the same for all quenches discussed in the following .
we find a polarization of @xmath198 at the impurity which is partially screened ( e.g. , @xmath199 ) so that the total magnetization amounts to @xmath200 .
we first discuss the non - interacting case , i.e. , a purely magnetic quench where @xmath201 . here , cpt predicts the exact time evolution ( cf . the discussion below eq . ) since the cluster self - energies @xmath202 vanish .
our results are shown in fig .
[ fig : non_int ] . for short times ( @xmath203 $ ] ) the local magnetic moment at
the impurity @xmath195 ( blue line ) decays to a value slightly above zero . subsequently ( @xmath204 $ ] ) the dynamics is governed by collapse - and - revival oscillations caused by the finite system size . in particular we find that @xmath195 returns arbitrarily close to its initial value for large times .
this is readily understood from the fact that the system s dynamics is governed by the one - particle propagator @xmath205 where @xmath206 denotes the final hopping matrix ( i.e. , after the quench ) .
@xmath206 involves only a small number of different one - particle energy levels and thus @xmath207 returns arbitrarily close to the identity matrix over time . for the non - interacting system it is possible to directly access the long - time average of the one - particle density matrix .
one finds @xmath208 where we used that @xmath175 can be diagonalized by a fourier transformation involving the reciprocal lattice vectors @xmath209 ( @xmath210 denotes the lattice vector to site @xmath55 ) .
we then have @xmath211 and @xmath212 , where @xmath213 is the system size . in fig.[fig : non_int ] this prediction is compared with the numerical time average and indeed shows perfect agreement .
it is interesting to note that for non - degenerate energy levels @xmath214 one would have @xmath215 , where @xmath216 is the total number of particles with spin @xmath217 , and therefore @xmath218 .
we conclude that degeneracy of energy levels is required to find memory of the initial state encoded in the average local magnetic moments @xmath219 . for finite
@xmath173 cpt becomes an approximation and it is a priori unclear what kind of phenomena it is able to describe . in fig .
[ fig : uquench ] we show the long - time evolution for quenches to different @xmath173 . for weak @xmath220
we find a ( prethermalization - like ) separation into two different time scales .
initially the time evolution qualitatively follows the non - interacting case , i.e. , we see a fast decay of the local moment at the impurity site ( blue line ) followed by a quasi - stationary region of collapse - and - revival oscillations . for larger times these oscillations decay and
the system relaxes into a state characterized by quasi - periodic fluctuations around its long - time average ( dashed blue line ) which are driven by different frequencies . taking a look at the @xmath173 dependence of the dynamics we notice that the region of collapse - and - revival oscillations shrinks with increasing @xmath173 and finally vanishes for @xmath221 .
the system then directly relaxes into a state with fluctuations around its long - time average . for comparison , also the magnetic moment at the neighbouring sites @xmath190 is plotted . while its dynamics for short times must naturally be different from @xmath195 due to the inhomogeneous initial state , we would expect a qualitative agreement in the long - time limit if the system thermalizes .
however , this is not the case .
there remains a clear difference in the amplitude of the fluctuations around the long - time average up to the largest simulated times .
hence we conclude that the system still keeps memory of the initial state and thus does not thermalize
. having in mind the general discussion on prethermalization in sec.[sec : numerics_intro ] , one can give an intuitive interpretation of these observations based on the effective - medium approach : while the non - interacting system is isolated and its dynamics is constrained through many constants of motion , there is a large number of virtual orbitals coupled to the system in the interacting case .
these virtual orbitals act like a surrounding bath . for weak @xmath173 the virtual orbitals
are only weakly coupled to the system and their influence is delayed to large times , while initially the dynamics is constrained similar to the non - interacting case . for strong @xmath173 , on the other hand , the coupling is strong and affects the dynamics of the system considerably .
however , the number of virtual sites is still too small to allow for a complete dissipation of the information on the initial state into the bath . therefore , a thermalized state is not reached . for an exact calculation
the number of virtual sites would scale exponentially in system size .
for cpt , on the other hand , it scales exponentially only in cluster size but linearly in the number of clusters and thus in the system size .
memory of the initial state is therefore retained within the one - particle density matrix and leaves its traces in the magnetic moments as seen in our calculations .
cpt as an approximation lacks any kind of self - consistency and is thus unable to respect the fundamental continuity equations and their corresponding conservation laws .
@xcite therefore , one has to expect a violation of energy- or particle - number conservation , for example .
furthermore , in contrast to the equilibrium case where cpt interpolates between the exact limits @xmath50 and @xmath222 , it yields exact results only for quenches to @xmath223 .
the dynamics after a quench to the atomic limit @xmath224 ( with finite @xmath176 ) can not be described exactly due to the non - local entanglement of the initial state .
we thus generally expect that the quality of the cpt results degrades with increasing interaction strength .
the numerical results for the total energy , see fig .
[ fig : energy ] , confirm this expectation .
energy conservation is respected for @xmath223 , where cpt is exact . with @xmath225 and increasing , however , a significant time dependence of the total energy sets in earlier and earlier . for @xmath226 energy conservation is violated already for @xmath227 .
similar results are found for the total magnetization @xmath228 , cf .
[ fig : total_density ] . while the magnetization should be constant for all times since neither hopping nor interaction ( cf .
eqs . and ) involve spin - flip terms , we find such behavior only for short times . for longer times
oscillations arise and the conservation of total magnetization is violated . for increasing @xmath173 the oscillations set in earlier indicating again that the quality of cpt is best for values of @xmath173 close to zero .
we note that the total particle number @xmath229 , however , is conserved during the time evolution .
this holds true for a half - filled and homogeneously charged system and is due to the fact that cpt preserves particle - hole symmetry .
this can easily be understood as follows : each cluster hamiltonian is particle - hole symmetric and since each cluster is solved exactly within cpt the corresponding effective hamiltonian @xmath182 is also particle - hole symmetric .
the cpt hamiltonian is now given by eq . which additionally includes the inter - cluster hopping
however , the inter - cluster hopping is clearly particle - hole symmetric and so is the final cpt hamiltonian . .
energy conservation is respected for @xmath230 where cpt is exact ( blue line ) .
an increasingly significant violation of energy conservation is seen for larger @xmath173 . ] by cpt .
the numbers indicate the value of @xmath173 .
curves for @xmath231 are only partially plotted for better visibility . ]
concluding , we have shown that the nonequilibrium self - energy of an interacting lattice - fermion model can uniquely be decomposed into a superposition of noninteracting , isolated modes .
this decomposition is a direct analog to a well - established decomposition of equilibrium green s functions , called the _
lehmann representation_. our proof not only provides a direct scheme to construct the lehmann representation of the self - energy , and thus allows for a deeper theoretical understanding of the self - energy complementary to its diagrammatic definition , but also proves useful for practical applications . as a proof of concept we investigated the time evolution of local magnetic moments in the fermionic hubbard model after an interaction quench using nonequilibrium cluster - perturbation theory .
our formalism allowed to avoid the solution of an inhomogeneous dyson equation on the keldysh contour and we were able to propagate the one - particle density matrix up to times @xmath232
. on the physical side , quenches to weak @xmath173 turned out to be most interesting . in agreement with the predictions of general perturbative considerations , @xcite we found a separation of the dynamics into two time scales . while the system qualitatively follows the constrained dynamics of the non - interacting @xmath233 limit , the constraints are broken up for large times due to the interaction and the system shows signs of relaxation .
however , memory of the initial state persists in the density matrix up to the largest simulated times clearly indicating the absence of thermalization .
while the simple treatment of correlations by nonequilibrium cpt has shown to be enough to cover the mentioned two - stage relaxation dynamics , it also leads to a violation of the fundamental conservation laws of energy and total magnetization .
this could be fixed by additionally imposing a self - consistency condition as it is done in nonequilibrium dmft or in self - energy functional theory . due to the significant , additional complexity of these approaches , however
, simulations would again be restricted to short time scales . a simpler , more pragmatic approach might thus be preferable where , for example , local continuity equations are enforced to ensure energy , total magnetization and particle - number conservation .
@xcite such a `` conserving cluster - perturbation theory '' could allow for a complete dissipation of initial perturbations and thus total loss of the memory of the initial state .
work along these lines is in progress .
we thank roman rausch for providing an exact - diagonalization solver for the hubbard model , felix hofmann for a reference implementation of nonequilibrium cpt , and martin eckstein and karsten balzer for helpful discussions . this work has been supported by the excellence cluster `` the hamburg centre for ultrafast imaging - structure , dynamics and control of matter at the atomic scale '' and by the sonderforschungsbereich 925 ( project b5 ) of the deutsche forschungsgemeinschaft .
numerical calculations were performed on the physnet computer cluster at the university of hamburg .
we assume that a small cluster is solved using exact diagonalization and that all time derivatives @xmath234 of the hamiltonian are known analytically .
the numerical evaluation of eq . for the @xmath39-matrix is straightforward within exact diagonalization .
its @xmath235-th derivative can be obtained as follows .
we have @xmath236 for the propagator @xmath207 .
the @xmath235-th derivative @xmath237 can then be calculated iteratively as it only depends on @xmath238 with @xmath239 .
using further that @xmath240^\dagger \ : , \end{aligned}\ ] ] one finds the @xmath235-th derivative @xmath241 of the annihilation operator and thus of @xmath242 , see eq . .
in the following we will assume that @xmath242 is available to arbitrary order .
we start by constructing @xmath243 , i.e. , a basis for the virtual sector .
it is easy to verify that @xmath244 defines a projector .
diagonalization of @xmath245 yields the eigenvalues @xmath164 and @xmath246 .
eigenvectors corresponding to @xmath246 are given by @xmath247 itself , eigenvectors corresponding to @xmath164 form the desired matrix @xmath248^\dagger$ ] .
initially , the effective medium is in equilibrium and thus explicitly given by eq .
at @xmath6 .
however , since we picked the completing basis vectors arbitrarily , we will have @xmath249 for @xmath250 , i.e. , generally @xmath93 will not be diagonal in the virtual sector
. explicit diagonalization of @xmath93 in the virtual sector yields a unitary transform @xmath251 @xmath252 replacing @xmath253 , we get @xmath254 , i.e. , we have found a completing basis so that @xmath93 is diagonal in the virtual sector .
assume that @xmath256 and @xmath257 are known for an arbitrary time @xmath15 .
this is at least the case for @xmath6 as we have seen so far .
we recall that we required @xmath68 to be constant in the virtual sector ( cf .
discussion below eq . )
@xmath258 i.e. , all time derivatives vanish in the virtual sector . only the hybridization elements and the physical sector yield non - trivial elements .
they follow from eq .
as @xmath259 [ o^{(n - k)}(t)]^\dagger]_{\alpha y}.\end{aligned}\ ] ] @xmath260 on the other hand only depends on @xmath261 , and @xmath262 , for @xmath263 , as readily follows from @xmath264 it is thus possible to iteratively calculate @xmath260 and @xmath255 .
we assume that @xmath62 and all derivatives of @xmath255 are known at some time @xmath15 and we want to propagate the @xmath111-matrix to @xmath265 .
analytically this can be written as @xmath266 using the magnus expansion,@xcite the propagator can be systematically expanded in @xmath267 and @xmath255 .
assuming that @xmath268 lies within the convergence radius of the magnus expansion ( this is generally expected to be the case for sufficiently small @xmath268 ) , we can reduce the propagation error arbitrarily by increasing the order . in practice ,
an evaluation of the magnus expansion using commutator - free exponential time propagators @xcite ( cfets ) allows for an efficient numerical propagation which takes advantage of the sparse form of the effective hamiltonian .
having found @xmath265 , we get @xmath269 from @xmath270_{\alpha y } \ : , \end{aligned}\ ] ] and can thus proceed by calculating @xmath271 and @xmath272 completing the circle .
we emphasize that the whole procedure is numerically exact , i.e. , the error is below machine precision , if @xmath268 is chosen sufficiently small . | it is shown that the nonequilibrium self - energy of an interacting lattice - fermion model has a unique lehmann representation .
based on the construction of a suitable non - interacting effective medium , we provide an explicit and numerically practicable scheme to construct the lehmann representation for the self - energy , given the lehmann representation of the single - particle nonequilibrium green s function .
this is of particular importance for an efficient numerical solution of dyson s equation in the context of approximations where the self - energy is obtained from a reference system with a small hilbert space .
as compared to conventional techniques to solve dyson s equation on the keldysh contour , the effective - medium approach allows to reach a maximum propagation time which can be several orders of magnitude longer .
this is demonstrated explicitly by choosing the nonequilibrium cluster - perturbation theory as a simple approach to study the long - time dynamics of an inhomogeneous initial state after a quantum quench in the hubbard model on a @xmath0 square lattice .
we demonstrate that the violation of conservation laws is moderate for weak hubbard interaction and that the cluster approach is able to describe prethermalization physics . |
we know that nearly all massive stars form in clusters ( see , e.g. de wit , these proceedings ) .
therefore , if we wish to understand massive star formation in general , we must understand the clustered mode of star formation .
super star clusters ( sscs ) are the most extreme type of stellar cluster ( and also one of the most extreme star forming environments in the universe ) , with stellar densities exceeding @xmath0 stars pc@xmath1 in their cores . as `` extreme '' objects , one might be tempted to simply think of them as cosmic curiosities .
however , these objects are not only fascinating because of their extreme nature , but also because they hold clues to an important mode of star formation during the time of galaxy assembly .
specifically , most present - day research strongly supports the idea that sscs are the adolescent precursors to the ancient globular clusters that are ubiquitous around massive galaxies in the local universe today .
however , one should also bear in mind that the infant mortality rate of sscs in likely to be quite high ( * ? ? ? * ( possibly as high as 99% , fall & zhang 2001 ) ) , and only the most robust of these objects will have survived to the ripe old age of present - day globular clusters . the low survival rate of sscs suggests that they must have been formed prodigiously in the early universe , making this mode of star formation even more important for understanding the early evolution of of today s massive galaxies .
sscs are also important components of galaxies in the local universe .
these clusters can have an incredible impact on their surrounding interstellar medium ( ism ) , and in some cases even the intergalactic medium ( igm ) .
each of these clusters can host up to thousands of massive stars that have powerful stellar winds , and these stars will all die in a violent way within a few million years of each other . the collective effect of these massive stars ( both in their lives and deaths ! ) can cause tremendous outflows , ionize large volumes of the ism and igm , and enrich vast amounts of interstellar material ( * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ( e.g. johnson et al .
2000 , heckman 2001 , martin , kobulnicky , & heckman 2002 ) ) moreover , sscs are also important tools for probing the universe .
they are the most luminous type of `` simple stellar population '' ( the constituent stars share the same age and metallicity ) , which makes them very useful `` test particles '' for investigating local galaxy properties .
in addition , sscs will be among the brightest point sources visible in galaxies out to large distances , which makes them very attractive targets for observations that must worry about surface brightness issues .
research on sscs is rapidly approaching a crossroads , and it is therefore worth briefly reviewing the tremendous advances that have been made in this field over roughly the last decade .
perhaps most importantly , we have learned that globular cluster formation continues today .
this discovery came to the forefront of cluster research with the _ hubble space telescope _ , and caused a major revision of the prevailing theories of globular cluster formation and galaxy evolution .
in particular , the previously accepted theories proposed that globular clusters were among the first bound structures formed in the universe , formed directly out of primordial material ( * ? ? ? * ( e.g. peebles & dicke 1968 ) ) .
this idea was particularly compelling because of the similarity between the predicted jean s mass for primordial conditions , and the observed mass of globular clusters .
the fact that globular clusters continue to form at the present time provides us with a window into the earlier universe when their formation was prevalent .
not only did the _ hubble space telescope _
spark the flurry of research in this area , but it has also been the work - horse ; the vast majority of super star cluster studies have utilized _ hubble _ in some way . because of _ hubble _ there is a sample of globular clusters spanning almost every galaxy type , with a range of ages and metallicities .
large statistical samples of sscs have been compiled , and their luminosity function ( and presumably mass function ) is well constrained ( a power - law of @xmath2 ) and roughly universal .
the effective radii of many sscs have been directly measured ( typically @xmath3 pc ) , and these radii in turn provide constraints on cluster dynamics and evolution .
these are only a few of the many results .
the advances in this field due to _ hubble _ have been truly remarkable .
however , at the time of this writing , there are no missions in the foreseeable future that will have capabilities comparable to _ hubble _ in the optical wavelength regime . given this prognosis ,
it is clear that further progress in ssc research will depend critically on utilizing other tools . with new observational capabilities becoming available in the infrared to radio wavelength regimes ,
it is likely that the next decade of ssc research will be dominated by studying their earliest stages of evolution , which these long wavelengths are particularly well - suited to probing .
for this reason , understanding the details of individual massive star formation in the milky way has become all the more important for extragalactic research .
there is , of course , a rich history of studying star formation in our own galaxy , as is highlighted in much of this volume . while there are clearly many important issues remaining in regard to massive star formation , a great deal has also been learned .
much of this knowledge can be applied to extragalactic objects , which can not be observed with as much detail and precision as their galactic counterparts .
observations of natal sscs suggest that the early stages of massive star cluster evolution may parallel those of individual massive stars observed in the milky way .
in particular , natal sscs appear to have spectral energy distributions similar to ultra compact hii regions ; these infant clusters typically have been identified at radio wavelengths as compact thermal ( often optically thick ) free - free sources , and they have copious amounts of thermal infrared emission . like individual massive stars , sscs also appear to spend @xmath4% of the massive star lifetimes in a stage analogus to ultra compact hii regions ( * ? ? ?
* ; * ? ? ?
* ( kobulnicky & johnson 1999 , vacca , johnson , & conti 2002 ) ) .
these similarities suggest that we can apply many of the tools , techniques , and ideas regarding massive star formation in our own galaxy to objects that are much more distant .
figure [ cartoon ] illustrates an analogy between the evolution of an individual massive star and a ssc . while the analogy between individual massive stars and sscs is a good starting point , the extent to which this analogy holds true will depend on a variety of factors .
for example , if star formation within a ssc is not `` instantaneous '' to within @xmath5 myr , then the constituent stars in a cluster will not all be in the ultracompact hii region stage simultaneously , which may result in different global seds . extrapolating from what we know about individual massive stars to sscs could also get us into trouble because of the extreme proto - stellar densities present in natal sscs .
it seems likely that we will gain the most physical insight not necessarily from the extreme cases alone , but also from trying to understand how the properties of star formation scale between individual massive star formation and ssc formation .
to date , the attempts to constrain the physical properties of natal clusters have been fairly crude , largely due to the limited observations available . as a result , we are forced by occam s razor to the most simple solution , one which requires the least free parameters .
our current concept is the following : when an ssc is born , the massive stars ionize the surrounding natal material , creating an extremely dense hii region , similar to a galactic ultracompact hii region , but scaled up in mass and luminosity .
this dense hii region is in turn surrounded by a dust cocoon that is presumably quite warm on the inner boundary .
the constituent ionizing stars may be surrounded by individual hii regions and dust cocoons in the earliest stages , but this is currently unknown .
clusters that are more sparse are likely to resemble ultracompact hii region complexes in the milky way ( such as w49a ) , and contain spatially discrete hii regions and cocoons , but the existence of individual cocoons becomes less tenable in the extreme stellar densities of the cores of massive sscs . these components of a natal cluster have separate observational signatures .
the dense hii regions are characterized by an `` inverted '' radio spectral energy distributions due to optically thick free - free emission ( @xmath6 , where @xmath7 ) .
the frequency at which the radio emission becomes optically thick is largely dependent on the density of the hii region . in the earliest stages of cluster evolution ,
most or all of the stellar luminosity is reprocessed by the surrounding dust cocoon , and this cocoon is observable at thermal infrared to sub - millimeter wavelengths .
the nature of the resulting spectral energy distribution depends on whether the dust in the cocoon is smooth or clumpy ( as will be discussed in [ models ] , but this is not currently constrained by observations .
existing observations have provided _ estimates _ for some of the physical properties of natal clusters , including size , electron density , pressure , stellar mass , hii mass , dust mass , and age . for the most nearby clusters , their size can be directly measured using high spatial resolution observations . for more distant clusters ,
model hii regions can be used to fit the radio spectral energy distribution and infer both the radius and electron density .
the embedded sscs that have been analyzed in this way ( * ? ? ? * ( e.g. those in he 2 - 10 , johnson & kobulnicky 2003 ) ) have hii region radii of only a few parsecs , and the electron densities have global values of @xmath8 cm@xmath1 , with peak values in excess of @xmath10 cm@xmath1 .
assuming temperatures of @xmath11 k , these densities imply global pressures of @xmath12 k cm@xmath1 and peak pressures reaching values of @xmath13 k cm@xmath1 .
the masses of the natal clusters , hii regions , and dust cocoons can be inferred from radio and infrared observations . the optically - thin thermal flux ( typically measured with the highest available radio frequency )
can be used to infer a total ionizing luminosity , and the ionizing luminosity can be translated into an embedded stellar mass by assuming a stellar initial mass function . using this technique , clusters with a range of masses
have been detected ( spanning modest ob - associations to massive sscs ) .
the hii mass can be estimated given the radii and densities for the ionized regions . in at least some cases ,
the hii mass has anomalously low values ( * ? ? ?
* ( @xmath14% of the stellar mass , johnson & kobulnicky 2003 ) ) , which we tentatively interpret as a sign of the youth of the hii regions .
the mass of the dust cocoons around embedded sscs has also been estimated in a few cases using infrared observations ; these estimates have suggested dust masses of @xmath15 for massive sscs in sbs0335 - 052 and he 2 - 10 ( * ? ? ?
* ; * ? ? ?
* ( plante & sauvage 2002 , vacca , johnson , & conti 2002 ) ) .
a critical and unresolved issue in the study of massive star clusters is whether sscs are just the statistical tail of a continuous mass distribution that extends all the way down to individual massive stars , or whether they can only be formed in special environments .
while there is some evidence that supports each side of this debate , the answer has been clouded by observational and evolutionary issues . in typical optical studies , a cluster s age , extinction , and mass are all contributing factors to its observed luminosity and inferred mass .
moreover , as clusters evolve , the lower mass clusters will tend to be disrupted on time - scales as short as a few million years ( * ? ? ? * ( portegies zwart et al .
2002 ) ) , causing further confusion when interpreting cluster luminosity and mass functions .
radio observations of natal clusters are immune to reddening and sample only the youngest clusters before they have had time to evolve . as a result , only radio observations are capable of directly sampling the cluster mass function at the time of their birth .
the starburst galaxy henize 2 - 10 is one of the nearest galaxies ( 9 mpc ) known to host multiple natal sscs detected in radio and infrared observations , as well as numerous sscs that have already emerged to be visible in optical light ( * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ( kobulnicky & johnson 1999 , johnson et al .
2000 , vacca , johnson , & conti 2002 , johnson & kobulnicky 2003 , cabanac , vanzi , & sauvage 2005 ) ) .
these properties make henize 2 - 10 a very important case study for examining the properties of natal sscs . in our previous radio observations , four natal sscs were detected , all with inferred masses in the range of @xmath16 .
although the observations were sensitive enough to detect sscs with somewhat lower masses , less massive natal clusters masses were not apparent . given the linear resolution of the observations used for these previous radio studies of @xmath17 pc , it seemed to be likely that some lower mass clusters were missing due to crowding and confusion . in order to address the question of whether henize 2 - 10 is only forming sscs with masses of @xmath18 , or whether lower mass clusters are present , but
had not yet been detected or disentangled from the more luminous sources , we obtained new high - resolution observations with the very large array using the additional pie town antenna .
a comparison between the old and new radio observations is shown in figure [ he210 ] ( * ? ? ? * ( biswas & johnson in prep ) ) . in the new observations ,
it is clear that many of the previously known radio sources are composed of multiple components .
although these observations are also roughly an order of magnitude more sensitive than the previous observations , it is not clear that any additional natal sscs exist outside of the main star forming region apparent in the previous radio observations .
the analysis is still underway at the time of this writing , but it is clear that these new high - resolution and extremely sensitive radio observations will shed light on many issues related to ssc formation .
with the _ spitzer space telescope _ providing a wealth of new observations , and _ herschel _ , _ sofia _ , and alma on the horizon , appropriate physical models of natal sscs are required in order to analyze data from these facilities . while the simple conceptual picture discussed in
[ properties ] is a good place to start , we know that the universe is complicated , and sometimes nice well - behaved uniform spheres just will not fit the data
. one of the main difficulties that must be overcome in order to properly model star forming environments is that the ism is not smooth , and therefore 1-d models can not reproduce physical structures that are consistent with the observed ism . we have produced a suite of models of natal sscs using the 3-d monte carlo radiation transfer code described by ( * ? ? ?
* whitney et al .
( 2003 ) ) .
these models allow for arbitrary density distributions , and we have invoked a clumpy fractal structure , consistent with observations of the ism .
the differences in the spectral energy distribution ( sed ) between smooth and clumpy models can be very significant , as shown in figure [ sed ] .
one of the main differences is that clumpy dust allows more near - infrared light ( both scattered and direct from the central source ) to escape .
more reprocessing of the photons also occurs in smooth dust cocoons , which changes the shape of the resulting sed .
spectral features , such as the silicate line at @xmath19 m can also vary dramatically depending on the clumpy fraction for model with otherwise identical parameters ( the silicate feature is much more strongly absorbed in models with a larger smooth fraction ) .
one important result is that the mid - infrared sed of the same object can vary significantly depending on the viewing angle if the dust cocoon is moderately clumpy .
this is very bad news from an observational perspective ; this degeneracy makes it impossible , regardless of the number of data points that sample a given sed , to determine a unique set of physical parameters that will fit the observations .
the good news is that the degeneracy introduced by clumping is less significant at far - infrared wavelengths , although these wavelengths also suffer from poor spatial resolution .
these results should serve as a strong warning that can not be overstated to investigators with thermal - infrared observations : _ using 1-d models could lead to a significant misinterpretation of the data .
_ there is a vast amount of parameter space to explore , even without allowing for 3-d structure .
the current study is limited to investigating the effects of a varying the inner and outer radius of the dust cocoon , total initial dust mass of the cocoon ( which directly reflects the star formation efficiency ) , and the fraction of the dust that is clumpy . within this limited parameter space
, we have simulated an evolutionary sequence for sscs as they go from deeply embedded to only having a thin dust shell .
the results from this evolutionary sequence can be directly compared to observed colors of ultra compact hii regions .
figure [ wc ] shows a comparison between the colors of natal sscs from our pseudo - evolutionary sequence produced by our 3-d models and the observed _ iras _ colors of ultra compact hii regions as well as `` field '' objects ( * ? ? ?
* ; * ? ? ?
* ( johnson et al . in prep , kurtz , churchwell , & wood 1994 ) ) .
overall , there is excellent agreement between our model colors and the observed colors of ultra compact hii regions . in general , as a natal ssc moves through our pseudo - evolutionary sequence ( its inner radius becomes larger and therefore the dust becomes cooler ) , its _ iras _ colors become more red .
the most deeply embedded objects lie in the lower left region of the spread of model points , which reflects the extremely hot dust in the inner portions of the dust cocoons .
if these objects with such hot dust exist , the lifetime of this state is likely to be extremely short .
therefore , embedded star forming regions with such hot dust should be quite rare , possibly accounting for the lack of ultra compact hii regions detected with colors as blue as the models can produce .
the model colors also extend further red - ward than observed for ultra compact hii regions , but this is not surprising ; as an embedded star forming region evolves and begins to emerge from its cocoon , its thermal - infrared luminosity will fall precipitously . as a result
, we expect that the existing samples of ultra compact hii regions become incomplete at these later stages ( or perhaps objects that are this evolved are no longer considered to be ultra compact hii regions ! ) .
there are no lack of questions regarding the birth of sscs , and given the new observatories that are becoming available , it is an exceptionally good time to be asking these questions .
a few major questions keep recurring that i want to highlight here : \(1 ) how does star formation vary between individual massive stars and sscs ( and is this reflected in the stellar initial mass function ) ?
\(2 ) what are the environmental requirements necessary to form a ssc , and how do they differ from individual stars ?
there are also a variety of properties that it would be nice to know more about .
for example , it would be instructive to _ directly _ measure densities , pressures , and temperatures using various line diagnostics at ir to radio wavelengths .
another quantity that might be particularly useful for theorists ( that is almost completely unconstrained ) is the star formation efficiency in these objects .
we also have no idea how much ionizing radiation could be escaping from the natal cocoons ( and it could be quite a lot if the dust is clumpy ) .
in fact , we know almost nothing about the natal dust cocoons ( including the temperatures , grain compositions , and geometry ) . this list is certainly not exhaustive .
an understanding of these quantities , and how they relate to the properties of the resulting cluster , will eventually provide a great deal of insight into the requirements for massive star cluster formation .
over the next decade ( or so ) , a number of powerful new facilities are going to become available at infrared to radio wavelengths .
these observational capabilities are likely to bring about a renaissance in the study of the formation of massive clusters .
these facilities will be incredibly well - suited to observing natal sscs ( figure [ obs ] ) , and will allow us to study natal sscs in amazing detail .
surely the new observations will bring about many new discoveries as well as many more new questions .
the work has benefited from many fruitful discussions with my collaborators , including i. biswas , p. conti , m. goss , l. hunt , r. indebetouw , h. kobulnicky , w. vacca , b. whitney , and k. wood .
i also gratefully acknowledge support for this work provided by the nsf through and astronomy and astrophysics postdoctoral fellowship , and by nasa through the hubble fellowship grant # 01173.02-a through the space telescope science institute , which is operated by the associated universities for research in astronomy , inc .
, under nasa contract nas5 - 26555 . | super star clusters are one of the most extreme star forming environments in the universe , and the most massive and dense of these may be proto globular clusters . like individual massive stars , the earliest stages of super star cluster evolution are deeply obscured , and therefore our knowledge about their birth environments is currently very incomplete .
however , the study of natal super star clusters has become somewhat of a cottage industry in recent years , and the sample of such objects has been growing rapidly with high - quality long - wavelength data now available from a number of observatories .
the natal super star clusters identified in thermal - infrared and radio observations represent the youngest stage of massive star cluster evolution yet observed .
their properties appear to be similar to those of ultracompact hii regions in the milky way , but scaled up in total mass and luminosity .
i will overview what we think we know about these objects based on existing observations , discuss their relationship to ultracompact hii regions , present new models of their spectral energy distributions based on 3-d simulations , and outline some of the most significant gaps in our current understanding . |
the standard model of an active galactic nucleus ( agn ) consists of a supermassive black hole being fed via an accretion disk .
the model also postulates the presence of a dusty molecular torus surrounding this inner region , and blocking it from view in agns where the line - of - sight intercepts the torus @xcite . for several years
, the only plausible signature of the accretion disk was the `` big blue bump , '' observed in the ultraviolet ( uv ) spectra of active galaxies , which was supposed to be the thermal emission from the disk @xcite .
a more telling feature , such as the kinematic signature of rotating gas , in the form of double - peaked line emission as observed in cataclysmic variables , was only discovered in optical spectra in the 1980s @xcite , and systematic searches were later published by @xcite and by @xcite .
broad , double - peaked emission lines from the nucleus of ngc1097 were reported by @xcite , and similar lines were found in other nearby galaxies with the _ hubble space telescope ( hst ) _ @xcite .
ngc1097 is particularly interesting because the double - peaked line originates in a low - luminosity agn and is a transient phenomenon , as it was not seen in previous observations .
these characteristics indicate that we would have a good chance of observing temporal variations in a reasonable ( not too long ) time interval .
indeed , monitoring of the double - peaked emission over the years 19922001 @xcite has put further constraints on the accretion - disk model and provided evidence against other possible origins for the double - peaked emission in this particular object , such as double jets and a binary black hole @xcite . in this _ letter _ , we report the presence of absorption features in small - aperture @xmath7 uv spectra of ngc1097 , and argue that they are characteristic of young o and b stars and of a low - ionization interstellar medium , all within a radius of at most 9pc from the nucleus .
such dimensions are typical of those of star clusters and also of those expected for obscuring tori in the standard model of agns .
besides the absorption features , the uv continuum also reveals the spectral energy distribution of a young starburst extinguished by a dusty medium .
uv and optical spectra of ngc1097 were obtained with the space telescope imaging spectrograph ( stis ) aboard @xmath7 on 2001 february 7 and 11 ut , using a @xmath8 slit and the g140l , g230l , g430l , and g750l gratings .
these gratings have resolving powers @xmath9 , which allow the separation of absorption features in the galaxy from those in the milky way .
the exposure times were 8302s for each of the two uv gratings , 1008s for grating g430l , and 936s for grating g750l . in order to obtain the best possible angular resolution achievable with the observations
, we extracted a nuclear spectrum within a window of @xmath10 , corresponding to a 9pc radius region centered on the nucleus .
we then combined the various spectral segments to construct one uv / optical spectrum spanning the spectral range 11009000 .
the combined spectrum was corrected for the foreground reddening @xmath11 mag from @xcite .
the spectrum , in the rest frame of the galaxy , is shown in figure [ fig1 ] , where we have edited out the geocoronal ly@xmath12 and [email protected] emission lines .
the main features readily observed in this spectrum are a `` hump '' in the uv continuum peaking near 2500 , and several double - peaked emission lines , identified in the figure : mgii@xmath132800 , h@xmath14 , hei@xmath135876 , and h@xmath12 .
broad components are also observed in ly@xmath12 , civ@xmath131550 , and ciii]@xmath131909 , although the double - peaked nature is not clear because of the lower signal - to - noise ratio .
the uv spectra of nuclei with broad double - peaked emission lines ( hereafter dubbed `` doublepeakers '' ) show a number of common absorption features in the spectral range 20003000 , mostly metastable feii absorption lines @xcite .
the absorption spectrum of ngc1097 in the above spectral range is similar to that of the prototypical doublepeaker , arp102b @xcite , which @xcite have proposed to originate in thin sheets or filaments embedded in an outflowing wind that overlays the accretion disk ( and thus is close to it , within roughly 5000 gravitational radii ) . @xcite
propose that many doublepeakers may have these filaments , and ngc1097 probably also fits within this scenario . here
we will focus on the range 11001600 of the nuclear spectrum of ngc1097 . in this spectral range
, ngc1097 shows a number of absorption lines , some of which are superimposed on the broad ly@xmath12 and civ emission lines , as illustrated in the top panel of figure [ fig2 ] .
the absorption features are identified at the bottom of the figure .
we include in this figure , for comparison , the stis spectrum of the seyfert galaxy akn564 ( representative of seyfert galaxy spectra ) and of the starburst nuclei of ngc7552 and ngc5253 , all in the rest frame of the galaxies . upon inspection of the uv spectra of many other seyferts and starbursts ,
we conclude that the absorption features in the uv spectrum of ngc1097 are not common in agns , but are frequently observed in the spectra of starburst galaxies .
we also include in figure [ fig2 ] a synthetic spectrum of a starburst obtained using the code starburst99 @xcite , with the following parameters : instantaneous burst of mass 10@xmath15 , age 10@xmath2yr , salpeter initial mass function ( imf ) , lower and upper limits for the imf of @xmath16 and @xmath17 ( respectively ) , and solar metallicity .
the spectra of figure [ fig2 ] have been normalized at 1350 to allow a better comparison among them , and have been shifted vertically for clarity . in order to further emphasize the similarity of the nuclear uv spectrum of ngc1097 with those of starbursts
, we show in the lower panels of figure [ fig2 ] the ngc1097 spectra on top of those of ngc7552 and ngc5253 .
we decided to keep in the spectra the geocoronal ly@xmath12 and o i @xmath131302 emission lines , in order to avoid introducing artificial discrepancies .
the following stellar absorption features , characteristic of the spectra of early - type stars ( o , b ) , are observed in the spectra of ngc1097 , ngc7552 , and ngc5253 @xcite : [email protected] , [email protected],1242.8 , [email protected],1402.8 , and [email protected],1550.8 .
the last three are resonance lines and can also originate in the interstellar medium . other interstellar lines present in the spectra are [email protected] , 1193.3 , [email protected] , [email protected],1260.4 , [email protected] , and [email protected] ( the last two partially filled by the geocoronal o i emission line ) , [email protected],1335.7 , and [email protected] .
there are also in the nuclear spectrum of ngc1097 a few absorption lines from our galaxy , but which can be separated from the absorption lines of ngc1097 because of the redshift of the latter . among the stellar absorption features ,
only ciii@xmath131175 is not a resonance line and must originate in the atmosphere of young stars .
this line is sharp and deep in the starburst synthetic spectrum , but is broader and shallower in the spectra of the actual starbursts ngc5253 and ngc7552 . in the ngc1097 spectrum
, it has a similar profile to those observed in the starburst galaxies , as can be seen in fig .
the presence of young stars is also supported by the p - cygni profile of the line nv@xmath131240 , which can be clearly observed in ngc1097 , ngc5253 and in the synthetic starburst spectra , although only a hint of it appears in the ngc7552 spectrum .
notice , in particular , the similarity between the emission portion of this line in ngc1097 and ngc5253 ( fig [ fig2 ] ) . a p - cygni profile is also observed in the civ@xmath131550 line of the ngc5253 and synthetic starburst spectra , but not in ngc7552 . in ngc1097
, there is only a hint of a p - cygni profile in the civ line , superimposed on the broad emission - line profile .
the large depth of the absorption , which reaches practically zero flux , indicates that this line probably originates not exclusively in the stellar atmospheres of young stars . a large depth
is also observed in the mgii@xmath132800 absorption .
these two deep absorption lines suggest that some of the absorbing gas is along our line of sight to the agn .
we thus conclude that at least part of the absorption must come from a region very close to the agn
. the absorber could be the starburst superwind whose filaments cross our line of sight to the nucleus [ e.g. @xcite ] , or filaments embedded in an accretion - disk wind [ as is the case favored for the other double peakers @xcite ] , or both .
is the spectral energy distribution ( sed ) compatible with the presence of a young starburst in the nucleus of ngc1097 ? in order to answer this question , we have used the code starburst99 @xcite to construct model seds of continuous and single bursts of star formation of different ages .
inspection of the nuclear continuum of ngc1097 in figure [ fig1 ] shows that the flux increases from the optical to the uv , as expected for a young starburst , but then , at @xmath192500 , decreases to shorter wavelengths .
the only way to have this flux decrease in the uv is through reddening .
we have tried different combinations of starburst models and reddening laws . for the starbursts , we have also experimented with different slopes for the initial mass function ( imf ) , and our conclusions are as follows .
continuous bursts provide too much flux in the optical , as do single bursts older than a few times 10@xmath2 yr . when combined with any of the known reddening laws , they fail to produce the observed shape of the uv sed .
the most suitable starburst model is the one of a single burst of age 10@xmath2yr .
the seds of younger bursts or bursts with different imfs are only slightly different from that of a single burst of age 10@xmath2 yr , and do not provide significant improvement to the fit of the observed spectrum .
we have also concluded that models including the contribution of the gas produce a strong balmer jump in emission which is not observed in the data .
the best fit to the ngc1097 continuum we could obtain from the available synthetic spectra includes only the stellar contribution to the continuum . in order to produce the observed flux decrease in the uv
, we have tried the calzetti reddening law @xcite , the milky way ( mw ) reddening law @xcite , the large magellanic cloud ( lmc ) reddening law @xcite , and the small magellanic cloud ( smc ) reddening law @xcite .
the calzetti reddening law did not have enough uv extinction to reproduce the data , both the mw and lmc laws introduced the characteristic 2200 absorption feature which is not present in the data , while the smc reddening law produced better agreement with the data for @xmath20 mag and @xmath21 .
although it is surprising that the best reddening law is that of the smc ( as ngc1097 is a much more luminous and a more metal - rich galaxy than the smc ) , we point out that extinction by an `` smc - like '' reddening law has also been found to best explain the seds of red quasars in the sdss quasar survey @xcite .
not only the sed but also the flux of the reddened 10@xmath2yr starburst is in approximate agreement with the observations , indicating that the mass of the starburst is roughly the one adopted in the model , 10@xmath22 .
this is illustrated in figure [ fig3 ] , where we show the ngc1097 spectrum compared to the reddened starburst sed .
an inset in the figure shows the sed of the unreddened starburst in comparison with the galaxy spectrum in order to illustrate the effect of reddening .
figure [ fig3 ] also shows the contribution of the continuum emitted by the agn , borrowed from a paper in preparation @xcite , but with preliminary results already published @xcite . in @xcite
the nuclear sed from x - ray to radio wavelengths is found to be well reproduced by a model consisting of the emission from a radiatively - inefficient accretion flow ( riaf ) plus that of a thin accretion disk .
although this model produces a good overall description of the sed , it does not reproduce the excess in the uv , as illustrated in figure [ fig3 ] .
we point out that in @xcite , the thin - disk emission is self - consistently derived from the incidence of the riaf radiation on the disk .
the thin - disk continuum emission peaks in the optical / infrared , and can not reproduce the excess uv emission .
although the introduction of a reddened starburst continuum can approximately reproduce the uv and its decrease to shorter wavelengths , there is still a small remaining `` bump '' around 2500 which is not reproduced by the riaf + thin disk + reddened starburst ( fig .
[ fig3 ] ) . in this wavelength region
there is a blend of fe ii emission lines which is frequently present in the uv spectra of seyfert 1 galaxies and quasars @xcite . in order to investigate the possibility that the above bump is due to the blend of fe lines , we have used an archival spectrum of i zw 1 ( kindly provided to us by karen leighly ) as a template for the fe emission .
although this template covers only the 20003000 spectral region , it includes the most intense feii emission lines . [ the contribution of feii lines in the optical is much smaller with peak intensities @xmath1910% of that in the uv @xcite .
detailed modelling of the optical feii emission is beyond the scope of this work . ] under the assumption that the feii emission is coming from the accretion disk , we have convolved the template with the double - peaked profile of the emission - lines . combining this component with the others , we could successfully reproduce the remaining bump around 2500 .
this is also illustrated in fig .
[ fig3 ] , where we further include the feii template and a model ( thick line ) for the ngc1097 continuum comprising all the components : the reddened starburst , the riaf + thin disk model and the feii template , properly scaled to fit the data .
the reddened starburst model was scaled by 0.85 , corresponding to a uv + optical flux of 1.7@xmath110@xmath23ergs@xmath24s@xmath25 and luminosity of 1.5@xmath110@xmath5l@xmath4 for an adopted distance to ngc1097 of 17mpc . finally , we point out that it is not our goal to produce a detailed fit to the spectrum , as there are many uncertainties .
for example , the reddening law may be somewhat different from that of the smc , and the starburst properties may also differ from those of the starburst99 model .
we have shown evidence for an obscured starburst located within 9 pc of the active nucleus of the galaxy ngc1097 .
the importance of this finding is two - fold : ( 1 ) the proximity of a young starburst to the nucleus and thus possible association with it ; and ( 2 ) the fact that the starburst is obscured , but the agn is not .
the maximum distance between the starburst and the nucleus of ngc1097 ( 9 pc ) is comparable to the sizes of star clusters with masses @xmath19 10@xmath15 @xcite ; hence , the cluster is quite close to the nuclear black hole @xcite
. the disruption of a cluster star passing closer to the black hole than its tidal radius could be the origin of the transient accretion disk , as we have proposed in @xcite .
flares in the uv @xcite and x - rays @xcite suggest that we are beginning to witness such events . alternatively ,
mass loss from evolving stars in the cluster could be the origin of the accretion disk . in any case
, we may be seeing a rare example of the `` starburst - agn connection '' the starburst is responsible for feeding the agn ! another interesting characteristic of this starburst is that it is immersed in an obscuring medium .
its maximum size is also on the order of that predicted for the torus of the `` unified model '' of agns @xcite .
although we can not prove that the geometry is that of a circumnuclear torus , there is no evidence that the nuclear source is obscured , as we can observe the riaf + thin - disk continua and the double - peaked emission - lines with no signs of obscuration ; indeed , the mgii line in the uv is the strongest double - peaked line of the spectrum .
thus , the starburst _ may be _ within a dusty toroidal structure which leaves the nuclear source unobscured .
if this is the case , we are observing , for the first time , a starburst inside a torus , as has been proposed in the literature as a solution for the stability problem of the torus : the toroidal structure is supported by the pressure produced by the evolving starburst inside @xcite . finally , we point out that the starburst reddening , @xmath20 mag , is not very high .
an additional reddening of 23 mag would render the starburst undetectable .
since higher reddening values have been found in the nuclear regions of many agns , our results show that starbursts may be present , but most will not be detected , at least in the uv .
we acknowledge valuable discussions with e. bica , c. bonatto , g. kriss , c. leitherer , and p. hall .
we thank k. leighly for providing the feii spectrum and the referee for valuable suggestions .
this work was supported by nasa grant go-08684 from the space telescope science institute , which is operated by aura , inc .
, under nasa contract nas 5 - 26555 , and the brazilian institutions cnpq , capes , and fapergs .
filippenko is grateful for a miller research professorship at u.c .
berkeley , and eracleous acknowledges support from nasa ltsa grant nag5 - 10817 .
antonucci , r. 1993 , , 31 , 473 arp , h. c. 1976 , , 207 , l147 baldwin , j. a. , et al .
2004 , , 615 , 610 barth , a. j. , et al .
2001 , , 546 , 205 bouchet , p. , et al . 1985 ,
a&a 149 , 330 bower , g. a. , wilson , a. s. , heckman , t. m. , & richstone , d. o. 1996 , , 111 , 1901 calzetti , d. , kinney , a. l. , & storchi - bergmann , t. 1994 , , 429 , 582 calzetti , d. , et al .
1997 , , 114 , 1834 cardelli , j. a. , clayton , g. c. , & mathis , j. s. 1989 , , 345 , 245 chen , k. , halpern , j. , & filippenko , a. v. 1989 , , 339 , 742 cid fernandes , r. , & terlevich , r. 1995 , , 272 , 423 eracleous , m. 2002 , in mass outflows in agns , eds .
m. crenshaw et al .
( san francisco : asp ) , 131 eracleous , m. , & halpern , j. 1994 , , 90 , 1 eracleous , m. , & halpern , j. 2003 , , 599 , 886 eracleous , m. , halpern , j. , & charlton , j. 2003 , , 582 , 633 eracleous , m. , et al . 1997 , , 490 , 216 eracleous , m. , et al .
2004 , in the interplay among black holes , stars , and ism in galactic nuclei , eds .
t. storchi - bergmann , et al .
( cambridge : cambridge univ . press ) , 29 halpern , j. p. 1990
, , 365 , l51 halpern , j. p. , & filippenko , a. v. 1988 , nature , 331 , 46 halpern , j. p. , eracleous , m. , filippenko , a. v. , & chen , k. 1996 , , 464 , 704 halpern , j. , gezari , s. , & komossa , s. 2004 , , 604 , 572 heckman , t. , et al .
2000 , , 129 , 493 ho , l. c. , et al .
2000 , , 541 , 120 kinney , a. , bohlin , r. c. , calzetti , d. , panagia , n. , & wise , r. f. g. 1993 , , 86 , 5 koornneef , j. , & code , a. d. 1981 , , 247 , 860 koratkar , a. , & blaes , o. 1999 , , 111 , 1 leitherer , c. , et al . 1999 , , 123 , 3 livio , m. , & pringle , j. 1996 , , 287 , 35 lorre , j. j. 1978 , , 222 , l99 malkan , m. a. , & sargent , w. l. w. 1982 , , 254 , 22 nemmen , r. s. , 2004 , in the interplay among black holes , stars , and ism in galactic nuclei , eds .
t. storchi - bergmann , et al .
( cambridge : cambridge univ . press ) , 53 nemmen , r. s. , et al .
2005 , in preparation pei , y. c. 1992 , , 395 , 130 perry , j. j. , & williams , r. , 1993 , , 260 , 437 renzini , a. , et al .
1995 , nature , 378 , 39 richards , g. t. , et al .
2003 , , 126 , 1131 schlegel , d. , finkbeiner , d. p. , & davis , m. 1998 , , 500 , 525 shields , g. a. 1978 , nature , 272 , 706 shields , j. c. , et al . 2000 , , 534 , l27 sigut , t. a. a. , & pradham , a. k. 2003 , , 145 , 15 storchi - bergmann , t. , baldwin , j. a. , & wilson , a. s. 1993 , , 410 , l11 storchi - bergmann , t. , et al . 1995 , , 443 , 617 storchi - bergmann , t. , et al .
1997 , , 489 , 87 storchi - bergmann , t. , et al .
2003 , , 598 , 956 strateva , i. v. , et al .
2003 , , 126 , 1720 vazquez , g. a. , et al .
2004 , , 600 , 162 vestergaard , m. , & wilkes , b. j. 2001 , , 134 , 1 wada , k. , & norman , c. a. 2002 , , 566 , l21 wills , b. j. , netzer , h. , & wills , d. 1985 , , 288 , 94 | we report evidence for a recent burst of star formation located within 9pc of the active nucleus of ngc1097 .
the observational signatures of the starburst include uv absorption lines and continuum emission from young stars observed in a small - aperture _ hst _ spectrum .
the burst is @xmath0 a few @xmath1 10@xmath2yr old , has a mass of @xmath3m@xmath4 , an observed luminosity of 1.5 @xmath1 10@xmath5 l@xmath4 and is obscured by @xmath6 mag .
the importance of this finding is two - fold : ( 1 ) the proximity of the starburst to the active nucleus and thus possible association with it ; ( 2 ) its obscuration by and apparent association with a dusty absorbing medium , while the broad emission lines appear unobscured , suggesting that the starburst could be embedded in a circumnuclear torus as predicted in the unified model of active galactic nuclei . |
photoionization of two - electron atoms has been studied theoretically by different authors starting from the pioneering paper by @xcite .
review of early literature on this subject can be found in ref .
subsequently , a large number of computations of helium photoionization cross - sections was reported @xcite .
these calculations produced a collection of results varying typically by 5% from each other . on the experimental side , the benchmark set of data was reported by @xcite who measured the total photoionization cross - section of he in the photon energy range from the threshold to 120 ev .
agreement between the theoretical and experimental data was within the same margin of 5% . in the following years , the theoretical interest shifted towards calculation of differential characteristics of the photoionization process and to studies of double photoionization . here , several approaches have been advocated including the many - body perturbation theory @xcite , convergent close - coupling method @xcite , time - dependent close - coupling method @xcite , @xmath0-matrix approach @xcite , and methods based on the computation of the dipole response function @xcite or @xmath1-spline implementations of the exterior complex scaling @xcite . due to this shift of focus
, there have been no further attempt to produce a consistent set of photoionization cross - sections of he with an accuracy better than several percent .
@xcite combined measurements of @xcite at low energies and theoretical calculations at high energies to construct a set of photoionization cross sections of he that should be reliable at all energies . however , there was no consistency check applied to the experimental data
. in the meantime , accurate helium photoionization cross - sections would be highly desirable due to importance of he in astrophysics and its use as a standard gas in determination of the photoionizaiton cross - sections of other atomic and molecular species . in the present paper ,
we develop the complex rotation method ( crm ) for highly accurate calculations of the total photoionization cross - section of two - electron atomic targets .
one way of calculating the photoionization cross - section is to combine the crm technique with the perturbation theory with respect to interaction of the atom with the electromagnetic field .
in such a perturbation theory , the crm provides the basis of the field - free atomic states .
it was demonstrated by @xcite that relying on the spectrum of the crm eigenvalues , one can construct a representation of the complete green s function of the atom .
this , in turn , allows to write down a convenient representation for the projection operator corresponding to the continuous spectrum of the atom @xcite . using this projection operator ,
one can compute probabilities of transitions into continuum under the action of some perturbation , in particular , the interaction of the atom with the electromagnetic field . calculations of total photoionization cross - sections of the helium atom based on this technique have been reported in refs .
similar ideas were also used to determine static and dynamic polarizabilities of helium @xcite . in the present work ,
we use the crm procedure in a somewhat different , non - perturbative way by applying it to the whole system the atom plus the electromagnetic field .
thus , we are capable of going beyond the perturbation theory and considering very strong fields . in this respect ,
the present technique has certain features in common with refs .
@xcite , where an approach based on the configuration interaction procedure was developed for atoms with more than one electron .
we cast our formalism using the language of square - integrable functions with a finite norm .
this approach becomes feasible in the so - called floquet - fourie representation of the time - dependent schrdinger equation .
another key ingredient of the present work is the hylleraas basis functions which have long been used in various variational - type calculations .
an excellent review of applications of the hylleraas basis to calculations of energies of two - electron atoms is given by @xcite .
a well - known trademark of the hylleraas basis set is a very high accuracy of the atomic energies . in the present paper ,
we show that the same high accuracy which is achieved for field - free atomic states can also be attained when the atom is subjected to a monochromatic electromagnetic field .
in particular , the total photoionization cross - sections can be calculated with an accuracy of the order of a fraction of a percent .
thus generated cross - sections were compared with the experimental results of @xcite .
we discovered a systematic deviation from the experiment , especially in the region close to double ionization threshold at the photon energies of @xmath280 ev .
this deviation was confirmed by comparison with earlier results produced by the convergent close - coupling ( ccc ) method @xcite .
the true potential of the present method is realized in the strong field regime where the perturbation theory fails .
as demonstrated below , the floquet - fourie - hylleraas ansatz produces very accurate results in this regime as well .
the non - relativistic hamiltonian of the helium atom in the presence of the external monochromatic linearly - polarized electromagnetic field can be written as : @xmath3 where , @xmath4 is a kinetic term : @xmath5 @xmath6 potential energy term : @xmath7 and @xmath8 describes interaction of atom and the field . in the length gauge ( which will be used in the present paper ) , this operator assumes the form : @xmath9 with @xmath10 . unless stated otherwise , the atomic units are used throughout the paper .
we write the solution of the time - dependent schrodinger equation ( tdse ) using the floquet - fourie ansatz @xcite @xmath11 by substituting this expression into the tdse and equating coefficients with @xmath12 , we obtain a chain of coupled equations for the floquet - fourie coefficients @xmath13 : @xmath14 we solve this set of equations with the help of the complex rotation procedure @xcite .
formally , the cmr can be described as a complex transformation of radial variables @xmath15 , where @xmath16 is the rotation angle , the sole parameter defining the transformation . under this transformation ,
the chain of equations ( [ fl1 ] ) is converted into @xmath17 according to the general theory of crm @xcite , the set of equations ( [ fl2 ] ) can be solved by means of variational techniques if the rotation angle @xmath16 is properly chosen .
we introduce a basis set of square integrable functions @xmath18 where the index @xmath19 refers to the number of the floquet block and the index @xmath20 denotes a particular @xmath21 function in the subspace of the @xmath19-th block so that @xmath22 . with these notations , the set of eqs.([fl2 ] ) can be rewritten in a matrix form as : @xmath23 notations can be further simplified by introducing obvious shorthands : @xmath24 where it is understood that summation is carried over the repeated @xmath20-index .
here @xmath25 , and @xmath0 , @xmath26 and @xmath27 stand for the overlap , kinetic energy and potential energy matrices , respectively .
one could say here a few words about the choice of the basis allowed by the structure of the system ( [ fl5 ] ) .
suppose first , that in each of the subspaces corresponding to different floquet blocks we chose some compete set of functions , such that for any @xmath13 in eq.([fl2 ] ) we had : @xmath28 .
let the set of @xmath29 s be the same for all floquet subspaces . then , if we have retained @xmath30 floquet blocks in the system ( [ fl2 ] ) and keep @xmath31 terms in the expansion for each @xmath13 in eq.([fl2 ] ) we have altogether @xmath32 unknowns @xmath33 in the system ( [ fl2 ] ) . to get a correctly posed eigenvalue problem , we should have the same number of equations .
this number is provided by projecting each of the equations ( [ fl2 ] ) on one of the @xmath29 s with @xmath34 .
this way of reducing the set of equations ( [ fl2 ] ) to the form of matrix eigenvalue problem is correct , but too general for our purposes .
it can be seen , that one can considerably diminish the resulting dimension of the matrix eigenvalue problem by using certain symmetry properties of the system eq.([fl2 ] ) .
it is easy to see , that this system allows the following class of solutions : @xmath13 s with even @xmath19 are of even parity , while @xmath13 s with odd @xmath19 are of odd parity .
parity here is understood with respect to the spatial inversion .
of course , there is a class of solutions with the opposite property : @xmath13 s with even @xmath19 are of odd parity , while @xmath13 s with odd @xmath19 are of even parity .
the solution we are looking for ( which is to describe behavior of the even @xmath35 state of helium ) evidently belongs to the first class .
we can therefore , choose the basis set as follows . instead of choosing the same set @xmath29 for each floquet block ,
we choose two sets : a set @xmath36 , consisting of basis finctions of even parity , is used as a basis to represent @xmath13 s with even @xmath19 s .
another set @xmath37 , composed of odd parity functions is used as a basis to represent @xmath13 s with odd @xmath19 s .
suppose that in the expansions of @xmath13 s with even @xmath19 s we retain @xmath38 terms , and in the expansions of @xmath13 s with odd @xmath19 s - @xmath39 terms .
let the number of floquet blocks with even and odd @xmath19 s be respectively @xmath40 and @xmath41 . than we have @xmath42 unknown coefficients @xmath33 .
we obtain the same number of equations by projecting equations ( [ fl2 ] ) on @xmath43 for even @xmath19 and on @xmath44 for odd @xmath19 .
projection of equations with even @xmath19 on the @xmath37 and of equations with odd @xmath19 on the @xmath36 gives identically zero and does not add new equations .
more details about the basis functions @xmath36 and @xmath37 is given below .
according to the general theory of crm , some of the energy values ( generally complex ) for which system ( [ fl5 ] ) has a solution are related to the position and width of the resonance state via @xmath45 , where @xmath46 is position of the resonance and @xmath47 its width .
this leads one to solving a generalized eigenvalue problem .
effectiveness of finding eigenvalues of such a problem depends crucially on the choice of the basis used to represent the matrices in eq.([fl5 ] ) .
the basis set used in the present paper was constructed from the hylleraas type functions : @xmath48 where @xmath49,@xmath50 are some constants ( to be specified below ) , @xmath51,@xmath52,@xmath30 are integers and the angular part @xmath53 represents two angular momenta @xmath54 coupled to a state with a total angular momentum @xmath55 .
the basis functions ( [ hil ] ) must also be properly symmetrized with respect to exchange of the electron coordinates . when choosing parameters in eq .
( [ hil ] ) , we followed the following empirical rules @xcite .
all the basis functions with the parameters satisfying : @xmath56 were included in the calculation .
the parameter @xmath57 determines the overall size of the basis .
there is also a semiempirical rule for choosing angular momenta @xmath58,@xmath59 in the eq.([hil ] ) .
thus , for states of the natural parity @xmath58,@xmath59 are best chosen so that @xmath60 .
both these criteria help to avoid the numerical problems due to near - degeneracy of the basis set when its dimension becomes large .
in the present work , our main goal is to obtain accurate photoionization cross - sections from the ground state of neutral helium for not very large electromagnetic field intensities .
accordingly , our main interest is focused on the states of @xmath61 and @xmath62 symmetries .
therefore , our first goal is to choose such a basis that solution of the eigenvalue problem ( [ fl5 ] ) yields accurate energies for the ground @xmath63 and first excited @xmath64 state of the helium atom in the absence of the field .
this goal was achieved as follows .
we chose parameters @xmath65 , @xmath66 for the @xmath61-states and @xmath67 , @xmath68 for the @xmath62-states .
the reason for enlarging the basis set for the excited @xmath62-states is that the electrons in such states are generally on different distances from the nucleus .
this choice combined with restriction on angular momenta ( [ max ] ) resulted in @xmath69 basis functions for the @xmath61-states and @xmath70 basis functions for the @xmath62-states .
the next step was to solve the generalized eigenvalue problem for the field - free case . in eq .
( [ fl5 ] ) we put @xmath71 , @xmath72 , and limited ourselves to the blocks with @xmath73 , @xmath74 , the @xmath73 block being composed of the states of @xmath35 symmetry , and @xmath74 blocks composed of the states of @xmath64 symmetry . all the numerical results reported below
were obtained using the quadruple precision arithmetics .
we note , that in the presence of the weak electromagnetic field account of the blocks with @xmath74 corresponds to absorption and emission of one photon .
we shall use this fact below to extract the photoionization cross - section from our calculation . for the moment
, we are concerned with testing the accuracy of our basis .
diagonalization of the eigenvalue problem ( [ fl5 ] ) with @xmath71 , @xmath72 in the basis described above produced the following results for the complex energies : @xmath75 ( the ground state ) and @xmath76 @xmath77 state ) .
a small imaginary part which , in the absence of the field , should be zero could be taken as an indication of an accuracy of our basis set .
either this criteria or a direct comparison with the well - known results of highly accurate calculations @xcite shows that we have achieved an accuracy of the order of @xmath78 a.u .
this accuracy , as will be demonstrated below , is sufficient to obtain the photoionization cross - sections with at least three significant figures . to calculate the total photoionization cross sections we adopted the following strategy .
the eigenvalue problem ( [ fl5 ] ) was solved with the floquet blocks @xmath79 retained , the composition of each block was the same as described above for the field - free case .
diagonalization of the eigenvalue problem ( [ fl5 ] ) produced energy shift and total width for the ground state . by definition , the photoionization cross - section from this state is related to the total width @xmath47 via @xmath80 where @xmath81 is field strength , @xmath82 its frequency , @xmath83 is the fine structure constant .
we need therefore to extract from our calculation the coefficient with @xmath84 in the asymptotic law defining the weak - field behavior of the width : @xmath85 to implement this strategy , we need an extrapolation procedure since the calculation based on the system ( [ fl5 ] ) is performed for a non - zero field strength .
although finite , this field strength should not be too small to compute @xmath47 with sufficient accuracy .
the issue of accuracy can be addressed as usual in variational - type calculations , by merely increasing the basis size and verifying that the results do not change appreciably .
such a test was performed for a photon energy @xmath86 ev and a field strength @xmath87 a.u . by varying the parameter @xmath57 in eq.([max ] ) for the @xmath61 and @xmath62 states .
the diagonalization of the problem ( [ fl5 ] ) was performed with the floquet blocks @xmath79 retained .
all the remaining details of the basis ( nonlinear parameters etc . ) were the same as in the field - free case reported above .
the calculation was performed for the value of the rotation angle @xmath88 .
.[tabm1 ] results for the ground state eigenvalue of problem ( [ fl5 ] ) as functions of parameters @xmath57 in eq.([max ] ) , @xmath86 ev , @xmath87 a.u . [ cols="^,^,^,^,^ " , ] comparison of the results given in table [ taben ] supports the assertion we made in the previous section as to the accuracy of our results for the widths .
as one can see , for the field strengths @xmath89 a.u .
, inclusion of the additional floquet blocks and basis states of symmetries other than @xmath61 and @xmath62 produces relative variations in the widths on the order of @xmath90 percent .
this means that for such field values we are still within the domain of the validity of the perturbation expansion . for the frequencies presented in the table
the domain of the perturbation theory actually extends quite far in the region of large field strengths .
as one can see from the table ( [ taben ] ) , the ratio @xmath91 starts changing in a more or less appreciable manner only for field strengths as large as @xmath92 a.u .
we performed a calculation of the total photoionization cross - sections from the ground state of helium .
we employed a theoretical procedure based on the floquet - fourie representation of the solutions of the tdse describing the helium atom in the presence of the linearly polarized monochromatic electromagnetic field .
the resulting set of floquet equations was cast into a generalized eigenvalue problem by the complex rotation method .
our approach is essentially non - perturbative .
this is in contrast with other works @xcite where the crm was used to produce an accurate description of the field - free helium atom , thus giving the ground for application of the perturbation theory . in our approach
, we do not rely on any perturbation expansion to describe interaction of the atom and the electromagnetic field .
this interaction is included into the theory from the beginning .
we would like to emphasize the accuracy of the present results for the photoionization cross - sections which , we believe , is on the level of a fraction of a percent .
although only few selected photon energies were reported in the paper , far wider and denser energy grid was covered by the present calculation .
these results might serve as an accurate database and find their use in various astrophysics and atomic physics applications .
the authors shall gladly communicate these data on request .
the authors acknowledge support of the australian research council in the form of discovery grant dp0451211 . | we outline a non - perturbative procedure for calculating the total photoionization cross - section of two - electron atomic systems .
the procedure is based on the floquet - fourie representation of the solution of the time - dependent schrodinger equation .
the floquet - fourie ansatz produces a set of equations which is recast into a generalized eigenvalue problem by means of the complex rotation procedure . with the use of the hylleraas - type basis functions ,
the total photoionization cross - sections are obtained within the accuracy of a fraction of a percent .
the total photoionization cross - sections for neutral helium are in good agreement with the convergent close - coupling calculations of kheifets and bray [ phys .
rev .
a * 58 * , 4501 ( 1999 ) ] but deviate notably from the experimental data of samson _ et al . _ [ j. phys .
b * 27 * 887 ( 1994 ) ] .
= 8 mm |
based on density - functional ( df ) theory , we have derived in previous work a set of integral equations , which allows to calculate the ion - ion and the electron - ion correlations in a liquid metal or a plasma , consistent with the bound - electron structure of the ion using only the atomic number @xmath5 as input ; these integral equations are named the quantal hypernetted - chain ( qhnc ) equations,@xcite which are derived from the exact expressions for the ion - ion and electron - ion radial distribution functions ( rdf ) in an electron - ion mixture . up to now , we have applied this approach to liquid metallic hydrogen,@xcite lithium,@xcite sodium,@xcite potassium@xcite and aluminum@xcite obtaining ion - ion structure factors in excellent agreement with experiments . since a liquid metal can be considered as a very special type of a high - density plasma , we can expect from the successful application of the qhnc - method to liquid metals , that this approach is able to provide accurate results for a plasma .
in such a system , both the ionic valency @xmath6 and the electron - ion interaction @xmath7 may vary over a wide range as temperature and density are changed .
our method is in particular suited to treat a plasma , since it is able to calculate the ionization @xmath6 and the electron - ion interaction in a self - consistent manner using the atomic number of the system as the only input data .
recently , we have extended this formalism and have performed a first - principles molecular dynamics ( md ) simulation based on the qhnc theory for alkali metals near the triple point : in this study those small deviations which were still observed between experimental results and qhnc data for the structure factor disappeared completely.@xcite nevertheless , the calculation of the ion - ion rdf in an md simulation is rather time - consuming .
recently , rosenfeld@xcite has proposed a new bridge - functional for hard - spheres ; its construction requires only fundamental measures of the hard - spheres . in combination with the mhnc - approach by replacing the bridge-_function _ by a bridge-_functional _ of the reference system
it was found out that this method is able to give very accurate results for the structure and thermodynamics of a large variety of one - component and binary liquid systems.@xcite therefore , with replacement of the md simulations by the mhnc method we can obtain accurate results for the ion - ion and electron - ion rdf s for dense fluids via the qhnc method using the rosenfeld bridge - functional . in a plasma , density and temperature vary over a wide range .
therefore , in order to check the applicability of the approach presented here to a strongly coupled plasma ( where no reliable experimental data for particle - correlations are available ) , it is important to examine to what extent the qhnc equations can describe a liquid metal in a wide range of densities and temperatures .
recently , tsuji _ et al . _
@xcite measured the structure factors of liquid rb at high pressures : 0.2 , 2.5 , 3.9 and 6.1 gpa .
these pressure - values bring along a compression of liquid rb : the corresponding density values are estimated to be 1.07 , 1.41 , 1.56 and 1.95 times the normal density , respectively.@xcite tsuji _
_ observed that liquid rb is uniformly compressed with increasing pressure . in this context
it should be mentioned that shimojo _
@xcite performed first - principles molecular - dynamics simulation for liquid rb under pressure ( 0 , 2.5 , and 6.1 gpa ) and compared their results with the rdf s extracted from the structure factors measured by tsuji _
_ : they found that liquid rb is compressed uniformly at 2.5 gpa , but that some deviation from uniform compression is observed at 6.1 gpa . in the present work , using the qhnc method in combination with the rosenfeld bridge - functional for the reference hard - sphere system we have calculated the structure factors of compressed rubidium , the pressure ranging from 0 to 6.1 gpa and compared these data with the experimental results .
we find excellent agreement with these experimental data , although in two cases the density has to be slightly readjusted .
we confirm with our results the uniform compression model .
we find out that the effective ion - ion potentials are insensitive to this pressure - variation , a feature which is obviously typical for liquid alkali metals ( we also confirm these observations for the case of liquid - lithium ) .
we finally observe that the structure factors coincide in one single curve if @xmath1 is scaled in units of the wigner - seitz radius @xmath0 .
the paper is organized as follows : in the subsequent section we briefly outline the qhnc method and give a few details about the rosenfeld bridge - functional . in section iii
we discuss our results and compare them with experimental data .
the paper is concluded by a summary .
in the present section , we give a brief outline of the qhnc theory and the integration of the rosenfeld bridge - functional in an integral equation approach for a one - component fluid .
let us consider a liquid metal or a plasma as a mixture of electrons and ions interacting through pair potentials @xmath8 [ @xmath9 . in this mixture
, the ion - ion and electron - ion rdf s @xmath10 are identical with the ion- and electron - density distributions under the external potential caused by a fixed ion at the origin respectively , since the ions behave as a classical fluid in a liquid metal.@xcite in general , df theory enables us to provide exact expressions for the density distributions in an inhomogeneous system caused by an external potential .
hence , by applying df theory to this mixture with densities @xmath11 , we can derive exact expressions for the ion - ion and electron - ion rdf s in terms of direct correlation functions ( dcf s ) @xmath12 and bridge - functions @xmath13 as follows:@xcite @xmath14\;,\\ g_{\rm ei}(r ) & = & n_{\rm e}^{0f}(r|u_{\rm e}^{\rm eff})/n_0^e\;,\end{aligned}\ ] ] with @xmath15d{\bf r } ' + b_{i\rm i}(r ) \right ] \\
& = & v_{i\rm i}(r)-\gamma_{i\rm i}(r)/\beta \;.\end{aligned}\ ] ] the wave equation for an electron under the external potential @xmath16 is solved to provide the total electron - density distribution @xmath17 around a nucleus , which is divided into the bound - electron and free - electron parts : @xmath18 by the criterion whether their eigenfunctions belong to bound- or free - levels . in eq .
( 2.2 ) , the free - electron part of the density distribution , @xmath19 , becomes identical to the electron - ion rdf ; the bound - electron part @xmath20 is taken to form an ion and contributes to generate the electron - ion interaction @xmath7 .
these expressions for @xmath10 can be rewritten in the form of a set of integral equations for a one - component fluid.@xcite one of them is a usual integral equation for the dcf @xmath21 of a one - component fluid : @xmath22 - 1-{\it\gamma}(r)\;,\ ] ] with an interaction @xmath23 defined by @xmath24 the other is an equation for the effective interaction @xmath23 , that is expressed in the form of an integral equation for the electron - ion dcf @xmath25 : @xmath26 in these equations , @xmath27 is the density response function of the noninteracting electrons with an electron density @xmath28 and @xmath29d{\bf r}'$ ] .
furthermore , @xmath30 denotes an operator defined by @xmath31 \equiv \chi_q^0\!\int e^{i{\bf q}\cdot{\bf r } } f(r)d{\bf r } \;.\ ] ] it should be kept in mind that the electron - ion dcf in eq .
( [ eq : veffii ] ) plays the role of a nonlinear pseudopotential , which takes into account nonlinear electron - accumulations around an ion and the influence of other ions in the form of a linear - response expression ; if the electron - ion dcf @xmath25 is replaced by a usual pseudopotential @xmath32 , then @xmath33 in eq .
( [ eq : veffii ] ) becomes an effective ion - ion interaction in pseudopotential theory based on the linear - response formalism . from the above formal and exact expressions ( [ eq : qhncii])-([eq : qhncei ] ) , the qhnc equations are obtained by introducing the following five approximations:@xcite + 1 .
the electron - ion bridge - function is neglected in eq .
( [ eq : qhncei ] ) : @xmath34 ( the hnc approximation ) .
the bridge - function @xmath35 of the ion - electron mixture is approximated by that of a one - component hard - sphere fluid ( modified hnc
approximation@xcite ) .
the electron - electron dcf @xmath36 in the ion - electron mixture is approximated by that of the jellium model : @xmath37 $ ] , which is written in terms of the the local - field correction ( lfc ) @xmath38 . in our calculation
we use the lfc proposed by geldart and vosko.@xcite 4 .
an approximate @xmath7 is obtained by treating a liquid metal as a nucleus - electron mixture@xcite in the form : @xmath39 -\mu_{\mbox{\tiny xc}}(n_0^{\rm e})\;.\ ] ] here , @xmath40 is the bound - electron distribution and @xmath41 the exchange - correlation potential in the local - density approximation , for which we have adopted the gunnarsson - lundqvist@xcite formula .
actually , the bound - electron distribution @xmath40 is determined as the bound - electron part of @xmath18 , when the electron - ion rdf in eq .
( [ eq : qhncei ] ) is calculated from the wave equation for an electron under the external potential @xmath42 .
the bare ion - ion interaction is taken as pure coulombic : @xmath43 . under these approximations , a set of integral equations ( [ eq : qhncii])-([eq : qhncei ] ) can be solved ; its solution allows the determination of the electron - ion and ion - ion correlations together with the ionization and the electron bound - states . for the mhnc closure relation we have used in this work the parametrization for the bridge - function of a
suitably chosen hard - sphere reference system that was proposed recently by rosenfeld.@xcite in this version of the mhnc the universality hypothesis of the bridge-_function _ @xcite ( which justifies the mhnc ) is generalized to the level of the bridge-_functional _ of the reference system .
this functional can be calculated very easily for the general case of an inhomogeneous system of hard - spheres @xcite involving only fundamental measures ; the functional is then specialized as required for our case to a system of homogeneous hard - spheres .
this fundamental measure bridge - functional is given in terms of characteristic quantities of the _ individual _ spheres and involves only integrations over known functions .
furthermore , in this approach the functional can be optimized by imposing the test - particle ( or source - particle ) self - consistency , which is realized by the transition from an inhomogeneous system to a homogeneous one if the source of the external potential becomes a particle of the liquid.@xcite the ornstein - zernike equation is then solved for the structure function of the homogeneous system along with the closure relation where the bridge - function is calculated by means of the above _ functional _ , assuming that the universality hypothesis is valid .
the obtained structure function is then fed into the bridge - functional yielding a new , improved bridge - function .
this procedure is iterated until numerical self - consistency is obtained in a sense that the structure function of the preceding step differs only marginally from the present step .
this method , which we denote by qhnc - mh , is in fact able to produce accurate data for the structure factors of rb as demonstrated in fig .
[ fig : sq313 ] in the direct comparison with experiments@xcite at the triple point ( 313 k ) ; the experimental data are denoted by the open and black circles .
in this section , we apply the qhnc - mh method to calculate the structure factors for compressed liquid - rubidium under high pressure ( 0.2 , 2.5 , 3.9 , and 6.1 gpa ) , i.e. , exactly the same pressure - values under which tsuji _ et al . _
@xcite performed their experiments to measure the structure factors .
the corresponding densities are estimated by these authors to be 1.07 , 1.41 , 1.56 and 1.95 times the normal density ; the temperatures are 370 , 520 , 540 and 570 k , respectively . in a first step
we apply our method to the 3.9 gpa state and examine how accurately the mhnc equation ( in combination with the rosenfeld bridge - function ) is able to reproduce the rdf obtained in an md simulation : the comparison is shown in fig . [
fig : rdfmd ] , where the qhnc - mh result ( full curve ) is found to be undistinguishable from the md rdf ( open circles ) .
this comparison confirms that the qhnc - mh method is able to produce reliable structure - data that are as accurate as those obtained in computer - experiments even for compressed liquids at high densities ; hence , md simulations are no longer required in this study .
we can therefore proceed to compare the qhnc - mh results with _
data for all high - pressure states : the structure factors calculated for pressures ranging from 0.2 to 6.1 gpa are plotted by full curves in fig .
[ fig : sqall ] in comparison with the experimental results shown by full circles .
we have found that our method gives structure factors in good agreement with experiment for 0.2 and 6.1 gpa .
however , in the case of 2.5 and 3.9 gpa , the agreement between the theoretical and experimental results is not so convincing : a systematic deviation between the data sets is observed . in their analysis of their experiment ,
tsuji _ et al . _
@xcite estimated the density of liquid rb under pressure from the measured lattice constant of crystalline rb including corrections for the thermal expansion and the volume jump at melting . however , according to tsuji,@xcite there remains an uncertainty in the evaluation of the density , that might be responsible for these deviations .
therefore , we have modified the density for the 3.9 gpa case : when decreasing the ion - sphere radius ( wigner - seitz radius ) @xmath0 [ @xmath44 and @xmath4 being the specific volume per ion ] by a factor of 1/1.05 we can find indeed a good agreement between the theoretical and experimental results for the 3.9 gpa state , as shown in fig .
[ fig : sqrees ] ; a similar good agreement is found also for the 2.5 gpa case if the wigner - seitz radius @xmath0 is decreased by 5 % .
it should be mentioned that the concept of our qhnc - method is for high pressures as reliable and valid as for the room pressure , since all the approximations entering this method remain valid as the pressure is increased .
this fact can be seen from the result for the electronic structure , which will be discussed in the following .
the bound levels of the ion in compressed liquid rb are almost the same as those at room pressure . as a consequence ,
the electron - ion rdf s remain as displayed in fig .
[ fig : rdfei ] almost unchanged for the five states considered .
the electron - ion dcf @xmath25 is determined by eq .
( [ eq : qhncei ] ) .
figure [ fig : cei ] illustrates the pressure - variation of the electron - ion dcf @xmath25 , which as noted above plays the role of a non - linear pseudopotential in the expression for the effective ion - ion interaction [ cf . eq .
( 2.5 ) ] ; also the electron - ion dcf does not change significantly under these pressure - variations . here , note that the electron - ion structure factor @xmath45 , the fourier transform of the electron - ion rdf , is written in the following form : @xmath46 where @xmath47 [ 1-n_0^e c_{ee}(q)\chi_q^0 ] -n_0^in_0^e| c_{ei}(q)|^2\chi_q^0 . \end{aligned}\ ] ] hence , eq .
( [ e : sei2 ] ) can be represented in @xmath48-space as @xmath49 which states that the free - electron distribution @xmath50 around an ion can be described exactly by the superposition of surrounding neutral pseudoatoms " .
each ion carries a screening electron - cloud @xmath51 [ with @xmath52 , and makes it thus electrically neutral ( including the core - electrons ) as if it were an atom .
therefore , in this formalism a liquid metal can be considered to be composed by neutral pseudoatoms . using eq .
( [ e : qrho ] ) , the free - electron density distribution of a pseudoatom is calculated for liquid rb under high pressures ( cf .
[ fig : pseudatom ] ) : the results indicate that the electron - density distribution of a pseudoatom suffers no significant change outside of the core - region ( where the bound - electron density is large ) under this pressure - variation .
summarizing , we can conclude that a pseudoatom remains almost unchanged in comparison to room pressure even when the density is increased by a factor of nearly 2 ( high pressure - state 6.1 gpa ) . on the other hand , tsuji _
et al_. expected that a hard - sphere model@xcite might be successfully applied to describe the structure of liquid alkali metals and that the effective diameter of the hard - sphere should vary with pressure due to the change in the screening effect ; that is , the size of a pseudoatom is assumed to be changed with pressure .
based on this model , the effective hard - sphere radius @xmath53 is considered to vary under pressure - variation keeping the packing fraction @xmath54 to be constant at a value of 0.45 ; this leads to a uniform - compression model in which the position of first peak @xmath2 in the structure factor should increase proportionally to @xmath55 with increasing pressure .
we also find in our approach that the effective ion - ion interaction for liquid rb is practically invariant under this pressure - variations ( cf .
8) , i.e. , it remains almost the same as the one at room pressure .
in contrast to the hard - sphere model , where the essential repulsive part of the effective ion - ion potential should be scaled in units of @xmath0 , no scaling feature is observed for the effective ion - ion interaction in units of the wigner - seitz radius @xmath0 . nevertheless , all the structure factors for all these five pressure - values almost coincide in one single curve when scaling the wavenumber in units of the wigner - seitz radius @xmath0 ( as shown in fig .
[ fig : scaledsq ] ) .
even if the liquid - rubidium potential under these pressure - variations can be considered as invariant with respect to the one at room pressure ( neglecting small deviation ) , we have also observed the above mentioned scaling property in the structure factor ( cf .
[ fig : scaledsq ] ) ; this means that the liquid - rubidium potential has a very special feature . for
a _ general _ liquid ( the lennard - jones potential for example ) , it is impossible to display the structure factor of two different states , scaled in length - units in such a way that they practically coincide . for a liquid alkali metal
such a scaling property _ can _ be observed : to examine this , let us consider liquid li as a further example . at
first we calculate the effective ion - ion interaction of liquid li at room pressure ( 470 k , @xmath56 with @xmath57 ) , and assume as for rb that the effective li potential is unchanged with respect to the one at room pressure .
then , we can easily find the following three sets of temperatures and densities : ( 470 k , @xmath56 ) , ( 600 k , 1.34@xmath56 ) and ( 750 k , 2.37@xmath56 ) , i.e. states for which the structure factors can be scaled almost in a single curve in unit of @xmath0 , as demonstrated in fig [ fig : lisq ] . for alkali liquid metals ,
this scaled structure factor is not very sensitive to states specified by the plasma parameter @xmath58 and @xmath59 in the sufficiently high - density region .
this is the main reason why structure factors of all alkali liquids ( from li to cs ) near the triple point coincide almost in a single curve when scaling @xmath1 in units of the wigner - seitz radius @xmath0 , as was observed experimentally@xcite and theoretically.@xcite this scaling property in the structure factors of liquid rb signifies that the first peak of the structure factor appears almost at the same position @xmath60 ( in scaled units ) for all these pressures ; this means that the position of the first peak @xmath2 in the structure factor ( taken in absolute values ) should increase proportionally to @xmath61 , where @xmath62 is the specific volume at room pressure . in our calculation , this peak - position in reduced units is estimated from fig .
[ fig : scaledsq ] to be @xmath63 , from which we obtain the relation @xmath64 because of @xmath65 ; this linear relation is plotted in fig .
[ fig : q1unif ] . in this figure , the full and the open circles denote the experimental points obtained for several states by tsuji _ et al._;@xcite in particular ,
the five full circles represent those states which we have investigated in our theoretical study , i.e. , for pressure - values 0 , 0.2 , 2.5 , 3.9 , and 6.1 gpa , respectively .
these _ experimental _
@xmath2-points in fig .
[ eq : qhncei ] are close to the linear relation which was determined from the _ calculated _ structure factors ; thus , this figure demonstrates that the uniform - compression model ( corresponding to a linear relation denoted by the full line ) is indeed valid , although there is some uncertainty due to difficulties in estimating the density and to problems in experiments under high pressure , in general .
the qhnc method in combination with the rosenfeld bridge - functional has been shown to reproduce the experimental structure factors of liquid rb under high pressures ( ranging from 0 to 6.1 gpa ) very accurately ; for the case of 2.5 and 3.6 gpa the experimentally estimated densities have to be readjusted in terms of a 5 % variation of the wigner seitz radius @xmath0 .
furthermore we observe that the structure factors coincide almost in one single curve if wavenumber are scaled in units of @xmath0 ( fig .
[ fig : scaledsq ] ) .
this indicates clearly that liquid rb is uniformly compressed as the pressure is increased : this , in turn signifies , that the first peak - position @xmath2 of the structure factors increases proportional to @xmath66 ( fig .
[ fig : q1unif ] ) .
it must be mentioned that in contrast to our result , shimojo _
et al._@xcite conclude from their result obtained in the first - principles md simulations that some deviation from uniform compression exists for the 6.1 gpa state , though liquid rb is compressed uniformly at 2.5 gpa .
their experimental rdf s for 2.5 and 6.1 gpa are obtained by a fourier transform of the experimental structure factors of tsuji _
et al_. on the basis of the rdf s they observe a different behaviour than on the basis of the structure factor , although they have used the same experimental structure factors and rb states as we did : ( i ) the first - peak position in the rdf at 2.5 gpa follows the uniform compression model , while the first - peak position in the rdf at 6.1 gpa shows a deviation from the uniform compression model .
thus , they have asserted that their calculated result agrees with the experiment .
( ii ) in contrast to their rdf s , the experimental structure factor at 2.5 gpa shows a substantial deviation from the uniform compression compared to the 6.1 gpa state ; the position @xmath2 of the first peak in @xmath67 lies far away from the uniform compression line in fig .
[ fig : q1unif ] while on the other hand , the experimental structure factor at 6.1 gpa shows that the uniform compression model is still valid : the @xmath2-point for 6.1 gp in fig .
[ fig : q1unif ] is very close to the uniform compression line . according to tsuji,@xcite this discrepancy between the conclusions based on the rdf s and the structure factors is possible , since there is some experimental ambiguity in the value of the peak height in the structure factor while the peak - position is accurate and reliable . nevertheless , from all the experimental data displayed in fig .
[ fig : q1unif ] , we can conclude that liquid rb is compressed uniformly up to a pressure of 6.1 gpa ; however , a more detailed discussion if deviations from the uniform compression model ( cf . fig .
[ fig : q1unif ] ) have any physical meaning is not very conclusive , due to the uncertainty in the evaluation of the experimental density and the difficulties encountered in experiments under high pressure .
it is interesting to notice that the structure factors of compressed liquid rb coincide practically in one single curve if wavenumbers are scaled in units of the wigner - seitz radius @xmath0 , despite of the fact that the effective ion - ion interaction remains under pressure unchanged with respect to room pressure ; this means that interaction potentials of liquid alkali metals have a special characteristic property , as we have demonstrated and confirmed in addition for the case of liquid li in fig .
[ fig : lisq ] .
the neutral pseudoatom in compressed liquid rb remains almost unchanged under pressure - variation , which , in turn , is the reason why the effective ion - ion interaction remains practically invariant , similar to state-_independent _ interactions , such as the one for liquid argon .
we have demonstrated in this contribution that the qhnc method can treat the outer - structure problem ( i.e. , calculation of the ion - ion and electron - ion rdf s ) and the inner - structure problem ( i.e. , calculation of the electronic structure of the ion ) in a self - consistent way using the atomic number as the only input data : therefore , this method can be considered to be very useful to treat a plasma , where the ionization @xmath6 is not known beforehand and where there is no way of constructing a pseudopotential to give the effective ion - ion interaction . from the successful results for compressed liquid rb presented here
we can conclude that the qhnc method is expected to be nicely applicable for plasma states in a wide range of densities and temperatures .
we thank professor tsuji and dr .
morimoto for sending us numerical experimental data and for discussions on the analysis of the experimental data .
gk acknowledges financial support by the sterreichische forschungsfonds under project no .
p11194-phy .
j. chihara , phys .
rev . a * 33 * , 2575 ( 1986 ) . j. chihara , j. phys . : condens . matter * 3 * , 8715 ( 1991 ) .
j. chihara , phys .
rev . a * 40 * , 4507 ( 1989 ) .
m. ishitobi and j. chihara , j. phys . : condens .
matter * 4 * , 3679 ( 1992 ) .
m. ishitobi and j. chihara , j. phys . : condens .
matter * 5 * , 4315 ( 1993 ) .
j. chihara and s. kambayashi , j. phys . : condens .
matter * 6 * , 10221 ( 1994 ) .
s. kambayashi and j. chihara , phys .
e * 53 * , 6253 ( 1996 ) .
y. rosenfeld , j. chem .
* 98 * , 8126 ( 1993 ) .
g. kahl , b. bildstein , and y. rosenfeld , phys .
e * 54 * , 5391 ( 1996 ) .
k. tsuji , in _ elementary processes in dense plasmas _ , edited by s. ichimaru and s. ogata , ( addison - wesley , new york , 1995 ) ; + k. tsuji , y. katayama , y. morimoto , and o. shimamura , j. non - cryst .
solids * 205 - 207 * , 295 ( 1996 ) .
y. morimoto , ( private communication ) .
f. shimojo , zempo , k. hoshino , and m. watabe , phys .
b * 55 * , 5708 ( 1997 ) .
y. rosenfeld and n. w. ashcroft , phys .
a * 20 * , 1208 ( 1979 ) .
d. j. w. geldart and s. h. vosko , can . j. phys .
* 44 * , 2137 ( 1966 ) . o. gunnarsson and b. i. lundqvist , _ phys .
b * 13 * , 4274 ( 1976 ) .
j. chihara , j. phys .
c : solid state phys .
* 18 * , 3103 ( 1985 ) .
y. rosenfeld , phys .
letters * 63 * , 980 ( 1989 ) .
j. r. d. copley and s. w. lovesey , int .
30 * , 575 ( 1979 ) ; + y. waseda , _ the structure of non - crystalline materials _ , ( mcgraw - hill , new york , 1980 )
. k. tsuji , ( private communication ) .
n. w. ashcroft and j. lekner , phys .
145 * , 83 ( 1966 ) .
the experimental data for the @xmath68 relation are provided by tsuji in the numerical table .
percus , phys .
* 8 * , 462 ( 1962 ) ; j.k percus , in : _ the equilibrium theory of classical fluids _
( ed . : h.l . frisch and j.l .
lebowitz ) ( benjamin , new york , 1964 ) , ii-171 .
huijben and w. van der lugt , _ acta cryst .
_ a * 35 * , 431 ( 1979 ) .
y. waseda and k. suzuki , _ proceedings of the second international conference on on the properties of liquid metals _ , s. takeuchi , ed , taylor & francis , london , 1973 , pp .
. h. minoo , c. deutsh and j. p. hansen , _ j. phys.(paris ) _ * 38 * , l-191 ( 1977 ) .
u. balucani , a. torcini , and r. vallauri , phys .
b * 47 * , 3011 ( 1993 ) . | we have applied the quantal hypernetted - chain equations in combination with the rosenfeld bridge - functional to calculate the atomic and the electronic structure of compressed liquid - rubidium under high pressure ( 0.2 , 2.5 , 3.9 , and 6.1 gpa ) ; the calculated structure factors are in good agreement with experimental results measured by tsuji _
et al .
_ along the melting curve .
we found that the rb - pseudoatom remains under these high pressures almost unchanged with respect to the pseudoatom at room pressure ; thus , the effective ion - ion interaction is practically the same for all pressure - values .
we observe that all structure factors calculated for this pressure - variation coincide almost into a single curve if wavenumbers are scaled in units of the wigner - seitz radius @xmath0 although no corresponding scaling feature is observed in the effective ion - ion interaction .
this scaling property of the structure factors signifies that the compression in liquid - rubidium is uniform with increasing pressure ; in absolute @xmath1-values this means that the first peak - position ( @xmath2 ) of the structure factor increases proportionally to @xmath3 ( @xmath4 being the specific volume per ion ) , as was experimentally observed by tsuji _
et al_. this scaling property comes from a specific feature characteristic for effective ion - ion potential of liquid rubidium : even if the effective liquid - rubidium potential is invariant under this pressure - variation with respect to the potential at room pressure , we can nevertheless observe this scaling property for the structure factors .
this property is obviously characteristic for the potential of alkali metals and we have examined and confirmed this feature for the case of a liquid - lithium potential : starting from the liquid - lithium potential at room pressure we can easily find two sets of densities and temperatures for which the structure factors become practically identical , when scaling @xmath1 in units of @xmath0 . |
the coulomb excitation of nuclei at relativistic energies is commonly calculated in first - order perturbation theory , using the formalism developed in ref .
the derivation is restricted to distant collision , where the intrinsic radial coordinates are smaller than the minimum distance between the colliding nuclei . in calculations of the coulomb dissociation of proton halo nuclei ,
it is of interest also to consider close collisions , since the density of the valence proton can extend to very large distances .
moreover , higher - order processes may also play a role , as suggested by the non - relativistic calculations of the @xmath0b breakup reported in ref .
@xcite . in order to calculate the coulomb dissociation of halo nuclei at energies where relativistic and higher - order effects can not be ignored
, one would need realistic multipole form factors for the electro - magnetic interaction .
such form factors have recently been studied @xcite but they were only determined in the long - wave - length limit for distant collisions .
we therefore find it timely to study the multipole decomposition of the interaction at relativistic energies , both for close and distant collisions .
the form factors we derive can be applied in higher - order dynamical calculations of two - body breakup reactions , such as the continuum discretized coupled - channels calculations @xcite , or in numerical methods that evolve the two - body wave function in coordinate space @xcite .
we base our study on the so - called linard - wiechert potential and include the effect of the convection current .
the magnetization current , on the other hand , is ignored for simplicity .
we assume that the relative velocity of the interacting nuclei is a constant , so that the retardation effect associated with a time - dependent velocity can be ignored .
the multipole expansion of the interaction contains both diagonal and off - diagonal multipole components .
the off - diagonal components appear because of the lorentz contraction .
our approach is similar to the method developed by baltz et al .
@xcite in their study of the @xmath1 pair production in relativistic heavy - ion collisions .
we extend their study and derive the multipole fields that are relevant to coulomb excitation . the multipole expansion of the linard - wiechert potential is presented in sect .
we also discuss how to calculate the interaction associated with the convection current .
we propose in sect .
iii a simple truncation in the the sum over off - diagonal multipole components .
we test this approximation in sect .
iv in first - order perturbation theory for distant collisions by comparing to the exact results of ref .
v contains our conclusions .
the electro - magnetic interaction of a fast charged particle ( with atomic number @xmath2 ) and a weakly bound proton in a target nucleus has the form @xcite @xmath3\bigr).\eqno(1)\ ] ] the charged particle is assumed to move with constant velocity * v * in the z - direction on the straight - line trajectory @xmath4 , with impact parameter * b * with respect to the target nucleus .
the coordinates of the proton in the target nucleus are denoted by * r * , and @xmath5 is the conjugate momentum operator .
the potential @xmath6 is the linard - wiechert potential @xmath7 where @xmath8 , and @xmath9 .
the second part of the interaction ( 1 ) , which contains the momentum operator @xmath5 , is due to the convection current .
we ignore the magnetization current but it could be included by replacing the momentum operator by @xmath10 , where @xmath11 is the magnetic moment of the proton . in the two - body breakup of a target nucleus , into a proton and a core nucleus , one would actually have to consider the effect of two interactions with the projectile , namely , the ` direct ' interaction with the proton and the ` recoil ' interaction with the core .
we will not discuss this complication here , since the multipole expansion of the two interactions can easily be generated from the formulation given below .
the multipole expansion of the linard - wiechert potential @xmath6 can be obtained from the fourier representation @xcite @xmath12 except for the last factor , this is just the ordinary coulomb potential . to proceed , we introduce the multipole expansion of the last term in eq .
( 3 ) @xmath13 where @xmath14 and @xmath15 are legendre polynomials of the second kind ; cf .
eq . 8.825 of ref .
@xcite or eq .
( 8.8.3 ) of ref .
explicit expressions and recursion relations for @xmath16 are given in ref .
@xcite .
we can now insert the plane wave expansion @xmath17 for the two plane waves in eq .
( 3 ) and obtain @xmath18 where @xmath19 and @xmath20 the above expressions were derived for a trajectory with constant velocity in the z - direction .
the matrix ( 5b ) is therefore diagonal in _ m_. let us also give a more general result which does not refer to any specific coordinate system .
this can be done by noticing that the angle @xmath21 in eq .
( 5b ) is actually the angle between * q * and the velocity * v*. thus by inserting @xmath22 = @xmath23 we obtain @xmath24 where the last two clebsch - gordan coefficients is the matrix element @xmath25 .
we note that the expansion ( 5 ) contains diagonal ( i.e. @xmath26 ) as well as off - diagonal multipole components ( i. e. @xmath27 .
the off - diagonal components are caused by the lorentz contraction which destroys the spherical symmetry of the potential @xmath6 .
the radial dependence is determined by the integral ( 5a ) , and analytic expressions and recursion relations are derived in app .
a. in numerical applications one can include the finite size of the projectile simply by multiplying the integrand in eq .
( 5a ) by the fourier transform of the charge distribution .
however , we find it instructive to show the analytic expressions for point - particles
. the radial dependence of the diagonal components is determined by ( see eq .
( a.5 ) ) @xmath28 where @xmath29 and @xmath30 . this is identical to the dependence one has in the non - relativistic limit .
in fact , the well - known multipole expansion of the coulomb interaction is recovered from eq .
( 5 ) , when we insert the non - relativistic limit of eq .
( 5b ) , viz .
@xmath31 = @xmath32
. the off - diagonal multipole contributions to @xmath6 are more complicated .
the radial form factors have the symmetry property @xmath33 according to eq .
( 5b ) , we only need expressions for even values of @xmath34 .
we show in app .
a that @xmath35 the explicit , non - zero expressions can be obtained from eq .
for @xmath36 we obtain in particular ( see eq .
( a.5 ) ) @xmath37 , \ \
{ \rm for } \ \ r < r'.\eqno(8)\ ] ] one can exploit the properties ( 7a ) and ( 7b ) and write expressions for @xmath6 that are valid for distant and close collisions , respectively .
thus we obtain for distant collisions ( @xmath38 ) @xmath39 for close collisions ( @xmath40 ) we can use the expression @xmath41 the sum over @xmath42 is finite for close collisions , since @xmath43 must be non - negative .
this is a very nice feature , which makes it feasible to include all terms .
for distant collisions , on the other hand , one would have to make a truncation in the sum over @xmath42 .
the contribution to the interaction ( 1 ) that originates from the convection current is @xmath44 the operator @xmath45 is effectively of dipole nature , as shown in app .
b. in practical numerical calculations it will operate on an expression of the form @xmath46 . > from eq .
( b.4 ) we then obtain @xmath47 @xmath48\bigr).\eqno(11)\ ] ] we can also give an explicit expression for the term @xmath49 , which appears in second version of eq .
( 10 ) . assuming that * v * points in the z - direction , we obtain from eq . ( 3 ) @xmath50 we can now repeat the derivation that lead to eq .
( 5 ) and obtain @xmath51 where @xmath52 and @xmath53 the last expression is similar to eq .
( 5b ) but the sum is now over odd values of @xmath54 .
the dependence on the radial coordinates can be derived from eqs .
( a.7 - 8 ) .
we note that matrix elements of @xmath55 can be calculated in a much simpler way in first - order perturbation theory , as we shall see in sect .
in order to illustrate the significance of relativistic corrections at intermediate energies , we show in fig .
1 the functions @xmath56 defined in eq . ( 4b ) .
they are seen to decrease rapidly as function of @xmath54 when @xmath57 .
this suggests that one would only need a few terms in the sum over @xmath42 in eq .
( 9a ) . in the next section
we determine the range of velocities where a truncation to @xmath58 is reasonable for distant collision .
thus we will apply the approximation @xmath59 @xmath60.\eqno(13)\ ] ] we assume a straight line trajectory with constant velocity in the z - direction and calculate the coulomb excitation in first - order perturbation theory .
we can then test the accuracy of the truncation , eq .
( 13 ) , by comparing to the exact results from ref .
in first - order perturbation theory for distant collisions one needs the fourier integral of the multipole fields , @xmath61 where @xmath62 is the excitation energy and @xmath63 are the so - called orbital integrals , @xmath64 for non - relativistic coulomb excitation and a straight - line trajectory , these integrals are @xcite @xmath65 below we derive the orbital integrals for the truncated interaction ( 13 ) .
the fourier transform ( 14a ) of the multipole field ( 13 ) can be expressed in terms of various combinations of the non - relativistic orbital integrals ( 15 ) .
this is obvious for the first term in ( 13 ) , and the second term is calculated in app . c. a slight complication is the very last term in eq .
( 13 ) which contains the factor @xmath66 .
it turns out to be a factor of @xmath67 smaller than the dominant term so let us ignore it .
thus we obtain @xmath68,\ ] ] _ -^dt
e^it y_l+2,m(r)r^l+1.(16)@xmath69 the latter integral is given in eq .
( c.5 ) .
we also need to calculate the contribution from the convection current , eq .
( 10 ) . here
we can again make the substitution ( b.1 ) .
assuming that the single - particle potential commutes with @xmath70 we can go a step further and replace the momentum operator by @xmath71 $ ] , where @xmath72 is the single - particle hamiltonian . thus we obtain @xmath73\phi + \phi [ h , z]\bigr ) = -{i\beta\over 2\hbar c } \bigl([h , z\phi ] + \phi hz -zh\phi\bigr).\eqno(17)\ ] ] we note that in first - order perturbation theory one can effectively replace the commutator @xmath74 $ ] by @xmath75 .
the other term , @xmath76 , has to be considered more carefully .
let us write the multipole expansion of @xmath55 as in eq .
( 14a ) , @xmath77 since @xmath78 in eq .
( 17 ) is a dipole operator , we see that @xmath79 can receive contributions from both @xmath80 and @xmath81 .
the contribution from the latter is smaller ( by a factor of @xmath67 ) so let us ignore it .
let us first consider the contribution from the commutator @xmath74 $ ] .
this leads to the expression @xmath82 ) \approx
-{i\beta^2\over 2 } \ { \omega\over v } \
s_{l-1,m } \ { \sqrt{l^2-m^2}\over 2l-1}.\eqno(18)\ ] ] for the dipole field
we see that the second term in eq .
( 17 ) , namely @xmath83 , is identical to the first term because @xmath84 commutes with the hamiltonian _
thus we obtain @xmath85 for the quadrupole field we can neglect the second term of eq . (
17 ) it is identical to zero when we insert @xmath86 , whereas inserting @xmath87 leads to a magnetic type transition ( @xmath88 ) .
thus we will use the approximation @xmath89 in order to illustrate the orbital integrals we obtain from the approximation ( 13 ) , it is useful to plot the reduced values @xmath90 the results we obtain for dipole and quadrupole excitations are shown by the solid symbols in figs . 2 and 3 , respectively , as functions of the adiabaticity parameter @xmath91 .
here we have chosen the beam velocity @xmath92 the full relativistic expression for electric transitions @xcite , which also includes a contribution from the convection current , is @xmath93 where @xmath94 can be extracted from @xcite , @xmath95 this result is shown by the solid curves in figs . 2 and 3 .
the dashed curves represent the non - relativistic result @xmath96 , c. f. eq .
it is seen that relativistic effects are significant and that the truncated calculation is in very good agreement with the full result .
if we choose a larger velocity , say @xmath97 = 0.8 , some discrepancy starts to occur , so it will be necessary to include higher values of @xmath42 in the sum ( 9a ) .
the multipole expansion of the electro - magnetic interaction in relativistic heavy - ion collisions that we have presented is exact for a straight - line trajectory with constant velocity .
it has not previously , as far as we know , been used in calculations of the coulomb excitation .
the formulation is , in particular , applicable to reactions where close collisions play a role , and it is also well suited for calculations of higher - order processes in the coulomb dissociation of halo nuclei at intermediate energies .
the multipole expansion contains diagonal as well as off - diagonal components , associated with the intrinsic coordinates and the coordinates for the relative motion of the colliding nuclei .
the off - diagonal multipole components are caused by relativistic effects , essentially by the lorentz contraction of the coulomb field ; only the diagonal components survive in the non - relativistic limit .
the number of off - diagonal multipole components is finite for close collisions but it is , in principle , infinite for distant collisions .
however , it is sufficient to include just a few off - diagonal terms at intermediate energies .
thus we find that the diagonal and first off - diagonal terms ( i. e. , @xmath58 in eq .
( 9a ) ) give excellent results in first - order perturbation theory for distant collisions and for velocities up to @xmath98 , i. e. up to about 200 mev / nucleon .
* acknowledgments * this work was supported by the us department of energy , nuclear physics division , under contracts w-31 - 109-eng-38 and de - ac02 - 98ch10886 , and by a fellowship ( c.a.b . ) from the j.s .
guggenheim memorial foundation .
here we study the integral in eq .
( 5a ) @xmath99 where @xmath100 and @xmath101 is an even number , according to parity selection of eq .
this integral is given in many textbooks but we shall mainly refer to abramowitz @xcite . we note that it has the symmetry @xmath102 so it is sufficient to determine ( a.1 ) for all values of @xmath103 and @xmath104 .
let us first assume that @xmath105 . from eq .
11.4.34 of ref .
@xcite we obtain ( with @xmath106 and @xmath107 ) @xmath108 the hypergeometric function in eq .
( a.3 ) is a polynomial of degree _
m _ , which can be obtained directly for the series eq .
15.1.1 of ref .
@xcite . inserting into that expression the pochhammer s symbols
( 6.1.22 ) of ref .
@xcite ) we obtain @xmath109 for @xmath110 = 0 , 1 and 2 we obtain in particular @xmath111,\ ] ] @xmath112 \
\bigl[\mu+1-(\mu+3)(a / b)^2\bigr].\eqno(a.5)\ ] ] using the symmetry property ( a.2 ) we see that @xmath113 is consistent with eq . ( 6 ) . when @xmath114 and @xmath110 is non - zero we find that @xmath115 this result follows from eq .
11.4.33 of ref .
@xcite because that expression will contain the factor @xmath116 which is zero for @xmath117 the functions @xmath118 defined in eq .
( 12a ) can be obtained from the relation @xmath119 this expression follows directly from the definitions in eqs .
( 5a ) and ( 12a ) and the recursion relation for spherical bessel functions , @xmath120 = @xmath121 . for @xmath122
we find explicitly ( c.f .
6.575 of ref .
@xcite ) @xmath123 while it is zero when @xmath40 .
the radial dependence can now in general be derived from the recursion relation ( a.7 ) .
here we derive an expression for the operator @xmath45 , which appears in the contribution from the convection current .
it can be expressed in terms of a commutator with the laplace operator @xmath124 as follows @xmath125.\eqno(b.1)\ ] ] here we can insert @xmath126 and express the laplace operator in spherical coordinates @xmath127 where @xmath128 is the angular momentum operator .
thus we obtain @xmath129\bigr).\eqno(b.4)\ ] ] here we derive an expression for the integral @xmath130 which appears in eq ( 16 ) .
this integral can be calculated in a simple way by observing that @xmath131 where ( use eq .
( 14 ) of app .
@xcite ) @xmath132 @xmath133 @xmath134 the trajectory we consider falls in the x - z plane , with @xmath135 and @xmath136 .
the integral ( c.1 ) can therefore be expressed in terms of the non - relativistic orbital integrals ( 15 ) as follows @xmath137 @xmath138.\eqno(c.3)\ ] ] a. winther and k. alder , nucl .
* a319 * , 518 ( 1979 ) . h. esbensen and g. f. bertsch , nucl . phys . *
a600 * , 37 ( 1996 ) . c. a. bertulani and v. yu .
ponomarev , phys . rep . * 321 * , 139 ( 1999 ) .
f. m. nunes and i. j. thompson , phys .
c * 59 * , 2652 ( 1999 ) .
h. esbensen , g. f. bertsch , and c. a. bertulani , nucl .
* a581 * , 107 ( 1995 ) .
t. kido , k. yabana , and y. suzuki , phys .
c * 53 * , 2296 ( 1996 ) . v. s. melezhik and d. baye , phys .
c * 59 * 3232 ( 1999 ) .
a. j. baltz , m. j. rhodes - brown , and j. weneser , phys .
a * 44 * , 5569 ( 1991 ) . i. s. gradshteyn and i. m. ryzhik , _ tables of integrals , series and products _ , ( academic press , new york , 1965 ) .
m. abramowitz and i. a. stegun , _ handbook of mathematical functions _ ( 4th printing , december 1965 ) . k. alder and a. winther , _ electromagnetic excitation _ ( north - holland , 1975 ) . | we derive a general expression for the multipole expansion of the electro - magnetic interaction in relativistic heavy - ion collisions , which can be employed in higher - order dynamical calculations of coulomb excitation .
the interaction has diagonal as well as off - diagonal multipole components , associated with the intrinsic and relative coordinates of projectile and target .
a simple truncation in the off - diagonal components gives excellent results in first - order perturbation theory for distant collisions and for beam energies up to 200 mev / nucleon .
= 10000 |
the jacobi shape transition , an abrupt change of nuclear shape from an oblate ellipsoid non - collectively rotating around its symmetry axis , to an elongated prolate or triaxial shape , rotating collectively around the shortest axis has been predicted to appear in many nuclei at angular momenta close to the fission limit @xcite .
signatures of the presence of elongated shapes can be found , among others , in the @xmath0-decay of the giant dipole resonance ( gdr ) , in the giant back - bend of the @xmath52 @xmath0-transition energies @xcite and in the angular distribution of the emitted charged particles . until now , however , a firm experimental evidence for the jacobi shape transition in nuclei could not be established .
some indications of such elongated shapes were obtained in the studies of the gdr decay from @xmath6sc@xmath7 @xcite and @xmath3ti@xmath7 @xcite compound nuclei . in the case of @xmath6sc@xmath7 ,
the gdr measurement was inclusive , while for @xmath3ti@xmath7 it was more exclusive , namely it was associated with various @xmath0-multiplicity values .
however , in both studies other explanations , as for example contribution from fission could not be completely ruled out . in order to address in more detail
the problem of the jacobi shape transition focusing on the gdr @xmath0-decay , a new highly selective experiment for @xmath3ti@xmath7 was performed . to obtain high energy @xmath0-ray spectra associated with fusion - evaporation reaction and free from contribution of fission products and/or other non - fusion reactions , the restrictive conditions of gating on well - known discrete lines of the evaporation residual nuclei
were applied . in this paper
we report on some new results concerning the measurement of the high energy @xmath0-rays with very stringent conditions selecting the highest part of the spin distribution .
the following section describes the experiment and the data analysis .
the discussion of the results presented here focuses for the first time on the importance of the coriolis splitting of the gdr components at the high rotational frequencies that can be reached only in light nuclei .
the experiment was performed at the vivitron accelerator of the ires laboratory , strasbourg ( france ) , using the euroball iv ge - array coupled to the hector array @xcite and the charged particle detector euclides .
the @xmath3ti compound nucleus was populated in the @xmath1o+@xmath2si reaction at 105 mev bombarding energy .
the excitation energy of the @xmath3ti nuclei was 86 mev and the maximum angular momentum @xmath8 35 @xmath4 .
for this experiment , the euroball consisted of 26 germanium clover and 15 cluster detectors ( all with the bgo anti - compton shields ) , and 75% of the inner - ball ( 83 bgo crystals ) which together with the germanium detectors resulted in 65% efficiency for the multiplicity determination .
the 8 large volume baf@xmath9 detectors of the hector were placed in the forward hemisphere , together with 4 small baf@xmath9 detectors which provided a good time reference signal .
the euclides consisted of 40 silicon telescope detectors covering approximately 90% of the solid angle .
only events having at least 2 compton suppressed ge signals and one high - energy @xmath0-ray were accepted .
a total number of 10@xmath10 events were collected , in which the @xmath0-ray energy in any baf@xmath9 detector was larger then 4 mev .
-ray spectra measured with the hector baf@xmath9-detectors : one is gated by the innerball folds 1120 ( histogram ) and the other is additionally gated by the discrete transitions in @xmath11ca measured in the euroball ge - detectors ( filled squares ) .
the inset shows the ratio of these two spectra . ]
figure [ fig:1 ] shows two measured high - energy @xmath0-ray spectra .
the first , displayed by the histogram , corresponds to the selection of @xmath0-fold in the interval 1120 , as measured in the innerball .
the second spectrum , showed with filled squares , was obtained with the additional simultaneos selection of clean and well resolved low energy @xmath0-ray transitions of @xmath11ca in the ge - detectors of the euroball array .
this particular residual nucleus was expected to be populated only in the decay from the highest spin region of the compound nucleus and , therefore , by gating on it we obtained data corresponding to the highest spins and free from fission contaminations .
the two spectra , normalized at @xmath12 6 mev , show differences in the two regions : around @xmath12 9 mev and around @xmath12 17 mev , where evident yield excesses were found in the germanium gated spectrum .
these excesses are better evidenced in the inset to figure [ fig:1 ] , which shows the ratio of the two measured spectra .
the statistical model analysis was made for the high energy @xmath0-spectrum gated by the discrete transitions in @xmath11ca .
this spectrum , being associated with a selected evaporation channel , was analyzed using the cascade code based on the monte - carlo technique .
this , in fact , allows to select only those decay chains which lead to the population of a specific residual nucleus . the used monte carlo version of the cascade code , previously employed for the analysis of the heavier mass data @xcite ,
was implemented for this mass region by choosing the most appropriate values of the statistical model parameters ( e.g. level density was assumed to be in accordance with reisdorf prescription @xcite , the yrast line was taken from the experiment and extrapolated by the liquid drop values ) . in the left panel of the figure [ fig:2 ] , the best fit to the experimental data
is shown with the full line together with the data ( filled squares ) .
it corresponds to a gdr strength function consisting of 3 components at @xmath13 10.8 , 18 and 26 mev exhausting approximately the entire ewsr strength .
-ray spectrum gated by the @xmath11ca transitions and by high fold region ( already shown in fig .
[ fig:1 ] ) , in comparison with the best fitting statistical model calculations ( full drawn line ) assuming a 3-lorentzian gdr lineshape with energies : @xmath13 10.8(1 ) , 18(1 ) and 26(2 ) mev ; widths : @xmath14 4.0(5 ) , 10(1 ) and 16(3 ) mev ; and strengths : @xmath15 0.35(5 ) , 0.38(10 ) and 0.7(2 ) .
right panel : the deduced experimental gdr strength function ( full drawn line , see text for explanation ) together with best fitting 3-lorentzian function and its individual components . ] in particular , the need of a low energy component at @xmath16 10 mev is consistent with the fact that this spectrum has an yield excess as compared to a more inclusive spectrum ( see the inset to figure [ fig:1 ] ) .
the quality of the fit can be judged more clearly by inspecting the right panel of fig .
[ fig:2 ] .
this figure shows the gdr strength function data , represented by the quantity @xmath17 . in that expression @xmath18 and @xmath19
are the experimental and calculated spectra , respectively , shown in the left panel of fig .
[ fig:2 ] . the best fit gdr strength function , @xmath20 , consists of 3-component lorentzian function with parameters ( energy , width and strength ) taken from the best fit to the experimental spectrum .
@xmath20 and its individual components are also plotted in the figure . it should be noted , that the best fit line @xmath20 corresponds to the effective nuclear shape probed by the gdr oscillations .
the most striking feature of our results is the presence of the excess of the @xmath0-yield at @xmath16 10 mev , identified for the first time in this mass region .
a possible explanation can be related to a mechanism such as the coriolis splitting of the gdr lineshape controlled by the rotational frequency of the rotating nucleus .
in fact , only in those light nuclei one can reach the highest values of the rotational frequency ( 23 mev in the spin interval 2030 @xmath4 ) and in this particular experiment we are able to select them much better than in the previous rather inclusive works . therefore , we have investigated in detail the effect of the coriolis splitting in the gdr lineshape as a function of the spin ( rotational frequency ) and the quadrupole deformation type ( @xmath0 ) and the size ( @xmath21 ) .
figure [ fig:3 ] shows the calculated gdr lineshapes for @xmath3ti to emphasize how they evolve as the functions of @xmath36 and @xmath21 for the cases of prolate , oblate and triaxial shapes .
one can clearly note that the coriolis splitting is not affecting the oblate shapes ( @xmath37 60@xmath38 ) .
in contrast , it produces strong effects , increasing as a function of spin , in the case of triaxial ( @xmath37 30@xmath38 ) and especially prolate shapes ( @xmath37 0@xmath38 ) .
in particular , it produces a rather strong and narow peak at @xmath16 10 mev for prolate and triaxial deformations with @xmath39 0.4 which is similar to that we have observed in our experimental data . in order to interprete
the obtained gdr strength function data in the entire measured @xmath0-ray energy region ( 530 mev ) , we use the same approach as has been adopted in almost all experiments concerning the gdr in hot and rotating nuclei @xcite .
this is based on the termal shape fluctuation model which predicts the gdr strength function corresponding to the shape ensemble probed by this collective excitation mode . in the present case
we have obtained the shape ensemble distributions ( proportional to the the boltzmann factor @xmath40 , where @xmath41 is the free energy and @xmath42 is the temperature ) for the relevant total excitation energy ( 85 mev ) and for different spins using the new parametrization of the liquid drop model lsd @xcite .
those calculations show that the equilibrium shape of @xmath3ti undergoes the jacobi transition from an oblate to a triaxial and a prolate shape for @xmath43 28 @xmath4 . for the comparison with the experimental data at the highest spins populated in the reaction , we select the calulations for the spin interval 2834 @xmath4 ( this corresponds to an average temperature @xmath44 2 mev and rotational frequency @xmath45 2.8 mev ) .
the calculations neglect shell effects , which are not expected to be important in this temperature region .
the resulting gdr strength function , obtained as a weighted average ( with the weight given by the boltzmann factor ) of strength functions calculated at each @xmath46 point ( including also coriolis effects ) , is displayed with the full line in fig .
[ fig:4 ] and compared with the experimental data .
the remarkable good agreement between the theoretical predictions and the present experimental results is very much in favour of the presence of the jacobi transition .
more evidence for this transition is also given by the fact that the peak at @xmath16 10 mev can originate only in the presence of a fast rotation in nuclei of prolate or triaxial shapes ( see discussion at the beginning of the section ) . in fact , for oblate shapes , typical for the equilibrium deformation at rotational frequencies lower than the critical value for the jacobi transition , the coriolis splitting is always absent .
@xmath16 2 mev and in the spin region 2834 @xmath4 ) of the gdr lineshape in @xmath3ti obtained from the thermal shape fluctuation model based on free energies from the lsd model calculations @xcite .
the filled squares are the experimental data shown also in the right panel of figure [ fig:2 ] . ]
a further signature for the jacobi transition is the presence of two other broad components in the strength function at higher energies .
one component , at @xmath16 17 mev , is clearly seen in the present data , while the other which is very broad ( at 2030 mev ) can only barely be identified because of the very limited statistics at these high energies in the present very exclusive spectrum . in summary
, the present work on the gdr @xmath0-decay in the hot rotating nucleus , @xmath3ti , shows evidence for the expected jacobi shape transition .
it is based on the observation of two particular features in the measured gdr strength function .
the first , already found in the previous works @xcite , is the presence of a high energy component related to large deformations .
the second feature , identified for the first time in the present experiment , is the aperance of a gdr component at @xmath16 10 mev ( in region where the statistics is very high ) , which is interpreted as due to the coriolis splitting of the lowest vibrational frequency ( which corresponds to the dipole vibration along the long axis of the well deformed prolate or triaxial shape ) and consequently shifting down a part of the strength .
this shows the importance of the selection of high rotational frequencies and of a good tagging on fusion - evaporation events in the investigations of nuclear shapes through the gdr @xmath0-decay
. 9 w.d .
myers , w.j .
wiatecki , acta .
phys . pol .
* b32 * ( 2001 ) 1033 k. pomorski and j. dudek , phys . rev .
* c67 * ( 2003 ) 044316 ; nucl - th/0205011 d. ward et al . ,
phys . rev .
* c66 * ( 2002 ) 024317 - 1 m. kiciska - habior et al .
, phys . lett . *
b308 * ( 1993 ) 225 a. maj et al .
, nucl . phys .
* a687 * ( 2001 ) 192 a. maj et al .
a571 * ( 1994 ) 185 f. camera et al .
, phys . rev .
* c60 * ( 1999 ) 014306 m. kmiecik et al .
j. * a12 * ( 2001 ) 5 m. kiciska - habior et al .
, phys . rev .
* c36 * ( 1987 ) 612 k. neergrd , phys . lett .
* 110b * ( 1982 ) 7 m. gallardo et al . , nucl .
* a443 * ( 1985 ) 415 j.j .
gaardhje et al . , acta phys .
* b24 * ( 1993 ) 139 ; t. dssing , private communication p.f .
bortignon , a. bracco , r.a .
broglia , giant resonances : nuclear structure at finite temperature , gordon breach , new york , 1998 j.j .
gaardhje , annu .
nucl . part .
* 42 * ( 1992 ) 483 k.a .
snover , annu .
nucl . part .
* 36 * ( 1986 ) 545 | the @xmath0-rays from the decay of the gdr in the compound nucleus reaction @xmath1o+@xmath2si at bombarding enery of 105 mev have been measured in an experiment using a setup consisting of the combined euroball iv , hector and euclides arrays .
the shape of the rotating compound nucleus , @xmath3ti , is expected to undergo the jacobi transition around spin 28 @xmath4 .
a comparison of the gdr lineshape data with the predictions of the thermal shape fluctuation model , based on the most recent rotating liquid drop lsd calculations , shows evidence for such jacobi shape transition .
in addition to the previously found broad structure in the gdr lineshape region at 1825 mev caused by large deformations , the presence of a low energy component ( around 10 mev ) , due to the coriolis splitting in prolate , well deformed shape has been identified for the first time . |
in @xcite , pierce and tian studied the self - similar solutions of the whitham equations which describe the zero dispersion limits of the kdv hierarchy . the main feature of the whitham equations for the higher members of the hierarchy , of which the kdv equation is the first member , is that these whitham equations are neither strictly hyperbolic nor genuinely nonlinear .
this is in sharp contrast to the case of the kdv equation whose whitham equations are strictly hyperbolic and genuinely nonlinear [ lax2 ] . in this paper , we extend their studies to the case of the complex modified kdv equation , which is the second member of the defocusing nonlinear schrdinger hierarchy .
the whitham equations for the defocusing nls equation are strictly hyperbolic and genuinely nonlinear , and they have been studied extensively ( see for examples , @xcite ) . however , for the mkdv equation , the whitham equations are neither strictly hyperbolic nor genuinely nonlinear .
let us begin with a brief description of the zero dispersion limit of the solution of the nls equation @xmath0 with the initial data @xmath1 here @xmath2 and @xmath3 are real functions that are independent of @xmath4 .
writing the solution @xmath5 , and using the notation @xmath6 , @xmath7 , one obtains the conservation form of the defocusing nls equation @xmath8 the mass density @xmath9 and momentum density @xmath10 have weak limits as @xmath11 @xcite .
these limits satisfy a @xmath12 system of hyperbolic equations @xmath13 until its solution develops a shock . system ( [ nlshydro0 ] ) can be rewritten in the diagonal form for @xmath14 , @xmath15 where the riemann invariants @xmath16 and @xmath17 are given by @xmath18 as a simple example , we consider the case with @xmath19constant . system ( [ nls0 ] ) reduces to a single equation @xmath20 the solution is given by the implicit form @xmath21 where @xmath22 is the initial data for @xmath23 .
one can easily see that if @xmath24 decreases in some region , then @xmath25 develops a shock in a finite time , i.e. , @xmath26 becomes singular .
after the shock formation in the solution of ( [ nlshydro0 ] ) or ( [ nls0 ] ) , the weak limits are described by the nls - whitham equations , which can also be put in the riemann invariant form @xcite @xmath27 where @xmath28 are expressed in terms of complete hyperelliptic integrals of genus @xmath29 @xcite . here
the number @xmath29 is exactly the number of phases in the nls oscillations with small dispersion .
accordingly , the zero phase @xmath30 corresponds to no oscillations , and single and higher phases @xmath31 correspond to the nls oscillations .
system ( [ nls0 ] ) is viewed as the zero phase whitham equations .
the solution of the whitham equations ( [ nlsw ] ) for @xmath31 then describes the averaged motion of the oscillations appearing in the solution of ( [ nls ] ) ( see e.g. @xcite ) .
let us discuss the most important @xmath32 case in more detail .
we note that it is well known that the kdv oscillatory solution , in the single phase regime , can be approximately described by the kdv periodic solution when the dispersive parameter is small @xcite .
it is very possible to use the method of @xcite to show that the solution of the nls equation ( [ nls ] ) for small @xmath4 can be approximately described , in the single phase regime , by the periodic solution of the nls equation .
the nls periodic solution has the form @xmath33 with @xmath34 and the velocity @xmath35 . here
s are determined by the equation obtained from ( [ nlshydro ] ) @xmath37 with @xmath38 , and @xmath39 is the jacobi elliptic function with the modulus @xmath40 .
we can also write @xmath36 s as @xcite @xmath41 with @xmath42 .
the velocity @xmath43 is then given by @xmath44 for constants @xmath45 , @xmath46 , @xmath47 and @xmath48 , formula ( [ periodicsolution ] ) gives the well known elliptic solution of the nls equation .
to describe the solution @xmath49 of the nls equation ( [ nlshydro ] ) , the quantities @xmath45 , @xmath46 , @xmath47 and @xmath48 are instead functions of @xmath50 and @xmath51 and they evolve according to the single phase whitham equations ( [ nlsw ] ) for @xmath32 . the weak limit of @xmath49 of nls equation ( [ nls ] ) as @xmath11 can be expressed in terms of @xmath52 , @xmath53 , @xmath54 and @xmath55 @xcite @xmath56 where @xmath57 and @xmath58 are the complete elliptic integrals of the first and second kind , respectively .
this weak limit can also be viewed as the average value of the periodic solution @xmath59 of ( [ periodicsolution ] ) over its period @xmath60 . in order to see how a single phase whitham solution appears
, we consider the following step initial data for system ( [ nls0 ] ) @xmath61 where @xmath62 , @xmath63 , @xmath64 .
the solution of ( [ nls0 ] ) develops a shock if and only if @xmath65 ( cf .
( [ nlsbeta ] ) ) . after the formation of a shock , the whitham equations ( [ nlsw ] ) with @xmath32 kick in .
for instance , we consider the whitham equations with the initial data @xcite @xmath66 now notice that the whitham equations ( [ nlsw ] ) for @xmath32 with the initial data ( [ regular ] ) can be reduced to a single equation @xmath67 .
the equation has a global self - similar solution , which is implicitly given by @xmath68 .
the @xmath50-@xmath51 plane is then divided into _ three _ parts @xmath69 where @xmath70 ( see ( [ bc3 ] ) and ( [ bc4 ] ) below for the derivation ) .
the solution of system ( [ nls0 ] ) occupies the first and third parts , i.e. , * for @xmath71 , @xmath72 * for @xmath73 , @xmath74 the whitham solution of ( [ nlsw ] ) with @xmath32 lives in the second part , i.e. , * for @xmath75 , @xmath76 where the solution @xmath47 can be obtained as a function of the self - similarity variable @xmath77 , if @xmath78 indeed , it has been shown that the whitham equations ( [ nlsw ] ) are genuinely nonlinear @xcite , i.e. , @xmath79 for @xmath80 . in figure
[ fig0 ] , we plot the self - similar solution of the whitham equations ( [ nlsw ] ) with @xmath32 for the nls equation , and the corresponding periodic oscillatory solution ( [ periodicsolution ] ) for the initial data ( [ nlsinitial ] ) with @xmath81 and @xmath82 .
the oscillations describe a dispersive shock of the nls equation under a small dispersion .
note here that the oscillations have a uniform structure , which is due to an almost linear profile of the whitham solution @xmath47 .
this will be seen to be in sharp contrast to the case of the mkdv equation , which we will discuss later ( cf .
figure [ fig1 ] ) .
[ fig0 ] ) of @xmath32 and the corresponding oscillatory solution ( [ periodicsolution ] ) of the nls equation with @xmath83 .
the dark line in the middle of the oscillations is the weak limit @xmath84 given by ( [ average ] ) .
the initial data is given by ( [ nlsinitial ] ) with @xmath85 , @xmath86 and @xmath82.,width=453 ] the figures in this paper all have the same form : on the left hand side is a plot of the solution of the whitham equations as a function of the self - similarity variable @xmath77 , which is exact , other than a numerical method used to implement the inverse function theorem .
on the right hand side is the oscillatory solution given by ( [ periodicsolution ] ) ( respectively ( [ periodicsolution2 ] ) for mkdv ) at @xmath87 , while the dark plot is the weak limit ( [ average ] ) of the oscillatory solution at @xmath87 , both plots on the right are also exact . in the first two figures we demark the region where the whitham equations with @xmath32 govern the solution , place a dashed - dotted line where the behavior of the oscillatory solution changes , and label the four functions @xmath80
the demarcation and labeling are similar in the other figures and we will leave them off for brevity . although we do not include any numerical simulations , we would like to mention that e. overman showed us his numerical simulation of the nls and mkdv equations which captures the features of the oscillatory solutions plotted in this paper .
the defocusing nls equation is just the first member of the defocusing nls hierarchy ; the second is the ( defocusing ) complex modified kdv ( mkdv ) equation @xmath88 we again use @xmath5 and notation @xmath89 , @xmath90 to obtain the conservation form of the mkdv equation @xmath91 } \ , \end{array } \right.\ ] ] where @xmath92 the mass density @xmath93 and momentum density @xmath94 for the mkdv equation also have weak limits as @xmath11 @xcite .
as in the nls case , the weak limits satisfy @xmath95 until the solution of ( [ mkdvhydro0 ] ) forms a shock .
one can rewrite equations ( [ mkdvhydro0 ] ) as @xmath96 where the riemann invariants @xmath97 and @xmath23 are again given by formula ( [ nsfii ] ) .
let us again consider the simplest case @xmath98 , where a is constant , to see how the solution of system ( [ mkdv0 ] ) develops a shock . in this case
, the system reduces to a single equation @xmath99 as in the nls case , we consider the initial function given by @xmath100 for @xmath101 and @xmath102 for @xmath103 .
we recall that , in the nls case , the zero phase solution of ( [ nls0 ] ) develops a shock if and only if @xmath65 .
however , the solution in the mkdv case develops a shock for @xmath104 if and only if @xmath105 .
in addition , if @xmath106 , the solution in the mkdv case develop a shock if and only if @xmath107 .
these differences between the mkdv and nls cases are due to the weak hyperbolicity of the system ( [ mkdv0 ] ) ( note that for the eigenspeed @xmath108 for @xmath23 , we have @xmath109 which can change sign ) . as will be shown below , this leads to an additional structure in the dispersive shock for the mkdv case . as in the case of the nls equation ,
immediately after the shock formation in the solution of ( [ mkdvhydro0 ] ) , the weak limits are described by the mkdv - whitham equations @xmath110 where @xmath111 s can also be expressed in terms of complete hyperelliptic integrals of genus @xmath29 @xcite . in this paper
, we study the solution of the whitham equations ( [ mkdvw ] ) with @xmath32 when the initial mass density @xmath112 and momentum density @xmath113 are step functions . in view of ( [ nsfii ] ) , this amounts to requiring @xmath97 and @xmath23 of system ( [ mkdv0 ] ) to have step - like initial data .
we are interested in the following two cases : * @xmath114 is a constant and @xmath115 * @xmath116 is a constant and @xmath117 in the case of the nls equation , the genuine nonlinearity of the single phase whitham equations ( see ( [ eq11 ] ) ) warrants that the solution is found by the implicit function theorem .
however the mkdv - whitham equations ( [ mkdvw ] ) , in general , are not genuinely nonlinear , that is , a property like ( [ eq11 ] ) is not available ( see lemma [ lem2.1 ] below ) .
our construction of solutions of the whitham equation ( [ mkdvw ] ) with @xmath32 makes use of the non - strict hyperbolicity of the equations . for the nls case ,
it has been known in @xcite that the whitham equations ( [ nlsw ] ) with @xmath32 are strictly hyperbolic , that is , @xmath118 for @xmath80 . for
the mkdv - whitham equations ( [ mkdvw ] ) with @xmath32 , the eigenspeeds @xmath119 may coalesce in the region @xmath80 ( we will discuss the details in section [ whithamsec ] ) .
let us now describe one of our main results ( see theorem [ thm1 ] ) for the single phase mkdv - whitham equations with step - like initial function ( [ step1 ] ) for @xmath85 , @xmath120 and @xmath121 .
in this case , the space time is divided into _ four _ regions ( see figure [ fig1 ] ) instead of _ three _ in the case of the nls equation ( cf .
figure [ fig0 ] ) @xmath122 where @xmath123 , @xmath124 and @xmath125 are some constants . in the first and fourth regions ,
the solution of the @xmath126 system ( [ mkdv0 ] ) governs the evolution : * for @xmath127 , @xmath128 * for @xmath129 , @xmath130 the whitham solution of the @xmath131 system ( [ mkdvw ] ) with @xmath32 lives in the second and third regions ; * for @xmath132 , @xmath133 * for @xmath134 , @xmath135 note that , in the second region , we have @xmath136 on a curve in the region @xmath137 .
this implies the non - strict hyperbolicity of the mkdv - whitham equations ( [ mkdvw ] ) for @xmath32 .
it is again possible to use the method of @xcite to show that the solution of the mkdv equation ( [ mkdv ] ) can be approximately described , in the single phase regime , by the periodic solution of the mkdv when @xmath4 is small .
the periodic solution has the same form as ( [ periodicsolution ] ) of the nls , i.e. , @xmath138 however , @xmath139 is now given by @xmath140 with the velocity @xmath141 ( see e.g. @xcite ) @xmath142 where @xmath143 and @xmath144 are the elementary symmetric functions of degree one and two , respectively .
the functions @xmath145 and @xmath54 are also given by formula ( [ rhos ] ) .
if @xmath45 , @xmath46 , @xmath47 and @xmath48 are constants , formula ( [ periodicsolution2 ] ) gives the periodic solution of the mkdv equation .
to describe the solution @xmath49 of the mkdv equation ( [ mkdv ] ) , the quantities @xmath45 , @xmath46 , @xmath47 and @xmath48 must satisfy the single phase mkdv - whitham equations ( [ mkdvw ] ) for @xmath32 .
the weak limit of @xmath49 of the mkdv equation is also given by formula ( [ average ] ) .
in figure [ fig1 ] , we plot the self - similar solution of the whitham equations ( [ mkdvw ] ) for @xmath32 and the corresponding periodic oscillatory solution ( [ periodicsolution2 ] ) .
we note here that the pattern of the oscillation in this case has two distinct structures : one corresponds to the region ( 2 ) , @xmath132 , and the other corresponds to the region ( 3 ) , @xmath134 , we also note that the weak limit @xmath84 is not @xmath146 smooth at the boundary point @xmath147
. self - similar solution of the mkdv - whitham equation ( [ mkdvw ] ) with @xmath32 and the corresponding oscillatory solution ( [ periodicsolution2 ] ) of the mkdv equation with @xmath148 .
the initial data is given by ( [ step1 ] ) with @xmath85 , @xmath86 and @xmath82 .
there are two distinct structures in the oscillations and they are separated by @xmath149 ( cf .
figure [ fig0]).,width=453 ] as we will show below , for other values of @xmath150 , @xmath151 and @xmath152 , the solutions of ( [ mkdv0 ] ) and ( [ mkdvw ] ) with @xmath32 will be seen to be quite different from the above .
the whitham equations ( [ mkdvw ] ) for the mkdv equation are analogous to the whitham equations for the higher members of the kdv hierarchy @xcite . for step like initial data , the single phase whitham solutions for the higher order kdv
are also constructed using the non - strict hyperbolicity of the equations . in the case of strictly hyperbolic whitham equations for the kdv ,
the oscillations ( dispersive shock ) have uniform structure . however , in the case of non - strict hyperbolic whitham equations for the higher order kdv , an additional structure has been found in the dispersive shocks .
this new structure is similar to the one found here in the dispersive shocks of the mkdv equation .
the organization of the paper is as follows . in section 2
, we will study the eigenspeeds , @xmath153 and @xmath154 of the whitham equations ( [ mkdvw ] ) for @xmath32 . in section 3
, we will construct the self - similar solutions of the single phase whitham equations for the initial function ( [ step1 ] ) with @xmath155 . in section 4
, we will construct the self - similar solution of the whitham equations for the initial function ( [ step1 ] ) with @xmath156 . in section 5
, we will briefly discuss how to handle the other step - like initial data ( [ step2 ] ) .
in this section we define the eigenspeeds @xmath28 s and @xmath111 s of the whitham equations ( [ nlsw ] ) and ( [ mkdvw ] ) with @xmath32 for the nls and the mkdv equations . for simplicity
, we suppress the subscript @xmath32 in the notation @xmath28 s and @xmath111 s in the rest of the paper .
we first introduce the polynomials of @xmath157 for @xmath158 @xcite : @xmath159 where the coefficients , @xmath160 are uniquely determined by the two conditions @xmath161 and @xmath162 the coefficients of @xmath163 can be expressed in terms of complete elliptic integrals .
the eigenspeeds of the whitham equations ( [ nlsw ] ) with @xmath32 for the nls equation are defined in terms of @xmath164 and @xmath165 of ( [ eq15 ] ) @xcite , @xmath166 which give @xmath167 here @xmath168 , and @xmath169 is given by a complete elliptic integral @xcite @xmath170 the function @xmath171 can be rewritten as a contour integral .
hence , @xmath172 since the integrand satisfies the same equations for each @xmath173 .
this contour integral connection also allows us to give another formulation of @xmath171 @xmath174 it follows from ( [ lambda ] ) , ( [ i ] ) and ( [ i2 ] ) that @xmath175 for @xmath176 .
this implies the strict hyperbolicity of the nls - whitham equation ( [ nlsw ] ) for @xmath32 .
the eigenspeeds @xmath177 s have the following values @xcite : at @xmath178 , we have @xmath179 and at @xmath180 , @xmath181 notice that the eigenspeed @xmath182 at @xmath180 is the same as the velocity of the periodic solution ( [ periodicsolution ] ) , i.e. @xmath183 . the eigenspeeds of the mkdv - whitham equations ( [ mkdvw ] ) with @xmath32 are @xcite @xmath184 they can be expressed in terms of @xmath185 , @xmath186 , @xmath187 and @xmath188 of the nls - whitham equations ( [ nlsw ] ) with @xmath32 .
[ lem2.1 ] the eigenspeeds @xmath189 s of ( [ eq18a ] ) can be expressed in the form @xmath190 where @xmath143 and @xmath191 is the solution of the boundary value problem of the euler - poisson - darboux equations @xmath192 also the @xmath193 s satisfy the over - determined systems @xmath194 we omit the proof since it is very similar to the proof of an analogous result for the kdv hierarchy @xcite .
the boundary value problem ( [ eq20 ] ) has a unique solution .
the solution is a symmetric quadratic function of @xmath45 , @xmath46 , @xmath47 and @xmath48 @xmath195 where @xmath196 is the elementary symmetric polynomial of degree two .
notice that @xmath197 gives the velocity of the periodic solution ( [ periodicsolution2 ] ) for the mkdv equation , i.e. @xmath198 .
for nls , @xmath177 s satisfy @xcite @xmath199 for @xmath200 .
similar results also hold for the mkdv - whitham equations ( [ mkdvw ] ) with @xmath32 .
@xmath201 for @xmath176 .
we use ( [ eq19 ] ) and ( [ 34 ] ) to obtain @xmath202 and @xmath203 where we have used equation ( [ eq20 ] ) @xmath204 it follows from formula ( [ q ] ) for @xmath197 that @xmath205 which , along with with ( [ eq20a ] ) and ( [ eq20b ] ) , proves ( [ ine1 ] ) .
inequality ( [ ine2 ] ) is proved in the same way .
the following calculations are useful in the subsequent sections .
using formula ( [ eq19 ] ) for @xmath206 and @xmath207 and formulae ( [ lambda ] ) for @xmath187 and @xmath188 , we obtain @xmath208 { \nonumber}\\ & = & { i \over ( { \partial}_{u_3 } i ) ( { \partial}_{u_4 } i ) } \left[{{\partial}q \over { \partial}u_4 } ( { { \partial}i \over { \partial}u_3 } - { { \partial}i \over { \partial}u_4 } ) - ( { { \partial}q \over { \partial}u_3 } - { { \partial}q \over { \partial}u_4 } ) { { \partial}i \over { \partial}u_4 } \right ] { \nonumber}\\ & = & { 2 i ( u_3 - u_4 ) \over ( { \partial}_{u_3 } i ) ( { \partial}_{u_4 } i ) } \ m \ , \label{m}\end{aligned}\ ] ] where @xmath209 here we have used equations ( [ epd ] ) for @xmath171 and equations ( [ eq20 ] ) for @xmath197 in equality ( [ m ] ) . since @xmath197 of ( [ q ] ) is quadratic , we obtain @xmath210 we note that another expression for @xmath211 is @xmath212 hence , we get @xmath213 we next evaluate @xmath214 when @xmath215 . using the integral formula ( [ i ] ) for the function i and applying the change of variable @xmath216 , we obtain @xmath217 the two integrals can be evaluated exactly as @xmath218 for @xmath219 .
we finally get @xmath220^{5 \over 2 } } \ , \ ] ] where @xmath221 ( u_1 + u_2 + 4 \xi ) { \nonumber}\\ & & \quad - \ 8 ( u_2 - \xi ) [ 2(u_2 - \xi)^2 + 3 ( u_2 - \xi)(u_1 - u_2 ) { \nonumber}\\ & & \quad\quad\quad + ( u_1 - u_2)^2 ] \label{u } \ .
\end{aligned}\ ] ] similar to ( [ m ] ) for @xmath206 and @xmath207 , we have @xmath222 where @xmath223 since
@xmath197 of ( [ q ] ) is quadratic , we obtain @xmath224 finally , we use ( [ bc3 ] ) and ( [ eq19 ] ) to calculate @xmath225 { { \partial}q \over { \partial}u_2 } - { 1 \over 2 } [ \lambda_3 - 2(u_1 + u_2 + 2u_4 ) ] { { \partial}q \over { \partial}u_3 } { \nonumber}\\ & = & { ( u_2 - u_4 ) \over 2(u_1 + u_2 - 2u_4 ) } \ v(u_1 , u_2 , u_4 ) \label{v } \ , \end{aligned}\ ] ] where @xmath226
in this section , we construct self - similar solutions of the whitham equations ( [ mkdvw ] ) with @xmath32 for the initial function ( [ step1 ] ) with @xmath155 .
the case with @xmath156 will be studied in next section .
the solution of the zero phase whitham equations ( [ mkdv0 ] ) does not develop a shock when @xmath227 .
we are therefore only interested in the case @xmath228 .
we first study the @xmath157-zero of the cubic polynomial equation @xmath229 where @xmath230 is given by ( [ u ] ) .
it is easy to prove that for each pair of @xmath150 and @xmath151 satisfying @xmath62 and @xmath231 , @xmath232 has only one simple real root .
denoting this zero by @xmath233 , we then deduce that @xmath234 is positive for @xmath235 and negative for @xmath236 . since @xmath237 in view of ( [ u ] ) , we must have @xmath238 for initial function ( [ step1 ] ) with @xmath155 and @xmath231 , we now classify the resulting whitham solutions into four types : * @xmath239 with any @xmath240 * @xmath241 with @xmath242 * @xmath243 with @xmath244 * @xmath241 with @xmath245 we will study the second type ( ii ) first . here
we consider the step initial function ( [ step1 ] ) satisfying @xmath246 and @xmath247 .
( see figure [ fig1 ] . )
[ thm1 ] for the step - like initial data ( [ step1 ] ) with @xmath155 , @xmath248 and @xmath246 , the solution @xmath249 of the zero phase whitham equations ( [ mkdv0 ] ) and the solution @xmath250 of the single phase whitham equations ( [ mkdvw ] ) with @xmath32 are given as follows : * for @xmath251 , @xmath252 * for @xmath253 , @xmath254 where @xmath255 is the unique solution @xmath47 of @xmath256 in the interval @xmath257 . * for @xmath258 , @xmath259 * for @xmath260 , @xmath261
the boundaries @xmath262 and @xmath263 are called the trailing and leading edges , respectively .
they separate the solutions of the single phase whitham equations ( [ mkdvw ] ) with @xmath32 and the zero phase whitham equations ( [ mkdv0 ] ) .
the single phase whitham solution matches the zero phase whitham solution in the following fashion ( see figure 1 . ) : @xmath264 at the trailing edge ; @xmath265 at the leading edge .
the proof of theorem 3.1 is based on a series of lemmas : we first show that the solutions defined by formulae ( [ ws1 ] ) and ( [ ws2 ] ) indeed satisfy the whitham equations ( [ mkdvw ] ) for @xmath32 @xcite . * the functions @xmath45 , @xmath46 , @xmath47 and @xmath48 determined by equations ( [ ws1 ] )
give a solution of the whitham equations ( [ mkdvw ] ) with @xmath32 as long as @xmath47 and @xmath48 can be solved from ( [ ws1 ] ) as functions of @xmath50 and @xmath51 .
* the functions @xmath45 , @xmath46 , @xmath47 and @xmath48 determined by equations ( [ ws2 ] ) give a solution of the whitham equations ( [ mkdvw ] ) with @xmath32 as long as @xmath47 can be solved from ( [ ws2 ] ) as a function of @xmath50 and @xmath51 .
\(1 ) @xmath45 and @xmath46 obviously satisfy the first two equations of ( [ mkdvw ] ) for @xmath32 . to verify
the third and fourth equations , we observe that @xmath266 on the solution of ( [ ws1 ] ) . to see this , we use ( [ a23 ] ) to calculate @xmath267 the second part of ( [ dia ] ) can be shown in the same way .
we then calculate the partial derivatives of the third equation of ( [ ws1 ] ) with respect to @xmath50 and @xmath51 @xmath268 which give the third equation of ( [ mkdvw ] ) with @xmath32 .
the fourth equation of ( [ mkdvw ] ) with @xmath32 can be verified in the same way .
\(2 ) the second part of lemma 3.2 can easily be proved .
we now determine the trailing edge .
eliminating @xmath50 and @xmath51 from the last two equations of ( [ ws1 ] ) yields @xmath269 since it degenerates at @xmath215 , we replace ( [ m34 ] ) by @xmath270 therefore , at the trailing edge where @xmath178 , equation ( [ f1 ] ) , in view of formulae ( [ m ] ) and ( [ m44 ] ) , reduces to @xmath271 noting that @xmath272 is the unique solution of ( [ cubic ] ) , we then deduce that @xmath273 .
equation ( [ f1 ] ) has a unique solution satisfying @xmath178 .
the solution is @xmath274 .
the rest of equations ( [ ws1 ] ) at the trailing edge are @xmath275 , @xmath276 and @xmath262 .
having located the trailing edge , we now solve equations ( [ ws1 ] ) in the neighborhood of the trailing edge .
we first consider equation ( [ f1 ] ) .
we use ( [ m ] ) to write @xmath277 of ( [ f1 ] ) as @xmath278 we note that at the trailing edge @xmath274 , we have @xmath279 because of ( [ m44 ] ) and ( [ u2 ] ) .
we then use ( [ pm ] ) and ( [ pm ] ) to differentiate @xmath277 at the trailing edge @xmath280 \ { { \partial}^3 i \over { \partial}u_3 ^ 2 { \partial}u_4 } > 0 \ , \end{aligned}\ ] ] where we have used the expression ( [ q ] ) for @xmath197 in the last equation and ( [ xx ] ) in the inequality .
these show that equation ( [ f1 ] ) or equivalently ( [ m34 ] ) can be inverted to give @xmath48 as a decreasing function of @xmath47 @xmath281 in a neighborhood of @xmath274 .
we now extend the solution @xmath282 of equation ( [ m34 ] ) in the region @xmath283 as far as possible .
we first claim that @xmath284 on the extension . to see this
, we first observe that inequalities ( [ p>0 ] ) are true at the trailing edge @xmath285 .
this follows from ( [ q ] ) and ( [ xx ] ) .
therefore , inequalities ( [ p>0 ] ) hold in a neighborhood of the trailing edge .
to prove that ( [ p>0 ] ) remains true on the extension , we use formula ( [ eq19 ] ) for @xmath206 and @xmath207 to rewrite equation ( [ m34 ] ) as @xmath286 { { \partial}q \over { \partial}u_3 } = { 1 \over 2 } [ \lambda_4 - 2(a + b + u_3 + u_4 ) ] { { \partial}q \over { \partial}u_4 } \ .\ ] ] since the two terms in the two parentheses are both negative in view of ( [ strict ] ) and since @xmath287 in view of ( [ q ] ) , neither @xmath288 nor @xmath289 can vanish on the extension .
this proves inequalities ( [ p>0 ] ) .
we deduce from lemma 2.2 that @xmath290 on the solution of ( [ m34 ] ) .
because of ( [ dia ] ) and ( [ dia2 ] ) , solution ( [ a ] ) of equation ( [ m34 ] ) can be extended as long as @xmath283 .
there are two possibilities : ( 1 ) @xmath47 touches @xmath151 before or simultaneously as @xmath48 reaches @xmath152 and ( 2 ) @xmath48 touches @xmath152 before @xmath47 reaches @xmath151 .
it follows from ( [ bc4 ] ) , ( [ eq19 ] ) and ( [ q ] ) that @xmath291 where we have used @xmath292 in the inequality .
this shows that ( 1 ) is unattainable .
hence , @xmath48 will touch @xmath152 before @xmath47 reaches @xmath151 .
when this happens , equation ( [ m34 ] ) becomes @xmath293 [ u**l ] equation ( [ u * * ] ) has a simple zero in the interval @xmath294 , counting multiplicities .
denoting the zero by @xmath255 , then @xmath295 is positive for @xmath296 and negative for @xmath297 .
we use ( [ m ] ) and ( [ pm ] ) to prove the lemma . in both formulae , @xmath298 , @xmath299 and @xmath300
are all positive functions . by ( [ pm ] ) ,
@xmath301 we claim that @xmath302 the second inequality follows from ( [ m ] ) and ( [ u30 ] ) .
the first inequality can be deduced from formula ( [ m44 ] ) @xmath303^{5 \over 2 } } < 0 ~~ \quad \mbox{for $ c <
\xi(a , b)$ } \ .\ ] ] therefore , @xmath304 has a zero in the interval @xmath305 .
the uniqueness of the zero follows from ( [ pm1 ] ) in that @xmath304 increases or changes from decreasing to increasing as @xmath47 increases .
this zero is exactly @xmath255 and the rest of the theorem can be proved easily . having solved equation ( [ m34 ] ) for @xmath48 as a decreasing function of @xmath47 for @xmath283 , we turn to equations ( [ ws1 ] ) . because of ( [ dia ] ) and ( [ dia2 ] ) , the third equation of ( [ ws1 ] ) gives @xmath47 as an increasing function of @xmath77 , for @xmath253 .
consequently , @xmath48 is a decreasing function of @xmath77 in the same interval .
the last two equations of ( [ ws1 ] ) can be inverted to give @xmath47 and @xmath48 as increasing and decreasing functions , respectively , of the self - similarity variable @xmath77 in the interval @xmath253 , where @xmath255 is given in lemma [ u**l ] .
we now turn to equations ( [ ws2 ] ) .
we want to solve the third equation when @xmath306 or equivalently when @xmath296 . according to lemma [ u**l ] , @xmath307 for @xmath308 . in view of ( [ p>0 ] ) , @xmath309 is positive at @xmath310 and hence , it remains positive for @xmath296 . by ( [ ine1 ] )
, we have @xmath311 hence , the third equation of ( [ ws2 ] ) can be solved for @xmath47 as an increasing function of @xmath77 as long as @xmath308 . when @xmath47 reaches @xmath151 , we have @xmath263 .
we have therefore proved the following result .
the third equation of ( [ ws2 ] ) can be inverted to give @xmath47 as an increasing function of @xmath77 in the interval @xmath312 .
we are ready to conclude the proof of theorem 3.1 .
the solutions ( [ bs1 ] ) and ( [ bs2 ] ) are obvious . according to lemma 3.5 ,
the last two equations of ( [ ws1 ] ) determine @xmath47 and @xmath48 as functions of @xmath77 in the region @xmath313 . by the first part of lemma 3.2 , the resulting @xmath45 , @xmath46 , @xmath47 and @xmath48 satisfy the whitham equations ( [ mkdvw ] ) with @xmath32 .
furthermore , the boundary conditions ( [ tr1 ] ) and ( [ tr2 ] ) are satisfied at the trailing edge @xmath314 .
similarly , by lemma 3.6 , the third equation of ( [ ws2 ] ) determines @xmath47 as a function of @xmath77 in the region @xmath315 .
it then follows from the second part of lemma 3.2 that @xmath45 , @xmath46 , @xmath47 and @xmath48 of ( [ ws2 ] ) satisfy the whitham equations ( [ mkdvw ] ) for @xmath32 .
they also satisfy the boundary conditions ( [ le1 ] ) and ( [ le2 ] ) at the leading edge @xmath263 .
we have therefore completed the proof of theorem 3.1 .
a graph of the whitham solution @xmath316 is given in figure [ fig1 ] .
it is obtained by plotting the exact solutions of ( [ ws1 ] ) and ( [ ws2 ] ) .
here we consider the initial function ( [ step1 ] ) satisfying @xmath317 with @xmath318 and @xmath231 .
self - similar solution of the whitham equations ( [ mkdvw ] ) with @xmath32 and the corresponding oscillatory solution ( [ periodicsolution2 ] ) of the mkdv equation with @xmath319 .
the initial data is given by ( [ step1 ] ) with @xmath320 , @xmath321 and @xmath82 of type i.,width=453 ] we will only present our proofs briefly , since they are , more or less , similar to those in section 3.1 .
the main feature of this case is that the @xmath157-zero point does not appear in the solution @xmath47 , and the whitham equations ( [ mkdvw ] ) with @xmath32 are strictly hyperbolic on the solution .
( see figure [ fig2 ] . ) for the step - like initial data ( [ step1 ] ) with @xmath155 , @xmath231 and @xmath322 , the solution of the whitham equations ( [ mkdvw ] ) with @xmath32 is given by @xmath323 for @xmath324 . outside this interval ,
the solution of ( [ mkdv0 ] ) is given by @xmath325 and @xmath326 it suffices to show that @xmath327 is an increasing function of @xmath47 for @xmath305 . substituting ( [ q ] ) for @xmath197 into ( [ pm ] ) yields
@xmath328 \ { { \partial}^3 i \over { \partial}u_3 ^ 2 { \partial}u_4 } \geq { 1 \over 4 } [ a + b + 4 \xi(a , b ) ] \ { { \partial}^3 i \over { \partial}u_3 ^ 2 { \partial}u_4 } > 0\ ] ] for @xmath305 , where we have used @xmath329 in the first inequality and ( [ xx ] ) in the second one .
we now use formula ( [ m44 ] ) to calculate the value of @xmath330 at @xmath331 @xmath332^{5 \over 2 } } \geq 0 \quad \mbox{for $ \xi(a , b ) \leq c$ } \ , \ ] ] because @xmath333 for @xmath334 . therefore , @xmath335 for @xmath305 .
it then follows from ( [ m ] ) that @xmath307 . since @xmath336
because of ( [ xx ] ) , we conclude from lemma 2.2 that @xmath337 for @xmath305 . here
we consider the step initial function ( [ step1 ] ) satisfying @xmath338 with @xmath339 .
self - similar solution of the whitham equations ( [ mkdvw ] ) with @xmath32 and the corresponding oscillatory solution ( [ periodicsolution2 ] ) of the mkdv equation with @xmath340 .
the initial data is given by ( [ step1 ] ) with @xmath341 , @xmath86 , @xmath82 of type iii.,width=453 ] ( see figure [ fig3 ] . ) for the step - like initial data ( [ step1 ] ) with @xmath342 , @xmath338 and @xmath343 , the solution of the @xmath32 whitham equations ( [ mkdvw ] ) with @xmath32 is given by @xmath344 for @xmath345 .
outside the region , the solution of equations ( [ mkdv0 ] ) is given by @xmath346 and @xmath347 it suffices to show that @xmath47 and @xmath48 of @xmath348 reaches @xmath151 and @xmath152 , respectively , simultaneously . to see this
, we deduce from the calculation ( [ u30 ] ) that @xmath349 vanishes at @xmath350 . here
we consider the step initial function ( [ step1 ] ) satisfying @xmath338 with @xmath351 .
self - similar solution of the whitham equations ( [ mkdvw ] ) with @xmath32 and the corresponding oscillatory solution ( [ periodicsolution2 ] ) of the mkdv equation with @xmath352 .
the initial data is given by ( [ step1 ] ) with @xmath353 , @xmath86 and @xmath82 of type iv .
the solution in the region @xmath354 represents a rarefaction wave.,width=453 ] ( see figure [ fig4 ] . ) for the step - like initial data ( [ step1 ] ) with @xmath155 , @xmath246 and @xmath245 , the solution of the whitham equations ( [ mkdv ] ) is given by @xmath355 for @xmath356 . outside the region , the solution of equation ( [ mkdv0 ] ) is divided into the three regions : * for @xmath357 , @xmath358 * for @xmath359 , @xmath360 * for @xmath361 , @xmath362 by the calculation ( [ m44 ] ) , when @xmath47 of @xmath363 touches @xmath151 , the corresponding @xmath48 reaches @xmath364 , which is above @xmath152 .
hence , equations @xmath365 can be inverted to give @xmath47 and @xmath48 as functions of @xmath77 in the region @xmath366 . in the region ( 2 ) , the equations ( [ mkdv0 ] ) has a rarefaction wave solution .
in this section , we construct self - similar solutions of the @xmath32 whitham equations ( [ mkdvw ] ) for the initial function ( [ step1 ] ) with @xmath156 .
the solution of equations ( [ mkdv0 ] ) does not develop a shock for @xmath367 .
we are therefore only interested in the case @xmath368 .
we classify the resulting whitham solution into four types : * @xmath369 * @xmath370 with @xmath371 * @xmath372 with @xmath373 * @xmath374 with @xmath375 where @xmath376 is a quadratic polynomial given by ( [ pv ] ) . here
we consider the step initial function ( [ step1 ] ) satisfying @xmath377 .
self - similar solution of the whitham equations ( [ mkdvw ] ) with @xmath32 and the corresponding oscillatory solution ( [ periodicsolution2 ] ) of the mkdv equation with @xmath378 .
the initial data is given by ( [ step1 ] ) with @xmath379 , @xmath380 and @xmath381 of type v.,width=453 ] ( see figure [ fig5 ] . ) for the step - like initial data ( [ step1 ] ) with @xmath156 , @xmath382 , the solution of the whitham equations ( [ mkdvw ] ) with @xmath32 is given by @xmath383 for @xmath384 .
outside this interval , the solution of ( [ mkdv0 ] ) is given by @xmath385 and @xmath386 it suffices to show that @xmath387 is a decreasing function of @xmath47 for @xmath388 . by ( [ eq19 ] ) ,
we have @xmath389 { { \partial}^2 q \over { \partial}u_3 ^ 2 } \ .\ ] ] the second term is negative because of ( [ strict ] ) and @xmath390 .
the first term is also negative : its first factor is positive in view of ( [ eq11 ] ) ; while its second factor is @xmath391 for @xmath388 , as we have that @xmath392 .
here we consider the step initial function ( [ step1 ] ) satisfying @xmath393 with @xmath394 .
self - similar solution of the whitham equations ( [ mkdvw ] ) with @xmath32 and the corresponding periodic oscillatory solution ( [ periodicsolution2 ] ) of the mkdv equation with @xmath378 .
the initial data is given by ( [ step1 ] ) with @xmath395 , @xmath380 and @xmath381 of type vi .
the oscillations have two distinct structures , which are separated by @xmath396.,width=453 ] ( see figure [ fig6 ] . ) for the step - like initial data ( [ step1 ] ) with @xmath397 and @xmath398 , the solution of the whitham equations ( [ mkdvw ] ) with @xmath32 is given by @xmath399 for @xmath400 and by @xmath401 for @xmath402 , where @xmath403 is the unique solution @xmath47 of @xmath404 in the interval @xmath388 . outside the region @xmath405 ,
the solution of equation ( [ mkdv0 ] ) is given by @xmath406 and @xmath407 we first locate the `` leading '' edge , i.e. , the solution of equation ( [ ws1 ] ) at @xmath180 .
eliminating @xmath77 from the first two equations of ( [ ws1 ] ) yields @xmath408 since it degenerates at @xmath180 , we replace ( [ mu23 ] ) by @xmath409 in the appendix , we show that , at the `` leading '' edge @xmath180 , we have @xmath410 in view of ( [ g3 ] ) , which along with ( [ q ] ) gives @xmath411 .
having located the `` leading '' edge , we solve equation ( [ g ] ) near @xmath412 .
we use formula ( [ g3 ] ) to obtain @xmath413 these show that equation ( [ g ] ) gives @xmath46 as a decreasing function of @xmath47 @xmath414 in a neighborhood of @xmath415 .
we now extend the solution ( [ b ] ) of equation ( [ mu23 ] ) as far as possible in the region @xmath416 .
we use formula ( [ eq19 ] ) to obtain @xmath417 { { \partial}^2 q \over { \partial}u_2 ^ 2 } \ , \\ { { \partial}\mu_3 \over { \partial}u_3 } & = & { 1 \over 2 } { { \partial}\lambda_3 \over { \partial}u_3}{{\partial}q \over { \partial}u_3 } + { 1 \over 2 } [ \lambda_3 - 2(a + u_2 + u_3 + b ) ] { { \partial}^2 q \over { \partial}u_3 ^ 2 } \ .\end{aligned}\ ] ] in view of ( [ eq11 ] ) and ( [ strict ] ) , we have @xmath418 we claim that @xmath419 on the solution of ( [ mu23 ] ) in the region @xmath416 . to see this , we use formula ( [ eq19 ] ) to rewrite equation ( [ mu23 ] ) as @xmath420 { { \partial}q \over { \partial}u_2 } = { 1 \over 2 } [ \lambda_3 - 2(a + u_2 + u_3 + b ) ] { { \partial}q \over { \partial}u_3 } \ .\ ] ] this , together with @xmath421 for @xmath422 , and inequalities ( [ strict ] ) , proves ( [ cl ] ) . hence , the solution ( [ b ] ) can be extended as long as @xmath416 .
there are two possibilities ; ( 1 ) @xmath46 touches @xmath152 before @xmath47 reaches @xmath151 and ( 2 ) @xmath47 touches @xmath151 before or simultaneously as @xmath46 reaches @xmath152 .
possibility ( 2 ) is unattainable . to see this , we use ( [ v ] ) to write @xmath423 which is negative for @xmath424 since @xmath425 of ( [ pv ] )
is an increasing function of @xmath46 and since @xmath371 .
therefore , @xmath46 will touch @xmath152 before @xmath47 reaches @xmath151 .
when this happens , we have @xmath426 equation ( [ mu23e ] ) has a simple zero , counting multiplicities , in the interval @xmath388 .
denoting this zero by @xmath403 , then @xmath427 is positive for @xmath428 and negative for @xmath429 . the proof , which involves formulae ( [ n ] ) and ( [ pn ] ) , is rather similar to the proof of lemma [ u * * ]
. we will omit it .
we now continue to prove theorem 4.2 .
having solved equation ( [ mu23 ] ) for @xmath46 as a decreasing function of @xmath47 for @xmath430 , we can then use the middle two equations of ( [ ws1 ] ) to determine @xmath46 and @xmath47 as functions of @xmath77 in the interval @xmath431 .
we finally turn to equations ( [ ws2 ] ) .
we want to solve the third equation of ( [ ws2 ] ) , @xmath432 , for @xmath429 .
it is enough to show that @xmath387 is a decreasing function of @xmath47 for @xmath429 . according to lemma 4.3 , @xmath433 for @xmath429 .
using formula ( [ eq19 ] ) for @xmath434 and @xmath206 , we have @xmath435 { { \partial}q \over { \partial}u_2 } <
{ 1 \over 2 } [ \lambda_3 - 2(a + c + u_3 + b ) ] { { \partial}q \over { \partial}u_3 } \ .\ ] ] this , together with @xmath436 for @xmath437 , and inequalities ( [ strict ] ) , proves @xmath438 for @xmath429 .
hence , @xmath439 { { \partial}^2 q \over { \partial}u_3 ^ 2 } < 0 \ , \ ] ] where we have used inequality ( [ eq11 ] ) . here
we consider the step initial function ( [ step1 ] ) satisfying @xmath440 with @xmath441 .
self - similar solution of the whitham equations ( [ mkdvw ] ) @xmath32 and the corresponding oscillatory solution ( [ periodicsolution2 ] ) of the mkdv equation with @xmath442 .
the initial data is given by ( [ step1 ] ) with @xmath443 , @xmath444 and @xmath381 of type vii.,width=453 ] ( see figure [ fig7 ] . ) for the step - like initial data ( [ step1 ] ) with @xmath445 and @xmath446 , the solution of the whitham equations ( [ mkdvw ] ) with @xmath32 is given by @xmath447 for @xmath448 . outside the region ,
the solution of equations ( [ mkdv0 ] ) is given by @xmath449 and @xmath450 it suffices to show that @xmath46 and @xmath47 of @xmath451 reaches @xmath152 and @xmath151 , respectively , simultaneously . to see this , we deduce from equation ( [ 2 ] ) that @xmath452 vanishes when @xmath373 .
here we consider the step initial function ( [ step1 ] ) satisfying @xmath445 with @xmath453 .
( see figure [ fig8 ] . ) for the step - like initial data ( [ step1 ] ) with @xmath445 and @xmath453 , the solution of the whitham equations ( [ mkdvw ] ) with @xmath32 is given by @xmath454 for @xmath455 , where @xmath456 is the unique @xmath46-zero of the quadratic polynomial @xmath425 in the interval @xmath457 . outside the region ,
the solution of equations ( [ mkdv0 ] ) is divided into the following three regions : * for @xmath458 , @xmath459 * for @xmath460 , @xmath461 * for @xmath462 , @xmath463 by the calculation ( [ 7 ] ) , when @xmath47 of @xmath464 touches @xmath151 , the corresponding @xmath46 reaches @xmath456 , where @xmath465 .
obviously , @xmath466 .
hence , equations @xmath467 can be inverted to give @xmath46 and @xmath47 as functions of @xmath77 in the region @xmath468 . to the right of this region , equations ( [ mkdv0 ] )
has a rarefaction wave solution .
self - similar solution of the whitham equations and the corresponding oscillatory solution ( [ periodicsolution2 ] ) of the mkdv equation with @xmath469 .
the initial data is given by ( [ step1 ] ) with @xmath85 , @xmath444 and @xmath381 of type viii .
the solution in the region @xmath470 represents a rarefaction wave.,width=453 ]
we conclude the paper by showing how to handle the initial data ( [ step2 ] ) . inequalities ( [ 34 ] ) are replaced by @xmath471 for @xmath200 .
results similar to lemma 2.2 can then be easily proved .
the rest of calculations are also similar to those in sections 3 and 4 .
the function @xmath472 of ( [ i ] ) can be written in terms of the complete elliptic integral of the first kind @xmath57 , i.e. , @xmath473 where @xmath474 using the derivative formula @xmath475 where @xmath58 is the complete elliptic integral of the second kind , we calculate @xmath186 and @xmath187 of ( [ lambda ] ) @xmath476 we then use ( [ eq19 ] ) to write @xmath477 as @xmath478 \ .\ ] ] hence , @xmath479 of ( [ g ] ) becomes @xmath480 \ .\ ] ] * acknowledgments . *
we thank ed .
overman for showing us his numerical simulations of the nls and mkdv equations for small dispersion .
t. were supported in part by nsf grant dms-0404931 and v.p .
was supported in part by nsf grant dms-0135308 .
p. deift , s. venakides and x. zhou , _
new results in small dispersion kdv by an extension of the steepest decent method for riemann - hilbert problems _ , international math . research jour . , 4(1997 ) ,
285 - 299 .
forest and j. lee , _ geometry and modulation theory for periodic nonlinear schrdinger equation _ , in oscillation theory , computation and method of compensated compactness , j.l .
ericksen , d. kinderlehrer , and m. slemrod eds . , springer - verlag , 1986 , pp .
35 - 69 .
lax , c.d .
levermore and s. venakides , _ the generation and propagation of oscillations in dispersive ivps and their limiting behavior _ , in important developments in soliton theory 1980 - 1990 , t. fokas and v.e .
zakharov eds . , springer series in nonlinear dynamics , springer , berlin , 1993 , pp .
205 - 241 . | we study the whitham equations for the defocusing complex modified kdv ( mkdv ) equation .
these whitham equations are quasilinear hyperbolic equations and they describe the averaged dynamics of the rapid oscillations which appear in the solution of the mkdv equation when the dispersive parameter is small . the oscillations are referred to as dispersive shocks .
the whitham equations for the mkdv equation are neither strictly hyperbolic nor genuinely nonlinear .
we are interested in the solutions of the whitham equations when the initial values are given by a step function .
we also compare the results with those of the defocusing nonlinear schrdinger ( nls ) equation . for the nls equation ,
the whitham equations are strictly hyperbolic and genuinely nonlinear .
we show that the weak hyperbolicity of the mkdv - whitham equations is responsible for an additional structure in the dispersive shocks which has not been found in the nls case .
whitham equations , non - strictly hyperbolic equations , dispersive shocks * ams subject classifications .
* 35l65 , 35l67 , 35q05 , 35q15 , 35q53 , 35q58 |
the requirements of laser fusion targets for high power lasers with good laser beam uniformity has driven a quest for new techniques for smoothing the intensity variations on the target surface .
early attempts at beam smoothing[1 ] were not well characterized but more systematic techniques[2 - 6 ] have demonstrated significant improvements in beam uniformity .
all techniques involve introduction of phase nonuniformities which replace the normal beam pattern , typically containing substantial hotspots , with a smaller - length - scale speckle pattern .
further addition of bandwidth to the laser provides temporal smoothing of the speckle .
the primary motivation for investing in these smoothing schemes is to reduce the initial nonuniformities that can seed fluid instabilities such as rayleigh - taylor , yet there is also palpable interest in reducing the strength of laser plasma instabilities such as stimulated raman or brillouin scattering or two plasmon decay . supplying a theoretical framework for understanding these laser plasma interactions with smoothed laser beams is the task we undertake in this article . in a subsequent article[7 ] , the results obtained here will be applied to particular instabilities in geometries of interest . with the usual assumptions about slow variation of parameters with respect to the frequencies @xmath0 and wave vectors k@xmath1 of the wave
, the coherent and incoherent problems can be studied within the context of the coupled mode equations ( ii.10 or ii.7 ) . the wave group velocities and damping rates
are denoted by v@xmath1 and @xmath2 respectively ; the strength of the coupling between the waves , @xmath3 , proportional to the pump or laser wave amplitude , has the dimensions of frequency and is the rate at which the waves grow without damping in an infinite homogeneous plasma . as the reader will remember , for a coherent driver , both convective and absolute instability may occur provided the laser intensity , i.e. @xmath4 , exceeds certain threshold values set by losses [ 8 ] .
in an unbounded plasma , the convective coherent threshold is @xmath5 absolute instability requires that the decay waves be oppositely directed @xmath6 , and , the absolute coherent threshold is @xmath7 temporal and spatial incoherence introduce two additional parameters , the temporal bandwidth @xmath8 and the spatial bandwidth ( or spread in wavevectors ) @xmath9 which is inversely related to the correlation length @xmath10 . the effects of both temporal and stationary spatial pump incoherence on parametric instabilities in homogeneous and inhomogeneous plasma has been studied extensively both theoretically and experimentally over the past thirty years [ 9 - 68 ] .
the first approach was quite naturally to consider the purely temporal problem with coupled mode equations in homogeneous plasma .
zaslavskii and zakharov [ 9 ] considered the generic undamped decay instability with methods developed earlier for studying the relaxation of two level molecular systems driven by an external field [ 10 ] .
they found that the convective growth rate was reduced , in the limit @xmath11 , from @xmath3 to @xmath12 .
both valeo and oberman [ 11 ] and tamor [ 12 ] addressed this problem with different methods and obtained the same result .
tamor included a damping rate on the langmuir wave coupled to an acoustic wave and found a dispersion relation where the damping rate for the langmuir wave @xmath13 is replaced by @xmath14 .
he noted reasonably that the bandwidth was unimportant unless it exceeded the damping rate @xmath15 but made no comment on the fact that the bandwidth appeared asymmetrically in his dispersion relation ; that is , the apparent damping of the acoustic wave was unaffected . in a series of articles [ 13 - 15 ] , thomson made a number of important contributions .
first , he noted that for equal damping on both modes , the threshold was @xmath16 if @xmath17 and incorrectly speculated that , for unequal damping , the threshold was @xmath18 .
this guess was based on the assumption that , for growth to occur in either mode , the average amplitude of both must grow .
the correct solution was presented by thomson[15 ] later using an exactly solvable model [ 69 ] . for the purely temporal problem
, the ensemble - averaged mode amplitude equations showed these equations decouple and have @xmath19 thresholds , i.e. @xmath20 and @xmath21 thomson noticed and explained the asymmetry as due to averaging over the rapidly varying phase of the mode amplitude .
he obtained a more appropriate threshold by solving the equations for the ensemble averaged intensities .
the solution showed that the lower of the thresholds ( i-3,4 ) was appropriate .
we show , as did thomson , that the incoherent growth rate for the average intensity is @xmath22 provided @xmath23 .
this answer is expected , given that the average amplitudes grow with rate @xmath24 .
however , thomson did not point out the interesting fact ( found also by laval et al.[16 ] in the space - time problem when @xmath25 ) that , if @xmath26 , the incoherent growth rate is @xmath27 that is twice the expected rate .
we will comment more on this in secs .
thomson then considered the space time problem with the ensemble averaged mode amplitude equations and again found an asymmetry in the thresholds for the absolute instability when @xmath28 .
thomson naturally assumed that the appropriate intensity threshold , in analogy with the temporal problem , was the lower one .
later work showed this to be incorrect .
laval et al . reconsidered [ 70 ] the space -time problem for the evolution of the intensities both by using the bourret approximation and , in the case @xmath29 , the exactly solvable kubo - anderson process ( kap ) .
they showed that the intensity thresholds are in fact much lower than the amplitude thresholds when @xmath30 and , in usual cases of interest ( @xmath31 ) , the incoherent absolute threshold equals the product of the bandwidth and the damping on the slow wave @xmath32 in a later publication [ 17 ] , thomson applied the results to stimulated raman forward scatter and incorrectly concluded , as we discuss subsequently , that small amounts of bandwidth , comparable to its growth rate , can suppress this slowly - growing instability . during this time period ,
the nonlinear evolution of instabilities driven by broadband pumps was investigated by using particle - in - cell ( pic ) simulations [ 18 - 20 ] .
these studies demonstrated a reduction in the stimulated brillouin reflectivity from 60% to less than 10% with 5% bandwidth .
because these plasmas were strongly inhomogeneous , no reduction is expected for modest bandwidth , @xmath33 , as shown by thomson [ 15 ] . however , with @xmath34 , the line separations in these simulations are larger than an acoustic frequency and each line acts independently .
kruer et al .
[ 19 ] also suggest that bandwidth in an inhomogeneous plasma will be ineffective unless @xmath35 where @xmath36 is the interaction length set by plasma inhomogeneity .
this can be rephrased in terms of the correlation length , @xmath37 where @xmath38 and @xmath39 for @xmath40 ; @xmath41 is the condition for phase matching .
kruer s condition is difficult to satisfy for typical laser systems .
we return to this subject in our discussion of spatial incoherence effects .
the correct theory [ 15 ] for convective amplification in an inhomogeneous plasma showed that the growth rate was reduced but the ensemble - averaged mode amplitude convectively saturates at the same level as the coherently driven amplitude .
( thomson actually misstated his result , although his analysis was correct . later work [ 45,49 ] that solved the coupled mode equations
clarified the answer . )
estabrook et al . also considered , with pic simulations , the effect of laser bandwidth on stimulated raman scattering again in inhomogeneous plasma .
bandwidth was represented by a series of equally spaced laser lines .
a reduction in the reflectivity was found when the line separation was greater the growth rate .
each line acted as an independent pump , and the intensity was low enough that the instability was not strongly saturated .
direct comparisons between homogeneous plasma theory and pic simulations of srs in a plasma slab were made by forslund et al . [ 21 ] with good agreement for the dependence of the srs growth rate on the bandwidth of the frequency modulated laser .
these authors pointed out that , @xmath42 , despite a fourfold reduction in the growth rate , only a modest reduction in the power reflected was observed in the final state .
a further increase of bandwidth to @xmath43 brought the instability below threshold .
later pic work [ 24 ] with multiple lines also showed good agreement with theory for the raman backscatter growth rate when @xmath44 at several combinations of density and laser intensity .
these simulations also showed a monotonic decrease with bandwidth in the absorption into hot electrons due to srs and in the srs reflected power .
less effect on the forward srs was observed , consistent with the theory we discuss subsequently . other theoretical work on pump bandwidth effects concerned applications to specific targets [ 30 ] , and the application to induced spatial incoherence ( isi ) [ 44 ] .
the effects of both temporal and spatial bandwidth in a 1d inhomogeneous plasma [ 45 ] for a phase mismatch , @xmath45 , varying linearly , @xmath46 , or quadratically , @xmath47 , was treated by analytical and numerical methods . in reference
[ 47 ] , the green s function formalism of brissaud and fritsch [ 69 ] , used by other authors [ 15,16,38 ] was reformulated in the language of effective hamiltonian matrices . in a second paper[48
] , this formalism was used to investigate the competition between temporal bandwidth and inhomogeneity for a variety of parametric processes , including two plasmon decay .
early experiments [ 25,26 ] that attempted to observe laser bandwidth effects utilized plasmas that were too nonuniform to expect observable effects . later experiments used gas jet targets [ 27,29 ] or used microwave plasmas [ 28 ] .
clayton et al . and giles et al .
divided the laser power into two lines and found a striking reductions in reflected sbs power compared to a single line with the same total laser power .
the lines were separated by much more than a growth rate .
it was recognized early [ 31 - 33 ] that random spatial modulation of the phase mismatch , @xmath48 , could also reduce parametric instability growth rates . using methods similar to the temporal problem , kaw et al .
used the steady - state limit of the coupled - mode equations ( ii-10 ) to find a convective gain rate , @xmath49 , where @xmath50 . here
@xmath51 is the wavevector mismatch induced by the random variation in the plasma properties and @xmath52 is the correlation length for the random process .
further work [ 34 - 38,40 - 42 ] on random fluctuations of the phase concerned its effect on the growth in inhomogeneous plasmas where , for example , the stabilization of absolute modes by linear gradients in the phase could be undone by these fluctuations . in this article , we concentrate on the effects in uniform plasmas .
beam smoothing techniques use not only temporally incoherent but also spatially incoherent pump waves . in the focal plane , the laser wave can be considered a sum of randomly phased plane waves with a spread in wavevectors , @xmath9 . in the limit of no temporal bandwidth ,
the spatial interference pattern of the pump wave envelope is stationary and is related to the effect of stationary plasma turbulence on parametric instability treated by williams et al.[39 ] using a novel approach ( unrelated to the methods used in the present analysis ) , they considered the threshold and growth rate for absolute modes in a finite system of length l. this analysis finds that for sufficiently large spatial bandwidth , @xmath9 , the fastest growth rate is reduced from the coherent absolute rate , @xmath53 when @xmath54 to the spatially incoherent absolute rate @xmath55 ( to obtain i-9 from eq .
52 of williams et al .
, the reader is advised that the correct normalization for the growth rate is @xmath56 their eq .
3 is incorrect . )
the incoherent result in eq .
( i-9 ) must of course be less than the coherent one in eq .
the unexpected feature of ( i-9 ) is the length dependent logarithmic factor which is related to the fact that the rate ( i-9 ) is not the average rate sought in our analysis but the largest rate expected in a system of length l. except for the logarithmic factor , we recover the scaling of the result of eq .
( i-9 ) . in the present article ,
we derive average amplitude equations by using the bourret approximation[70 ] and average intensity equations by using the so- called random phase approximation ( rpa).[71 - 74 ] the major objective of this article is to provide a unified theory of spatial and temporal incoherence effects on parametric instabilities .
the approximations and assumptions necessary to arrive at the rpa equations ( v.32 ) and the bourret equations ( v.1 4 ) that form the basis for theoretical results in sec .
vi are carefully explored and systematically presented in sec .
iii through v. a comprehensive set of results is presented for the threshold and growth rate of both absolute and convective instability in the incoherent limit .
the domains of validity of the statistical approximations are explicitly noted and outside these domains the coherent results are shown to apply .
moreover the same analysis is done for the spatial amplification of convectively unstable waves . finally , we present in sec .
vii numerical solutions of the fundamental set of equations ( ii-10 ) for particular models of incoherence that illustrate the meaning of the averaging procedures and verify the main results .
the rpa dispersion relation ( vi.1 ) for an infinite system obtained from the rpa equations ( v.32 ) form the basis for the analysis of growth rates and thresholds in section vi .
there are two parameters that play a role in the ensemble average equations that measure the effective bandwidth , @xmath57 where @xmath58 in the general case of spatial and temporal incoherence , there can be a gap between the domain of validity for the rpa dispersion relation and the coherent one .
the coherent domain is the whole region where either spectral width @xmath59 is less than the corresponding damping rate or growth rate.in this intermediate region , the average amplitude dispersion relations ( vi.2 ) are valid , and the more unstable one agrees with the rpa dispersion relation within a factor of two .
it is on this basis that we argue that the rpa dispersion relation can be used in the whole domain ( denoted the incoherent domain ) complementary to the coherent domain . in the remainder of sec .
vi , a description of parametric instabilities is given including the conditions for absolute and convective instability .
both early time behavior , which is dominated by convective instability , and long time behavior , which is dominated by absolute instability ( if it exists ) or by spatial amplification are considered .
it is worth noting that , in the incoherent domain , the average amplitude dispersion relations ( vi.2 ) allow only convective solutions whereas the rpa dispersion relation ( vi.1 ) allows the possibility of absolute instability if a threshold can be exceeded .
however , the simpler average amplitude equations do provide the convective mode threshold ( vi.5 ) , growth rate ( vi.7 ) , and spatial amplification factor ( vi.13 ) and ( vi.16 ) ( within better than a factor of two ) in agreement with those obtained in sec .
vi using the rpa dispersion relation .
absolute instabilities are of particular practical interest because , above threshold , the only limit to their growth is the finite laser energy or other nonlinear effects . on the other hand , an absolute instability generally has a larger threshold to overcome losses to collisional or landau damping than a convective instability . for a coherent laser wave ,
the threshold laser intensity is determined by setting the spatial growth rate @xmath60 equal to one - half of the sum of the spatial loss rates @xmath61 for the decay modes ( eq .
i-2 ) . in addition , as given by eq .
( i-8 ) , the absolute instability growth rate is smaller than the growth rate @xmath3 by approximately the ratio @xmath62 in the usual case @xmath63 . the exact formula is also given by eq .
( vi.11 ) . with incoherent laser beams ,
the threshold is still determined by setting the effective spatial growth rate equal to the sum of loss rates but now the spatial growth rate is @xmath64 , that is , the maximum spatial growth rate for the average amplitude equations .
the general formula is given by ( vi.9 ) .
note in the special but interesting case that @xmath65 , temporal incoherence increases the absolute threshold if @xmath66 which exceeds the naive criterion by the square root of the group velocity ratio .
we observe that the statement @xmath67 is equivalent to requiring that the coherent spatial growth length @xmath68 exceed the larger coherence length @xmath69 . a rigorous application[46 ] of the bers and briggs criteria[75 ] arrives at this same condition as a necessary condition ( vi.10 ) for the incoherent limit to apply . above the incoherent threshold ,
the incoherent absolute growth rate is roughly @xmath70 , ( the exact formula is given by eq .
( vi.8 ) ) i.e. approximately the same expression as the incoherent convective growth rate but with a different domain of validity .
there is a limit where the effective bandwidths @xmath59 are determined only by spatial incoherence .
then @xmath71 is independent of the group velocity and the growth rate , @xmath72 agrees with the scaling found previously and given in eq . ( i-9 ) .
thus our general threshold and growth rate formulae indeed recover the correct limits of purely spatial and purely temporal incoherence derived previously .
the coherent convective threshold , always lower than that for absolute instability , is given by the requirement that the growth rate @xmath3 be greater than the mean loss rate @xmath73 .
the same criterion for an incoherent pump applies if the laser intensity is reduced by the ratio of line widths factor @xmath74 .
the effective bandwidth @xmath59 must be larger than the corresponding damping rate @xmath2 for each mode to increase the threshold
. otherwise the coherent threshold applies ( i-1 ) . far above the incoherent threshold ,
the incoherent convective growth rate is reduced by the factor @xmath75 . note that when @xmath76 the convective and absolute incoherent growth rates are equal in the overlapping domain of validity .
more exact expressions are given by ( vi.5 ) - ( vi.7 ) .
for many cases of practical interest @xmath77 because the spread in laser wavenumbers is sufficiently small and/or the group velocity is small enough that @xmath78 and either @xmath79 or @xmath80 . then temporal incoherence is the dominant stabilizing influence .
however two special cases deserve mention .
examination of the expression ( vi.8b ) for the absolute instability growth rate above threshold shows that if the maximum value of the coherence length @xmath81 occurs for the minimum group velocity , the growth rate is strongly reduced @xmath82 in the usual case @xmath83 .
this case can occur if there is a large spread in wavenumber accompanied by a weak temporal incoherence so that @xmath84 .
the transient or convective growth rate for the same case ( in the incoherent limit ) is larger , @xmath85 .
the other unusual case occurs in the case of temporal incoherence for forward scatter when @xmath86 .
then @xmath87 so that the coherent convective growth still occurs even if @xmath88 .
the case of practical interest is stimulated raman forward scatter.[22 - 24 ] the fate of absolutely unstable modes necessarily requires consideration of nonlinear effects but , in a bounded plasma , the maximum amplitude of convectively unstable waves may be found by computing the spatial amplification rate . in general
there are two roots that are most easily obtained by using the average amplitude equations ( vi.2 ) but are also accessible from the average intensity equations ( vi.1 ) .
a bandwidth @xmath89 is usually sufficient to reduce the spatial growth rate . for moderate intensities , i.e. , below any absolute threshold , the spatial gain coefficient @xmath90 ( where @xmath91 ) is reduced by @xmath92 in the weak damping case . at higher intensities , a second root with gain coefficient @xmath93 occurs provided @xmath94 and @xmath95 , that is , above the absolute mode threshold when @xmath96 .
therefore both a convective and absolute mode exist ; and , interestingly , the convective mode has a spatial gain rate that exceeds that of the absolute mode . a more detailed discussion and
exact formulae are given in eqs .
( vi.12 ) - ( vi.18 ) . in the general case ,
it is difficult to summarize the effects of incoherence on convective and absolute stabilities .
however , it is useful to consider a common case of practical interest where @xmath97 and @xmath98
. then diagrams showing stability regions can be constructed with axes @xmath99 and @xmath100 , i.e. moving towards more incoherence in one direction and towards more damping in the other .
figure vi.1 is such a diagram for convectively unstable modes .
there are four regions : coherently unstable , incoherently unstable with reduced growth rate , coherently stable , and incoherently stabilized . in fig .
vi.2 , the diagram for absolutely unstable modes is drawn with four analogous regions .
this figure graphically shows that , in only a small region of parameter space does incoherence reduce the growth rate but not completely stabilize .
figure vi.3 shows the different regions for spatial amplification . finally , in fig .
vi.4 , an overall diagram is shown for all regions for convective , absolute , and spatial amplification . to this point , we have presented results for infinite homogeneous plasma . for a large enough system ,
these results are a good guide to the behavior in finite systems . nonetheless , real plasmas are finite and it is well known that a threshold length is necessary for absolute instability . moreover in sec .
vii , numerical solutions of the coupled mode equations ( ii.10 ) are presented in a necessarily finite system .
thus partly as a guide to the numerical solutions , we find the threshold length for absolute instability in the coherent and incoherent limit as obtained by solving eq .
( ii-10 ) or eq .
( v.32 ) as appropriate .
it is an interesting feature that the normal modes in the slab , sinusoidal in the coherent limit , are exponential in the incoherent limit .
the threshold length given by eq .
( vi.30 ) increases as expected with bandwidth above the coherent threshold length ( vi.29 ) .
several features of the statistical description that provoked further analysis were the `` unexpected '' factor of two that appeared in the growth rate for the ensemble - average intensity , the question of the validity of the rpa description in the intermediate domain , and the applicability of these results to analysis of experiments using beam smoothing techniques .
these aspects were examined in sec .
vii by integrating directly the coupled mode equations ( ii.10 ) or ( vii.1 ) with particular choices of random processes to represent the pump wave incoherence . for the purely temporal case , an analytic solution for the distribution function of mode amplitude ( eq .
( vii.10 ) is obtained which is remarkably broad if the damping rates are negligible .
in fact , the width of the gaussian distribution is equal to the mean .
thus , higher powers of the amplitude , e.g. , the intensity grow at faster rates than the mean which gives rise to factor of two mentioned earlier .
we compare this distribution to a numerically generated one in fig . 6 for the same parameters . on the other hand , with sufficient damping , this factor of two does not appear ; that is , the average intensity and amplitude grow at corresponding rates . in this case , the distribution is strikingly narrowed as also shown in fig .
6 . in the space - time problem
this factor of two also occurs for absolute instability driven by an incoherent pump when the decay wave group velocities are equal in magnitude and opposite in sign .
we have verified our supposition by numerical integration that once again the distribution is broad for @xmath101 but is much narrower if @xmath102 as shown in fig . 7 .
the validity of the rpa description in the intermediate domain was examined numerically in sec .
vii.b , by considering the model of a spatially incoherent pump driving a pair of decay waves satisfying the intermediate domain inequalities ( vii.13 ) .
although we did show that the rpa equations appear valid , we also discovered that the distribution of mode intensities ( fig .
8) is unusual in that it consists of a slowly decreasing tail on a distribution with a peak at nearly zero growth .
we also show in sec . vii.c that the standard model of an isi beam , eq .
( vii.15 ) , which has both phase and intensity variation , can be treated as an incoherent pump wave provided the temporal and spatial bandwidth are large enough .
thus with the appropriate identification of experimental parameters with @xmath8 and @xmath9 , the formulae in sec .
vi can be applied to experiments.[51 - 65 ] a few remarks are in order regarding the derivation of the statistical equations that form the basics for the results outlined above .
the analysis begins in section ii with the completely nonlinear coupled - mode equations ( ii.2 ) appropriate for the case when the pump and decay waves are weakly coupled and weakly damped . in the linear analysis of this article , the pump wave is unaffected by the decay waves and the characteristic growth rate of the parametric instability is simply related to the pump amplitude at its mean wavevector ( ii.5 ) .
normalization of the decay wave amplitudes to the average pump wave energy yields the linearized coupled - mode equation ( ii.7 ) in fourier space . from these equations ,
the envelope equations ( ii.9 ) in fourier space or ( ii.10 ) in real space are obtained by expanding the mode amplitudes about the value at the mean wavevector .
these envelope equations are used in sections iii - v to obtain equations for the ensemble - averaged mode amplitudes and intensities .
if the pump wave has a distribution of wavevectors , then a given pair of decay waves will be frequency matched to only a portion of the pump wave spectrum .
thus one is naturally led to consider the frequency mismatches ( iii.1 ) for a general triplet of wavevectors or the mismatches ( iii.2 ) at the mean value of the decay wavevector .
it is assumed that there exist triplets for which there is exact matching so that the mismatch near the mean wavevectors is small .
the maximum value of the mismatches at the mean decay wavevector determines that the interaction is coherent if the mismatches are both small in the sense defined by iii.6 . two approaches to generating
equations governing the ensemble averaged behavior of the instability are used in this article .
the first employs the bourret approximation[70 ] to obtain the dispersion relation ( iv.5 ) and ( iv.7 ) for the stability of the ensemble - averaged mode amplitude @xmath103 .
each equation for @xmath103 is simple in that it does not involve the other but it does involve integrals over the spectral density of the pump wave and the frequency mismatch . in the incoherent limit or markov limit defined by ( iv.17 ) , the dispersion relation takes the particularly simple and well- known form given by eq .
( iv.13 ) which states that the coupling between the waves is reduced by the ratio of the characteristic growth rate to the maximum mismatch @xmath104 for @xmath105 . for the one dimensional case with a pump spectrum that is lorentzian , e.g. a kubo - anderson process
, the dispersion relations ( iv.22 ) are the familiar ones derived previously[15 - 16 ] and the mismatch @xmath104 is equivalent to the spectral widths @xmath106 defined by ( iv.20 ) that are , within at most a factor of two , equal to the pump wave bandwidth @xmath8 ( except for the special case of forward scatter where the @xmath107 ) . note in one dimension , the problem of spatial and temporal incoherence are not independent in fact so our restriction to one dimension must yield the results[16 ] of laval et al .
however in our numerical models , this connection is broken for convenience and tractability .
the average amplitude equations ( iv.5 and iv.7 ) are somewhat unsatisfying because @xmath108 and @xmath109 evolve independently . for the case of temporal incoherence alone , this untidiness was remedied by obtaining equations for the ensemble averaged intensities which are coupled and symmetric . nevertheless , the result obtained for convective instabilities with only temporal incoherence is just that obtained by consideration of the most unstable average amplitude . in section v.b
, the same procedure can be followed to obtain bourret equations for the ensemble - averaged spectral densities ( v.25 ) that are related to the mode intensities by ( v.2 ) . as with the amplitude equations ( iv.5 and ( iv.7 ) , these equations involve integrals over the pump spectral density and the frequency mismatch ; but , in addition , the integral contains the spectral density of the other decay wave at the pump shifted wavenumber .
two further approximations can be made : first , the markov assumption valid provided that the mismatch is larger than the maximum of the growth rate and damping rates , i.e. @xmath110 removes the spatial and temporal growth rate from the integrand ; second , the assumption that the decay wave spectral densities are slowly varying functions allows these densities to be evaluated at the mean wavenumber at exact matching .
the first approximation yields the random phase approximation ( rpa ) equations ( v.28 ) ; and the subsidiary second approximation yields the set ( v.32 ) with which the instability analysis is done in section vi .
as they must do , the rpa equations ( v.28 ) are shown to reduce to the correct intensity equations derived using the bourret approximation when the pump wave group velocity is infinite .
for the purely temporal incoherence modeled by the kubo - anderson process[69 ] , this latter result is exact .
our conclusions are presented in sec .
there , we also give some examples of current interest and consider the special case when the temporal bandwidth is larger than one of the mode frequencies .
in this article we limit our discussion to the coupling of three wave - packets which are both weakly coupled and weakly damped .
we assume that each fourier component @xmath111 of the wave packet i ( with i = 0 , 1 , 2 ) ) behaves in lowest order as a normal mode ; i.e. , it oscillates in time like @xmath112 , where @xmath113 is the real part of the frequency @xmath114 which characterizes the wave - packet i and which satisfies the dispersion relation @xmath115 .
the assumption of weakly coupled and weakly damped modes can be written quantitatively as @xmath116 where @xmath117 is the linear damping increment of the fourier component @xmath111 , and where @xmath118 denotes the inverse of the characteristic time for nonlinear evolution of the coupled modes . in this limit
the equations describing the coupling of the three wave packets are of first order in time . in fourier space
they have the following general form : @xmath119 a_{{\bf k}_0}^0 = - \int d{\bf k}_1 v_{{\bf k}_0,{\bf k}_1 } a_{{\bf k}_1}^1 ( a_{{\bf k}_0 - { \bf k}_1}^2 ) \eqno ( ii.2a)\ ] ] @xmath120 a_{{\bf k}_0}^0 = \int d{\bf k}_0 v_{{\bf k}_0,{\bf k}_1 } a_{{\bf k}_0}^0 ( a_{{\bf k}_0 - { \bf k}_1}^2 ) ^ * \eqno ( ii.2b)\ ] ] @xmath121 a_{{\bf k}_2}^2 = \int d{\bf k}_0 v_{{\bf k}_0,{\bf k}_0 } a_{{\bf k}_0}^0 ( a_{{\bf k}_0 - { \bf k}_2}^1 ) ^ * \eqno ( ii.2c)\ ] ] where @xmath122 is proportional to the fourier component of the electric field of wave i ; wave 0 refers to the pump wave and waves 1 and 2 to the decay waves .
the coupling constants @xmath123 are derived from the usual field expansion of the fluid or the vlasov equations .
the parametric approximation consists in neglecting the rhs of eq .
( ii.2a ) and in taking @xmath124 ) in the rhs of eqs . ii.2b and ii2c . in this limit
the latter equations describe correctly the parametric coupling of waves in the usual decay regime ( the so - called modified decay instability and the modulational instability[76 ] can not be described by these equations since they correspond to coupled mode equations in which the second order partial derivative in time must be retained ) . in this article
we derive the conditions for a reduction of the parametric growth due to the pump wave incoherence .
the incoherence of the pump wave may be either temporal , or spatial , or both . in the case of purely _ temporal _ incoherence ,
the wave - numbers @xmath125 of the different components of the pump wave all have the same direction ; if @xmath8 is the spectral width in frequency space of the pump wave and is small compared to the pump frequency , @xmath126 , the spread @xmath127 of the corresponding wave - number modulus @xmath128 is given by @xmath129 here @xmath130 is the characteristic group velocity of the pump wave and @xmath131 denotes @xmath132 . in the case of purely _ spatial _ incoherence , all the wave - numbers @xmath133 have the same modulus @xmath134 and are spread within a small cone , the _ half _ angle of aperture of the latter being denoted by @xmath135 . in the general case , the mean wave - number and frequency of the pump wave
are denoted by @xmath136 and @xmath137 respectively , and the incoherence of the pump wave is characterized by the spectral width @xmath138 and by the half - angle spread @xmath135 .
we will assume an azimuthal symmetry around @xmath136 and a cross section of the total domain of existence of @xmath133 is displayed in fig
. 1 . an assumption made implicitly when using eqs .
( ii.2 ) to investigate the incoherence effects is that the spread in frequency @xmath139 is much smaller than the frequencies @xmath140 of the various waves .
similarly , the spread in wave number @xmath141 is assumed to be small compared to wave numbers @xmath142 .
we denote by @xmath143 the angle between the propagation directions of the pump wave and of the wave 1 . for a given direction @xmath143 of the wave 1 ,
there is a unique couple @xmath144 satisfying the two resonance conditions , @xmath145 @xmath146 throughout this article we investigate the reduction of the parametric instability corresponding to a given direction @xmath143 and the capital letters @xmath147 will be used exclusively for this wavenumber of wave i which corresponds to the resonance conditions ( ii.4 ) , for i = 1 , 2 ; as stated before , @xmath136 denotes the mean wavenumber of the pump wave , and the symbol @xmath148 will denote a generic wavenumber of the pump wave ; lastly @xmath149 is the angle between the mean pump wave - number @xmath150 and the wave number @xmath151 corresponding to the resonance conditions ii.4 written for an angle @xmath143 .
we may now relate the characteristic growth rate of the parametric instability corresponding to the coherent case , @xmath3 , for a given angle of observation @xmath143 , to the mode coupling constants and the pump amplitude by @xmath152 where @xmath153 denotes the average energy of the pump wave in a sense defined in the next section . defining the dimensionless coupling constants and wave amplitudes by @xmath154 @xmath155
the dimensionless form of equations is simply @xmath120 a_{{\bf k}_1}^1
= \gamma_0 \int d{\bf k}_0 v_{{\bf k}_0 , { \bf k}_1 } a_{{\bf k}_0}^0 \left ( a_{{\bf k}_0 -{\bf k}_1}^1 \right)^ * \eqno ( ii.7a)\ ] ] @xmath121 a_{{\bf k}_2}^2 = \gamma_0 \int d{\bf k}_0 v_{{\bf k}_0 , { \bf k}_0 -{\bf k}_2 } a_{{\bf k}_0}^0 \left ( a_{{\bf k}_0 -{\bf k}_2}^1 \right)^*\eqno ( ii.7b)\ ] ] where @xmath156 is a slowly varying function of @xmath133 and @xmath157 , with @xmath158 . finally , the latter equations are supplemented by the following equation for the pump wave amplitude @xmath159 at this point we can make a connection between the coupled mode equations ( ii.7 ) written in their general form in fourier space and their so - called envelope equation form in real space .
the envelope approximation for the coupled mode equations corresponds to the first order expansion of @xmath160 in a power series of ( @xmath161 ) . by writing @xmath162 , and by setting @xmath163 or i = 1 and 2 , and @xmath164 t \right ) \eqno ( ii.8b)\ ] ]
the coupled mode equations can be written as @xmath165 @xmath166 where @xmath167 and we neglected the slow variation of @xmath168 and of @xmath169 with @xmath170 and @xmath171 . by taking the inverse fourier transforms of eqs .
ii.9 , one obtains the mode - coupling equations in their envelope approximation limit , namely @xmath172 @xmath173 where the quantity @xmath174 is defined by @xmath175 equations ii.10 are the envelope equations which have been investigated with the statistical methods described in the next section .
these equations are generalizations to three dimensions of one dimensional equations used in previous treatments of this subject[16 ] . for
the sake of simplicity the envelope approximation will henceforth refer to either the coupled mode equations written in their envelope form ii.9 , or the first order expansion of @xmath140 .
concerning our notation , @xmath176 will denote , in the following the group velocity corresponding to a generic wavenumber @xmath157 , namely @xmath177 ; the notation @xmath178 will be reserved to the case of exact resonance , i.e. , @xmath179 .
on the other hand , the quantity @xmath180 ( * x*,t ) will denote the fourier transform of the field @xmath181 in the case where one considers the mode coupling equations in their general form ( ii.7 ) ; the notations @xmath182 and @xmath183 will be used by contrast in the case where the fast time and space variations have been factorized ab initio as in eq .
( ii.8a ) .
the connection between the two sets is given by @xmath184 and @xmath185
in the two following sections , we derive equations describing the evolution of average quantities such as the average amplitude @xmath186 , or the average intensity @xmath187 of wave @xmath188 . here
@xmath189 denotes the statistical average of the physical quantity @xmath190 , and its fluctuation is written as @xmath191 .
the meaning of a statistical average can in some sense be understood as a time average , and the use of a statistical framework is justified for the following reasons : a reduction of the parametric growth can be expected whenever the spectral width @xmath192 of the pump wave is large enough ( due to a temporal or spatial incoherence ) for its correlation time - as seen by the decay waves , in a sense to be defined in this section - to be shorter than the other characteristic times , namely , the damping and the growth time , @xmath193 and @xmath194 . in this case , the statistical description is justified whenever the correlation length in k - space of the pump wave @xmath195 is small compared to its spectral width ( e.g. , when the number of isi echelons or phase elements is large ) . in this limit the pump wave electric field can indeed be regarded as a stochastic process with a short correlation time and the standard statistical techniques can be applied . in the case of a purely temporal incoherence , the statistical description is physically justified only if the pump wave bandwidth is caused by some stochastic process , e.g. , the spread over several independent lines of a non - monochromatic laser . similarly , in the case of a purely spatial incoherence ,
the natural spread in angle of the pump wave wavenumbers must follow from the sum of many independent beamlets .
such a statistical independence may result from a random phase shift given to the different beamlets by means of a transparent phase mask[3 ] ; it may also be due to the scattering of the incident pump wave upon static random fluctuations .
lastly such a spatial incoherence can be achieved in the so - called isi technique[2 ] by a combination of delay increments @xmath196 given to the beamlets by echelon structures and of the laser temporal incoherence in the case @xmath197 .
in these techniques , the laser beam is broken up into a number of statistically independent `` beamlets '' that are brought to focus by an optic of f - number , @xmath198 . in the focal plane
these beamlets overlap creating a spatially nonuniform pattern with coherence length across the beam @xmath199 where @xmath200 is the half - angle of the optic , i.e. @xmath201 .
we will henceforth restrict ourselves to physical situations where the statistical independence between the pump wave fourier components is satisfied ; the pump wave electric field will thus be regarded as a stochastic variable with zero mean ; its statistical properties will be assumed to be entirely determined by its spectral density denoted as @xmath202 , the latter being itself characterized by its spectral width @xmath203 , that is to say by @xmath204 ( temporal incoherence ) and @xmath205 ( spatial incoherence ) . before deriving the statistical equations for
the time evolution of @xmath206 or @xmath207 , let us first consider on a physical basis the reduction of the parametric growth caused by the pump wave incoherence .
suppose that we consider the parametric coupling corresponding to a given angle @xmath143 between the pump wave and the first decay wave direction ; this angle defines the resonant wavenumbers @xmath208 and @xmath209 of the decay waves corresponding to the exact resonance condition ii.3 with the mean pump wave number @xmath150 . if one considers now a generic wavenumber @xmath210 of the pump wave , the resonance conditions ii.3 can no longer be satisfied for the same wave - numbers @xmath147 and @xmath211
one is thus led to define the resonance mismatches @xmath212 and @xmath213 as @xmath214 and @xmath215 when discussing the effect of the pump incoherence in the case of a given geometry defining the resonant wave numbers @xmath208 and @xmath209 , it is sufficient to consider the quantities @xmath216 and @xmath217 . in order to simplify our notation , we will denote in our paper by @xmath218 the value of a quantity @xmath219 in the case of the exact resonance @xmath220 and @xmath221 .
accordingly we define @xmath222 and @xmath223 by @xmath224 @xmath225 in the case of an incoherent pump wave , the spectral width @xmath226 gives rise to a characteristic size for @xmath227 , which will be denoted as @xmath228 , i.e. @xmath229 for instance , in the case where the envelope approximation can be applied , one readily finds @xmath230 which can also be written as @xmath231 or @xmath232 more generally , it has been shown[7 ] that it is sufficient to expand @xmath233 as follows : @xmath234 with @xmath235 . for backscatter ,
@xmath236 vanishes and it is necessary to expand to second order in @xmath135 . the coefficient of @xmath237 can be computed in terms of the waves parameters @xmath238 , and @xmath239 , and the angles @xmath240 defining the geometry of the interaction .
the first term in ( iii.5 ) corresponds to temporal incoherence and the last two to spatial incoherence . in the present article
, we will mainly restrict ourselves to some particular examples convenient for numerical experiments , in which the 2d character of the pump wave incoherence is modeled by the introduction of two parameters @xmath8 and @xmath9 , playing the role of the temporal and spatial incoherence , respectively , so that @xmath241 is similarly written as @xmath242 .
regarding now under which conditions the pump wave incoherence modifies the parametric coupling , one may first easily admit that a sufficient condition for neglecting the pump wave incoherence is that the two following inequalities be both satisfied : @xmath243 @xmath244 in the latter inequalities , @xmath245 denotes the dispersion relation @xmath246 in the envelope approximation , @xmath245 reduces to @xmath247 where , @xmath118 and @xmath248 denote the time and space growth rates .
( it can be noted that our definitions of @xmath245 and @xmath249 are consistent with each other , in the sense that if one considers _
ab initio _ the coupled mode equations in their envelope form ( ii.10 ) , @xmath245 reduces identically to @xmath249 without any further approximation : in this case , one has indeed @xmath250 , from which the relation @xmath251 follows . ) concerning the inequalities ( iii.6 ) , we will see more generally that the coherent or incoherent character of the parametric coupling is controlled by inequalities between @xmath252 and @xmath253 . on the other hand ,
since @xmath118 and @xmath248 are related through the dispersion relation , the size of @xmath254 depends upon the nature of the instability which is considered , namely convective , absolute , or spatially amplifying .
it therefore follows that the coherent or incoherent character of the parametric coupling depends itself on the nature of the instability .
defining now as `` incoherent '' the domain where the pump wave incoherence induces a reduction of the parametric growth , one realizes that such a domain should be specified as `` convectively incoherent '' , or `` absolutely incoherent '' , or `` incoherent for spatial amplification '' .
for instance , in the case of a purely temporal incoherence , with @xmath255 , it has been shown in ( laval , et.al . )
that the pump wave incoherence reduces the convective growth for @xmath256 and @xmath257 , whereas in the case of an absolute instability , the absolute growth can be reduced only if @xmath8 satisfies : @xmath258 [ for simplicity we consider the case where @xmath259 and @xmath260 .
] one thus sees with this simple case that the reduction of the absolute growth is much more difficult to achieve than that of the convective growth .
we may recall here that the convective growth rate characterizes the instability in the early stage of its development , by contrast with its long time evolution which is characterized either by the absolute growth rate ( whenever the conditions for their existence are satisfied ) or by the spatial amplification growth length ( in the opposite case ) .
in this article we will derive the conditions under which the pump incoherence gives rise to a reduction of the convective and of the absolute growth rates and an increase in the spatial amplification length .
it will be found that generally the conditions for reducing the absolute instabilities are more severe than those for the convective ones .
for this reason we will concentrate mainly our discussions on the `` convectively incoherent '' domain , which may thus be regarded as the largest domain in which a reduction of the parametric growth may be expected in the early stage of the instability development . in order to simplify our terminology , the `` incoherent '' domain , without any other specification , will henceforth refer to the previously defined `` convectively incoherent '' domain .
restricting ourselves to the convective instabilities , one may rewrite inequalities ( iii.6 ) as : @xmath261 @xmath262 in which we used @xmath263 in the coherent regime . as said before
, these inequalities represent only a _ sufficient _
condition for the pump wave incoherence being ignorable ; the question is thus whether both of these two inequalities have to be satisfied , or only one .
in order to answer this question , we first introduce in a brief discussion the validity conditions of the statistical equations which describe the time evolution of the quantities @xmath206 and @xmath207 .
a statistical description for the evolution of the decay waves 1 and 2 is provided by the random phase approximation ( rpa ) equations for the evolution of the average intensities @xmath264 .
the latter provide equations for the evolution on a slow time - or slow space - scale of the spectral densities @xmath265(x , t ) of the decay waves [ the latter functions @xmath266(x , t ) ] will be defined in the general case by eq .
( v.2 ) ; at this point it is sufficient to say that for a purely temporal problem one has the usual relation @xmath267 .
the rpa equations are nonlinear and they may account for the pump depletion ; the interested reader is referred to standard textbooks[72 ] for their derivation in the general case . in our problem of stability analysis , we face a linear problem ; we could naturally use directly the rpa equations on which we would make _ a posteriori _ the linear approximation which consists in neglecting the pump depletion . in this article
we will follow a different route .
we will first take advantage of the linear character of the coupled mode equations ii.6 which will enable us to use the so - called bourret approximation[70 ] for the evolution of @xmath206 ; on the other hand , the bourret approximation for the evolution of the intensities @xmath268 is easily tractable within the envelope approximation only , and then only for the special case where the pump wave group velocity @xmath131 is infinite . in the general case
it is necessary to make supplementary approximations to the bourret equations , and these approximations are justified in a domain that will be defined subsequently as the `` rpa domain . ''
returning now to the condition defining the domain in which the parametric coupling takes place coherently , we first define the temporal and spatial growth rates @xmath269 for @xmath270 , and @xmath271 for @xmath272 , according to : @xmath273 \eqno ( iii.9)\ ] ] @xmath274\ ] ] a factor 2 has been included in the argument of the exponential defining @xmath275 and @xmath276 so that the equality @xmath277 holds in the coherent case .
more generally the schwartz inequality yields the following result @xmath278 for i = 1 and 2 , in the case of a purely temporal growth ( and @xmath279 in the case of a spatial growth ) .
admitting that the pump incoherence may only reduce the parametric growth ( in the case of a convective instability in a homogeneous plasma ) , one realizes that , whenever either @xmath280 or @xmath281 behaves coherently , the two quantities @xmath282 and @xmath283 must also behave both coherently . in the next section
, it will be found that there exists a growth reduction for @xmath108 only if the mismatch @xmath284 satisfies the condition @xmath285 . from the previous considerations
, it follows that the wave energies @xmath286 will behave coherently whenever at least one of the two following inequalities is satisfied : @xmath287 or @xmath288 these inequalities define what is referred to henceforth as the `` convectively coherent '' domain .
accordingly , the `` convectively incoherent '' domain - or simply incoherent domain - is defined by the two conditions @xmath289 and @xmath290 on the other hand , the validity conditions for the rpa are very stringent .
in section v they will be found to consist of two sets of inequalities ; the first one corresponds to the inequalities opposite to ( iii.6 ) , namely @xmath291 @xmath292 and a second set involving the cross inequalities @xmath293 @xmath294 since the rpa dispersion relation yields @xmath295 for the convective growth rate , the latter inequalities reduce essentially to @xmath296 @xmath297 @xmath298 @xmath299 which can also be recast into the more compact form @xmath300 the latter inequalities ( iii.14 ) or ( iii.15 ) , define what is called the ( convective ) rpa domain .
one can now see that the incoherent domain ( iii.12 ) is subdivided into 1 ) the rpa domain in which the cross inequalities ( iii.14b ) and ( iii.14d ) are both satisfied , 2 ) the intermediate domain corresponding to the regime where the two inequalities @xmath301 and @xmath302 are both fulfilled whereas at least one of the cross inequalities ( iii.14c ) and ( iii .
14d ) is not satisfied . in this intermediate domain
there is no theory which is easily tractable for an explicit computation of the growth rate of the average energy @xmath303 . on the other hand ,
the bourret approximation makes it possible to calculate the growth rate of the average amplitudes @xmath108 and @xmath109 : it will then be found in section v that the larger growth rate of @xmath304 and @xmath305 is given ( for the convective instabilities ) , by the rpa prediction @xmath306 for @xmath307 and @xmath308 within a numerical factor no larger than two . for this reason we will argue in section vi that although the rpa equations are not in their range of applicability in the intermediate domain , the convective growth rate @xmath275 for the intensities remains of the order of @xmath306
accordingly , in the _ whole _ incoherent domain defined by the two inequalities : @xmath289 and @xmath290 the convective growth rate @xmath275 can be approximated by the rpa prediction @xmath306 .
the latter approximation is one of the main results of our paper since it makes it possible to compute the reduction of the parametric growth induced by the pump wave incoherence from the simple calculation of the rpa coupling constants @xmath309 and @xmath310 .
the bourret approximation[69,70 ] is a well known equation in the context of propagation in random media ; it deals with stochastic linear multiplicative equations of the form @xmath311 where @xmath312 is a linear deterministic operator , s a linear stochastic operator with zero mean , @xmath313 , and a is the physical quantity of interest .
the bourret approximation can be simply derived as follows : * by averaging eq .
( iv.1 ) one first obtains the exact relation @xmath314 with @xmath315 .
* by subtracting the latter from eq .
( iv.1 ) , one obtains the equation for evolution of the fluctuation @xmath316 @xmath317\ ] ] * the bourret approximation consists then in neglecting the so - called mode coupling term @xmath318 in the latter equation ; by doing so and neglecting the initial conditions , one obtains for the fluctuation @xmath316 the relation @xmath319 , where ( @xmath320 represents the green s function of the operator ( @xmath321 ) ; inserting the latter result into eq .
( iv.2 ) , one obtains the bourret equation for @xmath322 , namely @xmath323 defining @xmath324 as @xmath325 , it is natural to introduce the effective bandwidth @xmath326 as @xmath327 , and the validity condition for the bourret approximation can be expressed as @xmath328 the interested reader is referred to ref .
69 for a detailed discussion of these results . by applying the method outlined just above ,
one easily obtains the equation for evolution of the average amplitude @xmath108 .
consistent with ( ii 12b ) , we set @xmath329 and the dispersion relation corresponding to the slow time and space evolution for @xmath330 reads @xmath331 with @xmath332 where @xmath118 and @xmath248 stand for @xmath304 and @xmath333 , according to ( iii.9 ) . in deriving the latter equation , we neglected for simplicity the slow dependence of @xmath334 and @xmath335 upon @xmath133 and @xmath157 .
for the sake of clarity , we also restrict ourselves here to the case of exact resonance @xmath336 ; the general case @xmath337 will be considered in the next section in the rpa context ; ( consistent with our other notation , @xmath338 denotes the coupling constant @xmath339 evaluated at resonance ) . in deriving eq .
( iv.6 ) we made the envelope approximation for the slow space dependence only , by using @xmath340 ; on the other hand the computation of the resonance mismatch @xmath213 corresponding to a fast space dependence is not restricted to the envelope approximation ; lastly , the symbol @xmath341 in eq .
( iv.6 ) represents the usual prescription for the laplace transform contour .
the dispersion relation corresponding to the evolution of the average amplitude @xmath342 can be written in a similar way as @xmath343 with @xmath344 defining the effective bandwidths @xmath345 as @xmath346 and @xmath347 the validity condition of the bourret approximation for the average amplitude @xmath108 reads[16 ] , @xmath348 and @xmath349 the same naturally holds for the evolution of @xmath109 with @xmath350 . at this point
let us consider the so - called markov limit of the coupling constant @xmath351 ; the latter consists in taking the limit @xmath352 , and in using @xmath353 . in this limit
the coupling constant @xmath351 is given by @xmath354 where the superscript m stands for `` markov ; '' the same expression holds for @xmath355 with @xmath350 .
the latter two expressions for @xmath356 and @xmath357 are identical to the rpa coupling constants @xmath358 and @xmath359 to be derived later . at this point
it is sufficient to remark that whenever the markov limit can be taken , the orders of magnitude of @xmath360 and @xmath355 are @xmath361 @xmath362 these estimates follow simply from the definition ( iv-12 ) for @xmath355 and from the normalization condition @xmath363 . on the other hand it appears to be convenient , when discussing the continuity between the coherent and incoherent results , to rewrite the inequalities ( iii-12 ) limiting the incoherent domain in terms of the spectral widths @xmath364 and @xmath365 defined by the following relations @xmath366
@xmath367 according to the estimates ( iv.13 ) the order of magnitude of @xmath368 is given by @xmath369 it can also be seen that the quantities @xmath106 can be expressed as @xmath370 , so that the validity condition ( iv-10 ) for the bourret approximation for @xmath108 becomes , in the markov limit , @xmath371 .
the spectral widths @xmath368 are given in section vii for the examples of the pump wave correlation function corresponding to numerical solutions .
their general expressions are computed in ref . [ 7 ] in terms of the spectral width parameters @xmath9 and @xmath205 for the case of interacting wave- packets ; it will be seen that they can be expanded in a similar way as @xmath252 , namely @xmath372 where the parameters
@xmath373 , @xmath374 and @xmath375 can be expressed in terms of the waves parameters @xmath376 and @xmath377 .
the latter parameters are all of the same order of magnitude as the corresponding @xmath373 , @xmath374 and @xmath375 of the expression ( iii-2 ) for @xmath378 , although they may differ nonetheless by a numerical constant of order unity resulting from the averaging procedure over @xmath125 involved in eq .
( iii-23 ) , in all cases however the ordering @xmath379 holds and the quantity @xmath378 will henceforth be replaced by @xmath106 in the inequalities defining the domain of validity for the incoherent results .
moreover the convective growth rate , well above threshold , is easily found to be @xmath380 so that the validity conditions iv-10 and iv-11 of the bourret approximations for @xmath108 and @xmath109 in their markovian limits become simply @xmath381 @xmath382 the latter inequalities justify the conditions ( iii-12 ) , written in the previous section in terms of @xmath345 for simplicity , defining the incoherent domain , that is to say the domain where the bourret approximation is correct for the two average amplitudes @xmath108 and @xmath109 .
concerning now the validity condition for the markovian limit itself , made when deriving the coupling constant @xmath383 , there is no general criterion , except in the special case where the envelope approximation is correct - in this case the validity condition iii-27a for the bourret approximation for @xmath108 in the convective regime justifies _ a posteriori _ the markov limit @xmath384 .
this result is discussed in the next subsection ; it is shown in particular that in the case where the envelope approximation is valid , there is no intermediate regime between the incoherent domain iv-17 and the coherent domain ( iii-11 ) ; namely , either the bourret approximation is correct and the coupling constants take their markov limits @xmath385 in the incoherent domain , or the coherent results apply .
in addition , the two sets of results are continuous from one domain to the other . on the other hand in cases where the envelope approximation is not correct
, there is the possibility for an intermediate regime where neither the markov limit of the bourret approximation nor the coherent result apply ; such a situation is discussed in ref .
[ 7 ] . in order to illustrate the continuity between the coherent and the incoherent domains in the case where the envelope approximation is correct , we consider in the next sub - section the special case of a kubo - anderson process .
the kubo - anderson process ( kap ) is an example of a stochastic process for which the bourret approximation is exact , whatever the spectral width @xmath386 is . in the case of the coupled mode equations ,
it makes it possible to compute the coupling constant @xmath351 ( and therefore the growth rate ) as a continuous function of the spectral widths @xmath139 .
the interested reader is referred to the references[14,15,69 ] for an introduction to the kubo anderson process . in a one - dimensional geometry for the pump wave , and in the limit of the envelope approximation ,
the spectral density of the pump wave , when it is modeled by a kap , is given by @xmath387 performing the integration over @xmath133 in eq .
( iv-6 ) , one obtains @xmath388 with @xmath389 where @xmath365 is given by @xmath390 the quantity @xmath391 generalizes the quantity @xmath392 defined previously in a similar framework.[16 ] in the latter expressions , @xmath393 has simply to be replaced by @xmath8 in the case of a pump wave with an infinite group velocity @xmath394 ; this limit corresponds to a model in which the pump wave fluctuates in time only , i.e. where the function @xmath395 in the envelope equation ii-9 is a function @xmath396 of time characterized by a correlation time @xmath397 .
lastly , in order to make a connection with the previously introduced definitions , the expression ( iv.20 ) for @xmath365 can be seen to correspond to @xmath398 , following our general expression ( iv-16 ) for @xmath365 ( in the case of a one - dimensional geometry for the pump , one has naturally @xmath399 ) .
the quantity @xmath400 is defined in a similar way by substituting ( @xmath401 ) . in order to discuss the continuity between the coherent and the incoherent domains , it is convenient to introduce as before the quantities @xmath249 defined by @xmath402 the dispersion relation for @xmath108 is thus @xmath403 where for @xmath109 it is @xmath404 as stated before , these two dispersion relations are exact for a kap and they make it possible to investigate the behavior of the average amplitude @xmath206 as a function of the spectral width @xmath405 .
the first remarkable point is , as first noticed by thomson and karush[14 ] , that the two dispersion relations ( iv-22 ) are not identical . due to phase mixing , @xmath206
does not necessarily behave like @xmath406 , nor @xmath108 like @xmath109 . for simplicity we now restrict ourselves to the convective instability in our discussion concerning the behavior of the average amplitudes @xmath108 and @xmath109 . after setting @xmath407
, one can check directly , using the exact dispersion relation ( iv-22a ) for @xmath108 , what has been announced just above .
first , the validity condition @xmath408 of the bourret approximation necessarily reduces to the condition @xmath409 , and justifies _ a posteriori _ the markov limit @xmath410 .
since the latter condition @xmath411 is itself equivalent to @xmath412 , it is natural to define the incoherent domain for @xmath108 as the domain @xmath413 .
( the incoherent domain for @xmath109 corresponds naturally to the domain @xmath414 . )
second , there is no intermediate regime between the coherent domain @xmath415 and the incoherent domain for @xmath108 in the sense that there is no discontinuity between the usual dispersion relation of the coherent case @xmath416 and the one corresponding to the markov limit of the bourret approximation @xmath417 ; consequently the growth rate and the threshold are continuous from one domain to the other .
the same naturally holds for @xmath109 .
what can we infer regarding the behavior of the intensities @xmath418 from the results concerning the average amplitudes ?
this problem has already been investigated by several authors[14 - 16 ] . as explained above , from the inequality @xmath419
, one may assert that the system behaves necessarily in a coherent way in the coherent domain @xmath420 or @xmath421 . on the other hand ,
in the complementary domain defined previously as simply `` the incoherent domain , '' one may only use the latter inequality : first , it follows that a lower bound for the threshold for the intensities is given by the lowest threshold for the average amplitude , namely @xmath422 \eqno ( iv.23)\ ] ] secondly the growth rate @xmath275 satisfies the inequality @xmath423 \eqno ( iv.24)\ ] ] which ,
well above threshold , reduces to @xmath424 in a domain where @xmath425 for i = 1 and 2 .
the question is thus whether these lower bounds are actually attained in which case it would be sufficient to simply consider the most unstable average amplitude . in order to answer the question we first have to derive statistical equations for the wave intensities , which we undertake in the next section .
the bourret technique can be applied exactly in the limit @xmath394 in order to derive statistical equations for the wave intensities . in the case of a kubo anderson process
these equations are exact for any spectral widths @xmath8 , so that they make it possible to investigate the limit in which they take the form of the rpa equations . for the general case @xmath426
additional approximations have to be made to the bourret - like equations in order to obtain the rpa equations .
let us consider the coupled mode equations in their envelope form ii-9 , in which the stochastic function @xmath428 is a kubo - anderson process @xmath396 ; the latter is characterized by a correlation function of the form @xmath429 , corresponding to a spectral density in frequency space given by @xmath430 in order to derive the bourret equations for the wave intensities , following ref .
[ 16 ] , we consider the two point correlation function @xmath431 with i and j = 1 , 2 .
the spectral densitites are defined as @xmath432 where we allow for a slow time and space variation of the spectral density @xmath433 . in order to be consistent with the previous definition ( iii.9 ) of growth rates ,
we set @xmath434 where @xmath118 and @xmath435 stand now for @xmath436 and @xmath437 .
the bourret approximation is easily derived for the set @xmath438 to give the following system @xmath439 \eqno ( v.4a)\ ] ] @xmath440 \eqno ( v.4b)\ ] ] where @xmath249 is defined as before as @xmath441 ; the coupling constants @xmath442 are given by @xmath443 and the superscript stands for `` bourret '' . * let us first consider the markov limits of the latter expressions and show that they exactly reproduce the rpa results .
the markov limit of @xmath442 corresponds to the domain @xmath444 , in which case the coupling constants @xmath442 take the following form @xmath445 + on the other hand it will be seen in the next subsection that the rpa equations take the general following form @xmath446 , \eqno ( v.7a)\ ] ] @xmath447 .\eqno ( v.7b)\ ] ] + the quantity @xmath448 , consistent with our notations for the primed quantities , is given by @xmath449 where @xmath450 has previously been defined as @xmath451 in the envelope approximation limit @xmath452 reduces to @xmath453 [ it may be noted again that these definitions of @xmath454 and @xmath455 are consistent with each other in the sense that , if one considers _
ab initio _ the coupled mode equations in their envelope form , @xmath450 reduces identically to @xmath456 without any further approximation where @xmath457 .
the same definition holds for @xmath458 with the substitution ( @xmath401 ) .
+ in the somewhat degenerate limit @xmath394 considered in this subsection , @xmath459 has to be replaced by @xmath460 , corresponding to the limit @xmath461 in @xmath462 ; accordingly , in the integral appearing in the rhs of eq .
( v.7 ) , the quantities @xmath463 and @xmath464 have to be replaced by @xmath465 and @xmath466 respectively where @xmath467 denotes @xmath468 in the limit @xmath469 with @xmath470 .
one easily finds @xmath471 and ( @xmath472 ) for @xmath473 .
it is also only in the same limit @xmath474 that the integration over @xmath475 in eq .
( v-7 ) does not involve the spectral densities @xmath476 and @xmath477 ; the rpa equations take then a simpler form , namely @xmath478\ ] ] @xmath479 \eqno ( v.11)\ ] ] where the coupling constants are given by @xmath480 @xmath481 + performing the integration over @xmath482 in the latter expression , one finds @xmath483 + the rpa equations correspond thus exactly to the markov limit of the bourret approximation for the intensities @xmath484 .
* we now consider the dispersion relation corresponding to the exact bourret equations ( v-4 ) .
taking @xmath485 in eq .
( v-4b ) , one obtains @xmath486 the latter dispersion relation generalizes to three dimensions the results obtained previously by laval et al.@xmath487 following their analysis , one may remark that the coupling constants @xmath488 reach their maximum @xmath489 for @xmath490 , corresponding to the exact resonance conditions @xmath491 and @xmath492 for the two wave packets considered here .
in this case the dispersion relation ( v-14 ) reads simply @xmath493 from which it follows that the dispersion relation @xmath494 of the coherent case is recovered for @xmath495 .
the opposite limit @xmath496 corresponds to the validity domain for the markov approximation , and therefore for the rpa equations .
there is thus no intermediate domain between the coherent and the rpa ones in the case @xmath394 .
this result can be tracked back to the fact that for @xmath394 , one has @xmath497 , as can be seen from the general definition ( iv-20 ) for @xmath368 , so that the cross inequities ( iii-14c ) and iii-14d ) are automatically satisfied in the incoherent domain ( iii-12 ) , and consequently the rpa equations are exactly applicable in the whole incoherent domain .
to summarize , the parametric coupling takes place coherently in the coherent domain @xmath498 whereas in the complementary - or incoherent - domain @xmath499 the coupling constants @xmath500 may be reduced to their rpa values @xmath501 for i = 1 and 2 .
the corresponding dispersion relation is then simply @xmath502 * regarding now the comparison between the average amplitudes and the intensities , one may easily obtain the threshold and the growth rate for the intensities from the exact dispersion relation iii-14 .
+ the threshold reads @xmath503 ; it reduces to the coherent result @xmath504 for @xmath505 and to the rpa result , namely @xmath506 in the opposite case .
these expressions can be easily checked to be identical to the ones obtained from consideration of the average amplitudes , i.e. by looking at the most unstable average amplitude , as explained in section iii-2 - 2 .
+ concerning now the growth rates , the growth rate for the intensities is given , well above threshold , by the usual coherent expression @xmath507 for @xmath508 and by the rpa one @xmath509 in the opposite case .
it is interesting to notice that the correct value for the growth rate ( @xmath510 ) of the intensities is exactly _ twice _ the lower bound ( iv-25 ) obtained from consideration of the average amplitudes .
the square amplitudes behave indeed as @xmath511 , whereas the rpa equations predict in the same incoherent regime @xmath512 .
this result is most easily understood by remembering that , in a stochastic process , such as this one , no two realizations are exactly the same , and there is probability associated with a given mode s history in amplitude space . given that probability distribution function , one can compute the average mode amplitude or the average mode intensity .
if the distribution is narrow , then the intensity growth rate will be close to twice the amplitude growth rate .
however , in this case , the fluctuations about the mean are comparable to the mean in the incoherent limit so that higher order moments of the mode amplitude will have a more than proportionately larger growth rate .
this facet of the problem is discussed more completely in sec .
this result is not unexpected and is actually a general feature of the fully incoherent regimes : in an incoherent regime , the growth rate of an average amplitude is indeed given quite generally by @xmath513 , where @xmath3 is the coherent growth rate and @xmath514 the correlation time of the stochastic process .
since the coherent growth rate of the quantity @xmath515 is @xmath516 , we deduce that the growth rate of the average intensity is given by @xmath517 , corresponding thus to twice the lower bound @xmath518 obtained from eq .
more generally it will be argued in section vi that the thresholds and convective growth rate regarding the intensities can be obtained quite accurately from the results corresponding to the average amplitudes , whenever one has @xmath519 or @xmath520 , and at worst within a factor of two in the case @xmath521 . in this subsection
we derive the rpa equations in the general case from the bourret equations for the intensities . following the definition ( v-2 ) for the spectral densities , we set @xmath522 and we laplace transform in space the slowly varying spectral density @xmath523 ( * x * , t ) by setting @xmath524 .
the quantity @xmath525 is easily checked to be given by @xmath526 the equation for the evolution of @xmath527 can be obtained from the coupled mode equation ( ii.7a ) and reads @xmath528 a_{{\bf k } _
1+i{\bf \kappa}}^{(1 ) } \left ( a_{{\bf k}_1-i { \bf \kappa}}^{(1 ) } \right)^*\ ] ] @xmath529 @xmath530 \eqno ( v.21)\ ] ] where we neglected as before the slow variations of @xmath531 and @xmath532 upon @xmath133 and @xmath533 .
similar equations can be written for the quantities @xmath534 so that the complete set for the quantities @xmath535 has the form of a stochastic linear multiplicative system ( where the random variable is the set @xmath195 ) , upon which the bourret approximation can be performed .
it is convenient at this stage to laplace transform in time the resulting equation by setting @xmath536 and similarly for the other quantities .
the bourret equations take the form @xmath537 \left [ n_{{\bf k}_1}^{(1 ) } \left ( { \bf \kappa } , \gamma \right ) + n_{{\bf k}_0-{\bf k}_1}^{(2 ) } \left ( { \bf k } , \gamma \right ) \right ] \eqno ( v.22)\ ] ] where the resonance function @xmath538 is given by @xmath539 and where the operator @xmath540 denotes the hermitian part of @xmath541 , i.e. @xmath542.\ ] ] in the general case @xmath543 is given by @xmath544 which reduces to @xmath545 in the envelope approximation . as a matter of fact
, the envelope approximation can be made for the _ slow _ variation associated with the variables @xmath118 and @xmath435 ; thus the @xmath546 and @xmath547 appearing in the definition of @xmath548 can be reduced to @xmath549 and and @xmath550 respectively , and similarly the operator @xmath543 can be replaced by @xmath551 ( by contrast the quantity @xmath552 contains the fast dependence on the integration variable @xmath133 and can not be reduced to its envelope approximation in the general case . ] making the envelope approximation for the slow dependence , one obtains the following set of equations .
@xmath554 where the `` diagonal '' coupling constants @xmath442 are given by @xmath555 with @xmath556 and ( @xmath557 ) for @xmath558 .
the set ( v.25 ) constitutes the bourret equations for the intensities @xmath559 and @xmath560 .
this set is still very difficult to solve since , in addition to the convolution integral on the rhs , it contains various quantities which are non - markovian .
the coupling constants @xmath442 can be further simplified in the markov approximation , denoted by @xmath561 . in this approximation ,
i.e. @xmath562 can be replaced by @xmath563 so that the system ( v.25 ) becomes @xmath564 \eqno ( v.28a)\ ] ] @xmath565 \eqno ( v.28b)\ ] ] the latter set constitutes the rpa equations for the coupled mode equations neglecting pump depletion and written in the laplace variables @xmath118 and @xmath435 . in the real space the operators @xmath566 have simply to be replaced by @xmath567 in order to make a closer connection with the special case considered in the previous subsection ( v.7 ) , one can easily check that the latter set reduced identically to the system ( v.7 ) , when dealing with primed quantities ( i.e. after factorization of the fast variation ) .
one may also remark that the mismatch @xmath568 reduces to @xmath569 in the envelope approximation , in which limit the set ( v.28 ) can be recognized to be the 3-d generalization of the laval et al .
results.@xmath487 the validity condition for taking the markov limit of the bourret equations ( v.25 ) is min @xmath570 . on the other hand , by using the fact that well above threshold the growth rate is of the order of @xmath571 , one easily finds that the latter inequality reduces to @xmath572 that is to say the condition ( iii.15 ) previously written in terms of @xmath573 .
this condition entails in turn the inequality @xmath574 which ensures the validity of the bourret approximation .
therefore , the condition ( v.29 ) defines entirely the rpa domain .
the rpa set ( v.28 ) remains a system difficult to solve due to the integration over @xmath125 on the rhs .
one can however make some additional approximations which are consistent with the validity condition ( v.29 ) . by assumption the spectral density @xmath575 of the pump wave peaks when @xmath133 is equal to the mean wave number of the pump wave @xmath136 ; consequently , for a for a given angle of observation defining the direction of the scattered wave , the rpa coupling constant @xmath576 is maximum for @xmath220 , corresponding to the exact resonance . in this case
@xmath309is given by @xmath577 corresponding to the result announced in section iv , namely @xmath578 .
if one now takes @xmath220 in eq .
( v.28a ) the relation @xmath579 defines a surface in @xmath148 space that contains the point @xmath580 would require @xmath581 uniquely ; as a matter of fact , the function @xmath202 attains its maximum at this point .
due to the large spectral width imposed by the rpa validity conditions ( v.29 ) , the function @xmath582 is a slowly varying function of @xmath133 , so that one does not make a significant error in the integral appearing in the rhs of eq .
( v.28a ) by replacing @xmath582 with @xmath583 ( an approximation which is actually exact in 1d ) , and similarily @xmath584 by @xmath585 in eqs .
( v28b ) , and obtaining @xmath586 @xmath587 where the rpa coupling constants @xmath588 are given by @xmath589 and @xmath590 according to our definition ( iv.14 ) .
the quantities @xmath591 denote , for the for the sake of simplicity in the notations , the spectral density @xmath592 at the point of exact resonance @xmath593 .
equations ( v.32 ) provide the desired system with which the stability of the intensities @xmath594 is analyzed in the remaining of the paper .
in this section we derive the expressions of the convective and absolute growth rates for the parametric instabilities as a function of the two quantities @xmath364 and @xmath365 characterizing the pump incoherence ; we also compute the rate of spatial amplification .
let us first recall the physical meaning of these various growth rates by considering the green function , that is to say the response of the parametric coupling to an infinitesimal impulse given to the system at t = 0 and x = 0 . for simplicity
we restrict ourselves to a one dimensional problem , and we denote as @xmath595 the point where the green function g(x , t ) is maximum . for an infinite system , the _ convective _ growth rate , denoted as @xmath596 , is defined to be the time growth rate of @xmath597 , i.e. @xmath598 quite generally the point @xmath595 moves in time according to @xmath599 , where @xmath600 is the group velocity of the maximum . for a finite system ,
the convective growth rate characterizes therefore the transient behavior only ; although the green function @xmath601 for a plasma slab of length l does not reduce to the green function @xmath602 of the infinite case , it is usually admitted that the infinite system convective growth rate @xmath596 does properly describe the early time behavior for t@xmath603 .
regarding now the long time behavior of the parametric coupling , one has first to look for the existence of absolute instabilities .
for an infinite system , when such instabilities exist , the _ absolute _ growth rate is the time growth rate of @xmath604 , i.e. @xmath605\ ] ] for finite values of @xmath606 ; in the latter expression @xmath607 denotes the space growth rate associated with the absolute instability . for a finite system , there exists in general a critical length labs above which there are unstable normal modes , so that the system will behave asymptotically in time as the most unstable normal mode .
similarly it is usually admitted that whenever the plasma length l satisfies the condition @xmath608 , the growth rate of the most unstable normal mode is well approximated by the absolute growth rate @xmath609 of the infinite case .
when absolute instabilities do not exist but the system is convectively unstable , the long term behavior is determined by spatial amplification .
the latter corresponds to setting a source @xmath610 at x = 0 and looking for the @xmath611 which maximizes the _ spatial amplification _ rate @xmath612 ; bers and b riggs
have indeed shown that in such a case the time asymptotic behavior of an infinite system in the presence of the source goes like @xmath613_{\mid x \mid \rightarrow \infty } \sim { \rm exp}~ \left(\epsilon \kappa ( \omega_s ) x\right)\ ] ] where @xmath614 , depending upon the direction for spatial amplification .
such behavior also indicates that a significant spatial amplification may be expected in a finite plasma slab whenever the condition @xmath615 is fulfilled , where @xmath616 denotes the maximum @xmath617 .
lastly there is also the possibility for a mixed situation where there is the coexistence of an absolute instability and spatial amplification .
indeed such a case will be seen to be provided by the r.p.a . dispersion relation . in this case
the short and long time behaviors are still given by the convective and absolute growth rates , for the reasons explained just above . in the case
however where the spatial amplification growth rate @xmath618 is significantly larger than the space growth rate @xmath619 associated with the absolute instability , one may expect the existence of an intermediate regime in the time evolution during which the spatial amplification feature dominates that of the absolute instability . to begin with , let us first justify our conjecture concerning the applicability of the rpa results in the intermediate domain where neither the coherent nor the rpa equations could _ a priori _ be used .
to do so we compare the rpa predictions with those for the average amplitudes . the dispersion relation corresponding to the rpa equations v.32 . is @xmath620 and the dispersion relation for the average amplitude @xmath621 eq .
( iv.5 ) is @xmath622 where @xmath623 denotes the rpa coupling constant @xmath624 at exact resonance , namely @xmath625 and @xmath626 .
the spectral widths @xmath59 are defined by eq .
( i-10 ) and ( iv.20 ) .
as explained in section iii , the intermediate domain is defined as the domain where equations ( vi.2 ) are correct for the two waves @xmath188 = 1 and 2 ( i.e. the diagonal inequalities @xmath627 are both satisfied ) whereas the rpa equations are a priori not applicable ( because one of the cross inequalities @xmath628 , with @xmath629 , is not satisfied ) . in this domain
one can only use the inequality ( iii.10 ) , namely max@xmath630 which states that the actual growth rate @xmath436 is no less than the most unstable average amplitude ( the latter being reduced as compared with the coherent growth rate since eq .
( vi.2 ) predicts a reduction of the parametric growth in its domain of applicability ) .
the question is thus whether this lower bound is attained or whether @xmath631 remains of the order of the coherent growth rate . to answer this question ,
it is natural to first compare the rpa predictions with those for the average amplitudes in their common domain of applicability , namely the rpa domain .
it will be seen in the next subsection that they are identical with regard to the convective threshold and convective growth rate to within a numerical factor no larger than two .
the reason for this identity can be tracked back as follows .
the rpa dispersion relation vi.1 can be approximated by @xmath632 if @xmath633 or by @xmath634 if @xmath635 , so that the two quantities @xmath636 have necessarily a growth rate @xmath637 which is of the order of max @xmath638 .
more precisely , denoting by @xmath639 the growth rate obtained from the dispersion relation vi.1 , and by @xmath640 the largest convective growth rate given by the bourret dispersion relation vi.2 , it is the matter of a simple calculation to show that the following inequalities @xmath641 hold formally , i.e. independently of the applicability of the rpa and bourret equations ( the upper inequality occurs in the case @xmath642 ) .
since in the rpa domain one has @xmath643 and max @xmath644 , one deduces that the latter inequality can be written @xmath645 in the rpa domain .
this result means that the phase mixing effects on @xmath646 are negligible , at least for the most unstable average amplitude .
since this property is satisfied in the rpa domain , which can be regarded as the most incoherent domain , i.e. the domain in which the phase mixing effects are expected to be the more important , we make the conjecture that the inequality ( vi.4 ) is satisfied in the intermediate domain as well . on the other hand , in this domain
one has max @xmath647 , so that this conjecture means @xmath648 ; on comparing this inequality with ( iv.3 ) , one realizes that one can approximate in the intermediate domain , the actual growth rate @xmath649 by the rpa prediction @xmath639 , with an error no larger than a factor of two .
as anticipated in section iii , these arguments allow us to apply the rpa results in the entire incoherent domain @xmath650 and @xmath651 including the intermediate domain .
the threshold @xmath652 for the convective instabilities is easily obtained from the rpa dispersion relation ( vi.1 ) ; it reads @xmath653 ^{1/2 } \eqno ( vi.5)\ ] ] and can be approximated by @xmath654 ^{1/2 } \eqno ( vi.6)\ ] ] it is easily checked that the latter approximation corresponds to what can be deduced from the behavior of the average amplitudes , [ namely from the condition @xmath655 .
we may also notice that the expression vi.6 for the threshold in the incoherent domain goes continuously into the usual coherent threshold expression @xmath656 at the boundary between the two domains , namely @xmath657 and @xmath658 , or @xmath659 and @xmath660 ( again within a numerical factor no larger than two ) . the convective growth rate , well above threshold ,
is given by @xmath661 which is of the order of @xmath662 , again within a factor no larger than two [ one has @xmath663 for @xmath664 only , which occurs for instance in the case @xmath665 encountered in section iii.2 ] .
one can again show that the expression for the growth rate in the incoherent domain goes continuously into the coherent one @xmath666 at the boundary @xmath667 , and @xmath668 and @xmath669 ( once more within a numerical factor no larger than two ) .
the absolute growth rate is obtained from the rpa dispersion relation vi.1 by looking for a double root for @xmath248 and by imposing the well - known bers and briggs criterion , i.e. tracing the roots in the k - plane to insure that the double root arises from roots coming from opposite sides of the real k axis as one varies the growth rate from @xmath670 to @xmath671 .
one finds that this occurs for @xmath96 and that the absolute growth rate is given by @xmath672 \left [ \left ( { v_1 \over \delta \omega_1 } \right)^{1/2 } + \left ( { v_2 \over \delta \omega_2 } \right)^{1/2 } \right]^2 \eqno ( vi.8a)\ ] ] where @xmath673 denotes the absolute threshold @xmath674 ^ 2 } .
\eqno ( vi.9a)\ ] ] well above threshold the absolute growth rate can be approximated by @xmath675 and the absolute threshold is of the order of @xmath676 concerning now the domain of validity of the rpa expressions for the absolute instabilities , let us first consider the case @xmath677 ; in this case the cross inequalities ( iii.13c - d ) reduce to @xmath678 , so that it would appear that the rpa validity domain is given by the inequalities @xmath679 with a = 1,2 , where @xmath680 is computed for @xmath681 ( with @xmath248 evaluated at its double root @xmath682 ) .
actually a detailed calculation[46 ] performed with the exact dispersion relation ( v.15 ) for the case @xmath394 , shows that the limiting rpa expressions are applicable in a smaller region , obtained by imposing the rpa validity condition not at the single point ( @xmath683 ) but on the entire path needed to apply the bers and briggs procedure : in fact one requires that @xmath684 for @xmath118 varying from the absolute growth rate @xmath670 to the value at which one of the two roots @xmath685 crosses the real axis .
we assume such a procedure is also required in the the general case @xmath686 .
one then finds @xmath687 and @xmath688 , so that the condition @xmath689 reduces to the simple inequality @xmath690 on the other hand , by looking at the behavior of the average amplitudes @xmath691 , one easily finds that in the complementary domain @xmath692 , the parametric system behaves coherently for the absolute instabilities and the growth rate reduces to the coherent limit given by @xmath693 @xmath694 for this reason we can define the incoherent domain for the absolute instabilities by the inequality ( vi-10 ) . with regard to the two cross inequalities @xmath695 and @xmath696 ,
it can be seen that , because they are not necessarily fulfilled when the condition vi.10 is satisfied , there is again the possibility for an intermediate domain in which neither the rpa nor the coherent results apply for the absolute instabilities .
one can verify however that the single validity condition vi-10 ensures the continuity ( within a factor 2 ) of the rpa expression ( vi-8 ) for the absolute growth rate with the usual coherent expression @xmath697 .
for this reason we conjecture that the rpa expression ( vi.8 ) is actually applicable in the whole incoherent domain .
as already noticed[16 ] , it is interesting to remark that the behavior of @xmath698 and @xmath699 are significantly different as far as absolute instabilities are concerned : in the incoherent domain ( vi.10 ) the average amplitudes are essentially stable , whereas the intensities are absolutely unstable as long as @xmath3 exceeds the threshold ( vi.9 ) . in conclusion , therefore , the absolute growth rate is given by the minimum of @xmath700 given by eq .
( vi.8a ) and the coherent growth rate @xmath701 given by eq .
( vi.11 ) .
the spatial amplification growth rate @xmath616 is obtained from the rpa dispersion by taking @xmath702 , with @xmath703 real , and by looking for a maximum of @xmath704 which satisfies the bers and briggs criterion[75 ] for spatial amplification namely , the root @xmath248 in the @xmath248-plane has to cross the imaginary axis as one varies @xmath118 from @xmath705 to @xmath706 .
although the procedure is straightforward , the general expressions of the different unstable roots are tedious due to the large number of the independent parameters @xmath707 . for the sake of simplicity we will restrict ourselves in this subsection to the cases where the following ordering is satisfied : @xmath708 @xmath709 @xmath710 such an ordering is always satisfied in the case of srs and sbs instabilities and quite often for two plasmon decay .
it is easy to find that in the above limit there are two roots corresponding to spatial amplification .
the spatial amplification growth rate for the first root is given by @xmath711 and its domain of existence corresponds to the inequalities @xmath712 @xmath713 @xmath714 @xmath715 the inequalities ( vi.14 ) are the rpa validity conditions , whereas the inequalities ( vi.15 ) represent the condition for satisfying the bers and briggs criterion , i.e. , the condition for spatial amplification ( this condition implies in particular the inequality @xmath716 ) . the spatial amplification growth rate for the second root is given by @xmath717 and its domain of existence corresponds to the following inequalities : @xmath718 @xmath719 @xmath720 @xmath721 again where inequalities ( vi.17 ) are the rpa validity conditions for this root , and as the inequality ( vi.18 ) is the bers and briggs condition for spatial amplification .
it is interesting to remark that the inequalities ( vi.17c ) and ( vi.18 ) , reduce exactly to the conditions ( vi.10 ) and ( vi.9b ) for the existence of absolute instabilities in the rpa regime , in the case @xmath96 and where the ordering ( vi.12 ) holds . in this case one
is thus in a mixed situation where an absolute instability and spatial amplification coexist together .
it can also be easily shown that the ordering ( vi.12 ) implies the inequality @xmath722 , so that spatial amplification may dominate the absolute instability for an intermediate stage in time or for an unbounded plasma .
lastly , it is worth mentioning that the two roots @xmath723 and @xmath724 can be interpreted _ a posteriori _ as corresponding to spatial amplification for the average amplitudes @xmath725 and @xmath281 respectively
. for instance the expression for @xmath724 follows simply from @xmath726 , the inequalities ( vi.17 ) are simply the markov condition @xmath727 , and the condition ( vi.18 ) is the condition for instability for the average amplitude @xmath281 .
we may therefore conclude that the main features corresponding to spatial amplification - and to the convective growth rate - for the intensities @xmath728 may be obtained from consideration of the average amplitudes @xmath646 , in contrast with the absolute instabilities the features of which can be found from the dispersion relation for @xmath729 only .
in this subsection we repeat the various stability domains obtained in the previous subsections in the special case @xmath394 , for which one has @xmath730 , in order to exemplify the different kinds of stabilization that can be expected from the pump wave incoherence .
to do so , it is convenient to introduce the following dimensionless quantities @xmath731 @xmath732 @xmath733 we assume the ordering ( vi.12 ) to hold , and in addition we suppose @xmath734 so that we restrict ourselves to the situations where the two inequalities @xmath735 @xmath736 are satisfied . in fig .
2 we reproduce the stability diagram corresponding to the reduction of the convective instability growth rate induced by the pump wave incoherence . this diagram has to be understood as representing the effect of pump bandwidth @xmath8 as a function of @xmath737 , for @xmath3 and @xmath734 given . as explained in the introduction to this section , the convective instability represents the early or transient time behavior of the parametric coupling . the domain of applicability of the statistical theory corresponds to the inequalities @xmath738 @xmath739 that is to say , @xmath740 @xmath741 the domain corresponding to these inequalitites is referred to as `` vr '' in fig . 2 , and stands for `` convective growth rate reduction ; ''
the solid line presents the boundary between the coherent and incoherent domains .
the threshold vi.5 for convective instabilities is approximately @xmath742 , so that the pump wave incoherence induces a complete stabilization of the convective growth rate in the domain referred to as `` vs '' in fig .
2 , @xmath743 the boundary of the convectively stable region is the dotted line as obtained from eq .
lastly the parametric system is coherently stable in the domain @xmath744 ( corresponding to the usual coherent condition for stability @xmath745 ) , the boundary of which is represented by the dashed - dot - dot line . in fig .
3 , we reproduce the stability diagram corresponding to the reduction of the absolute instability growth rate due to the pump incoherence , i.e. the growth reduction of the long time behavior of the parametric coupling in the case of existence of absolute instabilities .
the domain of applicability of the statistical theory of absolute instabilities ( vi.10 ) reads here @xmath746 ; on the other hand the threshold ( vi.9b ) is @xmath747 .
the intersection between the two lines @xmath748 and @xmath749 corresponds to @xmath750 , which is smaller by a factor two than the usual coherent threshold for absolute instabilities , namely @xmath751 .
since the applicability condition for statistical theory has been obtained from strong inequalities ( whereas the threshold condition @xmath752 results from solving the rpa dispersion relation ) the two limits could be reasonably made continuous between each other by writing the rpa validity condition as @xmath753 for @xmath754 . by doing so
, the domain corresponding to a reduction of the absolute growth rate ( henceforth referred to as the `` ar '' domain ) would be approximated by the inequality @xmath755 and the domain for complete stabilization of the absolute growth ( referred to as the `` as '' domain ) would be given by : @xmath756 actually a more detailed calculation[46 ] based upon the exact dispersion relation ( vi.15 ) shows that the ar domain corresponds to the inequality @xmath757 for @xmath758 , and that the complete stabilization occurs when inequality ( vi.23a ) is satisfied in the range @xmath758 , and when the inequality @xmath759 is fulfilled in the range @xmath760 . in fig .
3 the boundary of the ar domain is represented by a short - dashed line , and the boundary of the as domain , computed from the incoherent threshold vi.9a , by a long - dashed line .
one may see that the two boundaries almost coalesce at @xmath761 , @xmath762 ; this simply means that the ar domain shrinks to zero in the limit @xmath763 for @xmath764 ( for this reason the short - dashed line , the long - dashed line , and the verticle dashed- dot - dot line bound a very small region for @xmath765 ) .
the verticle line is the absolute coherent threshold eq .
vi.11b .
it can be easily checked that the domains corresponding to the approximate expression ( vi.22a ) and ( vi.23a ) do not significantly differ from the exact domains ( vi.22b ) and ( vi.23b ) , in the range @xmath766 , i.e. far from the absolute threshold .
this result justifies in particular all the approximate expressions obtained in the remainder of our paper . on the other hand ,
in the range close to the absolute threshold , namely for @xmath767 , this exact solution exhibits the additional as domain @xmath768 , corresponding to a regime where a small @xmath139 may reduce the absolute growth rate .
such an effect is not really surprising since it means that a small frequency bandwidth may have a large effect when the parametric coupling is close to the coherent threshold , namely it converts an absolute instability into a convective one ( which in turn gives rise to a spatial amplification , as it will be seen in vi.d3 ) , by decreasing the usual coherent absolute threshold @xmath769 into the absolute threshold ( vi.23b ) modified by incoherency effects , namely @xmath770 . in fig .
4 , we reproduce the stability diagram corresponding to the reduction of the spatial amplification growth rate @xmath616 . the first root ( vi.13 ) for the spatial amplification growth rate is given in terms of dimensionless quantities as @xmath771 where the dimensionless growth rates @xmath772 are defined as @xmath773 .
the domain of existence of this root , referred to in fig . 4 as `` ar1 '' ( for `` spatial amplification reduction '' ) , is defined by inequalities ( vi.14 ) and ( vi.15 ) , namely @xmath774 @xmath775
@xmath776 the boundary of this domain is represented by the dotted - dashed line in fig .
the threshold condition ( vi.15 ) can be written as @xmath777 the boundary of this domain `` as '' is given more exactly by the dotted line as in fig .
the spatial amplification growth rate for the second root is @xmath778 and its domain of existence , referred to in fig . 4 as `` ar2 , '' corresponds to the inequalities ( vi.17 ) and ( vi.18 ) , i.e. @xmath779 @xmath780 in addition to inequalities in ( vi.25a ) and ( vi.25b ) .
the boundary of this domain is represented by the decreasing dotted - dashed line and the dashed line in fig .
it is interesting to remark that in the domain @xmath781 , the coherent spatial amplification growth rate is @xmath782 so that @xmath772 varies continuously from its coherent value to @xmath783 , along the boundary @xmath784 ( the dotted - dashed line for @xmath785 ) , whereas @xmath786 is strongly reduced from @xmath787 to @xmath788 along the boundary @xmath789 ( the dotted - dashed line for @xmath790 ) .
similarly , in the domain @xmath791 , and @xmath792 varies continuously from its coherent expression @xmath793 to @xmath794 along the boundary @xmath795 between the coherent domain and the ar2 domain , whereas @xmath772 is strongly reduced from @xmath796 to @xmath788 along the boundary @xmath797 between the ar2 and ar1 domains . let us consider now the last case , corresponding to the domain @xmath798 for @xmath799 .
in this case there is no spatial amplification in the coherent regime , but an absolute instability characterized by a spatial growth rate @xmath800 .
we thus find the same continuity as the one observed in the previous case @xmath801 along the boundary @xmath802 between the spatial growth rate @xmath803 , corresponding to the coherent _ absolute _ instability , and the incoherent _ spatial amplification _ growth rate @xmath796 ; on the other hand @xmath772 is strongly reduced from @xmath796 to @xmath788 along the boundary between the ar2 and ar1 domains , as in the case @xmath801 .
we may thus conclude that in all cases the boundary @xmath797 , i.e. @xmath804 in physical units , corresponds to a large reduction of the spatial amplification growth rate .
lastly , we may give for the sake of completeness the expression of the space growth rate @xmath805 associated with the incoherent absolute instability in the domain ar2 .
as stated before , for @xmath799 , the domain ar2 corresponds to a mixed situation where there is the coexistence of a spatial amplification characterized by @xmath806 , ( for @xmath807 and @xmath808 ) and of an absolute instability the space amplification growth rate of which is @xmath809 ; one has thus @xmath810 . in fig .
5 we reproduce the three boundaries found in the previous stability diagrams at which there is a reduction of the parametric growth due to the pump wave incoherence ; the solid line corresponds to the boundary of the domain vr where the pump wave incoherence reduces the convective growth rate , i.e. where it modifies the short time behavior ; the short- dashed line represents the boundary of the domain ar where the pump wave incoherence reduces the absolute growth rate , and the dotted - dashed line ( which coincides with the solid line for @xmath811 ) where it reduces the spatial amplification growth rate , i.e. the two domains where the long time behavior is modified .
one immediately realizes that all the domains for which there is a growth reduction in the long time behavior are contained in the set for which there is a reduction of the initial growth , whereas the converse is not true .
thus the domain b in fig .
5 is a domain where the spatial amplification is not affected by the pump wave incoherence although the convective growth is reduced in the initial stage ; the same remark applies to the domain a for @xmath801 ; for @xmath812 this latter domain corresponds to a regime where the absolute instability growth rate , although much smaller than the convective growth rate , is not modified by the pump wave incoherence whereas the initial convective growth rate is reduced .
we may have therefore the surprising conclusion that the reduction of the parametric growth is more difficult to achieve in the long time limit than in the initial stage .
the thresholds set by the physical damping of the modes have been given for convective modes by ( vi.5 ) and for absolute modes by ( vi.9 ) in the incoherent limit or by ( vi11b ) in the coherent limit .
in addition to satisfying these thresholds , the solutions must satisfy a threshold length criterion . in the case of absolute instability
, the length criterion in the coherent limit is known to be @xmath813 which in the absence of damping , @xmath814 sets @xmath815 a similar condition is obtained by solving the set ( v.32 ) with @xmath54 and with the boundary condition @xmath816 for @xmath817 and @xmath818 .
this rpa threshold condition is given by the unusual expression , @xmath819 which , not unexpectedly , does not reduce to the coherent expression as @xmath820 . in the limit of purely temporal incoherence , ( vi.30 )
reduces to @xmath821 whereas , for purely spatial incoherence , ( vi.30 ) reduces to @xmath822 the main point is that the ensemble averaged absolute growth rate for the infinite system given by vi.8a ) in the rpa theory can be obtained only if the plasma size l sufficiently exceeds a length given by the maximum of ( vi.29 ) and ( vi.30 ) .
the statistical analysis has obtained dispersion relations from which growth rates , thresholds , and amplification rates in different domains of incoherence were obtained for the ensemble averaged behavior of parametric instabilities with incoherent pump waves . in this analysis , some questions arose such as the validity of the rpa dispersion relation in the intermediate domain that can be answered in principle at least by a direct numerical integration of the coupled mode equations ( ii-10 ) . not only do we test the assumptions and approximations of the statistical analysis but , in sec .
viia , we illustrate the meaning of the averaging procedure .
the fundamental assumption that the rpa equations apply in the whole incoherent domain is examined in section ( vii.b ) ; finally the relation of the statistical growth rates and thresholds to the growth rates and thresholds of a particular model of induced laser beam incoherence is examined in section ( vii.c ) .
as has been noted at several points in this article , the incoherent absolute growth rates for the average intensity can be twice the rate expected on the basis of the growth rate for the average amplitude if certain inequalities are satisfied .
for the purely temporal problem ( where no distinction between absolute and convective is made ) , this factor of two arises if the damping is weak , specifically if max @xmath823 . however if this inequality is reversed , the average amplitude and average intensity growth rates are equal ( we remind the reader that the definition of growth rates in eq . iii.9
removes the expected factor of two from the intensity growth rate ) . in the space - time problem considered in this article ,
one again notices a factor of two if @xmath101 in eq .
( vi.8a ) for purely temporal incoherence , @xmath824 .
if one realizes that these random processes lead to a distribution of mode amplitudes f(a ) after a time t , then it is clear that the first moment @xmath825 and the second moment @xmath826 will be related only in special cases by @xmath827 .
if the distribution remains narrow then these two moments are simply related .
however we will discover that , in some cases , that is just when there is this factor of two at issue , the distribution is broad .
the second moment then depends on finding the distribution of rare events where the wave amplitude grows to larger amplitude than the mean amplitude .
we anticipate the result that the width of the distribution increases in time as does the mean so that the task of computing the ensemble average @xmath828 is increasingly more difficult as the sampling time is increased .
these remarks are now illustrated by considering the coupled mode equations for the purely temporal problem which have the simple form given by @xmath829 @xmath830 in previous work , an exact solution for the ensemble average mode amplitude and intensity was found when @xmath831 is a kubo - anderson process ( section v.1 ) .
when @xmath832 the growth rates are found to be @xmath833 and @xmath834 . here , we are interested in the fluctuations of the amplitude and intensity about the means @xmath835 and @xmath836 . some useful insight can be gained into this question of fluctuations and its influence on the behaviors of @xmath837 and @xmath515 by considering the simplest possible model , that of two coupled undamped modes @xmath838 and @xmath839 described by eqs .
( vii.1 ) where @xmath840 and @xmath841 is a random phase with bandwidth @xmath8 . in this simple case
we will be able to calculate the evolution of the distribution function f(a ) of amplitudes . because @xmath842 is a constant of the motion ,
we can conveniently write : @xmath843 @xmath844 from which the equations of motion ( vii.1 ) imply @xmath845 @xmath846 where @xmath847 if we define a distribution function @xmath848 , it evolves according to the fokker - planck equation : @xmath849 in the large bandwidth limit @xmath850 , @xmath851 .
, we can use a chapman - enskog like procedure , expanding @xmath852 and to zero order obtain @xmath853 , so that @xmath854 independent of @xmath855 .
to first order we obtain @xmath856 and to second order @xmath857 averaging over @xmath855 , we obtain the consistency equation , @xmath858 we are interested in the evolution of f from an initial condition @xmath859 .
if we assume @xmath860 , since we expect @xmath861 to increase we can approximate @xmath862 in the above equation .
it is then readily solved obtaining @xmath863 that is , @xmath861 is gaussian distributed with mean @xmath864 and variance @xmath865 .
this distribution is shown in fig .
7.1 for @xmath866 and @xmath867 together with some numerically generated distributions .
clearly then @xmath868 and @xmath869 .
on the other hand , if we calculate @xmath870 , we obtain @xmath871 for the special case n = 2 we observe that the growth rate for @xmath836 is @xmath872 , containing the factor of two noted earlier .
in particular we see explicitly that the mean growth rate is definition dependent because the distribution of growth rates is so broad . to compare with these analytic results , eqs .
( vii.1 ) were numerically integrated for a large number of different independent realizations of a kubo - anderson process such that @xmath873 and @xmath874 . in fig .
6 , the distribution ( broken line ) of the intensity @xmath875 is shown for @xmath876 and @xmath877 realizations .
the `` measured '' value of @xmath833 is 0.095 and @xmath649 is 0.16 , and obviously the distribution at @xmath878 is not resolved .
moreover , the few events with @xmath879 that have grown at the rate greater than 0.165 @xmath3 have determined @xmath649 . in general , the accurate numerical computation of random multiplicative processes in which rare events determine the quantity of interest is nontrivial[77 ] . on the other hand , the distribution of events that have grown at the rate @xmath833 is well resolved .
when there is sufficient damping ,
i.e. max @xmath880 the exact theoretical results show the ensemble average growth rate @xmath881 .
the simulation results for @xmath882 also shown in fig .
6 ( dashed - dotted line ) , demonstrate that the damping has dramatically narrowed the distribution of @xmath875 so that @xmath883 and @xmath884 is zero for @xmath885 .
note that this behavior is not specific to the kubo - anderson process .
numerical simulations with a randomly diffusing phase or a k - a process with equally spaced elapsed times obtained similar distributions of @xmath886 . here , the incoherent absolute growth rate @xmath887 for temporal incoherence is twice the `` nominal '' incoherent growth rate , @xmath888 , if @xmath101 .
given the results just presented for the purely temporal problem , this factor of two suggests that we might expect a broad distribution to result from considering different realizations of a temporal random process with bandwidth @xmath8 . in this case , we remind the reader that no absolute instability is allowed by the average amplitude equations since the average amplitude equations decouple in the rpa or bourret approximation .
it is also interesting to observe that , unlike the purely temporal case , damping on either mode does not affect the factor that multiplies @xmath889 to obtain @xmath649 .
this factor , @xmath890 , depends on the ratio of the group velocities and approaches two when the group velocity ratio approaches one .
thus , in analogy with the purely temporal problem , we expect the distribution of intensities of growing modes to broaden as the group velocity ratio approaches one .
before we proceed to test this hypothesis , we must be certain that both the incoherent absolute growth rate @xmath891 ( vi.8a ) is less than the coherent rate ( vi.iia ) and that the plasma length exceed sufficiently the threshold lengths ( vi.31 ) and ( vi.29a ) .
in addition the threshold set by the physical damping and bandwidth must be exceeded .
these conditions are met by choosing the parameters : @xmath54 , @xmath892 , @xmath893 , and @xmath894 .
when the group velocity ratios @xmath895 and -100 , the predicted incoherent absolute growth rates @xmath887 are 0.147 and 0.12 @xmath896 , respectively . the predicted coherent absolute rates are 0.47 and 0.2 @xmath3 respectively . the distribution of @xmath886 for @xmath897 is shown in fig .
7.2 for each velocity ratio .
each event in the distribution represents the ratio @xmath898 where @xmath899 and @xmath900 is the initial position of a source at t = 0 in @xmath839 .
the time @xmath901 is chosen nonzero to avoid the influence of transients and was @xmath902 in all these cases .
this `` initial '' time is longer than the time for the convective pulse to move across the system , @xmath903 but not always longer than the time it takes the slow wave to transit the length l. however since the growth is measured at the source point , the fact that in the entire plasma , the waves are not growing at the asymptotic rate should not matter . in these simulations ,
the numerically calculated ensemble average growth rate @xmath904 was smaller than the rpa prediction in all cases as expected .
when the velocity ratio @xmath905 and -100 the computed @xmath906 was 60% and 75% respectively of the rpa prediction . however , we were most interested n the width of the distribution .
when @xmath907 was 16 and 100 , the full width at half maximum of the distribution was 7% and 2% of the peak respectively , and thus displayed the narrowing expected .
when @xmath908 , the computed distribution was not significantly broader than the distribution for @xmath905 .
thus the difference in growth rates between @xmath908 and -16 may arise from differences in the unresolved tail of the distribution .
we conclude from this study that the rpa growth rates for the ensemble average intensity are correct but in a given simulation , or perhaps a particular experiment , the measured growth rate may well be a factor of two smaller .
however when the group velocity is large , as is typically the case of physical interest , the rpa predictions are more representative of the typical behavior .
the question here is whether the rpa equations are applicable in the whole domain defined by the inequalities @xmath909 and @xmath910 or whether there is an intermediate domain when one of the cross inequalities @xmath911 or @xmath912 is not satisfied .
to examined this case we look at the spatial kubo - anderson ( k - a ) process analyzed previously@xmath913 with zero temporal incoherence and spatial correlation function with the property @xmath914 if we assume @xmath102 and @xmath799 , then the intermediate domain is described by the chain of inequalities , @xmath915 a physical example that might produce this set of inequalities is two plasmon decay in the presence of stationary ion acoustic turbulence where the shorter wavelength langmuir wave is more strongly damped and has a faster group velocity .
the choices in our numerical solution were : @xmath916 for which the rpa equation absolute instability growth rate @xmath917 and the absolute instability coherent growth rate ( vi.11 ) , @xmath918 .
the simulation region @xmath919 was larger than the threshold length given by the maximum of eq . vi.32 and
vi.29.the simulation was run until @xmath920 for 1000 independent realizations of this spatial k - a process .
an initial value was given to the undamped wave at t = 0 but the intensity at @xmath921 was measured relative to the intensity at @xmath922 because by that time the modes were observed to grow at this time asymptotic rate .
the ensemble averaged rate @xmath891 was found to be independent of whether its calculation was based on the intensity at the source point @xmath923 or the total mode `` energy '' @xmath924 .
the results support our conjecture that the rpa equations apply in the whole incoherent domain , because the measured intensity growth rate , @xmath925 agreed with the rpa rate @xmath926 . however , the distribution of intensity at @xmath927 ( relative to @xmath928 ) displayed in fig 8 showed that , in fully a third of the cases , no growth occurred and that the remaining distribution consisted of a slowly decreasing tail .
the maximum intensity in the distribution was achieved by a mode that grew at close to the coherent rate .
in addition , the faster growth rates were achieved in cases in which there was an abnormally long distance between phase changes ; in the maximum growth case , this distance was 2.18 @xmath929 . as we remarked at the outset of this section and in the introduction ( sec .
i ) , the growth of parametric instabilities with this model of spatial incoherence was studied by williams et al . using more sophisticated statistical methods ( quite different in nature to those used in this article ) with the result that the fastest growing mode grows at the rate @xmath930 , given by eq .
this rate @xmath930 depends on the size of the system , which is understood on the basis that as the system gets larger there is an increasing possibility that there will be a large region with no phase change . in our simulation
this size dependent factor @xmath931 turns out to be approximately equal to the numerical factor of 4 in the rpa growth rate ( eq .
vi.8a ) ; all other factors are identical and @xmath932 .
unfortunately it is not feasible to simulate a system that doubles the size of this logarithmic factor with a constant value of @xmath933 .
thus these simulations do not distinguish between the rpa theory and the results of williams et al .
however , the simulations do support the conjecture that the incoherent results apply in whole incoherent domain . one motivation for this work is the induced spatial incoherence technique ( isi ) for creating smoothed laser beam intensity distributions in the focal plane . in this technique , statistically independent beamlets overlap and
thereby produce both intensity and phase variations . only pumps with phase variation alone
have been considered up to this point in our numerical examples . here
we wish to show that the same statistical methods are valid for intensity and phase varying pumps if the conditions for the incoherent results to apply are met .
a model that exhibits both intensity and phase variation in one dimension is given by , @xmath934 when @xmath935 is uniformly distributed on @xmath936 and @xmath937 is a random variable of time .
each @xmath938 varies independently with the same mean coherence time , @xmath939 .
the spatial coherence length in this model , @xmath940 , is related to the focal length of the focussing optic . a shorter focal length results in a shorter coherence length .
at any given point and instant of time , the intensity @xmath941 , being the absolute square of the sum of n complex numbers of length n@xmath942 , is gaussian distributed with unit mean and variance . because the sum varies with position over a length scale @xmath943 , the instantaneous spatial intensity pattern is quite spikey with peak to average variations of more than two quite common .
however because the sum at a given spatial position also varies in time with coherence time @xmath944 , the time averaged intensity is a much smoother function of space . clearly if @xmath945 is constant in time , then the spikey intensity pattern of eq .
vii.15 is stationary which can lead in the coherent domain to growth enhanced relative to a uniform intensity . even in the incoherent domain without temporal bandwidth
, we saw in the previous section that there is a finite probability , especially in large systems , for the phase to remain constant for a large distance . with phase and intensity variations , large fluctuations in growth rates
are expected in the absence of temporal bandwidth . in the complementary limit
where @xmath946 , the results of our statistical analysis should apply . for simplicity ,
we choose to compute the convective growth for both temporal and spatial incoherence in the incoherent limit and for large group velocity ratios , @xmath947 . with a single term in the sum in eq . vii.15 and @xmath948 ,
the convective growth rate was in agreement with eq .
( vi.7 ) . with no temporal bandwidth , @xmath949
, the pump intensity has striking amplitude variation with peak to average variations of more than three to one .
nonetheless if @xmath950 , the convective growth rate , is reduced according to eq .
( vi.7 ) even if @xmath949 . as the spatial incoherence weakens and the boundary @xmath951 is approached for @xmath949 the measured growth rate in the simulations diverges from eq .
( vi.7 ) until , for @xmath952 , growth rates in excess of @xmath953 can occur with large fluctuations from case to case .
laser bandwidth , even if @xmath954 , reduces both the fluctuations and the growth rate to less than @xmath953 .
thus , some bandwidth is needed in this model to provide a smooth transition from the coherent to incoherent domain .
other models such as the one used in section vii.b have the property that @xmath955 so that the statement that pump incoherence can only reduce the growth rate is indeed true .
simulations in that case do show a smooth transition from the coherent to incoherent growth rate
a comprehensive treatment of the effects of pump wave temporal and spatial incoherence has been presented for a homogeneous plasma . eschewing specific models of incoherence that allow exact solutions , we have derived equations for ensemble averaged mode intensity and amplitudes in the incoherent limit . at each step in the process of deriving these approximate statistical equations ,
the conditions that must be satisfied are clearly stated .
thus , the meaning of incoherence in the context of parametric instability theory is carefully defined .
the set of inequalities that comprise this definition form a major contribution of this work .
of course , the primary contribution of this work is the set of thresholds , growth rates , and amplification rates for both coherent and incoherent , convectively and absolutely unstable parametric interactions .
however , the majority of our readers are interested in the more practical results concerning thresholds , growth rates , and amplification coefficients as they pertain to laser plasma interactions . in a subsequent work,@xmath956
we will apply the results presented here to particular instabilities , and include both backscatter and sidescatter geometries . here
, we briefly consider the examples of stimulated raman and brillouin backscatter for parameters of interest to laser fusion .
we choose the electron temperature , @xmath957 , the charge state z = 40 , the atomic number a = 80 , the laser wavelength , @xmath958 , the laser intensity i = 10@xmath959 w/@xmath960 and the electron density , @xmath961 for srs and 0.25 n@xmath962 for sbs . here
n@xmath962 is the critical electron density defined by @xmath963 . for almost all calculations
it is the ratio z / a and not z or a by themselves that is important .
the nominal sbs growth rate is @xmath964 whereas the srs coherent convective threshold , eq .
( i-1 ) is @xmath965 .
landau damping was neglected because for @xmath957 , an electron density can always be chosen to make it negligible .
thus srs is an order of magnitude above threshold .
the laser bandwidth sufficient to achieve convective stability can be found by using eq .
we find @xmath966 . on the other hand , the coherent threshold for absolute instability given by eq .
( vi.9a ) , @xmath967 , is exceeded by less than a factor of two .
however , as we have emphasized , in sec .
vi , relatively large laser bandwidth is required to reduce the absolute growth rate .
satisfying the validity conditions for using the incoherent formula , eq .
( vi.10 ) , imposes the condition , @xmath968 , which is , coincidentally , the same condition numerically that is required for absolute stability as determined by eq .
( vi.9a ) .
the sbs thresholds are dependent on the ion acoustic damping rate and thus the ion temperature , @xmath969 .
however , even if @xmath970 the ion landau damping is negligible for @xmath971 .
thus , the ion acoustic damping rate is determined by the electrons and is given by @xmath972 .
hence , the convective coherent threshold @xmath973 is easily exceeded by the nominal sbs growth rate @xmath974 .
in addition , the sbs absolute instability coherent threshold @xmath975 is exceeded . here , the bandwidth required to achieve convective stability is @xmath976 .
moreover , the bandwidth required for absolute stability is by accidental choice of parameters , the same as for convective stability because although @xmath977 , nonetheless @xmath978 and the incoherent absolute threshold eq .
( vi.9a ) reduces to the incoherent convective threshold eq .
( vi.6 ) . at this point , we should warn the reader that the results reported here assumed the validity of the coupled mode equations which requires the mode frequencies @xmath979 for j = 0,1,2 .
however , we have in the case of sbs a critical bandwidth for stability that exceeds the ion acoustic frequency @xmath980 . at the very least
, one might expect additional stability when @xmath981 .
for a partial answer to the effects that @xmath982 has , we considered sbs again for the purely temporal problem without using the assumption that @xmath983 .
we start with the equations , @xmath984 @xmath985 where @xmath986 , is the difference between the pump frequency and the scattered light frequency . here
a@xmath987 are the mode amplitudes for the scattered light and ion acoustic waves respectively .
two different dispersion relations are obtained , as expected , by obtaining equations for the evolution of average amplitude @xmath988 .
the ion acoustic mode @xmath281 dispersion relation is not affected by magnitude of @xmath989 so that the growth rate , @xmath990 is a maximum when @xmath991 . on the other hand , the dispersion relation for @xmath725
is modified so that the growth rate when @xmath992 , @xmath993 shows a stronger reduction with bandwidth than @xmath994 . here
the maximum growth rate for @xmath725 occurs for a bandwidth dependent real frequency @xmath995 .
thus , the threshold for mode @xmath996 is increased when @xmath997 by the ratio @xmath989 . the actual threshold for the coupled set viii.1 is , @xmath998
this equation ( viii.4 ) is a generalization of eq .
( vi.6 ) . in our example , the threshold is not increased when @xmath981 . in our example , chosen for its relevance to laser fusion , the critical bandwidth to reach convective or absolute stability has been found to be 2 - 5% of the laser frequency .
such a bandwidth is beyond the currently achievable with any icf laser system .
however , novel techniques are being pursued that are capable of achieving these bandwidths[6 ] .
in addition , there is the question of whether the average laser intensity or the larger hotspot intensity should be used in evaluating the thresholds .
the use of the larger intensity would make the bandwidth requirements infeasible
. however , if bandwidth is combined with a beam smoothing scheme , use of the mean intensity may be correct .
recent calculations[66,67 ] have shown that , for at least one such scheme , isi , the use of the average intensity is justified .
in addition to temporal bandwidth , there is also the effect of spatial incoherence produced by beam smoothing techniques that lead to a spread of wavevectors , @xmath9 . in the examples of backscatter considered previously in this section for temporal bandwidth
, spatial incoherence will have little effect because @xmath999 where the f number of the focussing optic , f # , is typically large in icf applications .
furthermore , the wavenumber spread will not increase the convective threshold because the effective bandwidth , @xmath1000 , is in general small compared to @xmath1001 , and @xmath3 .
however absolute instabilities with significant wavevectors perpendicular to the laser propagation axis such as two plasmon decay may be influenced significantly by such a spread since there @xmath1002 will be the effective bandwidth .
the work reported here is only a step in the application of a statistical theory to particular instabilities .
further progress may require modifications or extensions to our work in order to remove some of the limitations of our approximations .
we look forward to new developments in the theory and even more to the stimulus of experimental data .
the authors express their appreciation to their many colleagues who have given freely of their ideas and criticisms , specifically b. b. afeyan , j. r. albritton , b. i. cohen , w. l. kruer , g. laval and r. pellat .
d. pesme acknowledges the support of the centre national de la recherche scientifique and the plasma physics research institute of lawrence livermore national laboratory and the university of california at davis through doe contract w-7405-eng-48 .
r. l. berger was supported by doe contract de - ac03 - 87dp10560 and is grateful to lawrence livermore national laboratory for its hospitality during the course of this collaboration .
e. a. williams was supported by doe contract w-7405-eng-48 .
annick bortuzzo - lesne was supported by a grant from fondation singer polignac .
a. bourdier also acknowledges the interest of w. l. kruer and the hospitality of lawrence livermore national laboratory where this work began .
two of the authors ( r. l. berger and e. a. williams ) wish to thank m. haines and c. moser who organized the 1988 and 1989 cecam workshops on laser - plasma interactions which helped further the advance of this work .
this work was the subject of an internal lawrence livermore national laboratory report with report number ucrl - jc-105479 november 19 , 1990 . * k. moncur , _ applied optics _ , * 16 * , 1449 ( 1977 ) . *
r. h. lehmberg and s. p. obenschain , _ opt .
* 46 * , 27 ( 1983 ) . *
y. kato , k. mima , n. miyanaaga , s. arinaga , y. kitagawo , m. nakatsuka , and c. yamanaka , _ phys .
_ , * 53 * , 1057 ( 1984 ) ; y. kato and k. mima , _ applied physics _ * 329 * , 186 ( 1982 ) . *
s. skupsky , r. w. short , t. kessler , r. s. craxton , s. letzring , and t. m. soures , _ j. appl
_ * 66 * , 3456 ( 1989 ) .
* d. veron , h. ayral , c. gouedard , d. husson , j. lauriou , o. martin , b. meyer , m. rostaing , and c. sauteret , _ optics communications _ , * 65 * ( 1988 ) .
* d. m. pennington , m. a. henesian , r. b. wilcox , t. l. wieland , d. eimerl and h.t .
powell , `` a novel bandwidth source for laser experiments '' , technical digest series , vol 18 , cleo 94 anaheim , ca .
* d. pesme , `` effects of temporal and induced spatial incoherence of parametric instabilities in laser plasma interactions '' , 1987 annual technical report cnrs - luli , ecole polytechnique .
( available from ntis , springfield va 22161 : document pb 92 - 100312 . ) * k. nishikawa , in `` advances in plasma physics , vol 6 '' , edited by a. simon and w. b. thompson ( wiley n. y. 1976 ) * g. m. zaslavskii , v. s. zakharov , _ sov .
phys . tech .
_ , + * 12 * , 7 ( 1967 ) . * g. e. vekshtein , g. m. zablavskii , _ sov .
phys . doklady _ ,
* 12 * , 34 ( 1967 ) .
* e. j. valeo and c. r. oberman , _ phys .
lett . _ * 30 * , 1035 ( 1973 ) .
* s. tamor , _ phys .
fluids _ * 16 * , 1169 ( 1973 ) . *
j. j. thomson , w. l. kruer , s. e. bodner , and j. s. degroot , _ phys .
fluids _ , * 17 * , 849 ( 1974 ) . *
j. j. thomson and j. i. karush , _ phys .
fluids _ * 17 * , 1608 ( 1974 ) .
* j. j. thomson , _ nuclear fusion _ ,
* 15 * , 237 ( 1975 ) . *
g. laval , r. pellat , d. pesme , a. ramani , m. n. rosenbluth , and e. a. williams , _ phys .
fluids _ , * 20 * , 2049 ( 1977 ) . *
j. j. thomson , _ phys .
fluids _ , * 21 * , 2082 ( 1978 ) . *
w. l. kruer , k. g. estabrook , and k. h. sinz .
fusion _ , * 13 * , 952 ( 1973 ) . * w. kruer , e. valeo , k. estabrook , j. thomson , b. langdon , and b. lasinski , _ plasma physics and controlled nuclear fusion research _ , * vol .
ii * , 525 ( 1974 ) . *
k. estabrook , j. harte , e. m. campbell , f. ze , d. w. phillion , m. d. rosen , and j. t. larsen , _ phys .
lett . _ * 46 * , 724 ( 1981 ) .
* d. w. forslund , j. m. kindel and e. m. lindman , _ phys fluids _ , * 18 * , 1017 ( 1975 ) . *
k. estabrook , w. l. kruer , and b. f. lasinski , _ phys .
_ , * 45 * , 1399 ( 1980 ) .
* k. estabrook and w. l. kruer , _ phys . fluids _
, * 26 * , 1892 * g. bonnaud and c. reisse , _ nuclear fusion _ ,
* 26 * , 633 ( 1986 ) . * c. yamanaka , t. yamanaka , t. sasaki , and j. mizui , _ phys .
_ , * 32 * , 1038 ( 1974 ) .
* r. r. johnson , p. m. campbell , l. v. powers , and d. c. slater , in proceedings of the topical meeting of inertial confinement fusion , 7 - 9 february , 1978 , san diego ( unpublished ) ; d. c. slater and d. j. tanner , proceedings of the ninth annual conference on the anomalous absorption of electromagnetic waves . *
c. e. clayton , c. joshi , a. yasuda and f. f. chen , _ phys . fluids _ , * 24 * , 2312 ( 1981 ) .
* a. mase , n. c. luhmann , j. holt , h. huey , m. rhodes , w. f. divergilio , j. j. thomson , and c. f. randall , _ plasma physics and controlled nuclear fusion research _ ( iaea , vienna , 1981 ) . * r. giles , r. fedosejevs , and a. a. offenberger , _ physical review a _ , 2 * 6 * , 1113 ( 1982 ) . *
d. g. colombant , w. m. manheimer , and j. h. gardner , phys .
fluids 26 , 3148 ( 1983 ) . *
v. v. tamoikin and s. m. fainstein , _ sov .
jetp _ , * 35 * , 115 ( 1972 ) .
* s. a. akhmanov , yu . e. dyakov and l. i. pavlov , _ sov .
jetp _ , * 39 * , 249 ( 1974 ) . * p. kaw , r. white , d. pesme , m. rosenbluth , g. laval , r. varma , and r. huff , _ comments plasma phys .
controlled fusion 2 _ , * 11 * ( 1974 ) . *
d. r. nicholson and a. n. kaufman , _ phys .
_ , * 33 * , 1207 ( 1974 ) .
* m. y. yu , p. k. shukla and k. h. spatschek , _ phys .
rev a _ , * 12 * , 656 ( 1975 ) * k. h. spatschek , p. k. shukla and m. y. yu , _ phys .
, * 51 * , 183 ( 1975 ) .
* d. r. nicholson , _ phys .
fluids _ , * 19 * , 889 ( 1976 ) . * g. laval , r. pellat and d. pesme , _ phys .
_ , * 36 * , 192 ( 1976 ) .
* e. a. williams , j. r. albritton , and m. n. rosenbuth , _ phys .
fluids _ * 22 * , 139 ( 1979 ) .
* e. z. gusakov and a. d. piliya , _ sov . j. plasma phys .
_ , * 6 * , 277 ( 1980 ) .
* e. z. gusakov and a. d. piliya , _ sov . j. plasma phys .
_ , * 7 * , 733 ( 1981 ) .
* e. z. gusakov and a. d. piliya , _ sov . j. plasma phys .
_ , * 8 * , 324 ( 1982 ) .
* w. rozmus , a. a. offenberger and r. fedosejevs , _ phys .
fluids _ , * 26 * , 1071 ( 1983 ) .
* a. bourdier and e. a. williams , `` effects of induced spatial incoherence on raman scattering , '' _ laser program annual report 1986 _ , lawrence livermore national laboratory , ucrl-50021 - 86 ( 1986 ) . * e. a. williams , r. l. berger , and a. bourdier , _ laser program annual report 1987 _ , lawrence livermore national laboratory , ucrl-50021 - 87 ( 1988 ) , p. 2
* a. bortuzzo - lesne , g. laval , d. pesme , and m. casanova , `` coefficient de retrodiffusion brillouin stimulee lorsque londe laser est incoherente , '' annual internal report gilm , ecole polytechnique ( 1984 ) . * l. lu , _ phys .
fluids _ , * 31 * , 3362 ( 1988 ) . *
l. lu , _ phys .
fluids _ , * b1 * , 1605 ( 1989 ) . * p. n. guzdar , _ phys . fluids _
, * b3 * , 2882 ( 1991 ) . * m. n. rosenbluth , phys . rev . lett . * 29 * , 565 ( 1972 ) . * s. p. obenschain , j. grun , m. j. herbst , k. j. kearney , c. k. manka , e. a. mclean , a. n. mostovych , j. a. stamper , r. r. whitlock , s. e. bodner , j. h. gardner , and r.lehmberg , _ phys .
lett . _ * 56 * , 2807 ( 1986 ) . * a. n. mostovych , s. p. obenschain , j. h. gardner , j. grun , k. j. kearney , c. k. manka , , e. a. mclean , and c. j. pawley , _ phys .
_ , * 59 * , 1193 ( 1987 ) .
* s. p. obenschain , c. j. pawley , a. n. mostovych , j. a. stamper , j. h. gardner , a. j. schmitt , and s. e. bodner , _ phys .
_ * 62 * , 768 ( 1989 ) . *
o. willi , d. bassett , a. giulettii and s. j. kartunnen , _ opt . comm .
_ , * 70 * , 487 ( 1989 ) .
* s. coe , t. afshar - rad , m. desselberger , f. khattak , o. willi , a. giulietti , z. q. lin , w. yu , and c. danson , _ europhys .
_ , * 10 * , 31 ( 1989 ) . *
s. e. coe , t. afshar - rad and o. willi , _ europhys . lett .
_ , * 13 * , 251 ( 1990 ) .
* o. willi , t. afshar - rad , s. e. coe and a. guilettii , _ phys .
fluids _ , * b2 * , 1318 ( 1990 ) . * t. afshar - rad , l. a. gizzi , m. desselberger , f. khattak and o. willi , _ phys .
_ , * 68 * , 942 ( 1992 ) . * t. afshar - rad , s. e. coe , o. willi and m. desselberger , _ phys . fluids _ , * b4 * , 1301 ( 1992 ) .
* w. seka , r. e. bahr , r. w. short , a. simon , r. s. craxton , d. s. montgomery and a. e. rubenchik _ phys .
fluids _ , * b4 * , 2232 ( 1992 ) . * c. labaune , s. baton , t. jalinaud , h. a. baldis and d. pesme _ phys . fluids _ , * b4 * , 2224 ( 1992 ) . * j. d. moody , h. a. baldis , d. montgomery , k. g. estabrook , s. dixit and c. labaune , _ j. of fusion energy _ , * 12 * , 323 ( 1993 ) .
* t. jalinaud , s. d. baton , c. labaune and h. a. baldis , `` effets de lames de phases aleatoire sur les diffusions brillouin et raman stimules en plasma preforme '' , luli 1992 annual report , p56 . *
c. labaune , s. d. baton , t. jalinaud , e. schifano , n. renard , d. pesme , h. a. baldis , j. d. moody and k. estabrook , `` effet du lissage optique par lame de phase aleatiore sur les instabilities parametriques '' , luli 1992 annual report , p23 . * j. d. moody , h. a. baldis , d. s. montgomery , k. estabrook , r. l. berger , e. a. williams , w. l. kruer and s. dixit ( submitted to phys . plasmas ( 1995 ) ) .
* r. l. berger , _ phys .
_ , * 65 * , 1207 ( 1990 ) .
* p. n. guzdar , _ phys .
fluids _ , * b3 * , 776 ( 1991 ) .
* h. a. rose , d. f. dubois and d. russell , _ sov .
j. of plasma physics _ , * 16 * , 537 ( 1990 ) .
* a. brissaud and u. frisch , _
j. math phys .
_ , * 15 * , 524 ( 1974 ) ; p. kubo , ibid . , * 4 * , 174 ( 1963 ) . * r. bourret , _ nuovo cimento _ , * 26 * , 1 ( 1962 ) .
* r. z. sagdeev and a. a. galeev , _ nonlinear plasma theory _
( benjamin , new york , 1969 ) . * b. b. kadomtsev , _ plasma turbulence _
( academic press , new york , 1965 ) . * v. n. tsytovitch , _ nonlinear effects in plasma _ ( plenum , new york , 1970 ) . * r. c. davidson , _ methods in nonlinear plasma theory _
( academic press , new york , 1972 ) . * a. bers and r. j. briggs , _ quarterly progress report no .
71 _ , research laboratory of electronics , mit , p. 122,(1963 ) unpublished ) .
* v. p. silin , _ zh . eksp .
_ , * 48 * , 1679 ( 1965 ) [ _ sov .
* 21 * , 1127 ( 1965 ) ] ; v. e. zakharov , _ op .
cit . _ * 62 * , 1745 ( 1972 ) [ _ op .
cit . _ * 35 * , 908 ( 1972 ) ] . * s. redner , _ am .
j. phys . _
* 58 * , 267 ( 1990 ) .
are the mean wavenumbers of the pump wave and of the decay waves ; the pump wave vector @xmath1003 varies in magnitude an amount @xmath1004 and lies in a cone of angle @xmath205 about the direction of the mean , @xmath1003 . also shown
are the angles @xmath143 and @xmath149 which the mean wavevectors @xmath208 and @xmath209 make with the mean pump wavevector @xmath1003,width=566 ] on the vertical axis and @xmath1005 on the horizontal axis , represents the boundary of the domain ar where the pump wave incoherence reduces the absolute growth rate . here @xmath1006 .
the long - dashed line is the boundary of the domain as where the pump wave incoherence completely stabilizes the absolute growth . to the right of the rightmost vertical line
is the domain of coherent stability .
this diagram assumes @xmath1007.,width=566 ] for the purely temporal problem .
the solid curve is the analytic solution , and the broken curve is the numerically generated solution for undamped waves .
the dashed - dotted curve is the numerical solution when @xmath882 .
the bandwidth is @xmath897 and the abscissa is the base 10 logarithm of the intensity @xmath1009 after an interval @xmath1010.,width=566 ] for the space - time problem with purely temporal bandwidth @xmath874 after an interval @xmath1011 .
the solid curve has @xmath1012 and the dashed curve has @xmath1013 .
the abscissa is the base 10 logarithm of the intensity @xmath1014.,width=566 ] | the effect on parametric instability growth of pump wave incoherence is treated by deriving a set of equations governing the space - time evolution of the ensemble - average coupled - mode amplitudes and intensities .
particular attention is paid to establishing the regions of validity of the statistical description .
thresholds , growth rates , and amplification rates are given for both spatially and temporally incoherent pump waves . both absolutely and convectively unstable modes are considered .
the statistical results are verified where appropriate by numerical integration of the coupled - mode equations with different models of pump incoherence . |
solving lattice qcd to high precision requires the use of dynamical ginsparg - wilson light quarks .
however , this is computationally expensive . therefore , as a starting place , we take improved staggered quark configurations , which have the advantage of light sea quarks , and use an overlap valence quark action , which has the correct chiral and flavour symmetries .
this has its own disadvantages .
firstly , the overlap inversion is still relatively expensive and , secondly , because we have different actions for the sea and valence quarks , it is not straightforward to interpret the results .
mixed actions are inevitable in the improved staggered programme , because we do not yet have a local version of the sea quark action to use for the valence quarks . also , it is complicated to measure some quantities in the improved staggered formalism .
for example , in order to measure the mass of the nucleon one has a choice of o(100 ) different nucleon operators . without measuring a representative set ,
the effect of taste symmetry breaking is unquantified and uncontrolled .
the simulations were performed on twenty coarse ( milc ) dynamical configurations with 2 + 1 flavours @xcite .
ten configurations have a light isodoublet with mass @xmath0 and ten have @xmath1 .
both have a lattice spacing @xmath2 fm and linear size @xmath3 fm .
three iterations of hyp - smearing were applied to each configuration @xcite . the overlap operator from szin code @xcite
was then used to calculate propagators .
these were created with seven different valence quark masses using the overlap multi - mass solver : four light and three heavy @xcite . even for the baryon spectrum
we get a remarkably good signal on ten configurations . as a check
, we applied several iterations of hyp - smearing to 624 quenched ukqcd configurations at @xmath4 with a volume of @xmath5 .
planar wilson loops were used to extract the quark - antiquark potential .
hyp - smearing quickly alters the short - distance behaviour , while the medium - to - long distance behaviour remains relatively unchanged for a small number ( @xmath6 ) of iterations .
smearing the configuration helps to speed up the convergence of the overlap operator , renders the effective interaction more local , and makes the eigenvalue spectrum more like that for an overlap sea @xcite .
we perform simultaneous fits to three different correlators in order to extract the pseudoscalar meson mass ( see figure [ effective_mass ] ) .
the fluctuations in a@xmath7 are larger than the apparent statistical errors , but this is probably due to underestimation of the variance on ten configurations .
all the analysis is carried out partially quenched - we hold the sea quark mass fixed and vary the input valence quark mass . since we have multiple input valence masses , we can construct non - degenerate light meson correlators .
a two - dimensional fit was performed to @xmath8 versus valence masses @xmath9 and @xmath10 , which allowed evaluation of the average @xmath11 and @xmath12 quark mass , @xmath13 , from @xmath14 where @xmath15 is the physical pion mass squared .
this in turn allows us to evaluate the strange quark mass from @xmath16 where @xmath17 is the physical kaon mass and @xmath18 is the strange quark mass .
we define @xmath19 as @xmath20 we obtain @xmath21 from the axial ward identity @xmath22 which we can express in terms of the pseudoscalar correlator , @xmath23 , and the pseudoscalar axial correlator , @xmath24 .
@xmath25 cancels in eq.([fps ] ) and hence we only require @xmath23 in order to evaluate @xmath19 . once again we perform a 2-d linear fit to the light non - degenerate pseudoscalars to calculate @xmath19 ( see figure [ pseudoscalardecay ] ) and extract the ratio of @xmath26 ( see table [ table1 ] ) .
the value increases slightly with decreasing light sea quark mass in the right direction to agree with experiment .
this is also evident from the slight change of the gradient in figure [ pseudoscalardecay ] . versus @xmath27 for the two ensembles . ]
.pseudoscalar meson decay constants [ cols="<,<,<",options="header " , ] + we measure the masses of the nucleon and delta baryon .
it is remarkable that we can see a signal for the negative parity partner of the nucleon on as few as ten configurations .
figure [ nucleonlinear ] shows the nucleon mass versus the pseudoscalar meson mass squared .
the lines shown are uncorrelated linear fits to our data .
the values calculated by the milc collaboration @xcite on their corresponding full ensembles are also shown .
heavy quarks essentially come for free in the overlap propagator calculation due to the multi - mass solver .
however , lattice artefacts are @xmath28@xmath29 and the heaviest input valence quark mass used is @xmath30 and hence @xmath31 . with the lattice spacing of @xmath32gev
, we are at best on the limit of simulating charm . because of the rapid decay in euclidean time ,
we require double precision . however , this does not slow the solver down appreciably as we need less reorthogonalisations of the krylov subspace than in single precision .
these heavy quark propagators were used to calculate @xmath33 ( see table [ table1 ] ) .
the value of @xmath33 increases with decreasing light sea quark mass , in the direction of the experimental value , as can be seen from the change of gradients in figure [ fds ] .
vs inverse heavy - strange pseudoscalar meson mass . ]
we have shown that , even with very low statistics , it is possible to calculate light hadron and charm physics using an overlap valence operator on an improved staggered sea .
this could be an alternative to staggered valence quarks for probing sea quark effects , although for our data , the effect of small changes in the sea quark mass is minimal . | we present exploratory results for the hadron mass spectrum and pseudoscalar meson decay constants using mixed actions .
we use improved staggered sea quarks and hyp - smeared overlap valence quarks .
we obtain good signals on 10 configurations at one lattice spacing and two different sets of sea quark masses . |
the recent results from rhic ( see e.g. @xcite ) have attracted renewed interest in better understanding the dynamics of particle production , not only in nuclear collisions .
quite different approaches have been used to describe the particle spectra from the nuclear collisions @xcite . the model in ref.@xcite with an educated guess for ugd describes surprisingly well the whole charged particle rapidity distribution by means of gluonic mechanisms only .
such a gluonic mechanism would lead to the identical production of positively and negatively charged hadrons .
the recent results of the brahms experiment @xcite put into question the successful description of ref.@xcite . in the light of this experiment , it becomes obvious that the large rapidity regions have more complicated flavour structure .
i discuss the relation between ugd s in hadrons and the inclusive momentum distribution of particles produced in hadronic collisions .
the results obtained with different ugd s @xcite are shown and compared .
at sufficiently high energy the cross section for inclusive gluon production in @xmath2 can be written in terms of the ugd s `` in '' both colliding hadrons @xcite @xmath3 above @xmath4 and @xmath5 are ugd s in hadron @xmath6 and @xmath7 , respectively .
the longitudinal momentum fractions are fixed by kinematics : @xmath8 .
the argument of the running coupling constant is taken as @xmath9 . here
i shall not discuss the distributions of `` produced '' gluons , which can be found in @xcite .
instead i shall discuss what are typical values of @xmath10 and @xmath11 in the jet ( particle ) production .
average value @xmath12 and @xmath13 , shown in fig.[fig : x1_x2 ] , only weakly depend on the model of ugd . for @xmath14 0 at the rhic energy w = 200 gev one tests ugd s at @xmath15 = 10@xmath16 - 10@xmath17 .
when @xmath18 grows one tests more and more asymmetric ( in @xmath10 and @xmath11 ) configurations . for large @xmath18 either @xmath10 is extremely small ( @xmath19 10@xmath20 ) and @xmath21 1 or @xmath22 1 and @xmath11 is extremely small ( @xmath23 10@xmath20 ) .
these are regions of gluon momentum fraction where the ugd s is rather poorly known . the approximation used in obtaining ugd s are valid certainly only for @xmath24 0.1 . in order to extrapolate the gluon distribution to @xmath25 1
i multiply the gluon distributions from the previous section by a factor @xmath26 , where n = 5 - 7 . and @xmath13 for @xmath27 0.5 gev and at w = 200 gev .
[ fig : x1_x2 ] , width=302 ]
in ref.@xcite it was assumed , based on the concept of local parton - hadron duality , that the rapidity distribution of particles is identical to the rapidity distribution of gluons . in the present approach i follow a different approach which makes use of phenomenological fragmentation functions ( ff s ) . for our present exploratory study
it seems sufficient to assume @xmath28 .
this is equivalent to @xmath29 , where @xmath30 and @xmath31 are hadron and gluon pseudorapitity , respectively .
then @xmath32 where the transverse mass @xmath33 . in order to introduce phenomenological
ff s one has to define a new kinematical variable . in accord with @xmath34 and @xmath35 collisions
i define a quantity @xmath36 by the equation @xmath37 .
this leads to the relation @xmath38 where @xmath39 is given in ref.@xcite .
now we can write the single particle distribution in terms of the gluon distribution as follows @xmath40 in the present calculation i shall use only lo ff s from @xcite
. 0.2 gev .
the experimental data of the ua5 collaboration are taken from @xcite .
[ fig : eta_glue ] , width=302 ] let us analyze now how the results for pseudorapidity distributions depend on the choice of the ugd . in fig.[fig :
eta_glue ] i compare pseudorapidity distribution of charged pions for different models of ugd s . in this calculation ff from @xcite has been used .
in contrast to ref.@xcite , where the whole pseudorapidity distribution , including fragmentation regions , has been well described in an approach similar to the one presented here , in the present approach pions produced from the fragmentation of gluons in the @xmath41 mechanism populate only midrapidity region , leaving room for other mechanisms in the fragmentation regions .
these mechanisms involve quark / antiquark degrees of freedom or leading protons among others .
this strongly suggests that the agreement of the result of the @xmath41 approach with the phobos distributions @xcite in ref.@xcite in the true fragmentation region is rather due to approximations made in @xcite than due to correctness of the reaction mechanism . in principle , this can be verified experimentally at rhic by measuring the @xmath42 ratio in proton - proton scattering as a function of ( pseudo)rapidity in possibly broad range .
the brahms experiment can do it even with the existing apparatus . in fig.[fig
: pt_glue ] i compare the theoretical transverse momentum distributions of charged pions obtained with different gluon distributions with the ua1 collaboration data @xcite .
the best agreement is obtained with the karzeev - levin gluon distribution .
the distribution with the gbw model is much too steep in comparison to experimental data .
this is probably due to neglecting qcd evolution in @xcite .
, width=302 ]
i have calculated the inclusive distributions of gluons and associated charged @xmath1 s in the nn collisions through the @xmath43 mechanism in the @xmath44-factorization approach .
the results for several ugd s proposed recently have been compared .
the results , especially @xmath45 distributions , obtained with different models of ugd s differ considerably .
contrary to a recent claim in ref.@xcite , we have found that the gluonic mechanism discussed does not describe the inclusive spectra of charged particles in the fragmentation region , i.e. in the region of large ( pseudo)rapidities for any ugd from the literature .
clearly the gluonic mechanism is not the only one .
since the mechanism considered is not complete , it is not possible at present to precisely verify different models of ugd s .
the existing ugd s lead to the contributions which almost exhaust the strength at midrapidities and leave room for other mechanisms in the fragmentation regions .
it seems that a measurement of @xmath0 distributions of particles at rhic should be helpful to test better different ugd s . | the inclusive distributions of gluons and pions for high - energy nn collisions are calculated .
the results for several unintegrated gluon distributions ( ugd s ) from the literature are compared .
we find huge differences in both rapidity and @xmath0 of gluons and @xmath1 s in nn collisions for different models of ugd s .
the karzeev - levin ugd gives good description of momentum distribution of charged hadrons at midrapidities .
we find that the gluonic mechanism discussed does not describe the inclusive spectra of charged particles in the fragmentation region . |
when squeezed , the single photon ceases to be a single quantum particle and it becomes a superposition in an infinite - dimensional hilbert space . as such , it gains many useful abilities exploited in the various quantum protocols @xcite , but its description and , consequently , analysis becomes more involved .
however , we can greatly benefit from the well - known similarity between squeezed single photons and superposed coherent states @xcite .
if we are able to rigorously confirm that a single photon was transformed into a high - quality css and that a css was transformed back into a high - quality single photon , we can safely claim that the squeezing operation works as intended and that it should have its potential place in future applications . in order to quantitatively
assess the transition between the single photon and the css , we employ the following figures of merit .
first , note that the ideal initial and target states , the fock state @xmath45 and the coherent - state superposition @xmath46 , become identical when @xmath47 .
hence the usual measure of fidelity between the experimentally generated states and an ideal state @xmath48 , @xmath49 , is not a useful figure of merit .
for example , the overlap between the experimental @xmath50 and a target css , @xmath51 would optimally attain a value close to the maximal value of unity .
this can be achieved when the state @xmath50 approaches an ideal css with amplitude @xmath22 , but also when @xmath50 is close to @xmath45 for @xmath47 . in order to avoid such a confusion between our particle - like and wave - like states and to gain a better insight into
each generated state s properties , we divide the fidelity into two separate measures : the coherent - state distinguishability factor , @xmath52 , and the coherent - state interference factor , @xmath53 . here , @xmath54 describes the average overlap of the investigated state @xmath50 with an independent pair of coherent states with amplitudes @xmath55 , which form the ideal superposition .
when @xmath56 , this overlap vanishes for our odd - number css , while it would become unity for an even - number css , @xmath57 . in either case , as well as for an incoherent mixture of coherent states
@xmath58 , @xmath54 has , ideally , two separated symmetric peaks with maxima approaching @xmath59 when @xmath60 . on the other hand
, the single - photon state @xmath45 has a maximal distinguishability @xmath61 at @xmath62 .
an additional feature separating the css from the single - photon state is that the distinguishability of @xmath45 is phase - insensitive , @xmath63 , whereas for both the ideal and the experimentally generated csss , it is not , having a maximum at @xmath64 .
besides two separated coherent - state peaks in phase space , a css is characterized by interference between the coherent states . for the wigner function
, this interference gives a negativity at the phase - space origin .
however , a single - photon state exhibits the same feature , and if we want to distinguish a css from a single - photon state , we must evaluate the difference in this interference . for this purpose
, we introduce the interference factor @xmath65 , which is always positive for a balanced mixture of coherent states and always negative for both the single - photon and the css @xmath66 .
this makes it useful for demonstrating nonclassical interference effects .
however , this figure of merit achieves more .
while it has a clear minimum for the single - photon state , @xmath67 at @xmath68 , for the ideal css , its extremal value depends on the amplitude @xmath22 and this value monotonically decreases to @xmath69 as @xmath22 increases .
although both for the ideal single - photon state and the ideal css , the distinguishability and the interference are related by @xmath70 , in the presence of realistic noise , this symmetry relation is broken and we benefit from treating the two quantities independently .
two benchmarks can be derived to demonstrate that we observe nontrivial interference effects in comparison to the classical theory .
the first benchmark represents the interference achievable purely by a mixture of coherent states . passing it is a first witness of non - classical interference .
the second benchmark , obtained through optimization over all possible mixtures of gaussian states , then serves as a witness of higher - order non - classical interference .
once again we stress that these criteria are different from simply observing negativities , which is always a witness of non - classicality , but does not discriminate between a css and a single - photon state like our interference factor @xmath65 does . in order to verify the single photon - to - css transition in our experiment
, we employ @xmath54 in order to describe the emergence of the two coherent - state peaks in the css . at the same time , @xmath65 enables us to rule out incoherent mixtures and verify the presence of non - classical interference . for the css - to - single photon transition
, we introduce an additional figure of merit , since any loss of phase properties in @xmath54 and @xmath65 ( as a possible indication for a successful conversion into a single - photon state ) may as well be explained by an incoherent dephasing . in order to describe this reverse transition from the css back to the single - photon state
, we take advantage of an important feature of the ideal single - photon state it is indivisible at a beam splitter .
this property , which has been essential in many single - photon experiments , can be evaluated through the so - called anti - correlation factor @xmath71 ( ref .
this operational measure is obtained assuming that the state in question is split at a balanced beam splitter and the two output modes are measured by realistic single - photon detectors .
the anti - correlation parameter is then @xmath72 , where @xmath73 is the probability for both detectors registering a photon , while @xmath74 is the probability for only one detector registering a photon ( independent of the other detector ) .
for a mixture of a single - photon and a vacuum state , @xmath75 , the anti - correlation parameter is @xmath76 , while classical light always has @xmath77 . as the amplitude of the ideal odd css increases , @xmath78^{-2}$ ] grows from zero , approaching unity . for the ideal squeezed single - photon state ,
@xmath71 increases monotonously with the squeezing and for @xmath79 , @xmath71 becomes larger than unity , thus confirming the loss of the single - photon character .
the phase - insensitive superposition of coherent states present in a single - photon state , @xmath80 , is a typical manifestation of wave - particle duality even in the supposedly particle - like state @xmath45 , wave - like coherent states are present , but these form a continuous superposition with arbitrary phases and hence are not visible in the total state .
however , in a @xmath20 @xmath81 css conversion , the squeezer can break this rotational symmetry and amplify coherent states along a single axis , thus making them more distinguishable while simultaneously preserving their interference features .
this is demonstrated in suppl .
[ fig_suppl1 ] , where we can see that squeezing not only increases distinguishability ( which could as well be a consequence of gaussian noise ) , but also shifts the maximum of @xmath54 towards larger values of @xmath82 , as one would expect from a larger css . at the same time , the minimum of @xmath65 is also shifted , thus preserving the symmetry of an ideal css while remaining negative and distinct from a mixture of coherent states .
the weakest squeezing operation ( @xmath83 ) even exhibits interference properties going beyond those of any gaussian state . at the same time , in suppl .
[ fig_suppl2 ] , we can observe a phase modulation of @xmath65 which emerges as a consequence of the phase - sensitive squeezing process .
for the opposite conversion , css @xmath81 @xmath20 , the above criteria are ambiguous , because the transitions of @xmath84 and @xmath85 and their loss of the phase dependence can be as well explained by an incoherent dephasing effect . however , by looking at the anti - correlation parameter @xmath71 , which effectively serves as a measure of the ` single - photon ' quality of the state , we see that the values of this parameter change from @xmath86 to @xmath87 . recalling that for an ideal single - photon state we have @xmath88 , this shows that the quality of the single - photon state present in the total state is enhanced , even though the absolute single - photon fraction in the resulting mixed state is actually reduced .
as a fundamental light source , we utilize a continuous - wave ti : sapphire laser with a wavelength of @xmath89 nm .
some part of the fundamental wave is converted to a second harmonic wave in order to pump the optical parametric oscillators ( opo-1 and 2 ) . as a non - classical input state ,
we generate single - photon states , or , alternatively , coherent - state superpositions ( csss ) . in fig .
[ fig : setup ] of the main text , the setup for creating these input states is visualized by an orange background .
compared to the setups of refs .
@xcite and @xcite , in the present scheme , we can switch between the two different non - classical input states . in the case of a @xmath20 input ,
non - degenerate photon pairs are generated in opo-1 and one photon of every pair is then detected by an avalanche photo diode ( apd ) .
a click of the apd at this arm heralds the presence of a single photon in the other arm . the separating cavity ( sc ) behind opo-1 allows for a spatial separation of each pair of correlated photons .
the filtering cavities ( fcs ) extract a single frequency mode of opo-1 from hundreds of modes in every @xmath90 mhz of the hundreds ghz bandwidth of quasi - phase - matching , where @xmath90 mhz is the free spectral range of opo-1 . for the case of an initial css
, opo-1 generates degenerate photon pairs , i.e. , squeezed vacuum states .
every click of the apd then corresponds to the subtraction of a photon i.e. , @xmath91 .
our measurement - based squeezer works as follows ( visualized by a blue background in fig .
[ fig : setup ] of the main text ) @xcite .
an ancillary @xmath16-squeezed state and the input quantum state are coupled by a beam splitter , where the level of squeezing @xmath33 can be adjusted by the beam - splitting ratio @xmath2 with @xmath92 .
we demonstrate three different squeezing levels , @xmath33 @xmath93 @xmath94 , @xmath28 , and @xmath29 , corresponding to @xmath2 @xmath93 @xmath95 , @xmath96 , and @xmath94 , respectively .
then , one output of the beam splitter is measured with respect to the @xmath15 quadrature and the outcome is fedforward to the @xmath15 quadrature of the other output mode . a negative squeezing parameter _
e.g. _ @xmath36 is achieved by @xmath97 rotations of the optical phases , corresponding to an @xmath15-squeezed ancilla , a measurement with respect to @xmath16 , and a feedforward on @xmath16 .
the feedforward operation is performed as follows .
first , a phase modulation is added to an auxiliary beam by means of an electro - optic modulator ( eom ) , where the applied voltage is proportional to the measurement result .
then , the auxiliary beam is coupled to the signal beam by a highly asymmetric beam splitter .
an optical delay of @xmath3 m in free space is utilized for a broadband synchronization of the feedforward @xcite .
thus , we achieve broadband squeezing of up to @xmath98 mhz , which is @xmath99 times broader than the conventional squeezing @xcite .
note that the bandwidth limitation is that of the ancillary squeezed light .
the bandwidth is broad enough to operate a full @xmath100 mhz bandwidth of the heralded quantum states , where @xmath100 mhz is the half - width - half - maximum of opo-1 . in a preliminary experiment ,
the squeezing level of the ancilla squeezed vacuum state was @xmath101 db , while the antisqueezing level was @xmath102 db .
the finite squeezing level of the ancilla causes some excess noise at the output of the squeezing gate , while the antisqueezing level of the ancilla does , in principle , not have any effect on the output , as will be explained below .
phase fluctuations in the delay line can have the negative effect of degrading the squeezing level of the ancilla , because a part of the antisqueezed quadrature noise disturbs the conjugate squeezing quadrature .
however , in our experiment , the phase fluctuations are sufficiently suppressed by a feedback control to the level of a few degrees .
in addition , the beam pointing in the delay line is also actively controlled in the experiment via feedback , by which a significant decrease of the interference visibility can be prevented during the experiment . the beam pointing control is implemented through a combination of a four - quarter photodiode ( to detect misalignments ) and a mirror mount attached to three piezoelectric actuators ( to compensate misalignments ) .
as for the high - purity single - photon state with a single - photon component of @xmath43 used as the input state , among the remaining inefficiencies of @xmath103 , the inefficiencies due to the detection process are estimated to be about @xmath104 .
the measured optical propagation losses inside and outside the opo are @xmath105 in total .
the contribution of fake signals from the apd corresponds to @xmath106 .
higher photon number components are about @xmath107 . an input state to be squeezed
is represented by its quadrature operators @xmath108 , while an ancilla squeezed vacuum state is represented by @xmath109 .
a linear transformation of these quadrature operators in the heisenberg representation uniquely represents the gaussian gate operation .
if we want to squeeze the @xmath15 quadrature , we choose the ancilla state to be an @xmath15-squeezed state , @xmath111 , where @xmath112 is a quadrature operator of a vacuum state and @xmath113 is a squeezing parameter .
next , we measure @xmath114 , and then perform a displacement operation on the remaining state along the direction of the @xmath16 quadrature by an amount proportional to the measurement outcome , corresponding to the transformation @xmath115 and @xmath116 , leading to as can be seen above , the ancilla s antisqueezed noise is canceled by the feedforward operation , and only the squeezed quadrature of the ancilla remains as an extra term . in the limit @xmath118 , this extra term becomes negligibly small , and the overall transformation approaches an ideal squeezing operation @xmath119 with @xmath120 .
similarly , if we want to squeeze the @xmath16 quadrature , we choose the ancilla state to be a @xmath16-squeezed state , @xmath121 , where @xmath122 is a quadrature operator of a vacuum state .
this time we measure @xmath123 and then perform a displacement operation on the remaining state along the direction of the @xmath15 quadrature , corresponding to the transformation @xmath124 and @xmath125 , this time resulting in all the elements of this squeezing gate , i.e. , the creation of an ancilla squeezed vacuum state , the beam splitter interaction , the homodyne detection , and the feedforward displacement operation , are each realizable in an unconditional and deterministic fashion , and hence the squeezing gate as a whole also works unconditionally and deterministically .
the only imperfection is the excess noise in the squeezed quadrature due to the finite ancilla squeezing . | we implement the squeezing operation as a genuine quantum gate , deterministically and reversibly acting ` online ' upon an input state no longer restricted to the set of gaussian states . more specifically , by applying an efficient and robust squeezing operation for the first time to non - gaussian states
, we demonstrate a two - way conversion between a particle - like single - photon state and a wave - like superposition of coherent states .
our squeezing gate is reliable enough to preserve the negativities of the corresponding wigner functions .
this demonstration represents an important and necessary step towards hybridizing discrete and continuous quantum protocols . from a fundamental point of view
, quantum states of light can behave in a complementary fashion , showing both particle - like and wave - like behavior . with regards to an application such as quantum computing , an important proposal for universally processing photonic qubits
@xcite makes use of quantum particle detections ( i.e. , photon counting ) and quantum wave evolutions ( i.e. , quantum interferences through passive , energy - preserving linear optical circuits ) . in this scheme , however , the required ancilla states consist of many highly entangled photons and thus are out of reach of current experimental capabilities .
it is therefore reasonable to extend the toolbox of optical quantum operations in order to reduce the cost of the necessary quantum resources .
for instance , apart from discrete ` click by click ' measurements that rely on the particle - like nature of light @xcite , quantum light fields may be detected more naturally via continuous phase - space measurements exploiting their wave - like features @xcite .
in fact , two recent continuous - variable teleportation experiments on non - gaussian input states , using gaussian entanglement and gaussian homodyne measurements , demonstrate that not only a wave - like coherent - state superposition ( css ) @xcite , but also a particle - like photonic qubit @xcite can be transferred efficiently and reliably , preserving the negativity of the wigner function and exceeding the classical fidelity limits , respectively . in order to further investigate the potential of such _ hybrid _ schemes @xcite , which simultaneously exploit discrete and continuous techniques for encoding , measuring , and processing quantum states of light , it is desirable , besides gaussian measurements and resource states , to also add gaussian gate operations to the set of possible optical elements for processing photonic qubits . indeed , one feasible regime of single - photon operations has still remained unexplored : gaussian operations including _ active squeezers _ , still linearly transforming the mode operators , but no longer preserving energy .
the squeezing operation has been traditionally associated with continuous - variable quantum optics and information @xcite .
in fact , it can be considered the essential , elementary operation of this area , as it is a necessary component of all gaussian gates @xcite and even some non - gaussian gates require it for their implementation @xcite .
the construction of a genuine squeezing _ gate _ is fundamentally different from simply preparing a particular squeezed _
state_. this basic difference was previously addressed in an experiment employing a measurement - based protocol @xcite .
later , this scheme was extended to more advanced gaussian gates by using the squeezers @xcite as their fundamental building blocks : a quantum nondemolition sum gate @xcite , which may be understood as a continuous - variable version of the controlled - not gate for qubits , and a reversible phase - insensitive amplification ( two - mode squeezing ) gate @xcite , which also functions as an approximate cloner for coherent - state inputs . however , in all these previous demonstrations , only gaussian states have been used for the inputs .
one reason for this is of technical nature : the bandwidth of the squeezing gate is typically very narrow and there has been no way so far to generate highly non - classical , non - gaussian states in such a narrow frequency band .
moreover , these non - gaussian states tend to be extremely sensitive to losses , and thus , coupling them directly into an optical parametric oscillator will easily erase any signature of their strong nonclassicality such as the negativity of the wigner function .
our demonstration here was made possible by introducing a recent technique for bandwidth broadening as well as a mechanism for increased loss robustness to the squeezing gate @xcite .
we experimentally demonstrate for the first time a deterministic squeezing gate that operates on non - gaussian input states .
in particular , as what we believe to be a nice illustration , we use a particle - like single - photon state as the input state of the squeezing gate .
the resulting output state then is a wave - like css .
since single - mode squeezing corresponds to a unitary , noiseless amplification process along a certain phase - space direction , our single - photon squeezer can be also interpreted as a phase - sensitive amplifier acting on an optical field mode in its first excited state ( for a more detailed discussion of this interpretation , see the supplemental material @xcite ) .
furthermore , we also demonstrate the inverse operation of the squeezer , where a wave - like css is converted into a particle - like single - photon state . from a more fundamental point of view , what we demonstrate here can be considered an unconditional and reversible two - way conversion between a single quantum particle and a non - classical , continuous wave . unlike the previous probabilistic conversion from a photon number state to a css @xcite , the squeezing gate deterministically and reversibly transforms a single photon into a css .
the css , @xmath0 , where @xmath1 is a coherent state @xcite , is a highly non - classical quantum state sufficient for universal quantum computation @xcite .
it is worth noting that an all - optical , high - purity , almost - on - demand single - photon source was reported recently @xcite , while no such source has ever been demonstrated for a css state .
therefore , our unconditional conversion between these two types of states means that , in principle , all such quantum resources , including css states , are now available nearly on demand .
we believe that this experiment paves the way for quantum applications that combine discrete - particle and continuous - wave protocols in a so - called hybrid fashion .
the squeezing - gate operation when acting on non - gaussian states has also a number of direct applications , such as quantum state discrimination of optical coherent - state qubits @xcite , gaussian optimization in non - gaussian state preparation @xcite , improved quantum state transmission through a lossy channel @xcite , and preprocessing before a light - matter coupling for an efficient quantum memory interface @xcite .
recently , it has been also realized that squeezing is an extremely useful tool for manipulating and measuring individual photons , for instance , in squeezing - enhanced bell measurements of optical qubits @xcite or in squeezing - enhanced entanglement distillation protocols where optical gaussian states are locally transformed into qubit bell pairs @xcite . 1 ): beam splitter with transmittance @xmath2 determining the degree of squeezing , opo : optical parametric oscillator , sc : separating cavity , fc : filter cavity , apd : avalanche photo diode , hd : homodyne detector , lo : optical local oscillator , eom : electro - optic modulator , cc : classical channel , r : reflectivity
. an optical delay line of about @xmath3 m , traveling in free space , is used to match the propagation times of the two signals , one of which gets converted to an electrical signal and back , while the other one remains optical throughout .
, title="fig : " ] the schematic of our experimental setup is shown in fig .
[ fig : setup ] .
it consists of two parts : a source of non - classical states and an unconditional squeezer . for the former , via a small variation of the setup , we can choose the non - classical states to be either a single photon @xcite or a css @xcite .
the states will always emerge randomly in time , however , from a photon `` click '' at the avalanche photodiode ( apd ) , we know whenever a state arrives .
these `` heralded ''
non - classical states are localized in time around the detections of correlated photons @xcite .
therefore , our unconditional squeezer must have enough bandwidth to be applicable in the corresponding short time slots @xcite .
we extended our previous measurement - based squeezer @xcite to meet this requirement .
this squeezer avoids direct coupling of fragile input states to nonlinear optical media , which typically involves large optical losses . instead
, an ancillary squeezed state is utilized as a resource of nonlinearity @xcite ( see the supplemental material @xcite ) . in this scheme ,
the higher the squeezing level of the ancilla state becomes , the more closely the squeezing operation resembles a unitary , completely reversible squeezing gate .
although our squeezer is assisted by homodyne detection on the ancilla beam , the non - classical signal state is never directly measured ( see the quantum eraser @xcite ) when the squeezer is applied . 1 in order to verify the conversions , we perform quantum homodyne tomography on input and output states @xcite .
recall that wave and particle properties in quantum optics are formally connected via a pair of annihilation and creation operators for photons , @xmath4 and @xmath5 , respectively .
these non - hermitian operators are the quantized versions of the complex and complex conjugate amplitudes of an optical field mode , satisfying the bosonic commutation relation @xmath6 = 1 $ ] .
similarly , the quadrature operators , @xmath7 and @xmath8 , correspond to the quantized real and imaginary parts of the optical complex amplitudes ( up to a factor @xmath9 ) , where @xmath10 = i$ ] . through homodyne detection , the quadrature @xmath11 can be measured , which gives an hermitian part of the operator @xmath12 ; @xmath13 and @xmath14 then correspond to @xmath15 and @xmath16 , respectively .
the experimental results for converting single - photon states into several csss are shown in fig .
[ fig : results1tocss ] , and those for the reciprocal conversion are given in fig .
[ fig : resultscssto1 ] .
the top panels show the phase dependence of quadrature distributions obtained by a series of homodyne measurements . from these ,
wigner functions and photon - number density matrices are calculated , as shown in the lower panels . in fig .
[ fig : results1tocss ] , the leftmost column shows the input single - photon state , while the three right columns show the output csss for three different squeezing levels .
similarly , in fig . [ fig : resultscssto1 ] , the left column shows the input css , and the right column shows the output single - photon state . 1 of @xmath17 . ( a )
quadrature distributions over a period .
( b ) wigner functions .
( c ) photon number distributions and photon number representation of density matrices .
the minimum value of @xmath18 for the input wigner function becomes @xmath19 after the conversion .
, title="fig : " ] we shall first discuss the quadrature distributions ( top panels ) .
the fock state @xmath20 of a single photon , which is a typical carrier of discrete - variable quantum information , is a highly non - classical energy eigenstate of a quantized oscillator with a totally undetermined phase .
the phase - insensitivity of the quadrature distribution is a characteristic of a single - photon state , as can be seen in the leftmost panel of fig .
[ fig : results1tocss](a ) and the right panel of fig .
[ fig : resultscssto1](a ) . on the other hand , any coherent state is an eigenstate of the annihilation operator , @xmath21 .
this corresponds to a sinusoidal wave with mean complex amplitude @xmath22 and minimal quantum noise @xcite . by superimposing two coherent states , @xmath0
, the quadrature distribution corresponds to two sinusoidal waveforms with quantum interference at each intersection , like in the three right panels of fig .
[ fig : results1tocss](a ) and the left panel of fig .
[ fig : resultscssto1](a ) .
this quantum interference is a witness for a genuine quantum superposition of @xmath23 and @xmath24 and it would never occur for a stochastic mixture of coherent states .
the conversion is achieved by means of a squeezing operation , @xmath25 , where @xmath26 quantifies the amount of squeezing . in fig .
[ fig : results1tocss](a ) , the phase - dependent oscillations increase with larger squeezing .
three different squeezing levels @xmath27 , @xmath28 , and @xmath29 are demonstrated , resulting in three different amplitudes of csss , @xmath30 , @xmath31 , and @xmath32 , respectively .
the gap at the intersection of the waves becomes less pronounced for larger @xmath33 because of the finite squeezing of the ancilla mode ( about @xmath34 db relative to shot noise ) . in an opposite manner , in fig .
[ fig : resultscssto1](a ) , a phase - dependent oscillation is canceled by squeezing , resulting in a phase - independent distribution with a gap , like for a single - photon state .
the input css with @xmath35 is converted by squeezing with @xmath36 . in the corresponding wigner functions ( middle panels ) , where detector inefficiencies and losses are not corrected ,
the non - classicality of the input and output states becomes manifest in negative values .
these wigner functions are converted from rotationally symmetric to asymmetric [ fig .
[ fig : results1tocss](b ) ] and from asymmetric to symmetric [ fig .
[ fig : resultscssto1](b ) ] , while preserving their large negative values at the phase - space origin . in fig .
[ fig : results1tocss](b ) , the minimum value of @xmath37 at the input becomes , respectively , @xmath38 , @xmath39 , and @xmath40 , at the output . in fig .
[ fig : resultscssto1](b ) , @xmath18 at the input becomes @xmath19 at the output .
the density matrices ( bottom panels ) represent the particle picture . in the particle picture ,
the effect of the squeezing is an infinite superposition of photons added and subtracted in multiples of two . as a result , squeezing leads to a superposition of even photon number states when applied to the vacuum and to a superposition of odd photon number states when applied to the single - photon state .
being in such an odd - number superposition is also a distinct feature of the target css , @xmath41 and this is exactly the reason why squeezing achieves the desired conversion @xcite .
the diagonal elements of the density matrices represent photon number distributions , while the off - diagonal elements correspond to superpositions of @xmath20 and @xmath42 .
the input single - photon state in fig .
[ fig : results1tocss](c ) has a dominant single - photon component of @xmath43 ( without any corrections ) , while the input css in fig .
[ fig : resultscssto1](c ) has dominating one- and three - photon components compared to the and four - photon terms .
this also holds for the off - diagonal interference terms such as @xmath44 .
two - photon creations and annihilations are revealed by an increase [ fig .
[ fig : results1tocss](c ) ] and a decrease [ fig .
[ fig : resultscssto1](c ) ] of the three - photon components , respectively . in order to quantitatively
assess the experimental conversion processes , besides reconstructing the wigner functions and density matrices of the input and output states , we used two additional figures of merit .
these are specifically designed to reveal either the most distinct features of the particle - to - wave transition or that of the converse , wave - to - particle transition @xcite ( for details , see the supplemental material @xcite ) . from an experimental point of view , it has been considered notoriously hard to apply a quantum optical squeezing operation upon more exotic , non - gaussian quantum states such as discrete - variable single - photon states . in our experiment
we have succeeded in this difficult task . by demonstrating the efficient and deterministic squeezing and unsqueezing of a single photon ,
we have opened an entirely new optical toolbox for future quantum information applications . in principle
, such a unitary , phase - sensitive amplifier ( and attenuator ) will allow for making use of the entire fock space when processing single photons , which may help to construct quantum gates and error correction codes for logical qubits .
using our universal and reversible low - loss broadband squeezer , we have for the first time access to a complete set of deterministic gaussian operations applicable to non - classical , non - gaussian states .
these expand the toolbox for hybrid quantum information processing @xcite , and therefore our result will directly lead to applications in this area . at the same time , besides providing a completely new class of optical quantum processors , our experiment bridges two quantum - mechanically distinct regimes : that of particle - like quantum states such as single photons with that of more wave - like states such as coherent - state superpositions .
this work was partly supported by pdis , gia , g - coe , and apsa commissioned by the mext of japan , first initiated by the cstp of japan , scope program of the mic of japan , and , jsps and ascr under the japan - czech republic research cooperative program .
. acknowledges projects p205/12/0577 of ga r .
p.m. acknowledges projects gpp205/10/p319 of ga r and lh13248 of czech ministry of education .
p.v.l . acknowledges support from the bmbf in germany ( quorep and hipercom ) . 99 e. knill , r. laflamme , and g. j. milburn , nature * 409 * , 4652 ( 2001 ) .
r. ionicioiu and d. r. terno , phys .
rev .
lett . * 107 * , 230406 ( 2011 ) .
r. j. glauber , phys .
rev . * 131 * , 27662788 ( 1963 ) .
n. lee , h. benichi , y. takeno , s. takeda , j. webb , e. huntington , and a. furusawa , science * 332 * , 330333 ( 2011 ) .
s. takeda , t. mizuta , m. fuwa , p. van loock , and a. furusawa , nature * 500 * , 315318 ( 2013 ) .
a. furusawa and p. van loock , _ quantum teleportation and entanglement : a hybrid approach to optical quantum information processing _
( wiley vch , berlin , 2011 ) .
d. f. walls and g. j. milburn , _ quantum optics _
( springer - verlag , berlin heidelberg new york , 1994 ) .
s. l. braunstein and p. van loock , rev .
mod .
phys . * 77 * , 513577 ( 2005 ) . c. weedbrook , s. pirandola , r. garca - patrn , n. j. cerf , t. c. ralph , j. h. shapiro , and s. lloyd , rev .
mod .
phys .
* 84 * , 621669 ( 2012 ) .
s. l. braunstein , phys .
rev .
a * 71 * , 055801 ( 2005 ) .
p. marek , r. filip , and a. furusawa , phys .
rev .
a * 84 * , 053802 ( 2011 ) .
j. yoshikawa , t. hayashi , t. akiyama , n. takei , a. huck , u. l. andersen , and a. furusawa , phys .
rev .
a * 76 * , 060301(r ) ( 2007 ) .
j. yoshikawa , y. miwa , a. huck , u. l. andersen , p. van loock , and a. furusawa , phys .
rev .
lett .
* 101 * , 250501 ( 2008 ) .
j. yoshikawa , y. miwa , r. filip , and a. furusawa , phys .
rev .
a * 83 * , 052307 ( 2011 ) .
see the supplemental material for additional evaluation of the conversion and for further experimental methods .
a. ourjoumtsev , h. jeong , r. tualle - brouri , and p. grangier , nature * 448 * , 784786 ( 2007 ) . t. c. ralph , a. gilchrist , g. j. milburn , w. j. munro , and s. glancy , phys
. rev .
a * 68 * , 042319 ( 2003 ) .
j. yoshikawa , k. makino , s. kurata , p. van loock , and a. furusawa , phys .
rev .
x * 3 * , 041028 ( 2013 ) .
m. takeoka and m. sasaki , phys .
rev .
a * 78 * , 022320 ( 2008 ) .
d. menzies and r. filip , phys .
rev .
a * 79 * , 012313 ( 2009 ) .
r. filip , phys .
rev .
a * 87 * , 042308 ( 2013 ) .
p. marek and r. filip , phys .
rev .
a * 81 * , 042325 ( 2010 ) .
h. zaidi and p. van loock , phys .
rev .
lett . * 110 * , 260501 ( 2013 ) .
h. takahashi , j. s. neergaard - nielsen , m. takeuchi , m. takeoka , k. hayasaka , a. furusawa , and m. sasaki , nat .
photon . * 4 * , 178181 ( 2010 ) .
s. zhang and p. van loock , phys .
rev .
a * 84 * , 062309 ( 2011 )
. j. s. neergaard - nielsen , b. melholt nielsen , h. takahashi , a. i. vistnes , and e. s. polzik , opt . exp . *
15 * , 79407949 ( 2007 ) .
k. wakui , h. takahashi , a. furusawa , and m. sasaki , opt .
exp . * 15 * , 35683574 ( 2007 ) .
r. filip , p. marek , and u. l. andersen , phys .
rev .
a * 71 * , 042308 ( 2005 ) .
u. l. andersen , o. glckl , s. lorenz , g. leuchs , and r. filip , phys .
rev .
lett . * 93 * , 100403 ( 2004 ) .
a. i. lvovsky , j. opt .
b * 6 * , s556s559 ( 2004 ) .
a. i. lvovsky and m. g. raymer , rev .
mod .
phys . * 81 * , 299332 ( 2009 ) .
a. p. lund , h. jeong , t. c. ralph , and m. s. kim , phys .
rev .
a * 70 * , 020101(r ) ( 2004 ) .
p. grangier , g. roger , and a. aspect , europhys . lett .
* 1 * , 173179 ( 1986 ) . @sw * supplemental material * |
silicon , one of the most relevant materials from the technological point of view , has excellent electronic properties as a semiconductor . however , in the crystalline form , it does not exhibit good optoelectronic properties because of the indirect gap , in the infrared range , between the valence and the conduction bands , and of the low probability of radiative electron - hole recombination .
however it is possible to modify deeply such properties using a process of electrochemical etching @xcite which produces a material known as light emitting porous silicon ( leps ) .
this form , because of its luminescence properties , is presently the object of great scientific and technological attention @xcite .
in fact , inside this material one can observe some wire structures : the wires are strictly interconnected with each other , and have a mean diameter of a few tens of ngstrom .
quantum confinement of charge carriers seems to be at the origin of the optical characteristics of the material , which are quite different from those of bulk silicon . in the recent past many different theoretical studies @xcite
have been developed with the aim of analyzing the actual validity of the quantum confinement hypothesis in order to explain those new optical effects . according to the current experimental evidence , besides the presence of nanostructures of reduced dimensionality ( wires and/or dots ) , the key ingredients that are necessary to induce luminescence are crystallinity and surface passivation of dangling bonds . in the present work we concentrate on the quantum wire model proposed to explain the optical properties of leps .
we use the _ ab initio _ molecular dynamics ( aimd ) method , as proposed by car and parrinello @xcite , to study silicon quantum wires oriented along the ( 111 ) crystallographic direction .
the aimd method has been successfully applied to a variety of different systems and problems @xcite .
porous silicon samples grown on a silicon substrate in the ( 111 ) lattice direction were characterized both in the morphology and in the optical properties @xcite .
the study presented here follows a similar work developed for quantum wires oriented along the ( 100 ) crystallographic direction @xcite .
this study showed the presence of a direct gap in the visible range and an important enhancement of the dipole matrix elements , which are directly related to the transition probability between the valence and the conduction band and , in consequence of that , to the absorption coefficient .
this work , such as all the others quoted above , shows how the optical and electronic properties of silicon become deeply modified by quantum confinement .
these predictions , with the limitations due to the local density approximation ( lda ) used in the aimd framework , support the thesis that the origin of porous silicon photoluminescence lies in the localization of charge carriers inside ordered structures of reduced dimensionality .
the present work tries mainly to ascertain if the orientation may be relevant in determining the optical and electronic properties of the material since , _ a priori _ , it is not obvious that charge carriers confinement acts in a similar way along different symmetry planes . in section ii we describe the geometry of the model used and give some computational details . in section
iii the results on the structural , vibrational and electronic properties of the model are discussed .
the final section is devoted to the conclusions . 0.4truecm
crystalline silicon is characterized by a diamond structure , in which atoms are distributed in a face centered cubic ( fcc ) lattice , each lattice site being occupied by a pair of particles whose chemical bond , about 2.35 long , lies in the ( 111 ) direction .
the silicon lattice structure along the ( 111 ) direction can be characterized by three double layers periodically repeated with a periodicity @xmath1 , where @xmath2 is the lattice parameter of bulk crystalline silicon ( 5.43 ) .
[ f : struc ] shows the smallest wire that one can build , still preserving the crystalline order along the ( 111 ) direction .
the unit cell , indefinitely replied in space through periodic boundary conditions ( pbc ) , contains 26 silicon atoms , plus 30 hydrogen atoms which are necessary to saturate the dangling bonds at the surface of the wire .
the geometry of the model is such that it exhibits a non uniform size along the longitudinal direction . in a way
, this feature makes the model more similar to real samples of porous silicon .
in fact , the diameter of the model has a maximum value of 7.67 , and a mean value of 5.51 .
we have also studied a larger wire whose basic structure consists of 62 silicon atoms and 42 hydrogen atoms , with a maximum diameter of 11.72 , and a mean value of 10.37 .
we note that a system consisting of 104 atoms is rather demanding on the computational side for _ ab - initio _ car - parrinello simulations . for this reason , we limited our calculations for the larger wire to those features which may confirm some specific properties of the thinner wire . in both cases
the size of the supercell in the longitudinal direction corresponds exactly to the bulk silicon characteristic size of 9.41 along the ( 111 ) direction .
the lateral size of the supercell is much larger than the diameter of the wire so as to avoid interactions of the particles with their images .
the parameters needed in the aimd simulation were suggested by the previous experience on this material : we used a time step of 7 a.u .
, an energy cutoff for the plane - wave expansion of 8 ry , an electronic mass of 300 a.u . , and pseudopotentials of the bachelet - hamann - schluter @xcite type for describing the interaction between valence electrons and the nuclei with core electrons .
as a first step , we wanted to verify the structural stability of the wire .
to this aim , after the electronic ground - state relative to the initial ionic configuration was found , we let all the ionic degrees of freedom relax with a steepest descent technique . we notice a small increase of the distance of the most external atoms from the axis of the wire , of the order of 0.16 , analogously to what was observed in @xcite .
as expected , we observe no relevant variations of the characteristic distances , along the longitudinal direction .
then , we let the system evolve dynamically for a time of about 2 ps , a considerable time in aimd , so that one may be confident that the system is reasonably close to thermal equilibrium . during this evolution the model did not exhibit any pathological behaviour : the system oscillates about the equilibrium configuration with no substantial atomic rearrangements .
we observe that the total energy ( potential + kinetic ) fluctuates around an average value which remains constant , as should be in a microcanonical aimd simulation .
the average temperature during this dynamical evolution is about 70 k. we also computed the velocity autocorrelation function and , through its fourier transform , an indicative vibrational density of states ( vdos ) of the system .
we find a good qualitative agreement with the typical experimental data of the vibrational modes of si - si and si - h bonds @xcite . from a strictly quantitative point of view
, we remark that the relative intensity of the peaks should not be taken too rigorously .
however the positions of the peaks are in fair agreement with the experimental data . in particular we recognize in the spectra of fig .
[ f : vibra ] the characteristic signal of the stretching of the si - h bond at about @xmath3 @xmath4 .
the double peak has its origin in the presence of doubly hydrogenated silicon atoms , which give rise to symmetrical and antisymmetrical stretching of the bond at slightly different frequencies .
the analysis of the oscillations of single and twin hydrogen particles shows that the higher - frequency peak can be ascribed to the symmetrical mode , while the lower - frequency signal arises from antisymmetrical stretching vibrations .
we notice also , at lower frequencies , the typical signals of scissors mode ( @xmath5 @xmath4 ) of the twin hydrogen atoms bonded to one silicon atom , and of si - h bending ( around @xmath6 @xmath4 ) .
the weak structure centered at about @xmath7 @xmath4 is likely associated with another bending vibrational mode of hydrogen atoms as it shows up in the motion of both single and twin passivating hydrogens .
the low frequency noise produced by the finite time evolution renders more uncertain the identification of the modes below @xmath8 @xmath4 that are associated with the motion of silicon atoms .
it is important to remark that such characteristic vibrational modes are observed , by infrared spectroscopy , also in real freshly prepared leps samples @xcite . for the larger size sample
, we have also studied the relaxation towards the global energy minimum .
it is noteworthy that in this case the mean radial relaxation , in the equilibrium configuration , is negligibly small if compared with that estimated for the thinner wire .
this can be intuitively understood if one thinks that in the smaller - size model the ratio of surface to bulk atoms is definitely higher than in the larger wire .
such a small relaxation originates from the competition between the doubly hydrogenated silicon atoms , which move slightly away from the axis of the wire , and the singly hydrogenated ones , which instead move inward . once the minimum energy configuration has been determined , we computed the electronic band structure of the quantum wire along the one - dimensional brillouin zone ( bz ) .
we computed the kohn - sham eigenvalues for six different k - points of the bz , from the center ( @xmath9 ) to the border ( which would correspond to l in the bulk ) .
the resulting band structure is shown in fig .
[ f : band ] for the ( 111)-oriented wire with a mean diameter of 5.51 .
the main result is the prediction of a direct gap of 2.83 ev between the conduction and the valence band at the zone center .
it is also interesting to notice that the second conduction state still shows a signature of the band structure of bulk silicon , i.e. a minimum located at about 0.2 . as to the gap , we recall that buda and co - workers found , for a ( 100)-oriented quantum wire with a diameter of 7.60 , a value of 2.33 ev @xcite . the difference is to be ascribed , as we shall see later , to the different mean diameter of the two models , more than to the different symmetry of the wires .
we recall that in bulk silicon the minimum of the conduction band along the @xmath9-l direction falls just at l , with a gap of 2.15 ev .
for the larger wire we find an energy gap of 1.90 ev at the center of bz which , consistently with the quantum confinement hypothesis , is larger than that in bulk silicon , but smaller than the value for the thinner wire .
the values of the two energy gaps , together with those of the ( 100 ) models and of bulk silicon as obtained in lda , are plotted in fig .
[ f : egap ] .
it is of striking evidence the linear behaviour of the energy gap as a function of the inverse size of the models .
it is remarkable that both the ( 111 ) and ( 100 ) computed values lie on the same straight line , suggesting the independence of such property on the specific orientation of the wires . in the same figure
we also show the theoretical prediction coming from effective mass arguments ( e@xmath10 ) , which should hold close to the bulk limit .
it appears that the matching between these two different behaviours should occur in the region between 2 and 5 nm .
the presence of a direct gap , and thus the possibility of zero - phonon transitions , is not a sufficient condition for ensuring good photoluminescence properties .
the probability of an electronic transition as a result of the absorption of a photon is given by the fermi golden rule , which in the dipole approximation is @xmath11 where @xmath12 and @xmath13 are the conduction and valence states involved in the electronic transition , while * e * is the polarization vector and * p * the momentum operator .
we determined all the matrix elements between kohn - sham states of the conduction and valence bands whose difference in energy falls in the visible range , or slightly above .
table [ t : small ] and table [ t : big ] give the values of the most relevant of these dipole - matrix elements for the smaller and the larger ( 111)-oriented wires .
we note the strong difference between the values relative to the longitudinal and transverse polarizations .
moreover , the quantum confinement of the electrons mixes up the states , thus allowing for transitions that are forbidden in the bulk by selection rules .
the dipole - matrix element between the top of the valence band and the bottom of the conduction band is not zero because of the mixing , even though not very large .
we computed the imaginary part of the dielectric function , that is related to the absorption coefficient , through the formula @xmath14 where @xmath15 and @xmath16 are the length of the wire and the volume of the supercell , respectively .
the results for the smaller model are shown in fig .
[ f : eps2small ] . in the construction of this quantity
we have considered only the dipole - matrix elements corresponding to energy differencies up to 4 ev . we observe that the component related to an external field parallel to the wire direction is much stronger than the perpendicular one , and this difference becomes larger and larger going to lower energies .
probably , the most interesting result is the presence , in the parallel component , of a small signal at the band gap ( 2.83 ev ) which is completely absent in the perpendicular one .
this is a clear effect of the mixing of the bulk states induced by the reduced dimensionality .
this result seems to be peculiar of the ( 111 ) wire , since in the ( 100 ) models @xcite the dipole - matrix element between the top of the valence band and the bottom of the conduction band was negligible in both the components . the plot of the dielectric function for the larger - sized wire , that is shown in fig .
[ f : eps2big ] , exhibits also a different behaviour in the parallel and the perpendicular components .
we notice also in this case a clear difference in intensity between the parallel and the perpendicular components at higher energies .
the most remarkable feature is the presence , below @xmath17 2.7 ev , of three peaks in the longitudinal component that are not obseved in the transverse one .
in particular we observe again a significative signal in correspondence of the energy gap ( 1.90 ev ) , whose intensity is lowered with respect to the smaller wire .
this is consistent with the fact that in the bulk limit this matrix element should go to zero . 0.4truecm
we have presented an electronic structure calculation , within the lda , of two ( 111)-oriented silicon quantum wires of different size .
the comparison of the present results with those obtained for the ( 001)-oriented models shows a substantial analogy of the electronic and optical properties of silicon quantum wires oriented along different crystallographic directions .
in particular we find a direct energy gap at the zone center whose magnitude decreases for larger wires consistently with the quantum confinement hypothesis .
quite surprisingly we find that the values of the gap for all the ( 111 ) and ( 001 ) wires studied are excellently fitted as a linear function of the inverse diameter .
therefore this quantity seems to be independent on the orientation and geometrical details , and is affected only by the width of the wire .
again , the present calculations support the idea that the photoluminescence properties of porous silicon can be ascribed to charge carriers confinement in nanostructures .
we find , similarly to the ( 001 ) wires , a strong anisotropy in the imaginary part of the dielectric function .
these results indicate that , if the leps structure consists of a network of randomly oriented wires @xcite , all of them will contribute to the photoluminescence independently on their specific orientation .
* acknowledgements * : this work was supported by the ministero delluniversita e della ricerca scientifica e tecnologica ( murst ) and by the istituto nazionale per la fisica della materia ( infm ) , italy .
the authors thank the centro di calcolo elettronico delluniversita degli studi di messina for the allocation of computer time .
a. m. s. would also like to acknowledge the hospitality of the ibm - ecsec centre in rome in the early stage of this work .
very useful discussions with j. kohanoff , s. ossicini , and o. bisi are also acknowledged .
@xmath18 & @xmath19 + 68 & 65 & 0.50@xmath20 & 0.11@xmath21 & 3.33 + 68 & 66 & 1.00@xmath20 & 0.14@xmath21 & 3.32 + 68 & 67 & 0.23@xmath22 & 0.86@xmath23 & 2.83 + 74 & 67 & 0.42@xmath20 & 0.17@xmath21 & 3.42 + 76 & 67 & 0.94@xmath24 & 0.45@xmath21 & 3.67 + @xmath18 & @xmath19 + 147 & 138 & 0.17@xmath25 & 0.12@xmath26 & 2.84 + 148 & 139 & 0.18@xmath25 & 0.12@xmath26 & 2.84 + 146 & 145 & 0 & 0.41@xmath20 & 1.90 + 156 & 145 & 0.36@xmath22 & 0.14@xmath26 & 2.83 + 167 & 145 & 0.17@xmath25 & 0.57@xmath26 & 3.34 + 168 & 145 & 0.11@xmath24 & 0.23@xmath26 & 3.36 + 171 & 145 & 0.17@xmath26 & 0.11@xmath26 & 3.44 + | we analyze the influence of spatial orientation on the optical response of hydrogenated silicon quantum wires .
the results are relevant for the interpretation of the optical properties of light emitting porous silicon .
we study ( 111)-oriented wires and compare the present results with those previously obtained within the same theoretical framework for ( 001)-oriented wires [ f. buda , j. kohanoff and m. parrinello , _ phys .
rev .
lett . _ * 69 * , 1272 , ( 1992 ) ] . in analogy with the ( 001)-oriented wires and at variance with crystalline bulk silicon
, we find that the ( 111)-oriented wires exhibit a direct gap at @xmath0 whose value is largely enhanced with respect to that found in bulk silicon because of quantum confinement effects . the imaginary part of the dielectric function , for the external field polarized in the direction of the axis of the wires , shows features that , while being qualitatively similar to those observed for the ( 001 ) wires , are not present in the bulk .
the main conclusion which emerges from the present study is that , if wires a few nanometers large are present in the porous material , they are optically active independently of their specific orientation . = 10000 0.4truecm |
the minimum - mass solar nebula ( mmsn ) is built by spreading the mass of the planets in concentric annuli then fitting a function to the distribution .
the original mmsn models produced surface density @xmath5 profiles that scaled with orbital radius @xmath6 as @xmath7 @xcite .
newer mmsn analyses have found solutions of @xmath8 @xcite and @xmath9 @xcite .
the mmsn model almost certainly does not represent the initial conditions from which the planets formed .
it presumes that the current orbital architecture of the solar system reflects the properties of its parent protoplanetary disk . however , the giant planets are thought to have undergone multiple episodes of orbital migration ; first during the gaseous disk phase @xcite and later due to planetesimal - driven migration @xcite .
models that do not invoke re - shuffling of the giant planets orbits systematically fail to reproduce the inner and outer solar system ( e.g. , * ? ? ?
* ; * ? ? ?
* ) . thirty to fifty percent of main sequence fgkm stars host planets with masses @xmath10 , radii @xmath11 , and orbital periods @xmath12 days @xcite .
these planets tend to be found in multiple systems in compact orbital configurations ( e.g. * ? ? ?
* ; * ? ? ?
the origin of these `` hot super - earths '' is debated ( see table 1 in @xcite as well as @xcite ) .
the two most likely formation mechanisms are inward , type 1 migration of a population of planetary embryos @xcite , and _ in situ _ accretion of embryos from a massive disk @xcite .
the bulk density distribution of hot super - earths should in principle be able to differentiate between the models @xcite .
however , this requires a large number of precise radius and mass measurements of low - mass planets @xcite and a good understanding of the degeneracies between internal structure models @xcite .
extra - solar planets can be used to create a minimum - mass _ extrasolar _ nebula ( mmen ) .
@xcite built an mmen using 11 radial velocity - detected systems dominated by jupiter - mass planets .
@xcite built an mmen using kepler s small ( @xmath13 ) planetary candidates by calculating a corresponding surface density for each candidate and performing a fit to the ensemble .
both analyses produced steep density profiles : @xmath14 for the gas giants and @xmath0 for the super - earths .
@xcite s mmen is by construction the disk required to form hot super - earths _ in situ_. however , it is roughly a factor of seven more massive than the mmsn ( * ? ? ?
* ; see below ) , and it ties all hot super - earths to a single disk or at least to a universal disk profile . as we show in section 2 , the minimum - mass disks calculated from multiple - planet systems show a wide range of surface density slopes .
this broad distribution of slopes can not be explained by selection effects or measurement errors ( section 3 ) . if hot super - earths formed by _ in situ _ accretion the inferred disks must reflect the initial distribution of planetary embryos @xcite .
however , many minimum - mass disks have very steep or flat profiles that can not be reconciled with viscous disk models ( section 4 ) .
we argue that this diversity does not represent the gas disk s true properties but rather is the product of a separation in the distributions of gas and solids in the inner parts of the disk , perhaps due to inward type 1 migration of planetary embryos .
our sample of planets consists of the kepler planet candidates @xcite and radial velocity systems downloaded via the open exoplanet database @xcite .
we first reproduced the result of @xcite using the kepler candidates with @xmath15 and @xmath16 days .
we assigned planet masses by the simple relation @xcite : @xmath17 we assigned each planet a surface density by simply calculating @xmath18 , where @xmath19 is the orbital radius .
the best - fit power - law was : @xmath20 this is very close to the mmen of @xcite , which is reassuring .
we next turned our attention to multiple - planet systems . to include systems discovered by both radial velocity and transits , we only considered systems with at least three planets interior to 1 au .
we restricted ourselves to systems containing only planets smaller than @xmath21 or less massive than @xmath22 .
for kepler candidates smaller than @xmath23 we assumed @xmath24 @xcite and for larger candidates we again assumed @xmath25 ( * ? ? ?
* ; although we tested different m - r relations , see below ) .
this effectively assumes an earth - like composition for planets with @xmath26 ( consistent with observations ; see * ? ? ?
* ) . for radial velocity planets we used minimum masses .
= 8.5 cm for each system we calculated the width of the concentric annuli over which to spread the planets masses as follows .
we first chose the midpoints between each pair of planets as the geometric mean of the two planets orbital radii .
we then applied the same orbital radius ratios to extrapolate beyond the endpoints .
for instance , to extrapolate interior to planets 1 and 2 ( with orbital radii @xmath27 and @xmath28 , where @xmath29 ) , the inner edge of the innermost annulus is @xmath30 . figure [ fig : mmsn ] shows the profiles of 191 minimum - mass disks ( 181 kepler systems and 10 rv systems ) . the median disk profile is @xmath31 this is 2 - 3 times more massive than the universal mmen we calculated above and that of @xcite .
it contains @xmath32 inside 1 au and is 15 - 200 times more dense than the minimum - mass solar nebula ( mmsn ) in that range .
[ fig : mmsn ] includes two estimates for the mmsn .
the first was calculated as above considering only the solar system s terrestrial planets : @xmath33 .
the second is a profile shown by accretion simulations to reproduce the terrestrial planets @xcite : @xmath34 .
there is a diversity of minimum - mass extrasolar nebulae ( fig .
[ fig : mmsn ] ) .
the surface density slopes range from -6.3 to + 5.8 .
19 disks have surface density slopes that increase with orbital distance , and another 19 have slopes of @xmath35 or steeper .
figure [ fig : hist ] shows the full distribution of disk slopes ( excluding the widest outliers ) .
= 8.5 cm our results are only weakly sensitive to the chosen mass - radius relation .
we tested two @xmath36-@xmath37 relations in addition to our default , two - piece @xmath36-@xmath37 relation .
first , we applied the @xmath25 relation of @xcite to all planets regardless of mass .
second , we applied the more complex , stellar flux - dependent mass - relation from @xcite .
the mmen slopes changed modestly between the different m - r relations : the median disk profile was @xmath38 for both our fiducial and the @xcite relations , and @xmath39 for the @xcite one .
k - s tests showed that the differences were not statistically significant .
there exist biases in generating minimum - mass disks . to address what underlying distribution could produce the inferred distribution of disks , we performed simple simulations of the detection of a population of planetary systems with a range of underlying characteristics .
we generated synthetic planetary systems from disks with specified surface density profiles .
these systems were designed to roughly reproduce the underlying characteristics of the kepler planetary candidate systems .
each artificial system contained six planets .
the inner edge of the disk was assumed to be between 0.025 and 0.075 au .
the period ratios among adjacent planets were drawn randomly between 1.3 and 3 ( see e.g. * ? ? ? * ) .
the planets masses were calculated by assuming perfect accretion within a given region of the disk , with boundaries between adjacent planets chosen at the geometric means of their orbital radii .
the planet masses were in most cases between 1 and 15 @xmath40 .
the planets were given randomly - chosen inclinations of up to @xmath41 ( * ? ? ?
* ; * ? ? ?
* although we show the effect of the inclination distribution below ) and randomly - chosen longitudes of ascending node .
each batch of simulations included @xmath42 systems .
we simulated the transit detection of these artificial systems .
we assumed that the innermost planet was found to transit , and that the viewing angle was aligned with that planet s longitude of ascending node .
we then calculated whether each other planet would transit along that line of sight by testing whether its * z * distance was between @xmath43 and @xmath44 , assuming a sun - like host .
we did not transform the planetary masses into radii but rather simply attributed an error to the mass of each planet that was found to transit .
the detected mass was the true planet s mass times a factor that followed a log - normal distribution with @xmath45 values of @xmath46 and @xmath47 .
this is simpler than more detailed studies @xcite , but it is sufficient for our purposes .
we performed the same fitting procedure on the simulated detections as for the real kepler systems .
we only included systems with at least three planets detected within 1 au .
this included cases where one or more planets were missed because of chance orbital alignment .
we compared the simulated detections with the kepler systems .
we first tested a universal disk profile with an @xmath4 surface density slope . with no uncertainties in measured masses and assuming perfectly coplanar systems
, our algorithm retrieved the correct surface density slope , with a median of 1.5 and a standard deviation of 0.1 .
for our default transit simulations with inclined orbits and mass errors the simulated systems had a median slope of 1.6 with a dispersion of 0.5 .
the dispersion is determined in large part by the assumed mass uncertainty ( see below ) .
the steeper slope comes from the fact that , when some planets are not found to transit , the annular area over which the detected planets masses are spread to produce a surface density is increased such that the surface density is decreased .
outer planets are more often missed than inner ones , which has a net steepening effect on the inferred profile , albeit with a large scatter .
we found evidence of this steepening in the data .
kepler systems with packed orbital configurations defined such that the maximum orbital period ratio @xmath48 of adjacent planets is less than 2 have flatter minimum - mass disks than systems in which at least one pair of adjacent planets had @xmath49 .
the median surface density slope was -0.95 for packed systems and -1.77 for the more widely - spaced systems with enough dynamical space to harbor additional non - transiting planets .
the difference between the two was statistically significant ( @xmath50 from a k - s test ) .
this steepening sets in at orbital inclinations of just @xmath51 and the distribution is independent of the underlying inclination distribution . for large mutual inclinations
a smaller number of systems have at least three transiting planets , but the inferred properties are essentially unaltered .
in - situ accretion from a universal disk profile is statistically inconsistent with the observations .
although the @xmath4 disks had a similar inferred median slope to the true kepler systems , the distribution was far too narrow ( fig .
[ fig : bias ] ) .
a k - s test showed a probability of @xmath52 that the two populations were drawn from the same underlying distribution .
the only way that we found to make the two distributions consistent was to dramatically increase the uncertainties in the inferred mass estimates to @xmath45 values between @xmath53 and @xmath54 ( i.e. , a factor of 30 within @xmath45 ) .
such gigantic mass uncertainties are unrealistic .
all populations with assumed single power laws suffered the same problem of producing distributions that were far too narrow ( fig .
[ fig : bias ] ) .
the simulated detections reproduced the observed distribution if we assumed a broad underlying distribution .
figure [ fig : bias ] shows two such examples that were consistent with observations ( each with @xmath55 ) .
the first assumed a flat distribution of surface density slopes evenly spread between @xmath56=0 and @xmath56=-2.5 .
the second assumed a gaussian profile centered at @xmath56=-1.25 with a standard deviation of 0.8 . in principle , a fraction of the observed systems could originate from in - situ accretion with a universal profile , as a wedge carved from the broad total distribution in fig .
[ fig : bias ] .
the maximum in - situ contribution is 30 - 50% for our default assumptions but depends on the inferred slope distribution , which is dominated by errors in the inferred planet masses .
we interpret this upper limit as a significant overestimate because the remaining distribution would likely require multiple origins to explain its distorted bimodal shape .
it is far simpler to assume a broad distribution of underlying disks .
= 8.5 cm there are other issues .
the radii of kepler candidates are uncertain , especially for low - mass stars ( e.g. * ? ? ?
* ) . to ensure that this did not affect our results
, we verified that our results are not a function of the stellar effective temperature .
we also note that many of the derived minimum - mass disks can not be representative of the large - scale disk structure .
disks with positive slopes contain thousands of earth masses in solids if projected to au radii , and disks with steep negative slopes contain very little total mass . finally , we have only considered low - mass and small planets so our minimum - mass disks pertain only to the solid components of the disk . of course
, if planets form _ in situ _ then the distribution of solids must reflect that of the gas @xcite .
gaseous protoplanetary disks are thought to evolve under the influence of internal viscosity ( e.g. * ? ? ? * ) .
viscous disk models produce surface density profiles with characteristic shapes .
they tend to zero at the inner edge and have a relatively shallow equilibrium slope extending out to close to the outer edge @xcite . for @xmath57-viscosity disks @xcite the equilibrium profile
is coupled to the temperature profile as : @xmath58 a temperature profile dominated by stellar irradiation follows a @xmath59 profile ( e.g. * ? ? ?
* ) with an equilibrium surface density profile of @xmath60 .
disks in which the temperature is determined by viscous heating can have a range of profiles from @xmath61 ( constant radial surface density ) to @xmath62 depending on the local opacity regime @xcite . a disk with a profile steeper that @xmath63 must have a temperature that increases with radius , which is not physical in the inner parts of disks .
alternately , steep profiles can be a result of an _ outward _ flow of gas in the outer parts of externally photo - evaporated disks @xcite .
sub - millimeter observations of cold dust in the outskirts of protoplanetary disks consistently derive surface density profiles between @xmath64 and @xmath65(@xcite ; although @xcite measured a wider range of slopes that included positive exponents ) .
interferometric measurements of molecular properties produce a similar range of density profiles but that extends to values as steep as @xmath4 @xcite .
if hot super - earths formed _ in situ _ then minimum - mass disks must reflect the properties of the gaseous disks that spawned them .
the distribution of minimum - mass disks is much broader than the parameters for gaseous disks that can be easily explained by theory .
the underlying distributions that are consistent with the kepler disks are similarly broad ( section 3 ) .
how can we reconcile this diversity of minimum - mass disks with observations and theory ? a simple solution is to invoke a separate evolution of solids and gas .
the planets from which minimum - mass disks are built would then not represent the structure of the full gas - dominated disk .
differential solid - gas evolution can occur at m- or smaller sizes due to rapid aerodynamic drift @xcite .
mars - mass or larger planetary embryos undergo type 1 migration via angular momentum exchanges with the protoplanetary disk @xcite . this migration can be directed inward or outward @xcite but is generally inward for embryos and planets with @xmath66 @xcite .
a leading model for hot super - earth formation invokes accretion during the inward migration of a population of planetary embryos @xcite .
migration is halted by the steep density transition at the inner disk edge @xcite .
embryos pile into chains of mean motion resonances .
waves of inward - migrating embryos destabilize the resonant chains , trigger collisions and generate new resonant chains .
this increases the solid surface density in the inner parts of the disk and forms planets that follow a steeper profile than the underlying gas disk , although there is a range of inferred slopes ( cossou et al , in preparation ) .
resonant chains are a weakness of the type 1 migration model ( but see * ? ? ?
the observed systems of hot super - earth are rarely found in resonance @xcite .
the _ in situ _ accretion model can reproduce a number of features of the observed systems including their orbital spacings @xcite .
late destabilization of resonant chains can act as a transition to the _ in situ _ accretion regime .
destabilization can be triggered by the dissipation of the gaseous disk or chaotic dynamics @xcite .
we thus favor a two - phase origin for hot super - earth systems .
first , inward migration of planetary embryos increases the solid surface density in the inner disk .
second , the resonant chain is disrupted during a final phase of late - stage accretion .
this model reproduces current observations and predicts that a significant fraction ( but not all ) of hot super - earths should be volatile - rich @xcite .
to conclude , we re - emphasize that minimum - mass nebulae both solar and extrasolar do not represent the underlying properties of the protoplanetary disks in which those planets formed .
a broad swath of studies have shown that planets and their building blocks undergo orbital migration at a range of size scales .
minimum - mass nebulae have been sculpted by this migration .
.1 in we are grateful to referee brad hansen for a helpful review and to franck hersant and franck selsis for discussions .
we acknowledge the anr for funding ( project _ mojo _ ) . | it has been proposed that the observed systems of hot super - earths formed _ in situ _ from high - mass disks . by fitting a disk profile to the entire population of kepler planet candidates , chiang & laughlin ( 2013 ) constructed a `` minimum - mass extrasolar nebula '' with surface density profile @xmath0 .
here we use multiple - planet systems to show that it is inconsistent to assume a universal disk profile .
systems with 3 - 6 low - mass planets ( or planet candidates ) produce a diversity of minimum - mass disks with surface density profiles ranging from @xmath1 to @xmath2 ( 5th-95th percentile ) . by simulating the transit detection of populations of synthetic planetary systems designed to match the properties of observed super - earth systems ,
we show that a universal disk profile is statistically excluded at high confidence .
rather , the underlying distribution of minimum - mass disks is characterized by a broad range of surface density slopes .
models of gaseous disks can only explain a narrow range of slopes ( roughly between @xmath3 and @xmath4 ) .
yet accretion of terrestrial planets in a gas - free environment preserves the initial radial distribution of building blocks .
the known systems of hot super - earths must therefore not represent the structure of their parent gas disks and can not have predominantly formed _ in situ_. we instead interpret the diversity of disk slopes as the imprint of a process that re - arranged the solids relative to the gas in the inner parts of protoplanetary disks .
a plausible mechanism is inward type 1 migration of mars- to earth - mass planetary embryos , perhaps followed by a final assembly phase .
[ firstpage ] planetary systems : protoplanetary disks planetary systems : formation solar system : formation |
it has recently become possible to collect large samples of high resolution cloud particle images in real time , opening up the modelling of cloud dynamics to detailed comparison with nature .
figure 1 shows ice crystal aggregates from a cirrus cloud over the usa , captured by non - contact aircraft - based imaging .
such aggregates can be seen to be comprised of varied rosette ice crystal types , and detailed statistics have recently been published on both the cluster aspect ratios [ _ korolev and isaac _ 2003 ] and size distributions [ _ field and heymsfield _ 2003 ] in cirrus clouds .
such aggregation is a key feature of cloud development in the troposphere and can be quite crucial to the development of precipitation , whether it reaches the ground as snow or melts first to arrive as rain .
the openness of the aggregates significantly accelerates their growth .
two clusters ( labelled by @xmath0 ) pass approach with their centres closer than the sum of their radii @xmath1 at a rate proportional to@xmath2 where for each cluster the sedimentation speed @xmath3 varies inversely with its radius @xmath4 and mass @xmath5 as @xmath6 here @xmath7 and @xmath8 are the viscosity and density of the air , @xmath9 is the acceleration due to gravity , and we have assumed that only one geometrical radius is relevant : _
mitchell _ [ 1996 ] discusses an elaboration .
given the above , the rates of aggregation per unit time for fixed cluster masses vary linearly overall with the cluster radii , and openness of aggregate structure enhances aggregation rates despite lowering fall speed . for real aggregates
this is a significant factor : using data from _ heymsfield et al .
_ , [ 2002 ] , one finds that rosette aggregates 2 mm across ( which yield 0.5 mm droplets ) aggregate four times faster than when melted . for cloud particles it is also relevant to consider the rates of aggregation per unit of distance fallen ( rather than per unit time ) , which at fixed mass is proportional to the square of radius , leading to 16 times enhancement for ice over water in the example cited .
we have made computer simulations of ice aggregation based on equations ( [ eq : kernel ] ) and ( [ eq : fallspeed ] ) , tracing trajectories through possible collisions to obtain accurate collision geometries .
we assumed that all collisions led to rigid irreversible joining of clusters , as the openness of the experimentally observed clusters suggests little large scale consolidation of structure upon aggregation , and that cluster orientations were randomised in between collisions but did not change significantly during them .
we took the sedimentation speeds to be governed by inertial flow , for which the mass dependence function in equation ( [ eq : fallspeed ] ) is given by @xmath10 .
details of implementation are given in a longer paper [ _ westbrook et al .
_ , 2004 ] .
some representative computer aggregates are shown in figure 1 alongside the experimental ones .
our simulations used three dimensional cross shapes for the initial particles as a crude representation of the experimental bullet rosettes . figure 2 shows a quantitative comparison of aggregate geometry , in terms of the ratio of cluster spans perpendicular to and along the direction of maximal span , as measured from projections in an arbitrary plane of observation .
we find that different initial particle geometries ( rosettes , rods ) approach a common asymptotic average cluster value . the aspect ratio of cpi images have been similarly calculated [ _ korolev and isaac _ 2003 ] , the results of which have been overlayed onto figure 2 , and these appear to approach the same value .
this universality of aspect ratios provides direct support for our hypothesis of rigid cluster joining upon contact . a deeper indicator of universality is provided by the fractal scaling of ice crystal aggregates , where one tests the relation @xmath11 between aggregate mass @xmath5 and linear span @xmath12 . our simulations and experimental observations [ _ heymsfield et al .
_ , 2002 ] , rather accurately agree on the fractal dimension @xmath13 and in _ westbrook et al . _
[ 2004 ] we discuss theoretical arguments leading to @xmath14 .
our simulations conform well to dynamical scaling of the cluster size distribution .
this means that number of clusters per unit mass varies with mass and time together as a single scaling function,@xmath15 where @xmath16 is the weight average cluster mass .
this relationship is confirmed in figure 3 , where we rescale the mass distribution from different times in the simulations onto a universal curve . the scaling function which we observe in figure 3 exhibits power law behaviour with @xmath17 for @xmath18 with @xmath19 .
this is not intrinsically surprising ( and indeed it matches theoretical expectations [ _ van dongen and ernst _ , 1985 ] ) but it has forced us to abandon the way experimentally observed distributions of cluster linear size have hitherto been plotted .
the problem is that given equation ( [ eq : dyn.scalg ] ) and its observed power law form , we must expect that the distribution of clusters by linear span @xmath20 should at small @xmath20 take the form @xmath21 which diverges as @xmath22 using our observed exponents
. for small enough crystal sizes this behaviour will be modified by the role of growth processes other than aggregation , but that lies outside the scaling regime . because of the divergence one has to take great care in constructing a characteristic linear size @xmath23 , where the natural choices are @xmath24 and the lowest whole number @xmath25 for which the denominator is not dominated by the smallest clusters is @xmath26 .
the simplest natural scaling ansatz for the cluster span distribution is then found [ _ westbrook et al .
_ , 2004 ] to be@xmath27 where @xmath28 .
figure 4 shows that this scaling ansatz works acceptably for our simulation data and well for the experimental observations .
the latter are rich data because cluster span is one of the simplest automated measurements to take .
the rescaled distributions from simulation and experiment agree fairly well but not perfectly , as shown in figure 4 .
one experimental reservation is the fall - off of experimental observation efficiency at small sizes , where clusters can be missed .
however our scaling procedure itself is in effect expressly designed to avoid sensitivity to this , and the superposition of the experimental data down to small reduced sizes looks good .
indeed it looks so good that the transient flattening around @xmath29 , which is absent from the simulations , appears to be significant .
one suggestion for the flattening around the middle of the rescaled distribution is that it might be associated with the peculiar feature of the collision rate being zero between clusters of equal sedimentation speed .
our simulations include this feature but for low relative approach speeds , where each cluster has more time to adjust its momentum in response to the other perturbing the local airflow , our approximation of ignoring hydrodynamic interactions ( and hence phenomena such as wake capture ) is less accurate . in summary , we have a fairly complete understanding of the geometry of the atmospheric ice crystal aggregates , dominated by sticking upon encounter .
further details of the sticking mechanism ( which we did not include ) appear not to be important for the cluster geometry , and the excellent scaling superposition of the experimental cluster size distributions suggests sticking efficiency does not favour aggregation at particular sizes .
the simplest interpretation of these observations is that although the sticking probability might be low for a single microscopic event , many such contacts will be attempted during a cluster - cluster encounter so that eventual escape is unlikely .
the actual sticking mechanism between ice crystals remains an intriguing open question , particularly for the low temperatures of figure 1 .
the fact that the same evolution is seen for differing initial monomer populations ( rods and rosettes ) suggests that a single set of geometric relationships for ice aggregates can successfully be applied in a wide range of cloud conditions .
this would lead to greater accuracy in retrieving cloud properties such as precipitation rate and predicting the radiative affect of ice crystal aggregates upon the climate system .
c to @xmath30c ( @xmath319 km altitude ) , using a cloud particle imager ( cpi , spec inc .
the pictures shown are aggregates of rosette ice crystal types .
( b ) aggregates as simulated by our computer model which assumed rigid joining when clusters collide under differential sedimentation .
, width=317 ] divided into the span perpendicular to the longest @xmath32 .
grey lines show cloud data of _ korolev and isaac _ [ 2003 ] plotted against longest span in microns for a range of temperatures between @xmath33c and @xmath34c .
black lines show simulation data plotted against longest span in arbitrarily scaled units , where the initial particles were three dimensional crosses ( solid line ) and simple rods ( dashed ) .
, width=317 ] c ) to 6.6 km ( @xmath37c ) in the cirrus cloud of _ field and heymsfield _ [ 2003 ] obtained during an arm ( atmospheric radiation measurement program ) flight ( 9th march 2000 ) .
each experimental size distribution represents an in - cloud average over 15 km .
black lines show simulation data .
, width=317 ] | aggregation of ice crystals is a key process governing precipitation .
individual ice crystals exhibit considerable diversity of shape , and a wide range of physical processes could influence their aggregation ; despite this we show that a simple computer model captures key features of aggregate shape and size distribution reported recently from cirrus clouds .
the results prompt a new way to plot the experimental size distributions leading to remarkably good dynamical scaling .
that scaling independently confirms that there is a single dominant aggregation mechanism at play , albeit our model ( based on undeflected trajectories to contact ) does not capture its form exactly . |
eigenfunction and spectral statistics in quantum systems with quenched disorder were a subject of intense study @xcite in the context of mesoscopic fluctuations of conductance and density of states ( dos ) , in particular in quantum dots @xcite . for this application
the most relevant is the regime of weak deviation @xcite from the wigner - dyson statistics given by the conventional random matrix theory ( rmt ) @xcite .
disordered multi - channel quantum wires is the most important example of systems where single - particle eigenstates are all localized . here
the statistics of eigenstates require a non - perturbative treatment using the formalism of nonlinear sigma - model @xcite or banded random matrices @xcite .
a special class are systems with the critical , multifractal ( mf ) eigenstate statistics @xcite .
two - dimensional disordered metals fall in this class @xcite provided that effects of localization are suppressed by magnetic field .
otherwise , one can speak only on weak multifractality which turns to localization before being fully developed .
the true physical realizations of the critical , mf eigenstate statistics are systems at the critical point of the anderson localization transition @xcite and the integer quantum hall systems at the center of landau band @xcite .
importantly , the class of systems with mf eigenstate statistics also allows for a random matrix representation @xcite , in particular using the power - law banded random matrices ( plbrm ) @xcite .
another field of intense research is the interplay between disorder and electron interaction with the seminal results on quantum correction to the tunnel dos and conductivity @xcite of disordered two - dimensional metals and the correction to superconducting transition temperature due to a simultaneous effect of disorder and the coulomb interaction @xcite . in all those works disorder and interaction
are taken into account essentially perturbatively along the lines given in @xcite .
recently there was an attempt @xcite to consider the problem of superconductivity near the anderson transition in which disorder has been treated non - perturbatively by _ postulating _ the mf statistics of one - particle states @xmath5 that enter the matrix element of a phenomenological electron attraction : @xmath6 in particular , the simplest quantity of interest is the disorder average matrix element @xmath7 at a given energy separation @xmath8 between one - particle energies @xmath9 and @xmath10 . for real eigenfunctions ( orthogonal symmetry class )
it is proportional to the correlation function @xmath11 , where @xmath12 and @xmath13 is the spectral correlation function which is close to 1 for @xmath8 much greater then the mean level spacing .
the correlation function @xmath14 is the main subject of the present paper .
the correlation function defined by eq.([c ] ) is a measure of overlap of _ two different _ eigenfunctions . for truly extended normalized states ( e.g. in a quantum dot ) @xmath15 and thus @xmath16 , where @xmath17 is a system volume .
remarkably , @xmath16 is also valid for classical examples of localized states , e.g. in a quantum disordered wire . in this case
two states are typically not overlapping but with a small probability of @xmath18 ( where @xmath19 is the localization radius ) they are localized in the same place and then the integral in eq.([c ] ) is of the order of the inverse localization volume @xmath20 .
there are cases , however , when an eigenfunction @xmath21 does not occupy all the available volume or all the localization volume and the typical amplitude @xmath22 is not just the inverse volume ( for extended states ) or the inverse localization volume ( for localized states ) . in this case a non - trivial behavior of the correlation function @xmath23 is expected .
such situation is realized near the critical point of the anderson localization transition . in the vicinity of this point in the region of extended states
( _ multifractal metal _ ) or in the region of localized states ( _ multifractal insulator _ )
the system retains the characteristic features of the critical multifractal statistics of eigenstates which makes it qualitatively different from both a normal metal or a normal anderson insulator . in this paper
we will identify and quantify such characteristic features in the correlation function @xmath23 and give their interpretation in terms of the typical behavior of single - particle states . to attain this goal
we will combine new analytical results for the plbrm with numerics on the plbrm and the anderson model .
we specially focus on the dependence of @xmath23 on the energy difference @xmath8 in the cross - over region in the vicinity but not exactly at the anderson transition point which has not been studied so far .
the paper is organized as follows . in section
ii we give a brief introduction into the subject of multifractality of critical eigenstates focusing on the main effect of multifractality which is the critical enhancement of eigenfunction correlations . in section
iii we give a cartoon of the off - critical states in a multifractal metal and a multifractal insulator and introduce the random matrix theories which may describe them .
in sec.iv we present the results of an analytical theory of eigenfunction correlations for a class of almost diagonal gaussian random matrices which all the rmt s suggested to describe strong multifractality fall into .
in sec.v we consider the two - eigenfunction correlation function exactly at the critical point of the localization transition in the 3d anderson model and for the critical random matrix ensemble in the limit of strong multifractality .
we show that the dynamical scaling relationship suggested by chalker is not violated even in the limit when the fractal dimensions are very small . in section
vi we describe the new phenomenon of eigenfunction mutual avoiding and present a qualitative picture that simultaneously explains the enhancement of eigenfunction correlations at small energy separations and the eigenfunction mutual avoiding at large energy separations . in section vii we consider the properties of eigenfunction correlations in a multifractal insulator . in particular , we describe the new phenomenon of logarithmic enhancement of eigenfunction correlations at small energy separations in the 2d and 3d anderson insulators and show the absence of such enhancement in the quasi-1d case .
we also suggest a truncated critical rm ensemble that describes all the principal features of eigenfunction correlations in the 3d multifractal insulator .
section viii is devoted to the random matrix description of the multifractal metal .
we show that the sub - critical plbrm suggested in ref.@xcite gives a reasonable agreement with the 3d anderson model . by analytical treatment of this rm model we found the region of parameters where the eigenfunction correlations become effectively short - range in the energy space which may poit out on the existence of a new metal phase above some critical dimensionlity in the multi - dimensional anderson model . in the conclusion
we list all the principal results of this paper .
the `` standard '' model ( the anderson model ) for the anderson localization transition in @xmath24 dimensions is the tight - binding model with the hopping constant @xmath25 and random on - site energies @xmath26 characterized by the distribution function @xmath27 which is frequently chosen constant @xmath28 in the interval @xmath29 $ ] and zero otherwise .
there is a vast literature ( see e.g. @xcite and references therein ) on numerical investigation of the anderson localization transition in this model on a 3d lattice .
recently also higher dimensions @xmath30 become accessible to modern computers @xcite .
while the earlier studies of this model were focused on the critical behavior of the localization / correlation length @xmath19 near the critical disorder @xmath31 , the recent works were mostly related with the statistics of critical eigenfunctions .
the multifractality of critical eigenfunctions predicted in @xcite almost immediately after emergence of scaling theory of localization has been confirmed and quantified in detail .
the results obtained for the anderson model exactly at the critical point seem to be very well described @xcite by the critical plbrm model @xcite .
this model is defined as an ensemble of random hermitean matrices which entries @xmath32 fluctuate independently around zero with the variance : @xmath33 } , & i\neq j\cr \end{matrix } \right.,\ ] ] where @xmath34 for the dyson orthogonal , unitary , and symlectic symmetry classes @xcite and @xmath35 is the parameter that controls the multifractality exponents .
this model has been studied and its comparison with the anderson model in @xmath3 dimensions has been done predominantly for the statistical moments @xmath36 of _ a single _ eigenstate at a given energy @xmath37 : @xmath38 the best known example is the inverse participation ratio ( ipr ) given by the second moment @xmath39 .
the multifractal statistics of a single eigenstate is characterized by the moment @xmath40 that scales with the system volume @xmath41 or the total number of sites @xmath42 as : @xmath43 where @xmath44 is the fractal dimension corresponding to @xmath45-th moment .
the existence of the scaling law eq.([mf ] ) and the dependence of the exponent @xmath46 on @xmath45 are the principle features of _
eigenfunction multifractality_. the fractal exponents @xmath46 depend also on the symmetry class @xmath47 and the space dimensionality @xmath3 . for the critical plbrm eq.([pl ] ) the dependence on @xmath3 is modeled by the dependence of @xmath46 on the parameter @xmath35 .
the critical scaling eq.([mf ] ) with respect to the system size @xmath42 has its _ dynamical _ counterpart when instead of one single eigenfunction one considers the correlation function eq.([c ] ) of _ two _ eigenfunctions at an energy separation @xmath2 between them .
this scaling has been suggested by chalker @xcite many years ago : @xmath48 where @xmath49 @xmath50 is the mean level spacing and @xmath51 is the upper cut - off of multifractality .
numerics on the integer quantum hall systems and in the critical point of the 3d anderson model was consistent @xcite with this scaling .
an important feature of eq.([chanz ] ) is that the exponent @xmath52 in the @xmath8-dependence is smaller than 1 . even in the limit of infinitely small correlation dimension
@xmath53 the correlation function decays slowly as @xmath54 .
this implies that the sparse critical states separated by large energy distance are still well overlaping @xcite , in contrast to strongly localized states which typically do not overlap even for nearest neighbors in the energy space .
the reason for such a behavior and the physical meaning of the energy scale @xmath51 will be discussed in section vi . as the correlation function is equal to @xmath55 both for
the truly extended and the ideal localized states , eq.([chanz ] ) implies the _ critical enhancement _ of eigenfunction correlations for @xmath56 .
this enhancement is crucially important for electron interaction near the anderson localization transition , in particular for the superconducting transition temperature @xcite . to illustrate this point we present in fig.1 the result of numerical diagonalization of the critical plbrm , the classical wigner - dyson rm @xmath57 with extended eigenstates , and the ensemble of conventional banded random matrices @xcite with exponentially decreasing entries @xmath58 which describes strongly localized eigenstates in quasi-1d disordered systems .
the critical enhancement of eigenfunction correlations is evident from this plot .
, the banded random matrices with b=5 , and wigner - dyson rm are shown in red , blue and green , respectively , and are represented by squares ( n=200 ) , circles ( n=1000 ) and stars ( n=2000 ) . ,
width=302,height=302 ] the physical origin of the enhancement is two - fold : ( i ) a critical eigenfunction `` occupies '' only part of the available space which by normalization @xmath59 enhances its amplitude , and ( ii ) the supports ( the manifold of @xmath60 where @xmath61 is essentially non - zero ) of different critical eigenfunctions are strongly overlapping .
it is important that both conditions are fulfilled simultaneously .
for instance the condition ( i ) is fulfilled for localized states even better than for the critical ones but the lack of the condition ( ii ) levels off the gain in the correlation function @xmath23 . on the contrary , in a metal the condition ( ii ) is trivially fulfilled , but the eigenfunction amplitude is small .
gaussian random matrix models proved to be an efficient and universal theoretical tool for describing complex systems .
the success was partially due to the available analytical solutions @xcite and partially due to efficient algorithms of numerical diagonalization of matrices .
therefore it is highly desirable to have random matrix models that describe not only the critical mf eigenstates but also localized and extended eigenstates in the _ vicinity _ of the anderson transition .
the criterion to select such models is a qualitative and ( when possible ) a quantitative agreement with the results on the 3d anderson model .
as will be demonstrated below , the correlation function @xmath23 in the 3d anderson model contains the critical power - law behavior eq.([chanz ] ) well beyond the anderson transition point . as a matter of fact
the correlation function @xmath23 is indistinguishable from the critical one until the dynamic length @xmath62 ( @xmath63 is the mean dos ) exceeds the localization / correlation length @xmath19 . for @xmath64 , or @xmath8 smaller than the level spacing in the localized volume @xmath65
, the correlation function loses its critical features and shows typical features of a metal or an isulator .
this allows us to suggest the following cartoon of typical eigenfunctions in the vicinity of the localization transition shown in fig.2 .
namely , a typical localized state in a _ `` multifractal insulator '' _ can be viewed as a `` piece of multifractal '' of the size of the localization radius @xmath19 ( fig.2b . ) , in contrast to a conventional localized state where all the localization volume is more of less homogeneously `` filled '' ( fig.2a . ) . in the same way , typical extended states on the metallic side of the localization transition ( _ `` multifractal metal '' _ ) should look like a mosaic made of such `` pieces of multifractal '' ( fig.2c . ) . based on the persistence of the critical behavior beyond the critical region it is natural to assume that the random matrix model for the extended states near the critical point and the localized states on the other side of the transition should bear features of the critical rmt .
is shown in each case.,width=18,height=37 ] let us start by constructing a random matrix model for the multifractal insulator . given that the quasi-1d insulator is well described by the banded random matrices @xcite with exponentially decaying variance @xmath66
we suggest the following hybrid rm model as a model for the multifractal insulator : @xmath67 as compared with the critical plbrm model eq.([pl ] ) , the model eq.([tran ] ) contains an additional parameter @xmath68 which sets in a finite localization radius @xmath69 .
it also contains an exponent @xmath70 which depends on the space dimensionality @xmath3 of the disordered lattice model we would like to model by the rmt . in sec.viii
we give both analytical and numerical arguments in favor of the choice @xmath71 another candidate has been suggested in ref.@xcite : @xmath72 } , & i\neq j\cr \end{matrix } \right.\ ] ] in this case the localization radius @xmath19 is controlled by the variable exponent @xmath73 of the power - law . for a multifractal insulator @xmath74 .
the possible rm models for multifractal metal are also constructed as deformations of the critical plbrm .
the model eq.([alpha - ins ] ) for @xmath75 is believed @xcite to describe the multifractal metal .
one can also think that the gaussian rmt @xmath76}+(b / b)^2 , & i\neq j\cr \end{matrix } \right.\ ] ] which is a hybrid of the critical plbrm and the wd rmt , is also suitable for this purpose .
below we will study all those rm models in detail and compare the corresponding results for the correlation function @xmath23 with the results obtained by numerical diagonalization of the @xmath3-dimensional anderson model .
the characteristic properties of multifractal statistics of critical and off - critical states are best seen when the multifractality is strong .
this is the case where the parameter @xmath77 in eqs.([pl ] , [ tran ] , [ alpha - ins ] , [ anti - trank ] ) is small . on the other hand
, this is exactly the limit where the typical off - diagonal elements of @xmath32 are small compared to diagonal ones
. such matrices ( referred to as _ almost diagonal _ random matrices ( adrm ) ) may possess a non - trivial statistics of eigenfunctions which justifies their special study @xcite .
the idea of analytical treatment of adrm , first suggested in ref.@xcite and used in @xcite to compute the correlation dimension @xmath78 for the critical plbrm model eq.([pl ] ) , is similar to the virial expansion in dilute gases .
however , instead of taking into account two- , three- and multiple - particle collisions , one considers progressively increasing number of interacting resonance sites coupled by a small off - diagonal matrix element @xmath32 .
recently the virial coefficients for the gaussian adrm with an arbitrary ( but small ) variance @xmath79 were expressed through the supersymmetric field theory @xcite and the correlation function @xmath23 has been explicitly calculated in the two - state approximation for the _ unitary _ symmetry class @xmath80 . the result is the following : @xmath81 where @xmath82,\ ] ] and @xmath83 in eqs.([k],[r ] ) we denote @xmath84 . the result given by eqs.([k],[r ] ) is valid in the limit when : @xmath85 ) with @xmath74 which are suggested to describe the multifractal insulator , the sum over @xmath86 in eq.([cond])converges .
then the validity of eqs.([k],[r ] ) is independent of the matrix size @xmath42 in the limit @xmath87 and is controlled only by a small parameter @xmath77 . on the contrary , for the models of the multifractal metal described by eq.([anti - trank ] ) and eq.([alpha - ins ] ) with @xmath75 the sum in eq.([cond ] ) diverges at large @xmath42 .
then eqs.([k],[r ] ) are only valid for @xmath88 where we define the _ correlation radius _ @xmath19 as follows : @xmath89 we will show below that a good qualitative description of the metal phase in the limit @xmath87 can still be obtained from the above theory if one substitutes @xmath19 for @xmath42 in eqs.([k],[r ] ) .
eqs.([k])-([xi ] ) will be used throughout the paper to analize different random matrix ensembles suggested as possible models for critical eigenstates and the off - critical states in a multifractal metal and insulator .
it is not a priori clear that the critical power - law behavior eq.([chanz ] ) and the dynamical scaling relationship eq.([d - mu ] ) hold true for all systems where eq.([mf ] ) is valid . in particular
it is interesting to study the correlation function eq.([c ] ) in the limit of strong multifractality when @xmath90 .
below we will derive an _
formula for the critical plbrm in the limit @xmath91 which corresponds @xcite to @xmath90 and confirm the scaling law eq.([d - mu ] ) by numerical diagonalization of plbrm with very small @xmath35 .
one can easily see from eqs.([k])-([baro ] ) in which we plug in eq.([pl ] ) that in the interval @xmath92 ) with @xmath93 : @xmath94 applying eq.([cond ] ) to the critical plbrm eq.([pl ] ) gives the criterion of validity @xmath95 in which case one can not distinguish between @xmath96 and @xmath42 , or , correspondingly , between @xmath97 and @xmath54 .
thus the analytical formulae eqs.([k],[pl ] ) is consistent with the scaling relationship eq.([d - mu ] ) , given that @xmath98 as @xmath91 .
a comparison of the analytical results and the results of numerical diagonalization of the critical plbrm with @xmath80 and @xmath99 is shown in fig.3 .
the coincidence is very good for large energy separations .
the deviation at small energy separations is due to the difference in the values of @xmath100 .
the exponent @xmath93 for the analytical curve and @xmath101 for the numerical curve which is very close to the prediction of eq.([d - mu ] ) @xmath102 , where @xmath78 is found from the numerical data for @xmath103 and eq.([mf ] ) .
the scaling relationship eq.([d - mu ] ) is further checked in fig.4 where the numerical data for @xmath100 and @xmath104 is plotted as a function of @xmath35 .
the fulfillment of this relationship down to @xmath35 as small as 0.005 and an agreement with the theoretical prediction of ref.@xcite is spectacular . for the critical plbrm with @xmath80 , @xmath99 and @xmath105(square),1000(circle ) and 2000 ( star ) . the analytical curve at @xmath106 given by eqs.([c - k]-[r ] ) is shown by a solid line.,width=340,height=340 ] and @xmath78 for the @xmath80 critical plbrm .
the solid line is the prediction based on ref.@xcite @xmath107.,width=302,height=302 ] ) with @xmath108 and @xmath109 ( open symbols ) .
the energy difference @xmath2 is measured in units of mean level spacing .
the insert shows the mean density of states ; the mobility edge corresponds to @xmath110 .
the energies @xmath111 were taken from the window @xmath112 for the 3d anderson model and @xmath113 for the critical plbrm .
the slope of the critical power - law eq.([chanz ] ) is 0.52 in both cases which corresponds to @xmath114.,width=340,height=340 ] thus from the combination of analytical and numerical results we conclude that the chalker s scaling eqs.([chanz]),([d - mu ] ) is valid for arbitrary small @xmath35 and thus for arbitrary strong multifractality .
finally we demonstrate how well the critical plbrm eq.([pl ] ) describes the two - eigenfunction correlations in the 3d anderson model at the mobility edge . to this end
we modify the distribution of the on - site energies in the anderson model from the standard rectangular box distribution to the triangular distribution where the mobility edge corresponds to @xmath115 . the correlation function @xmath23 with @xmath111 near the mobility edge
is shown in fig.5 .
it coincides almost exactly with the corresponding curve resulting from numerical diagonalization of the critical plbrm ensemble with only one fitting parameter @xmath109 .
results of both numerical and analytical calculation presented in fig.3 reveal another unexpected feature of eigenfunction correlation which appears to be common to all adrm .
surprisingly it is also present for the 3d anderson model both in the metal and in the insulator phase ( see fig.6 ) .
this is the _ negative _ eigenfuncton correlations for @xmath116 .
indeed , one can see from fig.3 and fig.6 that for large enough @xmath8 the correlation function @xmath23 goes below the uncorrelated limit @xmath117 which corresponds to @xmath118 .
we denote by @xmath51 the value of @xmath8 where this limit is reached . for @xmath119
the correlation function @xmath120 decreases down to zero . .
points below the dotted line correspond to eigenfunction avoiding .
, width=302,height=302 ] such a behavior implies that two eigenfunctions separated by an energy difference @xmath119 try to _ avoid each other_. that is , if a site @xmath121 is occupied in one of the states it should be predominantly empty in the other .
to explain such a behavior the following cartoon is useful .
let us define a _ support _ of an eigenfunction as the manifold of sites @xmath122 where @xmath22 is essentially non - zero .
to construct such a support starting from a given site @xmath121 with the on - site energy @xmath123 we find all the sites in resonance with the site @xmath121 , i.e. such sites @xmath124 which on - site energies @xmath125 obey the relationship @xmath126 .
then the procedure should be repeated for all sites @xmath127 and so on .
it is important that so obtained manifold @xmath60 does not always include all the sites of the system .
if this is the case , the whole coordinate space is _ stratified _ into a set of mutually non - intersecting supports ( fig.7 ) .
smaller that the single - shell bandwidth @xmath51 and mutual avoiding of eigenstates for @xmath119.,width=151,height=151 ] once the support is defined , one can build a _
shell _ of states on this support by making a linear combination of on - site states , pretty much in the same way as in building the conduction band states out of the on - site states in the tight - binding model .
then by construction the eigenfunctions belonging to the same shell are well overlapping but those belonging to different shells do not overlap . from this cartoon
it is clear that the physical meaning of the scale @xmath51 is the width of the energy band corresponding to a single shell . indeed ,
if the energy separation @xmath8 greatly exceeds the typical single shell bandwidth , the two eigenfunctions must belong to different shells and thus do not significantly overlap in space . on the contrary ,
if @xmath8 is smaller than the single shell bandwidth , the two states typically belong to the same shell and thus overlap strongly no matter how sparse the shell support .
the new energy scale @xmath51 , which is the upper energy cut - off of the multifractal correlations , corresponds to a new length scale @xmath128 which has a meaning of the minimum length scale of the fractal texture . in the @xmath3-dimensional anderson model
the energy scale @xmath51 can be estimated as @xmath129 where @xmath130 is the total bandwidth . estimating the dos as @xmath131 we find @xmath132 , where @xmath133 is the lattice constant .
clearly the picture with a stratified coordinate space is possible for plbrm eq.([pl ] ) with small enough @xmath134 when the single shell bandwidth @xmath135 is small compared to the total bandwidth @xmath136 .
amazingly , the 3d anderson model which low - frequency critical features are well described by the critical plbrm with @xmath137 , also follows the predictions of the critical plbrm for high frequencies @xmath119 .
this is a consequence of a relatively large value @xmath138 of the critical disorder which results in @xmath51 considerably smaller than the conduction bandwidth @xmath130 . in particular its coordinate
space must be stratified to explain the observed ( see fig.6 ) mutual avoiding of eigenstates .
( circles ) , @xmath139 ( stars ) .
the inverse participation ratio is equal to 0.23 and 0.46 , respectively.,width=302,height=302 ] as has been demonstrated in sec.v critical eigenfunctions with @xmath140 are strongly correlated in space .
here we consider the case of multifractal insulator where the localization radius @xmath19 is large compared to relevant microscopic lengths ( the lattice constant or elastic scattering length ) but is much smaller than the system size @xmath141 .
we will identify a suitable random matrix model to describe this case and compare the properties of eigenfunction correlation in this model with those of the 3d anderson model .
we start by considering a limit of strong disorder when the localization length @xmath142 and the multifractal nature of eigenstates does not show up .
a common wisdom is that in the strongly localized regime the positions of the localization centers are completely uncorrelated .
as it is shown in the introduction , this leads to @xmath143 which we will refer to as the _ ideal insulator limit_. fig.1 shows how this limit is reached in the ensemble of banded random matrices .
( green ) , @xmath144 ( blue ) , @xmath145 ( red ) , @xmath146 ( purple ) .
the system size is @xmath147 for filled symbols and @xmath148 for open symbols .
the inverse participation ratio for the four insulating systems is @xmath149 , @xmath150 , @xmath151 , @xmath152 which corresponds to @xmath153 , @xmath154 , @xmath155 , @xmath156 according to @xmath157 .
the change of the slope occurs at @xmath158 .
the slope for larger energy separations @xmath159 progressively increases with increasing @xmath160 remaining smaller than 1 .
the insert shows the result for @xmath144 , @xmath147 for the periodic ( upper blue curve ) and the hard wall ( lower red curve ) boundary conditions.,width=302,height=302 ] note that @xmath23 in this limit is much smaller than the self - overlap of @xmath61 given by the inverse participation ratio eq.([ipr ] ) . only for very small energy separations
( typically @xmath161 ) which we will not be considering here , the ipr limit can be approached .
now let us see how does the correlation function @xmath23 look like for the strong anderson insulator .
the corresponding plot for 1d anderson model is shown in fig.8 .
it coincides almost exactly with the ideal insulator limit eq.([st - loc - l ] ) .
the plot for the 3d anderson model is shown in fig.9 .
on can see that @xmath162 is significantly enhanced at small energy separations and does not resemble at all the correlation function in 1d anderson insulator . in order to understand why the ideal insulator limit is not reached in the 3d case despite the ratio @xmath164 we compute numerically the probability distribution ( pdf ) @xmath165 of the distance @xmath166 between the points @xmath167 and @xmath168 in _ real space _ ( _ centers of localization _ ) where @xmath169 has an absolute maximum , provided that the energy separation between the states @xmath170 is @xmath8 .
the results are shown in fig.10 .
it is seen that the function @xmath171 is far from being independent of @xmath172 ( which would imply the lack of correlations between centers of localization ) .
in fact , there is a _ repulsion of centers of localization _ at distances @xmath173 which shows up in the decreasing probability density to find two centers of localization close to each other .
note that @xmath174 is almost 10 times larger than the localization radius @xmath19 estimated from the inverse participation ratio @xmath39 . ) of having two centers of localization at a distance @xmath172 in real space and at a distance @xmath8 in the energy space computed for the 3d anderson model in the strong localization regime ( @xmath144 , @xmath175 , @xmath147 ) .
the repulsive core @xmath176 exceeds the `` hard ball '' limit @xmath177 by a factor of 5 - 6.,width=226,height=226 ] an explanation to this fact of repulsion between centers of localization @xmath178 and @xmath179 is based on the resonance interaction between states @xmath180 and @xmath181if the energy distance between them is smaller than the typical overlap integral @xmath182 .
the size of the repulsion core @xmath174 can be estimated from the equation : @xmath183 the characteristic energy scale @xmath184 is the mean level separation for states localized in the same volume @xmath185 .
thus the repulsion of centers of localization is a direct consequence of repulsion of energy levels for states confined in the same volume @xmath185 .
the energy scale @xmath184 depends on the strength of disorder and is of the order of the fermi energy for strongly localized states . at @xmath186
the size of the repulsion core @xmath174 may considerably exceed the localization radius .
the qualitative picture of repulsion of centers of localization can be quantitatively confirmed using the analytical theory eqs.([c - k]-[baro ] ) for the almost diagonal gaussian rmt . to this end
we look at the contribution of @xmath187 to the sum in eq.([k ] ) . replacing the summation over @xmath86 by integration
we obtain the contribution to @xmath162 : @xmath188 this equation can be easily interpreted using an elementary perturbation theory .
indeed , for strongly localized states @xmath189 the eigenfunction correlation function can be represented as follows : @xmath190 the amplitude at the tail of the wavefunction @xmath191 with the maximum at a point @xmath167 can be computed from the elementary perturbation theory in which the wavefunction of the zero - th approximation corresponding to the energy @xmath192 is @xmath193 : @xmath194 the fluctuating _ on - site _
energy @xmath192 is the main part of the eigenvalue @xmath195 for a sufficiently strongly localized state .
thus we come to a conclusion that the amplitude of the wavefunction @xmath196 at a center of localization of the wavefunction @xmath197 is inversely proportional to @xmath198 and thus is strongly _ enhanced _
when @xmath186 .
at the first glance this is in a contradiction with the common wisdom that @xmath199 which is apparently @xmath8-independent .
the point is that the quantity @xmath200 has many accidental spikes due to resonances between on - site energies .
the measure of such resonance points is small and for some ( but not all ) purposes one can neglect them and to approximate @xmath201 .
the best known example when the two - spike eigenfunction makes the main contribution is the low - frequency conductivity in the localized phase @xcite . as we will see below , here we deal with a very similar phenomenon . now
the correlation function @xmath23 can be computed just by averaging over disorder and the distance @xmath202 : latexmath:[\[\begin{aligned } \label{c - finger }
c(\omega)&=&\int f_{\omega}(r)\,\langle c_{nm}(r)\rangle\,d^{d}r \\ \nonumber & \approx & \int f_{\omega}(r)\,\frac{2\langle @xmath171 is the pdf defined by eq.([jpdf ] ) .
comparing eq.([c - finger ] ) for @xmath204 with eq.([hvost ] ) we see that : @xmath205 where @xmath174 is found from the condition @xmath206 similar to eq.([r ] ) . in the opposite limit @xmath207 , or @xmath208
, we have a resonance enhancement @xmath209 and @xmath210 .
then the comparison of eq.([c - finger ] ) with eq.([k ] ) yields : @xmath211 now let us consider eq.([c - finger ] ) for @xmath212 assuming that all states are exponentially localized and thus @xmath213 , where @xmath214 .
then one immediately obtains from eq.([c - finger ] ) that due to the phase volume factor @xmath215 the correlation of exponentially localized eigenfunctions depends crucially on the dimensionality of space .
namely , for @xmath204 the ideal insulator limit eq.([st - loc - l ] ) is reached for sufficiently small @xmath19 ( see fig.1 ) , while for @xmath4 and @xmath186 the correlation function acquires a logarithmic in @xmath8 enhancement factor : @xmath216 the physics behind this result is similar to the one which leads to the selebrated mott s law @xcite @xmath217 for the ac conductivity @xmath218 at @xmath186 . the difference is that the contribution to conductivity from the resonance states with the distance @xmath172 between the points of maximal amplitude is proportional to the square of the dipole moment @xmath219 , so that the phase volume factor @xmath215 gets multiplied by @xmath220 resulting in emergence of the logarithmic factor @xmath221 instead of @xmath222 in our case .
below we obtain this result for the _ truncated critical _ rmt defined by eq.([tran ] ) .
the phase volume factor @xmath215 can be formally taken into account in the random matrix formalism eq.([k ] ) if one assumes the following relationship between the @xmath3-dimensional vector @xmath223 and the difference of matrix indices @xmath224 : @xmath225 is the total solid angle in the @xmath3-dimensional space . in particular eq.([repl ] ) suggests that for exponential localization the correct truncating factor in eq.([tran ] ) has the form : @xmath226.\ ] ] this sets the exponent @xmath70 in eq.([tran ] ) equal to : @xmath227 then eq.([k ] ) can be used which is convenient to rewrite in the following form : @xmath228 where @xmath229 is found from the equation @xmath230 and @xmath231 for the truncated critical rmt eq.([tran ] ) with @xmath232 one finds : @xmath233 where @xmath234 the integral in eq.([ka ] ) is well convergent and thus mainly contributed by @xmath235 .
this makes it possible to obtain a simple analytical expression for @xmath162 : @xmath236 where @xmath237 @xmath238 the first line of eq.([c - ffin])is consistent with eq.([ench ] ) in which @xmath239 and @xmath240 . the power - law behavior in the second line of eq.([c - ffin ] ) is a remnant of the critical behavior eq.([chanz ] ) .
it exists only for considerably large @xmath241 where @xmath242 , i.e. only in the _ multifractal _ insulator . for a very strong insulator with @xmath243 ,
the localization radius is smaller than the minimal length scale @xmath244 of the fractal texture .
this is the region of an ordinary insulator where the entire localization volume is more or less homogeneously filled .
thus the eigenfunction correlation function for the truncated critical rmt eq.([tran])interpolates between the behavior given by eq.([ench ] ) at @xmath245 ( or @xmath246 ) and the critical behavior eq.([chanz ] ) which is valid for @xmath247 . at @xmath248
both asymptotic forms are apparently matching with each other . the physical picture that leads to such a behavior is the following .
there are two distinct regions @xmath249 and @xmath64 where physics of eigenfunction correlations is entirely different . in the case
@xmath249 the characteristic length @xmath250 has a meaning of the period of beating in the overlap of two fractal eigenfunctions inside the localization volume .
the regions where two fractal supports match well with each other alternate with the regions with a strong mismatch between them , very much like in the case of two grids with slightly different periods .
the regions of strong overlap make the main contribution @xmath251 to the eigenfunction correlation function @xmath23 which is of the order of the ipr of a multifractal metal ( see next section for more details ) with the system size equal to the localization radius @xmath19 and the correlation length equal to the size @xmath250 of the well overlapping regions . to obtain the correlation function
@xmath23 one has to multiply @xmath252 by the probability for the entire localization volumes to overlap .
this probability is @xmath253 , as for @xmath254 there is no correlations in the positions of the localization volumes .
thus we obtain the critical power - law eq.([chanz ] ) : @xmath255
. for @xmath64 physics of eigenfunction correlations changes drastically .
now localization volumes are statistically repelling each other and the overlap is only due to the tails . in this region
the length scale @xmath250 loses its physical meaning which is taken over by the length scale @xmath174 given by eq.([r ] ) . the overall shape of @xmath23 with the logarithmic enhancement factor @xmath222 obtained within the truncated critical rmt describes the numerical results on the 3d anderson model very well ( see fig.11 ) .
the absence of this factor at @xmath204 explains the qualitative difference between the case @xmath204 ( see fig.8 ) and @xmath256 ( see fig.9 ) .
this difference is essentially due to a competition between two effects ( i ) repulsion of centers of localization and ( ii ) resonance enhancement of overlap by tails .
the first effect tends to decrease the probability of the overlap of localization volumes .
the second effect increases the eigenfunction overlap by means of tails . in the 1d case
these two effects compensate each other and the result is the same as one would obtain for completely uncorrelated positions of localization volumes and the typical exponentially decreasing tails . in higher dimensions the enhancement of overlap in the tail region prevails because of the increased volume of those regions .
concluding this subsection we claim that the truncated critical rmt provides an excellent description of the 3d anderson insulator both in the strong localization region ( see fig.9 ) and in the region of _ multifractal insulator _ where the localization radius is large and the corresponding scale @xmath184 is small compared with the upper cutoff @xmath51 of multifractal correlations . because of the limited size of the 3d lattice this latter region is out of reach for numerical simulations on the 3d anderson model , and the random matrix theory is the only mathematical model which properly describes physics of the multifractal insulator .
note that there is another rmt eq.([alpha - ins ] ) suggested in ref .
@xcite as a candidate to describe eigenfunction correlations in the multifractal insulator .
below we show that this _ super - critical _ plbrm is principally flawed , as it corresponds to a power - law localization which is not the case in the 3d anderson model
. this can be best demonstrated by eq.([pt ] ) in which @xmath257 with @xmath74 .
accordingly , the typical scale for the repulsion of centers of localization is : @xmath258 in fig.12 we plot the results of numerical calculation of the pdf @xmath171 for the super - critical plbrm eq.([alpha - ins ] ) .
the characteristic scale @xmath259 where @xmath171 reaches its maximum , is well seen in this plot . for the super - critical plbrm eq.([alpha - ins ] ) with @xmath260 , @xmath261 , @xmath262 for @xmath263 ( black ) , @xmath264 ( blue ) , @xmath265 ( green ) .
the insert shows the @xmath8-dependence of @xmath259 where @xmath171 reaches its maximum . the finite slope of @xmath171 for @xmath266 is a finite size effect which was neglected in eq.([f - w]).,width=264,height=264 ] the analytical treatment based on eq.([ka ] ) yields for this model : @xmath267 where @xmath268 , and @xmath269 as well as the entire approach based on eq.([k ] ) , the above results are valid when @xmath35 is the smallest relevant parameter . in the problem of plbrm with @xmath73 close to 1
, there is a competition between the small parameters @xmath270 and @xmath35 , so that the validity of eqs.([r - w]),([c - mir ] ) requires also @xmath271 .
we see that in the infinite system @xmath87 the power - law eq.([c - mir ] ) in @xmath23 is not restricted at small @xmath8 , and no energy scale similar to @xmath184 emerges .
this can be explained only if we assume that the localization length for @xmath272 is of order one .
then for all energy separations @xmath273 the repulsion core @xmath274 , and no qualitative change in the correlation function occurs until @xmath259 hits the system size @xmath42 . for smaller @xmath8
the correlation function is almost a constant .
this quantitative analysis is illustrated by fig.13 . ) with @xmath272 calculated analytically using eqs.([c - k]-[baro ] ) .
the power law @xmath275 is valid for all energy separations corresponding to @xmath276 .
the onset of the plateau moves to @xmath277 in the limit @xmath87 .
the ideal insulator limit is reached by decreasing the slope with increasing @xmath73.,width=226,height=188 ] the region of @xmath73 that could describe the multifractal insulator with large @xmath278 corresponds to @xmath279 . in this case
an energy scale similar to @xmath184 should appear .
it can be found from the condition @xmath280 at @xmath158 the slope on the log - log plot of @xmath23 should change from the critical value at @xmath281 to a different ( but constant ) @xmath73-dependent value at @xmath282 .
this change of the slope is clearly seen in the numerical simulations on the super - critical plbrm presented in fig.14 .
it appears that in all cases studied the slope @xmath100 at @xmath282 is _ larger _ than that at @xmath159 .
this is in a clear contradiction with the results ( see fig.9 ) obtained in the 3d anderson insulator .
an important conclusion we can draw from the above analysis of the super - critical plbrm is that the correlation function @xmath23 for @xmath283 is the power - law in this model .
this can be traced back to the power - law character of localization in the super - critical plbrm which is not the case in the disordered lattice models ( such as the 3d anderson model ) with short - range hopping integrals .
this is the reason why the super - critical plbrm is not suitable to describe the insulating phase of the 3d anderson model .
surprisingly , the natural counterpart to the truncated critical rmt eq.([tran ] ) which is defined by eq.([anti - trank ] ) ( `` anti - truncated '' critical rmt ) does not describe extended states in the multifractal metal .
the reason is that this model possesses _ two _ low - frequency system - size independent energy scales instead of the single scale @xmath184 which is associated with the size @xmath19 of a multifractal cell in fig.2c . in order to see this we analyze the analytical formulae eqs.([c - k]-[baro ] ) with the variance defined by eq.([anti - trank ] ) . to this end
we expand the summand of eq.([k ] ) in @xmath284 to arrive at the formula similar to eq.([hvost ] ) but with the upper limit of integration equal to the correlation radius @xmath19 : @xmath285 where @xmath286 according to eq.([xi ] ) . substituting eq.([anti - trank ] ) for @xmath287 we arrive at : @xmath288\approx \frac{b}{\omega}+ 2\xi \frac{b^{2}}{\omega^{2}b^{2}}.\ ] ] eq.([kkk ] )
is valid at @xmath289 when the upper limit of integration is larger than the lower limit .
this sets the energy scale @xmath290 .
another scale @xmath291 gives the cross - over scale that separates the critical @xmath54 behavior and the @xmath292 behavior that takes place for indermediate frequencies @xmath293 .
while the scale @xmath294 ( similar to the scale @xmath295 in a 1d insulator ) determines the onset of the low - frequency plateau , the second relevant scale @xmath296 that appears in the model eq.([anti - trank ] ) seems to have no physical meaning .
indeed , the existence of this scale leads to a characteristic form of the correlator @xmath23 which log - log plot has a significant slope increase just before it drops to zero at the plateau ( see figs.15 , 16 ) .
is well seen both for small and large system sizes .
the solid line is a fit according to eq.([kkk]).,width=264,height=264 ] we did not find the behavior of such type in the 3d anderson metal ( see fig.17 ) .
the plot in fig.17 clearly shows a saturation @xcite at @xmath297 : @xmath298 however , there is no evidence of a maximum in the slope just above the onset of the plateau . ) ( red curve ) and for the sub - critical plbrm eq.([alpha - ins ] ) ( blue curve ) of unitary symmetry class computed analytically from eqs.([c - k]-[baro ] ) . for the anti - truncated critical rmt
the non - monotonous behavior of log - log slope is similar to the one in fig.15.,width=188,height=188 ] ( green ) , @xmath299 ( red ) , @xmath139 ( blue ) .
the system size is @xmath148 ( diamonds ) @xmath300 ( circles ) and @xmath147 ( squares ) .
the correlation length is estimated from the inverse participation ratio as follows @xmath301 ( @xmath302 , @xmath303 ) and is equal to : @xmath304 , @xmath305 , @xmath306 at @xmath139 , @xmath307 , @xmath308 , respectively . in the insert
we compare results for @xmath299 , @xmath147 for the periodic ( red filled squares ) and the hard - wall ( red open squares ) boundary conditions . for hard - wall boundary conditions the correlation function looks `` more critical '' , as the critical point @xmath309 in this case is closer to @xmath299.,width=302,height=302 ] now we consider the sub - critical plbrm ensemble defined by eq.([alpha - ins ] ) with @xmath310 . in this
case analytical arguments similar to eqs.([kk],[kkk ] ) predict only one relevant energy scale @xmath311 such that for @xmath312 the correlation function @xmath23 is constant and for @xmath313 ( but @xmath314 ) it is a pure power law @xmath315 .
thus the sub - critical plbrm is free from the drawback related with the unphysical second energy scale . for comparison we plotted the analytical results for the anti - truncated critical rmt and for the sub - critical plbrm in fig.16 .
it is seen that the overall shape of the blue curve for sub - critical plbrm is much closer to the results of 3d anderson model of fig.17 .
note however , that the power - law emerging in the analytical results for the sub - critical plbrm has an exponent @xmath316 which is _ larger _ than the critical exponent .
computer simulations ( see fig.18a ) on the sub - critical plbrm with very small @xmath35 confirm this analytical result as @xmath87 extrapolation and show that the slope increases with increasing the system size n. however , the slope of the corresponding curves for the 3d anderson model of fig.17 is almost independent of the system size and is equal or _
smaller _ than the critical slope .
limit @xmath317 , @xmath318 at @xmath262 ( squares ) and @xmath105 ( diamonds ) .
the critical correlations for @xmath317 , @xmath319 and @xmath262 are shown by stars .
the solid line corresponds to the power - law @xmath320 .
the slope of the sub - critical curve is larger than 1 ( correlations are short - range in the energy space ) .
this case is relevant for the metallic phase of the anderson model in very high dimensions .
right panel ( b ) : the case @xmath321 , @xmath322 at @xmath262 ( squares ) and @xmath105 ( diamonds ) is relevant for the anderson model in @xmath256 ( `` multifractal '' metal ) . the slope is less than the critical ( which in turn is less than 1 ) and is almost size - independent .
correlations are long - range in the energy space.,width=359,height=264 ] the reason for the discrepancy is that the analytical result for the slope @xmath323 corresponds to the limit @xmath324 . at a finite @xmath35
the slope decreases with increasing @xmath35 and at a sufficiently large @xmath35 may become smaller than the critical one .
it is reasonable to assume that at small @xmath325 and @xmath35 this happens at @xmath326 .
the relevance of the parameter @xmath327 is also seen from the expression for the correlation length @xmath19 @xmath328,\;\;\;\;a_{\alpha}=\ln\left(\frac{1-\alpha}{b^{\alpha}}\right)+{\rm const } .
% \frac{1-\alpha}{b^{\alpha}}\right)^{\frac{1}{1-\alpha}},\ ] ] which was found ( up to a constant of order one ) analytically from eq.([xi ] ) .
this expression is apparently meaningless for @xmath329 where @xmath330 may become negative .
numerical simulations on the sub - critical plbrm with @xmath331 ( e.g. for @xmath322 and @xmath321 relevant for the 3d anderson model ) show ( see fig.18b ) that the log - log slope of @xmath23 is somewhat smaller than the critical one and is almost independent of the matrix size @xmath42 .
thus the sub - critical plbrm shows exactly the same character of eigenfunction correlations as in the 3d anderson metal ( see fig.17 ) .
two parameters of the sub - critical plbrm allow to simulate the effect of the finite correlation length ( choice of @xmath75 ) and the dimensionality of space ( choice of @xmath35 ) .
note in this connection that for the disorder strength @xmath160 significantly smaller than the critical value @xmath31 , not only @xmath73 but also @xmath35 is @xmath160-dependent .
the point is that in the 3d anderson model the variance of the on - site energies fluctuations is proportional to @xmath332 , while the off - diagonal hopping integral is equal to 1 .
this implies that the ratio of a typical off - diagonal to a typical diagonal elements controlled in eq.([alpha - ins ] ) by the parameter @xmath35 should scale like @xmath333 . as the log - log slope of @xmath23 decreases with increasing @xmath35 , moving away from the anderson transition into the _
phase @xmath334 has an effect of _ decreasing _ the slope . on the insulator side of the transition
the situation is opposite and one should expect an _ increase _ of the slope ( for @xmath335 ) with increasing @xmath160 .
fig.9 shows that it is apparently the case .
another relevant note is that for the anderson model in _ higher dimensions _ the correlation dimension @xmath78 decreases .
this can be modeled by a decreasing parameter @xmath35 .
then the analogy with the sub - critical plbrm suggests that for sufficiently high dimensions @xmath336 the behavior in the @xmath3-dimensional anderson model should become similar to the one in fig.18a .
namely , the exponent @xmath100 in eq.([d - mu ] ) may become _ larger than 1_.
this changes qualitatively the eigenfunction correlations , as they become effectively short - range in the energy space .
in particular , the return probability @xcite which is proportional to the fourier transform of @xmath23 behaves in the time interval @xmath337 as @xmath338 for @xmath339 and is a constant for @xmath340 .
we believe that this qualitative change in the eigenfunction statistics ( if confirmed for a @xmath3-dimensional anderson model with @xmath336 ) should lead to dramatic physical consequences marking a transition to a new metallic state .
in conclusion we list the main results obtained above .
the most important of them is the persistence beyond the point of localization transition of the critical power - law in the dependence of the eigenfunction correlation function @xmath23 on the energy separation @xmath8 and the related enhancement of @xmath23 at @xmath341 , where @xmath184 is the mean level spacing in the localization / correlation volume and @xmath51 is the upper energy cut - off of multifractality .
this enhancement leads to an enhancement of matrix elements of local electron interaction which may result in , e.g. an enhancement of the superconducting transition temperature in the vicinity of the anderson localization transition @xcite .
another important observation is that the enhancement of correlations at @xmath342 is always accompanied by the depression at @xmath343 , both phenomena being the consequences of the stratification of the coordinate space into mutually avoiding supports of the fractal structure with well overlapping eigenfunctions living on each of them .
an independent but also important phenomenon is the logarithmic enhancement of @xmath23 in the 2d and 3d anderson insulator at @xmath283 ( and the absence of such enhancement in the quasi-1d disordered wire ) . it is a result of a competition of two simultaneous phenomena : the repulsion of centers of localization and the resonance enhancement of the eigenfunction overlap by tails .
both phenomena are studied quantitatively within the truncated critical random matrix model which is suggested as a universal tool to describe the localized eigenfunctions with a multifractal texture .
we also show that the sub - critical power - law banded random matrix ensemble suggested in ref.@xcite describes the multifractal metal reasonably well . from the analytical solution for this rmt
we conclude that a critical dimensionality @xmath344 may exist above which the @xmath3-dimensional anderson model has an unusual metal phase characterized by an effectively short - range correlation function @xmath345 with @xmath346 .
_ acknowledgement._ the authors are grateful to b.l.altshuler , m.v.feigelman , a.silva and v.i.yudson for stimulating discussions and especially to o.yevtushenko for a collaboration at an earlier stage of this work and for a help in preparing figures .
thanks the feder and the spanish dgi for financial support through project no .
fis2004 - 03117 . 99 a.d.mirlin , phys.rep .
* 326 * , 259 ( 2000 ) .
i.l.aleiner , p.w.brouwer , l.i.glazman , phys.rep . *
358 * , 309 ( 2002 ) .
v.e.kravtsov and a.d.mirlin , jetp lett .
* 60 * , 656 ( 1994 ) .
a.d.mirlin and y.v.fyodorov , jetp lett . *
60 * , 790 ( 1994 ) .
m.l.mehta , _ random matrices _ , academic press , san diego ( 1991 ) .
k.b.efetov , _ supersymmetry in disorder and chaos _
( cambridge , 1997 ) .
y.v.fyodorov , a.d.mirlin , phys.rev.lett . * 67 * , 2405 ( 1991 ) .
v.e.kravtsov , k.a.muttalib , phys.rev.lett . * 79 * , 1913 ( 1997 ) .
j.t.chalker , v.e.kravtsov , i.v.lerner , jetp lett . * 64 * , 386 ( 1996 ) .
v.i.falko , k.b.efetov , europhys.lett . * 32 * , 627 ( 1995 ) .
f.wegner , z.phys.b * 36 * , 209 ( 1980 ) .
j.t.chalker , physica a * 167 * , 253 ( 1990 ) .
j.t.chalker , g.j.daniell , phys.rev.lett .
* 61 * , 593 ( 1988 ) .
a.d.mirlin , y.v.fyodorov , f.m.dittes , j.quezada , t.h.seligman , phys.rev.e * 54 * , 3221 ( 1996 ) .
b.l.altshuler , a.g.aronov , p.a.lee , phys.rev.lett . * 44 * , 1288 ( 1980 ) . s.maekawa and h.fukuyama , j.phys.soc.jpn . *
51 * , 1380 ( 1982 ) ; a.m.finkelshtein , jetp lett . * 45 * , 46 ( 1987 ) .
l.p.gorkov , a.i.larkin , d.e.khmelnitskii , jetp lett . * 30 * , 228 ( 1979 ) .
m.v.feigelman , l.b.ioffe , v.e.kravtsov , e.yuzbashyan , phys.rev.lett .
* 98 * , 027001 ( 2007 ) .
t.ohtsuki , k.slevin , t.kavarabayashi , annalen der physik , * 8 * , 655 ( 1999 ) .
a.mildenberger , f.evers , a.d.mirlin , phys.rev.b * 66 * , 033109 ( 2002 ) ; a.m. garca - garca and e. cuevas , phys .
b * 75 * , 174203 ( 2007 ) .
a.d.mirlin and f.evers , prb , * 62 * , 7920 ( 2000 ) .
t.brandes , b.huckestein and l.schweitzer , ann.phys . , * 5 * , 633 ( 1996 ) .
l.schweitzer and h.potempa , physica a , * 266 * , 486 ( 1999 ) . y.v.fyodorov and a.d.mirlin , phys.rev.b * 55 * , 16001 ( 1997 ) .
o.yevtushenko and v.e.kravtsov , j.phys.a-math.gen . *
36 * , 8265 ( 2003 ) ; o.yevtushenko and v.e.kravtsov , phys.rev.e . , * 69 * , 026104 ( 2004 ) ; v.e.kravtsov , o.yevtushenko and e.cuevas , j.phys.a-math.gen . *
39 * , 2021 ( 2006 ) .
l.s.levitov , phys.rev.lett . ,
* 64 * , 547 ( 1990 ) .
o.yevtushenko and a.ossipov , j.phys.a -math.theor . * 40 * , 4691 ( 2007 ) .
n.f.mott , _ metal - insulator transitions _ , 2d edition , london taylor and francis , 1990 .
v.l.berezinsky , sov.phys.jetp * 38 * , 620 ( 1974 ) . the constant @xmath347 ( and the corresponding constant @xmath348 ) for insulator
were found from an independent numerical study of an exponential tail @xmath349 of the wave function , with @xmath350 counted from the maximum of the wave function amplitude . | we consider the correlation of two single - particle probability densities @xmath0 at coinciding points @xmath1 as a function of the energy separation @xmath2 for disordered tight - binding lattice models ( the anderson models ) and certain random matrix ensembles .
we focus on the parameter range close but not exactly at the anderson localization transition .
we show that even away from the critical point the eigenfunction statistics exhibit the remnant of multifractality characteristic of the critical states .
this leads to an enhancement of eigenfunction correlations and a corresponding enhancement of matrix elements of the local electron interaction at small energy separations .
this enhancement is accompanied by a depression of correlations at large energy separations , both phenomena being a consequence of the stratification of space into densely packed but mutually avoiding resonance clusters .
we also demonstrate that the correlation function of localized states in a @xmath3-dimensional insulator is logarithmically enhanced at small energy separations provided that @xmath4 .
a simple and general physical picture of all these phenomena is presented . finally by a combination of numerical results on the anderson model and analytical and numerical results for the relevant random matrix theories we identified the gaussian random matrix ensembles that describe the multifractal features both in the metal and in the insulator phases . |
the ferromagnetic kondo lattice model has been widely used as a minimal model to describe some features of the manganites @xcite la@xmath4sr@xmath5mno@xmath6 , la@xmath4sr@xmath7mno@xmath8 , and la@xmath9sr@xmath10mn@xmath11o@xmath12 .
the model contains one itinerant @xmath13 orbital and the @xmath2 core spin at each site .
the @xmath13 electrons are ferromagnetically coupled to the @xmath0 core spins generated by the fully occupied @xmath2 orbitals .
the large ferromagnetic hund s rule coupling leads to the formation of two bands , the lower and the upper kondo band , with the @xmath13 electrons predominantly parallel to the core spins in the lower band .
the core spins strongly influence the mobility of the @xmath13 electrons via double exchange ( de ) . at high hole density
, this leads to a ferromagnetic arrangement of the core spins , while antiferromagnetism is preferred for the completely filled lower kondo band . in the opposite case of an empty kondo band , the @xmath2 core spins are antiferromagnetically oriented due to superexchange . since the core spins have a fairly large spin , @xmath0 , they are frequently approximated by classical spins , greatly simplifying calculations .
furthermore , the on - site coulomb repulsion @xmath14 between the @xmath13 electrons is often neglected because double occupancy is already suppressed by the hund s rule coupling and because its treatment considerably increases the numerical effort.@xcite a review of these semi - classical simulations can be found in refs . and references therein .
several of these studies @xcite have found that phase separation into regions with either ferromagnetically or antiferromagnetically aligned core spins occurs when the lower kondo band is nearly empty ( @xmath15 ) or nearly filled @xmath16 . in previous work by some of the current authors , also treating the core spins classically
, similar numerical results were obtained.@xcite however , a closer analysis of the data revealed that the features which had been interpreted by other authors to indicate phase separation ( a discontinuity in the electron density versus the chemical potential , a pseudogap in the one - particle density of states ) were , in fact , due to small , independent ferromagnetic polarons .
likewise , ferromagnetic polarons have been found for the almost empty lower kondo band , for localized @xmath17 quantum spins.@xcite for the antiferromagnetic kondo model,@xcite small ferromagnetic droplets were predicted by kagan et al.@xcite the question arises whether the correct quantum mechanical treatment of the @xmath0 core spins would favor phase separation instead of independent ferromagnetic polarons , especially for @xmath18 . in this paper
, we address the influence of quantum spins on this issue .
the impact of a quantum mechanical treatment of spins on models for the manganites has been addressed in refs . .
in one dimension , quantum mechanical core spins with @xmath17 were employed in a number of studies conducted with the density matrix renormalization group ( dmrg).@xcite recently , garcia et al.@xcite presented a phase diagram for @xmath17 , which , however , was determined for only three values of the density and did not address the physically interesting region of doping , @xmath19 , treated in this paper .
quantum mechanical spins with @xmath0 have been briefly addressed in an exploratory study.@xcite the authors report phase separation when the lower kondo band is nearly filled . in this work ,
we present extensive calculations for the one - dimensional ( 1d ) ferromagnetic kondo lattice model with @xmath0 core spins using the dmrg .
we observe that quantum spins yield results in very good agreement with previous calculations@xcite for a model with classical core spins : for finite @xmath2 superexchange @xmath3 , polarons are favored over phase separation . in addition , we include an on - site coulomb repulsion @xmath14 between the @xmath13 electrons : accordingly , the superexchange parameter @xmath3 which favors antiferromagnetic alignment of the core spins is renormalized and the effective antiferromagnetic coupling @xmath20 is weakened .
this leads to an increased polaron size for large @xmath3 and phase separation for small @xmath3 .
the remainder of this paper is organized as follows . in sec .
[ sec : model ] , we define the model hamiltonian , for which we present dmrg results in sec . [
sec : results ] .
we discuss the ground state configurations in sec .
[ sec : ground ] . in sec .
[ sec : pol_rep ] we show that the polarons are actually repulsive at short distance .
we then discuss the impact of the antiferromagnetic @xmath2 superexchange @xmath3 and of the hubbard repulsion @xmath14 and show that quantum mechanical core spins are very well - approximated by classical spins ( sec .
[ sec : class ] ) .
we discuss the transition to the homogeneous ferromagnetic chain in sec .
[ sec : trans ] and present a phase diagram in sec .
[ sec : phase_diagram ] .
negative compressibility and the discontinuity in the density vs. the chemical potential are often taken as a sign for phase separation . in sec .
[ sec : compr ] , we argue that negative compressibility is an uncertain result when obtained from numerical methods such as the dmrg , quantum monte carlo , or exact diagonalization and show that a discontinuity in the density can equally well result from small independent polarons . finally , sec .
[ sec : conclusion ] summarizes and discusses the results presented in this work .
we study the ferromagnetic kondo lattice model with localized quantum core spins @xmath0 and one orbital per site , including a small heisenberg - like superexchange between the core spins as well as an on - site coulomb repulsion @xmath14 between the @xmath13 electrons : @xmath21 where @xmath22 ( @xmath23 ) creates ( destroys ) an @xmath13 electron with spin @xmath24 at site @xmath25 , @xmath26 is the corresponding density operator , @xmath27 is the core spin at site @xmath25 and @xmath28 the electron spin .
the first term describes the electron hopping ; the hopping integral @xmath29 will be used in the following as the unit of energy .
the second term describes the coulomb repulsion for the @xmath13 electrons ; we will treat @xmath30 and @xmath31 .
the third term describes the ferromagnetic hund s rule coupling between the @xmath13 electrons and the @xmath2 core spins . in this work , we take @xmath32 , which corresponds to @xmath33 in refs . and , if one compensates for the normalization of classical core spins to @xmath34 .
the last term describes an additional direct superexchange between the core spins . for manganites
, this effective interaction favors antiferromagnetic ordering of the core spins , i.e. , @xmath35 .
we vary @xmath3 from @xmath36 to @xmath37 .
note that @xmath38 corresponds to @xmath39 in the units of refs . and .
we employ the density matrix renormalization group , keeping up to 1000 states at each dmrg iteration and treating chains of length @xmath40 with open boundary conditions .
the discarded weight is at most of order @xmath41 for the results presented here .
our calculations indicate , however , that 48 states , as used in ref . , would lead to insufficient accuracy for the system sizes treated here .
figure [ fig : l24n21jh0.01u0 ] shows the on - site electron density and the nearest - neighbor core spin correlation @xmath42 for a dmrg ground state for a chain of length @xmath40 for @xmath38 and @xmath30 with three holes .
the holes clearly form three individual polarons , each extending over approximately three sites with ferromagnetically aligned core spins that are embedded in an antiferromagnetic background . and core spin - core spin correlations @xmath43 of the dmrg ground state for a chain of length @xmath40 with @xmath38 and @xmath30 .
three well - separated polarons can be clearly seen.,title="fig:",scaledwidth=45.0% ] + since the dmrg is a variational method , convergence to the true ground state is not guaranteed .
therefore , we have checked the consistency of the ground state by trying different system buildup strategies with respect to particle injection for many parameter sets .
thereby , we have examined whether there is degeneracy of states with respect to polaron positions and how far the existence of well - separated polarons and their positions are determined by the details of the dmrg algorithm . in addition , one can perform calculations with different @xmath44-components @xmath45 of the total spin @xmath46 : increasing @xmath45 reduces the size of the hilbert space .
this , in general , allows for more accurate results , while the smaller hilbert space still contains the ground state as long as @xmath47 . in an @xmath40 system with two holes ( @xmath48 )
, we always find two polarons regardless of how the system is built up .
all energies obtained are degenerate within the estimated numerical accuracy .
the polaron position , however , depends on where the particles are injected . in all cases , we find polarons that are separated by at least a few lattice sites with afm order .
even if we add both holes simultaneously during the system buildup , they separate into two polarons . therefore
, we conclude that well - separated polarons are effectively independent , while they seem to repel each other at short distances .
we estimate the energy connected to this repulsion by introducing small electrostatic potentials @xmath49 in order to trap the holes at sites @xmath50 and @xmath51 . for all distances @xmath52 , we obtain fm polarons covering three lattice sites with on - site densities symmetric with respect to @xmath50 and @xmath51 .
the ground - state energy as a function of the distance @xmath53 is shown in fig .
[ fig : epot ] , which corroborates the fact that polarons separated by two or more sites are effectively independent .
the configuration with only one intermediate site is @xmath54 higher in energy , and in order to obtain a state with adjacent polarons @xmath55 has to be paid . we therefore conclude that the holes actually have a _ repulsive _ interaction at short range and that separated polarons are energetically favored over phase separation into larger fm and afm regions near the half - filled kondo band for @xmath38 and @xmath30 .
the fact that we can obtain degenerate states with a localized polaron at different positions upon doping with one hole suggests that polarons are quasi - particles having a large effective mass . for physical reasons , we would expect that a single polaron delocalizes , forming a heavy fermi liquid , but it appears that polarons have a band - width that is too small to be resolved by the dmrg . as a function of the distance @xmath56 of two polarons on an @xmath40-site chain with @xmath57 .
the holes are centered at the position of electrostatic impurity potentials @xmath58 and exhibit a symmetric density distribution .
the energies for @xmath59 are degenerate within the estimated numerical accuracy as designated by the error bars .
[ fig : epot],scaledwidth=40.0% ] in ref .
, the energy for phase separation was compared to that for independent polarons as a function of an effective superexchange @xmath60 of classical core spins for the almost filled lower kondo band ( @xmath61 ) . at small values of @xmath60 ,
an increase in the optimal polaron size was found and bipolarons and phase separation dominated .
the effective exchange @xmath62 is defined in terms of the @xmath2 superexchange @xmath3 and the energy @xmath63 for virtual excitations into the upper kondo band : @xmath64 without coulomb repulsion , @xmath63 is given by the energy for the low - spin state @xmath65.@xcite for finite hubbard @xmath14 , the energy for the virtual excitations is higher , however , because they lead to doubly occupied sites for @xmath66 ; @xmath63 is then given by the energetically higher @xmath67 and @xmath68 states.@xcite taking into account coulomb repulsion should therefore have an effect similar to reducing @xmath3 . ) , for classical core spins with effective spinless fermions for @xmath36 , @xmath30 , inverse temperature @xmath69 and an @xmath40-site chain .
the ferromagnetic area grows with doping , indicating phase separation.[fig : hss_classical],scaledwidth=40.0% ] + the spinless fermion model with classical core spins@xcite forms polarons for @xmath38 ( corresponding to @xmath70 in the units of ref . ) , but for @xmath36 the polarons tend to attract each other and phase separate .
this can be seen by examining the dressed core spin correlation function @xmath71 where @xmath72 gives the hole density relative to the half filled chain , i.e. , it is only nonzero if the site is unoccupied .
we have evaluated this observable for the classical model with @xmath36 .
the result , depicted in fig . [ fig : hss_classical ] , shows that the ferromagnetic regions around the holes grow with doping , i.e. , that the polarons attract each other and tend to phase separate .
the dressed core spin correlation function @xmath73 is shown for quantum mechanical core spins and @xmath30 in fig .
[ fig : hss_qm_u0 ] .
for @xmath38 the size of the ferromagnetic regions around the holes does not grow with doping in accordance with our analysis in sec .
[ sec : pol_rep ] . in the case of @xmath36 , we observe phase separation also for the quantum model , which is reflected in the pronounced increase of fm correlations with doping in fig . [
fig : hss_qm_j0.00u0 ] . without an on - site coulomb repulsion @xmath14
, we find that the spinless fermion model@xcite agrees qualitatively very well with the present dmrg results , confirming that localized @xmath0 spins are well - approximated by classical spins .
if ones takes a closer look at the numerical values of the dressed core spin correlation function for both models , one has to keep in mind that states with double occupation have been projected out in the classical case , while such states are included in the dmrg calculations . in the latter case
, there is a finite hole density @xmath74 at sites @xmath25 within the afm background far from polarons , and still higher hole densities can be observed in the proximity of polarons .
these contributions of the afm background to the dressed core spin correlation function partly cancel the fm terms from within the polarons in the quantum model ; this accounts for the almost vanishing nearest - neighbor correlation @xmath75 for one hole shown in fig .
[ fig : hss_qm_j0.01u0 ]
. the inclusion of a finite hubbard @xmath31 decreases the effective exchange @xmath20 as discussed above and therefore increases the tendency to phase separation .
in fact , phase separation takes place even for @xmath38 which is reflected in the dressed core spin correlations in fig .
[ fig : hss_qm_u10 ] .
+ we first examine the polaronic case @xmath38 and @xmath30 , where the polarons extend over approximately 3 sites .
the ground state of an @xmath40 chain for @xmath76 ( 6 holes ) shown in fig .
[ fig : l24n18jh0.010u0 ] corresponds to a periodic arrangement of polarons , in agreement with the energy considerations of sec .
[ sec : pol_rep ] .
similar ` island phases ' have also been found for @xmath17 core spins at commensurate fillings:@xcite they consist of small ferromagnetic islands that are aligned antiferromagnetically or have one anti - aligned spin between them .
adding one more hole , we find that the polaronic configuration is destroyed and a large fm region forms with two afm arranged spins at each end , see fig .
[ fig : l24n17jh0.010u0 ] .
this could be a sign of phase separation , but much larger chains would have to be treated in order to clarify this issue . for dopings of @xmath77 ( 9 holes on an @xmath40 chain )
we observe complete fm polarization . when the core spin superexchange is increased to @xmath78 ( still neglecting coulomb repulsion @xmath30 ) , polaronic states become stabilized up to dopings of @xmath79 and we find another island phase of afm stacked polarons for 8 holes ; see fig . [
fig : l24n16jh0.020u0 ] .
the result for 9 holes in fig .
[ fig : l24n15jh0.020u0 ] suggests phase separation between an island phase and a fm region ; however , chains of length @xmath40 are too small to make definitive statements .
as the treatment of much larger systems is not feasible using currently available computational resources and the main focus of this paper is the almost - filled lower kondo band , we did not investigate the nature of the phase transition further .
we finally note that upon doping the @xmath40 chain with 10 or more holes , complete fm polarization is obtained ( not shown ) . for parameters
@xmath78 and @xmath31 , the polaronic regime extents up to @xmath76 ( 6 holes ) , so the transition to ferromagnetism is very similar to the one obtained for @xmath38 and @xmath30 .
we find an island phase as already shown in fig .
[ fig : l24n18jh0.010u0 ] and for 7 holes a configuration similar to the one shown in fig .
[ fig : l24n17jh0.010u0 ] . at dopings of 8 holes ( @xmath79 )
complete fm polarization is obtained for @xmath78 and @xmath31 .
@xmath3 plane for @xmath30 .
the symbols designate the following characteristics : p : polarons , ilp : island phase ( periodic arrangement of polarons ) , ps : phase separation , fm : ferromagnetic , fr : central ferromagnetic region with antiferromagnetically oriented sites at the ends of the chain .
the symbols ( letters ) are determined for an @xmath40-site chain . ,
title="fig:",scaledwidth=45.0% ] + we summarize the impact of the parameters @xmath3 and @xmath14 in phase diagrams in the @xmath80@xmath3 plane , where @xmath81 designates the filling and @xmath3 refers to the @xmath2 superexchange .
figure [ phases_l24u0 ] shows the phase diagram for an @xmath40-site chain without coulomb repulsion , @xmath30 .
one sees a polaronic region ( p ) near the filled lower kondo band , including periodic arrangement of polarons ( island phase , denoted as ilp ) for commensurate fillings . for large @xmath82
, this region extends to hole densities @xmath83 , while phase separation is found for vanishing @xmath36 .
some of our results indicate that there might be phase separation for intermediate fillings between polaronic and fm phases at @xmath35 .
however , much longer chains would be needed to clarify this issue ; see also our discussion in sec .
[ sec : trans ] .
corresponding states with a large central fm region and either some afm sites or polarons at both ends ( cf . figs .
[ fig : l24n17jh0.010u0 ] and [ fig : l24n15jh0.020u0 ] ) have been designated as `` fr '' in the phase diagram .
regions are labeled `` fm '' if _ all _ nearest - neighbor spin correlations are positive . for @xmath30 and @xmath36
, we find phase separation ( ps ) : by varying the system buildup , we are able to obtain both polaronic configurations and states with a larger fm region embedded in an afm background ; for these parameters , the phase separated states always have a lower energy .
while we expect macroscopic polarization in the thermodynamic limit near the empty kondo band , as also observed in 1d for the @xmath84 model@xcite and a cu - o chain@xcite , we did not investigate the extent of polarization for finite systems in detail .
we note that saturation should not be expected for quantum spins in the thermodynamic limit according to ref . . ,
but with @xmath31.[phases_l24u10],title="fig:",scaledwidth=45.0% ] + fig .
[ phases_l24u10 ] shows the phase diagram for coulomb repulsion @xmath31 . comparing with @xmath30
, we see that the polaronic phase is suppressed at lower doping and phase separation takes place even for @xmath38 .
this is consistent with the reduction in the effective parameter @xmath62 with increasing @xmath14 . for parameters
@xmath38 and @xmath31 , we find that two holes can gain @xmath85 forming a bipolaron instead of two separated polarons .
compared to the energy @xmath86 of the bipolaronic state , the energy difference is rather small and this is one reason for the sensitivity of the dmrg with respect to system buildup : if there are states very close in energy but far apart in phase space , numerical results will strongly depend on the initial configuration .
we have designated the corresponding parts with `` ps '' in fig .
[ phases_l24u10 ] .
the inverse compressibility of a system can be computed approximately by numerical differentiation of energies by @xmath87 where @xmath88 is the total energy of a chain with @xmath89 electrons on @xmath90 sites and @xmath91 is the difference in particle number .
negative values are sometimes taken to be an indication of phase separation.@xcite we want to argue that this condition is neither necessary nor sufficient : on the one hand , negative values only result from finite - size effects and should vanish in the thermodynamic limit leading to @xmath92 , as will be discussed below . on the other hand , we will show in fig .
[ fig : mu_n ] that polarons can cause @xmath93 as well .
such observations therefore have always to be complemented by an investigation of correlation functions , e.g. , the dressed core - spin correlation , eq .
( [ eq : hss ] ) , in order to show the existence of two distinct phases .
first , we note that numerical methods like the dmrg or monte carlo calculations which do not impose homogeneity , as for instance mean - field theory would , should yield separated phases in different spatial regions in a proportion that minimizes the free energy in effect , the system performs the maxwell construction by itself and should thus avoid negative compressibilities in the thermodynamic limit . for phase separation on finite systems , however , negative compressibilities @xmath94 can indeed arise , because the surface separating the two phases also contributes to the total energy .
( from the occurrence of ps , it can be inferred that the phase boundary is not energetically favorable because the system would otherwise tend to _ maximize _ instead of minimizing it and form many small droplets or a mixed phase . )
if this boundary has high enough energy and grows with doping , its contribution to the total energy leads to a negative compressibility . in the thermodynamic limit @xmath95 , such surface terms become negligible compared to the bulk contributions , implying @xmath96 .
moreover , the boundary surface does not have to grow with doping at all : in the present one dimensional case , the phase boundary always consists of just two bonds connecting the two phases regardless of their size . for this reason , one could obtain @xmath97 here even for finite systems .
in addition , one has to keep in mind that numerical differentiation is notoriously sensitive to even small numerical errors of the ground state energies entering eq .
( [ eq : compr ] ) .
a discontinuity of the density @xmath80 as function of the chemical potential @xmath98 is equivalent to the limit of infinite compressibility @xmath96 , and is likewise taken as an indication for phase separation .
we will here show that it can arise from independent polarons as well . in order to obtain @xmath99 from the dmrg calculations at fixed particle numbers @xmath89
, we set @xmath100 , where @xmath89 minimizes the grand canonical expectation value @xmath101 .
the results for chains with @xmath102 and @xmath31 , respectively , are shown in fig .
[ fig : mu_n ] .
we find a jump near @xmath103 both for @xmath31 , were ps is observed , as well as for @xmath30 , where we see _ polarons_. in the latter case , it can be accounted for by the independence of the polarons at low doping , i.e. , adding each polaron costs the same energy , which is balanced by the chemical potential @xmath104.@xcite this also corresponds to our observation that the ground state energy per site as a function of filling ( not shown ) lies practically on an straight line near @xmath105 with @xmath106 . in conclusion
, we have argued that a negative compressibility for a finite system is neither a necessary nor a sufficient condition for phase separation .
it is also important to note that a discontinuity of the density @xmath80 as function of the chemical potential @xmath98 is a necessary condition for phase separation , but is not a sufficient condition because other mechanisms such as independent polarons can induce a ` jump ' of @xmath99 as well .
we have presented an extensive numerical study of the one - dimensional ferromagnetic kondo lattice model with quantum mechanical @xmath0 core spins using the density matrix renormalization group method , treating a non - degenerate conduction band with and without on - site coulomb repulsion at low to intermediate doping . in particular , we have explored the similarities with the analogous model with classical core spins , where ferromagnetic polarons have been found to dominate over phase separation@xcite for parameters relevant to manganites .
we have investigated whether the inclusion of quantum fluctuations leads to an attractive interaction between the polarons and thus to phase separation .
we find that this is not the case : the polarons are in fact _ repulsive _ at short distances . in general , the results of the quantum model agree very well with the classical model and show that classical spins are indeed a very good approximation for manganites .
furthermore , we have investigated the influence of a relatively large local coulomb repulsion @xmath31 , which reduces the effective afm coupling @xmath62 relevant when @xmath107 . for small @xmath2 superexchange @xmath108
, phase separation dominates over separated polarons if on - site coulomb repulsion is taken into account , while polarons are favored upon increasing the superexchange parameter to @xmath78 .
we have summarized these findings in phase diagrams in the plane of doping and @xmath2 superexchange @xmath3 for @xmath30 and @xmath31 .
we have also argued that the observation of a ( small ) negative compressibility and of a jump in the density @xmath80 as a function of the chemical potential @xmath98 are not compelling indicators of phase separation .
while a discontinuity of @xmath99 is a necessary condition for phase separation , it is not a sufficient one : different mechanisms such as the formation of independent polarons can leave similar traces . | we present an extensive numerical study of the ferromagnetic kondo lattice model with quantum mechanical @xmath0 core spins .
we treat one orbital per site in one dimension using the density matrix renormalization group and include on - site coulomb repulsion between the electrons .
we examine parameters relevant to manganites , treating the range of low to intermediate doping , @xmath1 . in particular , we investigate whether quantum fluctuations favor phase separation over the ferromagnetic polarons observed in a model with classical core spins .
we obtain very good agreement of the quantum model with previous results for the classical model , finding separated polarons which are repulsive at short distance for finite @xmath2 superexchange @xmath3 .
taking on - site coulomb repulsion into account , we observe phase separation for small but finite superexchange @xmath3 , while for larger @xmath3 polarons are favored in accordance with simple energy considerations previously applied to classical spins .
we discuss the interpretation of compressibilities and present a phase diagram with respect to doping and the @xmath2 superexchange parameter @xmath3 with and without coulomb repulsion . |
recently , there has been a growing interest in the study of complex phenomena appearing in heterogeneous areas different from the traditional fields of research . in particular ,
one of those interesting fields turns to be the application of statistical physics methods to social phenomena . at present discussions of such heterogeneous aspects
can be found in several reviews @xcite .
one of the sociological problems that attracts much attention is the construction of consensus out of some initial condition and different models have been introduced in order to simulate and analyze the dynamics of opinion formation processes @xcite . mainly , opinion formation models are composed by interacting _
agents _ usually represented by the state adopted by certain variables .
agents interact between each other forming a network with a certain topology .
models differ to each other in three aspects : the type of variables used to represent the opinion of the agent , the type of interaction imposed between the agents and the topology adopted . here , in order to maintain certain degree of generality , the dynamics of consensus formation is studied using different topologies . each agent
is represented by a spin state variable representing the opinion in favor ( @xmath1 ) or against ( @xmath2 ) about a certain topic and the interaction between the agents is randomly changed during the evolution between two possible mechanisms , the galam majority rule @xcite and the sznajd updating rule @xcite .
this last model is a successful ising spin system describing a simple mechanism of making up decisions in a closed community in which the opinion influence flows outward a given group to the nearest individuals . in the sznajd model
a pair of nearest neighbors convinces its neighbors to adopt the pair opinion if and only if both members of the pair have the same opinion ; otherwise the pair and its neighbors do not change opinion .
the sznajd consensus model has rapidly acquired importance in the new field of computational socio - physics @xcite . additionally , it is meaningful to mention that consensus models are set up in complex networks with different topologies .
the statistical properties of real - world social networks vary strongly , for example , the degree distribution can be single - scale , broad - scale or scale - free @xcite . due to the lack of a single model covering the topological features of social networks
, we consider a few well established network models aiming to unveil the effect to different aspects of the topology : * standard one- and two- dimensional regular lattices ; * watts - strogatz small - world network @xcite ; * erds - rnyi random graph network @xcite ; * triad scale - free network ( barabsi - albert - like model ) @xcite .
on the system lattice , each site @xmath3 ( @xmath4 ; where @xmath5 is the total number of sites - agents ) carries a spin @xmath6 , which has two possible directions , @xmath7 or @xmath8 .
the state of these spin - like variables can be considered to represent an agent or individual carryng one of two possible opinions , against or in favor about a certain topic : @xmath9 represents a positive opinion and @xmath8 , a negative one .
initially the opinions are distributed randomly , @xmath7 with probability @xmath10 or @xmath11 , otherwise . as stated above , during the time evolution of the model ,
each agent can use two different strategies in order to update its opinion state variable @xmath12 , * with probability @xmath0 the agent opinion is updated according to a kind of sznajd rule @xcite : * * * square lattice and random graphs : * : for each site @xmath3 chosen , we select randomly one of its neighbors .
if this selected neighbor has the same opinion as the site @xmath3 , then all their neighbors follow the pair s opinion .
otherwise , nothing is done . * * * usual 1d regular chain : * two updating rules have been applied : a ) the original sznajd rule @xcite : @xmath13 and @xmath14 ; b ) 2 sites convincing rule . * with probability @xmath15
the agent opinion is updated using a galam - like majority rule without inertia in which individuals gather during their social life in _ meeting cells _ of different sizes where they discuss about a topic until a final decision , in favor or against , is taken by the entire group .
@xmath16 where the @xmath17 function takes the value @xmath1 if the argument is positive and the value @xmath2 , @xmath18 is the number of vertices in the neighborhood @xmath19 of site @xmath3 : the nearest neighbors in the usual one- and two- dimensional regular lattices and all nodes connected to @xmath3 in the random graphs . in the regular lattices
there is never a tie .
although the random graphs have been constructed considering a node connectivity to avoid such a tie , in the case it occurs , then the individual randomly assumes an opinion @xmath20 . in this subsection
we briefly describe the different topologies used , in addition to the standard one- and two- dimensional lattices .
social networks are far from being completely regular or completely random .
for instance , it has been found that most real - life networks display some common characteristics the most important of which are called small - world effect and scale - free distribution .
the recognition of small - world effect involves two factors : the clustering coefficient and the average shortest path length , a network is called a small - world network as long as it has small average shortest path length and large clustering coefficient @xcite .
one of the most well - known small - world models is watts - strogatz small - world network ( ws model ) @xcite , which can be constructed by the following algorithm : the initial network is a one - dimensional lattice of @xmath5 sites , with periodic boundary conditions ( i.e. , a ring ) , each site being connected to @xmath21 nearest neighbors .
we choose a vertex and the edge that connects it to its nearest neighbor in a clockwise sense .
with probability @xmath22 , we reconnect this edge to a vertex chosen uniformly at random over the entire ring , with duplicate edges forbidden ; otherwise we leave the edge in place .
this process repeats until one lap is completed and proceeding outward to more distant neighbors after each lap , until each edge in the original lattice has been considered once .
the construction of the erds - rnyi random graph @xcite starts with a set of @xmath5 isolated vertices , then successive edges are randomly added with a probability @xmath23 . in this way , the total number of edges is @xmath24 and the average number of neighbors of a node ( degree or connectivity ) is @xmath25 . in the limit @xmath26
, the mean number of bonds per site can be approximated by @xmath27 and a poissonian connectivity distribution is observed . in our simulations , the graph has been built in such way that each node has at least @xmath21 links , although the coordination number of any particular vertex can be bigger than @xmath21 links ( @xmath28 in our simulations ) .
the most fashionable network presenting both properties , scale - free and small - world aspects , is the barabsi - albert scale - free network @xcite .
although the ba network has successfully explained the scale - free nature of many networks , a striking discrepancy between it and real networks is that the value of the clustering coefficient varies fast with the network size @xmath5 and for large systems is typically several orders of magnitude lower than found empirically and it vanishes in the thermodynamic limit @xcite . in social networks , for instance
, the clustering coefficient distribution @xmath29 exhibits a power - law behavior , @xmath30 , where @xmath31 is the node degree ( number of neighbors ) and @xmath32 ( everyone in the network knows each other ) . ) ) and anti - ferromagnetically ( @xmath33 ) as a function of the probability @xmath10 for initial opinion @xmath1 for an one - dimensional lattice , @xmath34 and @xmath35 . on the left , original updating rule , and on the right , @xmath36 nodes convincing rule.,title="fig:",width=262,height=221 ] ) ) and
anti - ferromagnetically ( @xmath33 ) as a function of the probability @xmath10 for initial opinion @xmath1 for an one - dimensional lattice , @xmath34 and @xmath35 . on the left , original updating rule , and on the right , @xmath36 nodes convincing rule.,title="fig:",width=262,height=221 ] ) ) and
anti - ferromagnetically ( @xmath33 ) as a function of the probability @xmath10 for initial opinion @xmath1 for an one - dimensional lattice , @xmath34 and @xmath35 . on the left , original updating rule , and on the right , @xmath36 nodes convincing rule.,title="fig:",width=262,height=221 ] ) ) and anti - ferromagnetically ( @xmath33 ) as a function of the probability @xmath10 for initial opinion @xmath1 for an one - dimensional lattice , @xmath34 and @xmath35 . on the left , original updating rule , and on the right , @xmath36 nodes convincing rule.,title="fig:",width=262,height=221 ]
this problem has been surmounted and scale - free models with high clustering coefficient have been investigated , by adding a triad formation step on the barabsi - albert prescription @xcite . the barabsi - albert network starts with a small number @xmath21 ( @xmath28 in our simulations ) of sites ( agents , people ) all connected with each other .
then a large number @xmath37 of additional sites is added as follows : first , each new node ( node @xmath3 ) performs a preferential attachment step , i.e , it is attached randomly to one of the existing nodes ( node @xmath38 ) with a probability proportional to its degree ; then follows a triad formation step with a probability @xmath39 : the new node @xmath3 selects at random a node in the neighborhood of the one linked to in the previous preferential attachment step ( node @xmath38 ) .
if all neighbors of @xmath38 are already connected to @xmath3 , then a preferential attachment step is performed ( `` friends of friends get friends '' ) . in this model
, the original barabsi - albert network corresponds to the case of @xmath40 .
it is expected that a nonzero @xmath39 gives a finite nonzero clustering coefficient as @xmath5 is increased , while the clustering coefficient goes to zero when @xmath40 ( the ba scale - free network model ) @xcite . indeed , the clustering coefficient increases as the probability @xmath39 and @xmath21 increase .
at every time step @xmath41 , all the individuals @xmath5 are randomly visited and updated ( a random list of nodes assures that each node is reached exactly once ) by following the stated rules in section 2.1 . if a full consensus ( all individuals have the same opinion ) or if the maximum number @xmath42 of iterations is reached , the simulation ends . the curves presented here correspond to the results averaged over @xmath43 samples . in fig.[fig :
uni1 ] we present the mean opinion of the system as a function of the probability @xmath10 for the one dimensional model when different values of probability @xmath0 are considered . on the left side ,
the results are from using majority @xcite and the original sznajd @xcite rules .
as we can see , the mean opinion ( or final magnetization @xmath44 ) is given by @xmath45 when only the sznajd rule ( @xmath35 ) is applied and as @xmath0 increases @xmath44 quickly approaches a static value @xmath46 , that depends only on the initial magnetization @xmath47 . on the right side , it can be noticed that the system exhibits the same behavior for any value of probability @xmath0 .
moreover , from this plot , we find that in both updating rules , the 2 nodes convincing and the majority rule , the system always evolves to a ferromagnetic consensus ( either all up or all down ) and that they provide the same probability distribution for the system to reach a fixed point with all spins up or all spins down ( fig.[fig : uni2 ] , right ) , except in the anomalous case of an anti - ferromagnetic initial state ( @xmath48 ) .
nevertheless , when the system updates according to the original rule , besides the two ferromagnetic states , an anti - ferromagnetic fixed point is also reached ( fig.[fig : uni2 ] , left ) .
these results are in good agreement with previous studies on sznajd , majority and minority models @xcite .
very recently , a modification of the sznajd model in order to avoid the unrealistic anti - ferromagnetic fixed point was proposed in ref.@xcite . , but for the square lattice .
on the left , @xmath49 and @xmath50 , and on the right side , @xmath51 and @xmath52.,title="fig:",width=262,height=221 ] , but for the square lattice . on the left , @xmath49 and @xmath50 , and on the right side , @xmath51 and @xmath52.,title="fig:",width=262,height=221 ] in fig .
[ fig : sq ] , we observe the mean opinion density of the system as a function of the probability @xmath10 for initial opinion @xmath1 for a square lattice , as well as for different values of probability @xmath0 . from this figure , one can observe a phase transition : concentration @xmath53 lead to full consensus @xmath1 and concentrations @xmath54 to full consensus @xmath2 for large enough systems . since in a finite network , of course , phase transitions are never sharp , and thus the transition is indicated numerically by a slope ( fig.[fig : sq ] left ) becoming the steeper the larger the network size is ( fig.[fig : sq ] right ) . in an infinite network , one would get a sharp step function for the mean opinion of the systems versus the initial concentration @xmath10 of opinion @xmath1 . moreover , the results are the same independently of the value of the probability @xmath0 , i.e. , to update the system following sznajd recipe or majority rule provides the same behavior and consensus threshold @xmath55 . , but for erds - rnyi random graph . on the left , @xmath56 and @xmath57 , and on the right side , @xmath58 and @xmath52.,title="fig:",width=262,height=221 ] , but for erds - rnyi random graph . on the left , @xmath56 and @xmath57 , and on the right side , @xmath58 and @xmath52.,title="fig:",width=262,height=221 ] fig .
[ fig : erdos ] shows the mean opinion density of the system as a function of the probability @xmath10 for initial opinion @xmath1 for the erds - rnyi random graph . as we can see
, it presents the same pattern as in fig .
[ fig : sq ] , so that the consensus threshold is @xmath59 and there is no difference related to the chosen dynamics for updating the system : sznajd or majority rule . , but for watts - strogatz small - world network . on the left , @xmath56 and @xmath57 , and on the right side , @xmath58 and @xmath52 .
others parameters : @xmath60 and @xmath61,title="fig:",width=262,height=221 ] , but for watts - strogatz small - world network . on the left , @xmath56 and @xmath57 , and on the right side , @xmath58 and @xmath52 .
others parameters : @xmath60 and @xmath61,title="fig:",width=262,height=221 ] in fig .
[ fig : watts ] , we show the same as in fig.[fig : sq ] , but for the watts - strogatz small - world network . from this figure
, it can be observed that the consensus threshold is also @xmath59 and the chosen updating dynamics does not exert any influence on the results , as already noticed for the square lattice and the erds - rnyi random graph . , but for the triad scale - free network .
on the left , @xmath56 and @xmath50 , and on the right side , @xmath58 and @xmath52 .
@xmath62 , with @xmath28 , title="fig:",width=262,height=221 ] , but for the triad scale - free network .
on the left , @xmath56 and @xmath50 , and on the right side , @xmath58 and @xmath52 .
@xmath62 , with @xmath28 , title="fig:",width=262,height=221 ] fig .
[ fig : triad ] presents the mean opinion density of the system as a function of the probability @xmath10 for initial opinion @xmath1 for the triad scale - free network .
the same consensus threshold @xmath55 and results completely independent of the adopted updating dynamics are once more observed .
the phase transition at @xmath63 here observed for the square lattice and for the random graphs does not exist in one dimension @xcite or when a single site @xcite ( instead of a pair or plaquette ) on the square lattice @xcite convinces its neighbors , although it has been also found on the square lattice when a plaquette or a neighboring pair persuades its neighbors @xcite , on a correlated - diluted square lattice @xcite , on a triangular and simple cubic lattice if a pair convinces its @xmath64 ( or @xmath65 , respectively ) neighbors @xcite , on the barabsi - albert network @xcite , on a triad network @xcite and on a pseudo - fractal network @xcite .
in this article , we have investigated a simple model of opinion formation in which binary - state agents evolve due to the influence of agents in a local neighborhood by two different updating rules : with probability @xmath0 , the agents adopt the sznajd rule and the majority rule with probability @xmath66 . in one dimension , for @xmath67 the system always evolves to three fixed points : two of which refer to ferromagnetism ( `` all up '' or `` all down '' ) and one to anti - ferromagnetism . as @xmath0 increases
, the magnetization quickly approaches a fixed values that depends only on the initial magnetization . on the other hand
, we have also obtained that the 2 nodes convincing rules corresponds to a kind of majority rule in which only a ferromagnetic consensus is observed with similar probability distribution for a consensus + 1 or consensus -1 . moreover , taking into account more realistic topologies , the interplay between majority and sznajd rule leads the system to a phase transition as a function of the initial concentration of @xmath68 opinions at @xmath63 : for @xmath69 all samples end up with @xmath70 , and for @xmath71 they all end up in the other fixed point @xmath68 , for large enough systems . indeed , such consensus threshold @xmath72 of the opinion dynamics is not specific of the particular graph one uses to describe the social relationships between individuals , since it takes only one possible value @xmath73 .
moreover , no matter whether the direction of the information flux is inward ( @xmath74 , majority rule ) or outward ( @xmath75 , sznajd rule ) the system provides the same behavior .
in fact , our results represent a natural outgrowth of recent works @xcite on the interplay between the majority rule and sznajd model which claim that the sznajd model represents a particular case of a class of galam majority model , that shows the same threshold value @xmath72 .
we thank d. stauffer , s. galam and k. malarz for many helpful discussions , advice and a criticial reading of the manuscript .
s. galam , y. gefen and y. shapir , math .
* 9 * , 1 ( 1982 ) ; s. galam , b. chopard , a. masselot and m. droz , eur .
j. b * 4 * , 529 ( 1998 ) ; s. galam and j.d .
zucker , physica a * 287 * , 644 ( 2000 ) ; s. galam , eur . phys .
j. b * 25 * , 403 ( 2002 ) ; s. galam , physica a * 336 * , 49 ( 2004 ) ; s. galam , cond - mat/0409484 , a.o .
sousa , k. malarz and s. galam , cond - mat/0503702 ( int .
c * 16 * ( 2005 ) , in press ) .
g. deffuant , d. neau , f. amblard and g. weisbuch , advances in complex systems * 3 * , 87 ( 2000 ) ; d. stauffer , a.o .
sousa and c. schulze , j. artif .
( http://jasss.soc.surrey.ac.uk/7/3/7.html ) ; a.o .
sousa , cond - mat/0406766 .
d. stauffer , physica a * 336 * , 1 ( 2004 ) ; d. stauffer , journal of artificial societies and social simulations * 5 * , issue 1 , paper 4(jasss.soc.surrey.ac.uk ) ( 2002 ) ; computing in science and engineering * 5 * , 71 ( may / june 2003 ) ; and in _ the monte carlo method on the physical sciences _ , edited by j. e. gubernatis , aip conference proceedings , * 690 * , 147 ( 2003 ) ( cond - mat/0307133 ) .
krapivsky and s. redner , phys .
lett . * 90 * , 238701 ( 2003 ) ; m. mobilia , phys .
rev . lett.*91 * , 028701 ( 2003 ) ; m. mobilia and s. redner , phys . rev.e
* 68 * , 046106 ( 2003 ) . ;
d. makowiec , lecture notes in computer science , publisher : springer - verlag gmbh , 141 ( 2004 ) ( cond - mat/0407622 ) . a .-
barabsi and r. albert , science * 286 * , 509 ( 1999 ) ; r. albert and a .-
barabsi rev . of mod .
phys . * 74 * , 47 ( 2002 ) ; a.l .
barabsi , _ linked : the new science of networks _ , perseus publishing , cambridge ma , 2002 .
j. davidsen , h. ebel and s. bornholdt , phys.rev.letters * 88 * , 128701 ( 2002 ) ; p. holme and b.j .
kim , phys .
e * 65 * , 026107 ( 2002 ) ; g. szab , m. alava and j. kertsz , phys . rev .
e * 67 * , 056102 ( 2003 ) . | a simple model of opinion formation dynamics in which binary - state agents make up their opinions due to the influence of agents in a local neighborhood is studied using different network topologies .
each agent uses two different strategies , the sznajd rule with a probability @xmath0 and the galam majority rule ( without inertia ) otherwise ; being @xmath0 a parameter of the system . initially , the binary - state agents may have opinions ( at random ) against or in favor about a certain topic .
the time evolution of the system is studied using different network topologies , starting from different initial opinion densities . a transition from consensus in one opinion to the other is found at the same percentage of initial distribution no matter which type of network is used or which opinion formation rule is used .
social systems , opinion formation , structures and organization in complex systems , computer modeling and simulation . 89.65.-s , 89.75.-k , 05.10.-a |
the subject of gravitational lensing by black holes has received great attention in the last decade , basically due to the strong evidence about the presence of supermassive black holes at the center of galaxies , including ours @xcite .
the study of black hole gravitational lenses can be simplified by using the strong deflection limit , which is an approximate analytical method for obtaining the positions , magnifications , and time delays of the images .
it was introduced by darwin @xcite for the schwarzschild geometry , rediscovered by several authors @xcite , extended to reissner - nordstrm geometry @xcite , and to general spherically symmetric black holes @xcite .
kerr black holes were also analyzed adopting the strong deflection limit @xcite .
numerical studies @xcite were done too .
another related aspect that was considered , with the intention of measuring the properties of astrophysical black holes , are the shadows cast by rotating ones @xcite , which present an optical deformation caused by the spin , instead of being circles as in the non - rotating case .
this subject have been recently re - examined by several authors @xcite in the belief that the direct observation of black holes will be possible in the next years .
black holes coming from alternative theories @xcite , braneworld cosmologies @xcite and even naked singularities @xcite were considered as lenses . in the recent review article by bozza @xcite ,
these topics are discussed with more detail , including the observational prospects and additional references .
+ the problem of singularities in general relativity , where the theory breaks down , is a long standing one , involving both black holes and cosmological solutions .
regular black holes are solutions of the equations of gravity for which an event horizon is present , but the spacetime is free from singularities .
the first regular black hole was introduced by bardeen @xcite .
this solution has both an event and a cauchy horizons , but with a regular center , and it was shown by borde @xcite that the absence of the singularity is related to topology change . in bardeen model
the singularity is replaced by a de sitter core .
the bardeen black hole can be interpreted as a magnetic solution of the einstein equations coupled to nonlinear electrodynamics @xcite .
other spherically symmetric black holes with a regular center were studied by several authors , mainly in the context of non - linear electrodynamics @xcite ( for a recent review see ref .
@xcite ) . + in this article , we study gravitational lensing by a regular bardeen black hole , in the strong deflection limit .
the paper is organized as follows : in sec .
2 we find the strong deflection limit approximation for the deflection angle ; in sec .
3 we obtain the positions and magnifications of the relativistic images and we calculate the observables corresponding to the supermassive galactic black hole ; finally , in sec 4 we summarize the results obtained .
bardeen black holes are described by a spherically symmetric metric of the form @xmath0 where the metric functions , working in geometrized units ( speed of light in vacuum c=1 and gravitational constant g=1 ) , are given by @xcite @xmath1 with @xmath2 being the mass and @xmath3 a constant , which can be interpreted as a charge in the context of nonlinear electrodynamics @xcite .
for @xmath4 we recover the schwarzschild geometry , which has a singularity at its center . when @xmath5 this metric is regular everywhere , with a de sitter behavior for small values of the radial coordinate @xmath6 , and an asymptotically schwarzschild behavior for large @xmath6
. it is convenient to adimensionalize the metric ( [ m1 ] ) in terms of the schwarzschild radius @xmath7 by defining @xmath8 , @xmath9 , @xmath10 , so it becomes @xmath11 with @xmath12 it was shown @xcite that the existence of an event horizon for this metric is possible only in the case that @xmath13 .
for the strict inequality there are two horizons : the inner and the event ones ; and when @xmath14 , the horizons fuse into one . equating @xmath15
, we obtain @xmath16 which is a third degree polynomial equation in @xmath17 .
its largest positive solution corresponds to the event horizon radius @xmath18 .
+ the photon sphere radius @xmath19 is given by the largest positive solution of the equation @xcite @xmath20 where the prime represents the derivative with respect to @xmath21 .
for the bardeen metric this equation can be written in the form of a fifth degree polynomial equation in @xmath17 @xmath22 from which the value of @xmath19 is obtained by using standard software .
the deflection angle for a photon coming from infinity , as a function of the closest approach distance @xmath23 , has the form @xcite @xmath24 where @xmath25^{-1}-1}}. \label{i0}\ ] ] the deflection angle grows if @xmath23 approaches to @xmath19 , where it diverges .
following ref .
@xcite , we define the new variable @xmath26 by @xmath27 and the functions @xmath28 \sqrt{c(x_0 ) } , \label{rz}\ ] ] @xmath29c(x_0 ) [ c(x)]^{-1 } } } , \label{fz}\ ] ] where @xmath30 $ ] . by performing a taylor expansion of the function inside the square root in eq .
( [ fz ] ) one obtains @xmath31 where @xmath32 , \label{varphi}\ ] ] and @xmath33 ^{2}}{2[a'(x_0)]^{3 } [ c(x_0)]^{2}}\left\ { 2 [ a'(x_0)]^{2 } c(x_0 ) c'(x_0 ) - a(x_0 ) a''(x_0 ) c(x_0 ) c'(x_0 ) \right .
\nonumber \\ & & \left .
+ a(x_0 ) a'(x_0 ) \left [ c(x_0 ) c''(x_0 ) -2 [ c'(x_0)]^{2}\right ] \right\ } . \label{gamma}\end{aligned}\ ] ] with these definitions , the integral in eq .
( [ i0 ] ) can be separated into two parts @xmath34 where @xmath35 and @xmath36dz .
\label{ir}\ ] ] if @xmath37 we have that @xmath38 and @xmath39 , so the integral @xmath40 converges . instead , when @xmath41 , by using eq .
( [ psg ] ) we have that @xmath42 and @xmath43 , so @xmath40 has a logarithmic divergence .
therefore , @xmath40 is the term containing the divergence at @xmath44 and @xmath45 is regular because it has the divergence subtracted . + ( left ) and @xmath46 ( right ) as functions of @xmath47.,width=604 ] obtained numerically ( full line ) and to second order in @xmath47 ( dashed line ) .
it can be seen from the plot that the taylor approximation is good for small values of @xmath47 , as expected.,width=321 ] calculated numerically ( full line ) and the corresponding strong deflection limit values ( dashed line ) as functions of the impact parameter @xmath48 , for some values of @xmath47.,width=604 ] the impact parameter @xmath48 is more easily connected to the lensing parameters than @xmath23 . as a consequence of the conservation of angular momentum ,
@xmath48 is related to @xmath23 by @xmath49 bozza @xcite showed that the logarithmic divergence of the deflection angle for photons passing close to the photon sphere has the form @xmath50 where @xmath51 is the impact parameter evaluated at @xmath44 , @xmath52 and @xmath53 with @xmath54 this is the so - called strong deflection limit . for the bardeen black hole
we obtain that the strong deflection limit coefficients are given by @xmath55 } } , \ ] ] and @xmath56 where @xmath57 } { \left(-2q^{2}+x^{2}_{ps}\right)^{3}\left [ 1-x^{2}_{ps}\left(q^{2}+x^{2}_{ps}\right)^{-3/2}\right ] } .
\label{delta}\ ] ] in our case , @xmath58 can not be calculated analytically .
therefore , for small values of @xmath59 , a second order power expansion is suitable as a good approximation : @xmath60 where @xmath61 is the value of @xmath58 for a schwarzschild black hole and @xmath62 is the correction provided by the parameter @xmath63 .
we obtain @xmath64\approx 0.949603 , \label{cr0}\ ] ] @xmath65\right\}\approx -3.86568 .
\label{cr2}\ ] ] for large values of @xmath59 , this taylor expansion is not accurate , and @xmath58 should be obtained numerically .
the numerical evaluation of the integral that gives @xmath58 is rather cumbersome , and @xmath46 can be obtained more easily in a similar way as it was done in ref .
@xcite , from the limit @xmath66 .
\label{barblim}\ ] ] the strong deflection limit coefficients as functions of @xmath10 , with @xmath46 obtained numerically from eq .
( [ barblim ] ) , are presented in fig .
1 . for small values of @xmath59
the taylor expansion of @xmath46 in eq .
( [ cr2o ] ) gives good results , as shown in fig .
2 . for photons passing close to the photon
sphere the strong deflection limit is an excellent approximation , as it can be seen in fig .
we consider a black hole lens situated between a point source of light and the observer .
the bardeen space - time is asymptotically flat , and we take both the observer and the source far enough from the black hole to be in the flat region . in this section ,
we find the positions and magnifications of the relativistic images and we obtain the observables for the supermassive black hole at the center of our galaxy . the optical axis can be defined as the line joining the observer ( @xmath67 ) and the lens ( @xmath68 ) .
the angular positions of the source ( @xmath69 ) and the images , seen from the observer , are @xmath70 ( positive ) and @xmath71 , respectively . the observer - source ( @xmath72 ) , observer - lens ( @xmath73 ) and the lens - source ( @xmath74 ) distances ( measured along the optical axis ) are taken much greater than the horizon radius . in this approximation , the lens equation has the form @xcite @xmath75 the lensing effects are more important when the objects are highly aligned , so we will only study this case , in which the angles @xmath70 and @xmath71 are small , and @xmath76 is close to an even multiple of @xmath77 .
when @xmath78 two weak deflection primary and secondary images , which will be not analyzed here , and two infinite sets of point relativistic images are obtained @xcite .
the first set of relativistic images have a deflection angle that can be written as @xmath79 , with @xmath80 and @xmath81 . in this approximation ,
the lens equation results @xcite @xmath82 for the other set of images , @xmath83 , therefore @xmath84 should be replaced by @xmath85 in eq .
( [ pm2 ] ) . from the lens geometry
it is clear that @xmath86 , so the deflection angle given by eq .
( [ ad ] ) takes the form @xmath87 inverting eq .
( [ pm4 ] ) to obtain @xmath71 in terms of @xmath76 and performing a first order taylor expansion around @xmath88 , the angular position of the @xmath89-th image is given by @xmath90 with @xmath91 , \label{pm7}\ ] ] and @xmath92 from eq .
( [ pm2 ] ) , one has that @xmath93 , and replacing it in eq .
( [ pm6 ] ) @xmath94 then , using that @xmath95 and keeping only the first order term in @xmath96 , it is easy to see that the angular positions of the images are @xmath97 the second term in eq .
( [ pm14 ] ) is a small correction on @xmath98 , so all images lie very close to @xmath98 . with a similar treatment , the other set of relativistic images
have angular positions
@xmath99 when @xmath100 ( perfect alignment ) instead of point images we obtain an infinite sequence of einstein rings with angular radii @xmath101 it is well known @xcite that gravitational lensing conserves surface brightness and the quotient of the solid angles subtended by the image and the source gives the magnification of the @xmath89-th ] and , using that the angles are small and eq .
( [ pm14 ] ) , we have @xmath102 \frac { \zeta _ { n}d_{os}}{d_{ls } } , \label{pm18}\ ] ] which can be approximated to first order in @xmath103 by @xmath104 the same expression is obtained for the other set of relativistic images .
the magnifications decrease exponentially with @xmath89 , so the first image is the brightest one .
the magnifications are proportional to @xmath105 , which is a very small factor , then the relativistic images are very faint unless @xmath70 is close to zero ( nearly perfect alignment ) .
the magnification becomes infinite if @xmath100 , and the point source approximation is no longer valid in this case .
the outermost image is the first one , with angular position @xmath106 , and the others approach to the limiting value @xmath107 as @xmath89 increases . when the outermost image can be resolved from the rest of them ,
the lensing observables defined by bozza @xcite @xmath108 and @xmath109 are useful .
the observable @xmath69 is the angular separation between the first image and the limiting value of the succession of images , and @xmath6 represents the ratio between the flux of the first image and the sum of the fluxes of the other images . by using the strong deflection limit and for high alignment
, they can be written in the form @xmath110 and @xmath111 by measuring @xmath112 , @xmath69 and @xmath6 and inverting eqs .
( [ eqs ] ) and ( [ eqr ] ) , the strong deflection limit coefficients @xmath113 and @xmath46 can be obtained and their values can be compared with those predicted by the theoretical models to identify the nature of the black hole lens . +
the galactic center black hole , modeled using the bardeen metric , is analyzed here to provide a numerical example .
the black hole parameters , taken from ref .
@xcite , are the mass @xmath114 and the distance from the earth @xmath115 kpcs .
the observables , as functions of the ratio between the charge and the schwarzschild radius , are shown in fig .
we see that @xmath116 and @xmath6 lessen with @xmath47 while @xmath69 grows with @xmath47 .
( left ) , @xmath69 ( center ) , and @xmath6 in magnitudes [ @xmath117 , right ] as functions of @xmath47 for the supermassive galactic black hole.,width=604 ]
in this article we have obtained the strong deflection limit for the deflection angle corresponding to a regular bardeen black hole , in terms of the self - gravitating monopole charge @xmath3 in non linear electrodynamics @xcite .
this limit was calculated analytically by performing a second order taylor expansion in the case of small values of the quotient @xmath118 between the charge and twice the mass , while for any value it was found numerically . for close alignment , the resulting positions and magnifications of the relativistic images were obtained and the results were applied to calculate observables for the supermassive black hole situated at the center of our galaxy .
we have found that when @xmath118 increases , the relativistic images are closer to the black hole .
the angular separation @xmath69 also grows with @xmath118 , which means that the first relativistic image and the asymptotic limit of the angular positions of the images are more separated than in the case of the schwarzschild black hole .
the first relativistic image also loses intensity with respect to the others if @xmath118 increases .
then , for a bardeen black hole , the first relativistic image is less intense with respect to the others than in the schwarzschild geometry .
future instruments are expected to look into the region close to the galactic supermassive black hole ( see for example ref . @xcite and references therein ) .
gravitational lensing would serve then as a tool to compare different black hole models with observations .
this work has been supported by universidad de buenos aires and conicet .
luminet , astron . astrophys . * 75 * , 228 ( 1979 ) ; h.c .
ohanian , am .
* 55 * , 428 ( 1987 ) ; r.j .
nemiroff , am .
* 61 * , 619 ( 1993 ) ; v. bozza , s. capozziello , g. iovane , and g. scarpetta , gen .
. gravit . * 33 * , 1535 ( 2001 ) .
v.bozza , phys .
d * 67 * , 103006 ( 2003 ) ; v. bozza , f. de luca , g. scarpetta , and m. sereno , phys .
d * 72 * , 083003 ( 2005 ) ; v. bozza , f. de luca , and g. scarpetta , phys .
d * 74 * , 063001 ( 2006 ) .
a. de vries , class .
quantum grav .
* 17 * , 123 ( 2000 ) ; r. takahashi , astrophys .
j. * 611 * , 996 ( 2004 ) ; a.f .
zakharov , a.a .
nucita , f. depaolis , and g. ingrosso , new astron .
* 10 * , 479 ( 2005 ) ; a.f .
zakharov , f. de paolis , g. ingrosso , and a.a .
nucita , astron . astrophys . * 442 * , 795 ( 2005 ) ; k. hioki and u. miyamoto , phys .
d * 78 * , 044007 ( 2008 ) ; c. bambi , and k. freese , phys .
d * 79 * , 043002 ( 2009 ) ; k. hioki and k.i .
maeda , phys . rev .
d * 80 * , 024042 ( 2009 ) ; l. amarilla , e.f .
eiroa , and g. giribet , phys .
d * 81 * , 124045 ( 2010 ) .
a. bhadra , phys .
d * 67 * , 103009 ( 2003 ) ; e.f .
eiroa , phys .
d * 73 * , 043002 ( 2006 ) ; k. sarkar , and a. bhadra , class .
quantum grav .
* 23 * , 6101 ( 2006 ) ; n. mukherjee , and a.s .
majumdar , gen .
. gravit . * 39 * , 583 ( 2007 ) ; g. n. gyulchev , and s. s. yazadjiev , phys . rev .
d * 75 * , 023006 ( 2007 ) ; s. chen and j. jing , phys .
d * 80 * , 024036 ( 2009 ) .
v. frolov , m. snajdr , and d. stojkovic , phys .
d * 68 * , 044002 ( 2003 ) ; e.f .
eiroa , phys .
d * 71 * , 083010 ( 2005 ) ; r. whisker , phys .
d * 71 * , 064004 ( 2005 ) ; a.s .
majumdar , and n. mukherjee , int .
d * 14 * , 1095 ( 2005 ) ; e.f .
eiroa , braz .
* 35 * , 1113 ( 2005 ) ; c. r. keeton , and a. o. petters , phys
d * 73 * , 104032 ( 2006 ) ; s. pal and s. kar , class .
quantum grav . * 25 * , 045003 ( 2008 ) ; c.g .
boehmer , g. de risi , t. harko , and f.s.n .
lobo , class .
quantum grav .
* 27 * , 185013 ( 2010 ) ; a.y .
bin - nun , phys .
d * 81 * , 123011 ( 2010 ) ; a.y .
bin - nun , phys .
d * 82 * , 064009 ( 2010 ) .
e. ayn - beato and a. garca , phys .
lett . * 80 * , 5056 ( 1998 ) ; e. ayn - beato and a. garca , phys .
b * 464 * , 25 ( 1999 ) ; e. ayn - beato and a. garca , gen .
* 31 * , 629 ( 1999 ) ; k.a .
bronnikov , phys .
d , * 63 * , 044005 ( 2001 ) ; e. elizalde and s. r. hildebrandt , phys .
d * 65 * , 124024 ( 2002 ) ; i. dymnikova , int . j. mod . phys .
d * 12 * , 1015 ( 2003 ) ; i. dymnikova , class .
quantum grav .
* 21 * , 4417 ( 2004 ) ; e. ayn - beato and a. garca , gen .
relativ . gravit . * 37 * , 635 ( 2005 ) ; k.a .
bronnikov and j. c. fabris , phys .
lett . * 96 * , 251101 ( 2006 ) ; k.a .
bronnikov , h. dehnen , and v. n. melnikov , gen .
. gravit . * 39 * , 973 ( 2007 ) . | in this paper , we study a regular bardeen black hole as a gravitational lens .
we find the strong deflection limit for the deflection angle , from which we obtain the positions and magnifications of the relativistic images . as an example , we apply the results to the particular case of the supermassive black hole at the center of our galaxy .
pacs numbers : 98.62.sb , 04.70.bw , 04.20.dw keywords : gravitational lensing , black hole physics , regular solutions |
a comprehensive understanding of superdeformed ( sd ) bands requires knowledge of the quantum numbers ( spin and parity ) and excitation energies of the levels in the second well .
in particular , these quantities allow for stringent tests of configuration assignments and , more importantly , of the ability of theory to calculate shell - correction energies at large deformation .
however , although more than 250 sd bands are known in the a=150 and 190 regions@xcite , only a few sd bands in @xmath2hg@xcite , @xmath2pb@xcite , @xmath3pb@xcite and in @xmath4dy@xcite , have the spins and excitation energies determined through one - step linking transitions .
the yrast sd band in @xmath0hg was the first one to be discovered in the @xmath5 region @xcite .
it is especially interesting to obtain the spins and excitation energies for an odd - a sd band which , combined with data on neighboring even - even nuclei , can give information on the relative pair correlation energies in normal - deformed ( nd ) and superdeformed ( sd ) states . so far , the main information on single - particle configurations has come from detailed analyses of the @xmath6 dynamic moments of inertia of the sd bands . with a knowledge of the level energies and the associated quantum numbers ,
calculations can be put to more extensive tests and information can be obtained on properties such as particle alignment . the work of vigezzi _ _ et al.__@xcite and recent improvements by refs .
@xcite laid the theoretical foundation for treating the coupling of an isolated , cold sd state with a high - density of hot compound nd states , which forms the basis of the decay mechanism .
the decay of sd bands happens suddenly , typically out of one to two sd states in the mass 190 region .
one possible mechanism responsible for this sudden decay out of the sd band , proposed in ref .
@xcite , is chaos - assisted tunneling .
when the sd band decays , most of the strength is fragmented over numerous pathways , thus forming a quasi - continuum spectrum @xcite with sharp lines at high energy , which are due to direct decay to low - lying discrete nd levels .
the decay spectrum from sd states is similar@xcite to the spectrum following the decay of neutron - capture states@xcite . in both cases ,
the decay originates from a discrete point in excitation energy and spin and proceeds to a multitude of final states .
one way of determining the spins and excitation energies of sd bands is to analyze the quasi - continuum decay spectrum connecting sd and nd states .
the technique to extract the quasi - continuum decay spectrum was pioneered on @xmath3hg in the work of henry _ _ et al.__@xcite .
an improved method is described in detail by lauritsen _ _ et al.__@xcite and is the one used in this paper .
the method has been successfully tested in the case of sd band 1 in @xmath2hg@xcite , where it determined the same spins and excitation energies as those known from several one - step @xmath1-ray transitions connecting the sd band to known nd states@xcite .
the results from the quasi - continuum analysis are an important complement in cases where only one or two decay pathways are known . however , in most instances , one - step transitions are not observed , and it is then the only available option . in this work the quasi - continuum spectrum following the decay of the yrast sd band in @xmath0hg has been extracted . from this spectrum ,
we determine the excitation energy and spin of the sd band and also derive information on pairing in normal - deformed states .
we also present two 1-step decay pathways , which directly connect the yrast sd band in @xmath0hg with known yrast levels in the nd level scheme .
it will be shown that the results from the two methods agree very well and , thus , we can make a confident assignment of the spin and excitation energy of the yrast sd band in @xmath0hg .
the experimental results are compared with theoretical calculations based on the hartree - fock - bogoliubov ( hfb ) theory with several skyrme interactions .
we shall also extract and discuss information on pairing in sd states , by comparing the present results with those from the even - even hg nuclei .
superdeformed states in @xmath0hg were populated using the @xmath7yb(@xmath8ne,5n)@xmath0hg reaction .
the experiment was performed with the gammasphere array@xcite , which had 96 compton - suppressed ge detectors at the time of the experiment . the 120 mev @xmath8ne beam
was provided by the 88 cyclotron at lawrence berkeley national laboratory .
the 3.1 mg/@xmath9 @xmath7yb target had a 6.8 mg/@xmath9 @xmath10au backing to stop the recoiling nuclei .
the decay - out @xmath1 rays were emitted after the recoils came to rest in the backing , so that the transitions will correspond to sharp lines rather than doppler broadened ones .
a total of @xmath11 x @xmath12 triple- or higher - fold coincidence events were collected .
the @xmath1-ray spectrum , obtained with pairwise coincidence gates on sd transitions , is shown in fig . [
fig : link ] .
the lower panel presents the high - energy part of the spectrum , in particular , the two transitions at 2778 and 3310 kev , which will be shown to connect sd and known nd states@xcite .
the stronger 2778-kev transition has an area of 6@xmath13 , while the 3310-kev transition has an area of 3@xmath13 , where @xmath13 is the statistical uncertainty .
the intensities of the 2778 kev and 3310 kev lines are 0.8 % and 0.4 % , respectively , of the maximum sd intra - band intensity .
figure [ fig : link - sd ] gives the coincidence spectra gated on a sd line and either the 2778-kev ( upper figure ) or 3310-kev ( lower figure ) transition . even though the statistics are low , the intensities of the nd lines suggest that the 2778-kev transition feeds the known nd 35/2@xmath14 yrast level at 3222 kev .
the 3310-kev transition has been assigned to feed a 33/2@xmath14 known nd level at 2690 kev .
the deduced decay scheme is shown in fig .
[ fig : level ] . on the basis of the coincidence data , both one - step transitions place the deexciting sd level ,
i.e. the one fed by the 351 kev sd transition , at 6000 kev - see fig .
[ fig : level ]
. the angular distribution coefficient of the stronger one - step line ( 2778 kev ) , a@xmath15 0.57@xmath16 0.48 , is consistent with a @xmath17i = 0 dipole assignment , suggesting a 35/2 @xmath18 spin assignment for the level fed by the 351 kev sd transition .
we rule out the possibility of it being a stretched e2 transition , because that would require m3 multipolarity for the 3310-kev one - step transition .
the spin is consistent with a favored @xmath19 , @xmath20 configuration assignment , which is calculated to be yrast at high spin @xcite .
the experimental data do not allow for a parity assignment .
however , the @xmath20 configuration assignment requires the sd band to have negative parity , implying m1 multipolarity for the one - step transitions .
this m1 assignment is discussed later . the partial level scheme in fig .
[ fig : level ] shows the levels fed in the decay of the sd band .
the energy of the 13/2@xmath21 state is set at zero to facilitate the comparison of the experimental and theoretical sd excitation energies and to circumvent the 22-kev uncertainity in its excitation energy @xcite .
hence , the 3/2@xmath22 ground state , which is not populated by the sd band , has an energy of -128(22 ) kev in fig .
[ fig : level ] .
a method has been developed at argonne@xcite to isolate the quasi - continuum @xmath1-ray spectrum connecting the sd and nd states . to confirm the results from the one - step linking transitions , this method , which is described in detail in ref.@xcite , was followed here .
first , the data were sorted with double coincidence gates on sd transitions to obtain clean spectra . only double gates which produce clean sd spectra , without significant contamination by nd transitions ,
were used .
the background subtraction was done using the ful method@xcite .
corrections were carried out for @xmath1-ray summing@xcite and for neutron interactions in the detectors@xcite .
the spectra were unfolded @xcite to eliminate contributions from compton - scattered @xmath1 rays and corrected for the detector efficiency .
the area of the spectrum was then normalized to multiplicity by requiring that the sum of the intensities of transitions feeding the ground state is 1 .
the 390 kev line is a doublet composed of a sd transition and a transition feeding the nd ground state .
the 390-kev sd component is taken to have multiplicity 1 , suggested by the plateau in the intensity of the sd transitions@xcite .
the total normalized @xmath1 spectrum is shown on a logarithmic scale in fig .
[ fig : sdnd ] , together with the equivalent spectrum , obtained by gating on two nd yrast lines . above 1 mev ,
the spectra are contracted to 32 kev / channel and below that to 1.33 kev / channel .
there is clearly extra yield in the sd gated spectrum between 1 and 2.5 mev , which comes from the decay out of the sd band@xcite .
the discrete peaks below 800 kev are subtracted from the continuum spectrum .
they can be identified as transitions either along the yrast sd band or near the nd yrast line ( including previously unassigned transitions ) .
the remaining continuum spectrum contains contributions from components ( of statistical , quadrupole and m1/e2 nature ) that feed the sd band , in addition to the sought - after decay - out spectrum . to extract the decay - out spectrum , the feeding components , starting with the one of statistical nature have to be subtracted .
the feeding component of statistical nature can not be disentangled experimentally from the decay - out continuum . instead
, it is obtained from a monte carlo simulation of the feeding of the sd band in @xmath0hg .
the monte carlo code is described in refs .
one of the inputs in the calculation of the statistical spectrum was the shape of the entry distribution .
this was not measured for @xmath0hg in the experiment ; so the shape of the entry distribution for @xmath3hg from ref .
@xcite was used .
the distribution was shifted to have the right average entry spin and excitation energy ( 33.9(1.7 ) @xmath18 and 13.8(0.7 ) mev ) , as found from the analysis of quasi - continuum @xmath1 rays feeding all states in @xmath0hg .
the average entry point for cascades feeding only the sd band was 44.1(2.2 ) @xmath18 and 17.3(0.9 ) mev , as given in table [ tab : comp ] .
the feeding components of the spectrum are doppler shifted , since the speed of the recoiling compound nucleus is v / c = 0.019 ( for @xmath0hg nuclei formed halfway through the target ) . to take this into account
, the spectra were transformed into the center - of - mass system .
after the statistical feeding component was removed , the quadrupole and dipole feeding components could be separated based on the a@xmath23 coefficient of the angular distribution in the center - of - mass system - see fig .
[ fig : a2 ] . the low - energy component ( @xmath24 kev )
is characterized by large negative @xmath25 coefficients , indicating m1/e2 nature ( as seen also in @xmath26hg@xcite ) . after extraction and subtraction of the quadrupole e2 component ,
the dipole m1/e2 feeding component and decay - out component remain . a sharp drop around 850 kev in the m1/e2 spectrum ( fig .
[ fig : qc ] ) and the drop in the a@xmath23 coefficients in the same energy region indicates the presence of two components . the upper component is assigned to the decay - out of the sd band , following refs .
@xcite .
the different components of the continuum spectrum are presented in the upper part of fig .
[ fig : qc ] . the energy and spin removed , on average , by the different @xmath1-ray components
are listed in table [ tab : comp ] . for comparison ,
the different components of the spectrum feeding all states ( mostly of nd nature ) in @xmath0hg , from a similar analysis with two gates on yrast nd transitions ( 390.5 kev and 750.2 kev ) , is shown in the lower part of fig .
[ fig : qc ] .
there are two notable differences in the two spectra in fig .
6 . first , the quadrupole component feeding sd states has lower energy ( 0.70 _ vs. _ 0.77 mev ) , is narrower and has larger multiplicity ( 4.0 _ vs. _ 3.1 ) than that feeding nd states .
the differences in this component , which arise from excited @xmath1 cascades , are attributed to the larger collectivity in the sd well .
second , the top spectrum has an additional component , starting at 0.8 and extending to 3.3 mev , which arises from the decay - out quasicontinuum @xmath1 rays connecting sd and nd states . the decay - out spectrum connecting the sd and yrast nd states ( including statistical and discrete non - yrast transitions ) is given in fig .
[ fig : doyr ] . from monte carlo simulations@xcite of the statistical decay
, it is found that each quasi - continuum @xmath1 ray removes 0.5(1)@xmath18 of spin .
the energy and spin removed , on the average , by the decay - out components are found by : @xmath27 and @xmath28 where @xmath29 and @xmath30 are the average energy and spin removed per @xmath1 ray ( for a given component @xmath31 ) and @xmath32 is the average multiplicity of this component .
the total @xmath1-ray spectrum connecting sd and yrast nd states removes @xmath33 mev and @xmath34 . from the intensities of the nd yrast transitions in our sd gated spectrum , the average entry point into the nd yrast region
is found to be 2.24(15 ) mev and 14.7(4 ) @xmath18 .
the yrast transitions are taken from the nd level scheme of ref.@xcite .
the energy and spin of the level fed by the 351 kev sd transition is , therefore , determined to be e@xmath35=5.7(5 ) mev and i@xmath35=17.8(8 ) @xmath18 .
contributions to the uncertainty come from the calculated feeding statistical spectrum , the normalization to multiplicity , uncertainty in the spin removed by the quasi - continuum decay - out component and the multipolarities of the unknown lines .
the errors are added in quadrature . in fig .
[ fig : test ] , the experimental results from the 1-step linking transitions and from the quasi - continuum analysis are presented in a spin - energy diagram .
the filled circles represent the yrast nd and sd levels , as given by the level scheme ( fig .
[ fig : level ] ) , based on the one - step decay paths .
the filled diamond denotes the sd level , which is fed by the 351 kev transition , and the open diamond the average entry point into the nd yrast band , as obtained from the quasi - continuum analysis .
the box around the filled diamond shows the uncertainty in spin and energy from the quasi - continuum analysis .
clearly , the results from the 1-step linking transitions and from the quasi - continuum analysis are in good agreement with each other .
this gives confidence about the spin and energy assignments .
with the spins now known , the kinematic moment of inertia @xmath36 can also be determined .
figure [ j1j2 ] shows both @xmath36 and @xmath6 moments as a function of @xmath37 .
the dynamical moment of inertia @xmath6 can be expressed in terms of the harris expansion@xcite : @xmath38 integration of eq .
( [ khoo1 ] ) gives @xmath39 and @xmath40 here @xmath37 is the rotational frequency , given by @xmath41 ; @xmath42 is the spin perpendicular to the symmetry axis and @xmath31 , the integration constant , represents the quasi - particle alignment . for @xmath0hg , fits of @xmath6 and @xmath42 _ vs. _ @xmath37 with eqs .
( [ khoo1 ] ) and ( [ khoo2 ] ) give @xmath43 .
the behavior of @xmath36 of @xmath0hg is different from that of other a@xmath44190 sd bands , where the moments of inertia increase monotonically with @xmath37 .
the u - shaped curve of @xmath36 arises from the @xmath31/@xmath37 term ( which causes the unusual rise at low @xmath37 ) and provides a characteristic signature for a band exhibiting finite alignment .
knowledge of the spins allows configuration assignments to be made on a solid foundation . in the past
, the assignments were largely based on the variation with rotational frequency @xmath37 of the dynamical moment of inertia @xmath6 . in only a handful of cases ,
spins were extracted using a fit method [ using eqs .
( [ khoo1 ] and [ khoo2 ] ) ] , with the assumption that @xmath31 = 0 ] . for sd band 1 in @xmath0hg , which exhibits particle alignment ,
this method can not be used and spins were proposed by carpenter @xmath45 @xcite , based on the entry spin into the nd yrast line after decay from the sd band .
the present work firmly establishes the spins and confirms the assignment of ref .
@xcite , thus validating the interpretation that sd band 1 in @xmath0hg is based on the n=7 @xmath20 [ 761]3/2 configuration .
the alignment , @xmath46 , is an important ingredient in this assignment .
together , this work and ref .
@xcite add confidence about the single - particle orbitals calculated by theory .
the woods - saxon potential gives this orbital as the yrast configuration at large deformation and at high spin .
the neutron quasi - particle routhians for @xmath3hg @xcite suggest that this feature would also be given by hfb theory .
however , details are not always correctly predicted , e.g. , at the lowest frequencies , there is a discrepancy of 2 @xmath18 in alignment between experiment and hfb theory for sd band 1 in @xmath0hg @xcite . altogether , mean - field theories provide good descriptions of the general features of sd bands in the mass 190 region , such as the rise of @xmath6 with @xmath37 ( due partly to the n=7 orbital ) and the convergence at high @xmath37 for most nuclei .
this has been summarized in work by fallon _
et al . _
@xcite , which distills the main physics results .
the assigned spin of the band is consistent with a favored @xmath19 , @xmath20 particle configuration assignment , which is calculated @xcite to be yrast at high spin .
the experimental data do not allow for a parity assignment .
however the @xmath20 assignment requires the sd band to have negative parity , implying m1 multipolarity for both of the one - step transitions . from neutron - capture data ,
it is known that , at e@xmath47 @xmath44 8 mev , the decay is dominated by e1 transitions@xcite . however , in @xmath0hg the one - step transitions have significantly lower energy , e@xmath47 @xmath44 3 mev .
in fact , m1 transitions with similar energy have been observed to compete with e1 transitions in the decay out of the sd band in @xmath2pb@xcite .
in addition , in neutron capture experiments on @xmath48dy targets , the m1 strength was found to be comparable to the e1 strength at @xmath49 3 mev @xcite .
the scissor mode @xcite probably accounts for the enhanced m1 strength .
the @xmath1 spectrum of fig . [ fig : doyr ] shows a region with depleted yield between 2.3 and 3.3 mev .
following dssing _
et al . _
@xcite , this depletion can be explained by the reduction in level density in the interval from the nd yrast line up to the energy required to break the first pair of neutrons or protons . in ref .
@xcite it is seen that the width of the depleted region in the @xmath1 spectrum is around 1.5 times the average pairing gap .
the depleted region ( which is most clearly defined by the decay - out transitions with @xmath50 ) occurs between 2.3 and 3.3 mev , implying a pair gap of about 0.7 mev . for the non - rotating nucleus , @xmath51 ( or @xmath52 )
is approximately given by the five - point mass formula @xmath53 for a sequence of isotones ( or isotopes) see , for example , eq .
( 7 ) in ref .
@xcite . around @xmath0hg , @xmath54 mev (
if one neglects mean - field contributions to @xmath53 , which are discussed in ref .
although the information from the tail of the decay - out gamma - spectrum is quite uncertain , it yields a pairing gap similar to that given by @xmath53 . in table [ tab : theory ] , the experimental sd excitation energies are presented given for @xmath55hg .
the excitation energies of the sd levels of @xmath3hg are given by two tentative decay - out pathways @xcite from the 10@xmath21 level , combined with limits imposed by the quasi - continuum analysis @xcite , giving @xmath56 mev .
the sd bands are extrapolated to spin 2.9 and 0 @xmath18 , where the rotational frequencies are zero .
( for @xmath0hg , eq . ( [ khoo2 ] ) gives @xmath57 at @xmath37 = 0 , and @xmath58 . ) table [ tab : theory ] also presents the sd excitation energies from theoretical calculations based on the self - consistent hartree - fock - bogoliubov ( hfb ) approach with the effective skyrme interactions , skp , sly4 and skm@xmath59 , and the density - dependent zero - range interactions of ref .
@xcite for the particle - particle ( pairing ) channel .
the theoretical results were extrapolated in the same way as the experimental values in the case of @xmath0hg , while the theoretical value for the sd state in @xmath3hg was calculated directly for the ground state spin of 0 @xmath18@xcite .
the calculations with the sly4 interaction show the best overall agreement with the experimental data .
the excitation energies of sd states in odd - a and even - even nuclei give information on pairing in the sd well . in even - even nuclei
pairing is stronger so the ground state has lower energy than that of the neighboring odd - even nuclei .
this can be seen by comparing the nd and sd yrast bands of @xmath0hg and @xmath3hg after accounting for the difference in the mass excess of the two nuclei - see fig .
[ fig : mass ] .
the ground state of @xmath3hg is taken as a reference , i.e. set to zero .
one sees that the 0@xmath21 state in @xmath3hg has a more negative mass excess than the 13/2@xmath21 state in @xmath0hg , implying an extra binding of the even - even nucleus by @xmath601.5 mev .
the observed sd states are also shown in fig . [ fig : mass ] . here the energy of the sd `` ground '' state in @xmath3hg is lower than that in @xmath0hg by 0.8 mev . this smaller value ( compared to 1.5 mev for nd states ) , is consistent with reduced pairing in the sd well , as suggested before , e.g. ref .
@xcite , from the increase of the @xmath6 moment of inertia with frequency . however , in addition to pairing , mean - field effects ( e.g. a change in the fermi energy and a polarization energy ) contribute to the binding energy @xcite .
the convergence of the yrast lines of @xmath55hg around spin 10 and 25 @xmath18 , for nd and sd states , respectively , may be attributed to a reduction of pairing due to rotation .
an alternative but equivalent way to present the differences in binding energies in fig .
[ fig : mass ] is in terms of the neutron separation energies @xmath61 in the nd and sd wells .
[ note that @xmath61 = mass excess ( @xmath0hg ) - mass excess(@xmath3hg ) + mass excess(neutron ) ] .
table [ tab : theory ] compares the experimental and theoretical neutron separation energies s@xmath62 in the sd and nd wells .
the s@xmath62 values and sd excitation energies from hfb calculations with the sly4 force are compared with experimental results in figure [ fig:1nsep ] .
the experimental neutron separation energy in the sd well is found to be s@xmath63 mev , compared to s@xmath64 mev in the nd well ( to the 13/2@xmath21 state ) .
the difference of 0.7 - 1.0 mev means that it is easier to remove a neutron from the sd well than from the nd well .
as discussed above , part of the reduction in s@xmath62 in the sd well is due to a decrease in pairing , but other effects contribute as well . in order to gauge the reliability of extracting the pairing gap @xmath65 ( and the difference in the nd and sd wells ) from nuclear masses , we write the equations for 2- , 3- and 5-point mass differences in the form discussed by duguet @xmath45 @xcite . @xmath66
here n is the neutron number , @xmath65 the pairing gap , @xmath67 the polarization energy due to time - reversal symmetry breaking ( from the blocking of a single - particle level ) and @xmath68 the fermi level .
these equations show that , in order to deduce @xmath65 from experimental masses , theoretical values for each well are also required for @xmath67 , as well as for @xmath68 or @xmath69 if @xmath70 or @xmath71 are employed
. it would be best to use @xmath72 , since that requires a calculation of only @xmath73 .
( the value of @xmath73 is around @xmath16100 kev , but there is some uncertainty in its calculation @xcite . )
however , @xmath72 for sd states would need the sd excitation energies in 5 consecutive nuclides , @xmath74hg , and would require new experimental sd energies in @xmath75hg ( work on which is in progress@xcite ) , as well as in @xmath76hg . for the nd _ ground states _ , which have measured masses , @xmath77 yields @xmath78 excellent agreement of 1.1 mev
is obtained using the experimental @xmath79 , together with a theoretical @xmath80 = -8.4 mev ( obtained with the sly4 force ) .
this agreement provides validation of @xmath80 for hg nuclides around @xmath3hg , and is consistent with the reproduction of @xmath81 values ( within 0.2 mev ) for the nd ground states of nuclides @xcite in this region with the sly4 force . for the sd well
, @xmath79 and @xmath82 ( @xmath83 mev ) give @xmath84 where the errors do not include the uncertainty in @xmath82 .
this value of @xmath85 is a direct indicator of pair correlations in the sd well .
there appears to a reduction of this value with respect to that in the nd well , but the uncertainties do not allow for a definitive conclusion .
the spins and excitation energies of the yrast sd band in @xmath0hg have been determined from two single - step linking transitions and from the quasi - continuum spectrum that connects the sd and nd states .
the results from the two methods are in good agreement , within the error bars , providing confidence about the spins and excitation energies of the yrast sd band .
the sd level fed by the 351 kev sd transition has e@xmath86
= 6000 kev and i = 35/2 @xmath18 .
excitation energies and spins provide a stringent test of orbital assignments .
the spin is consistent with that expected for a @xmath20 orbital configuration , previously assigned to this sd band@xcite .
this is the first time that the excitation energies and spins have been determined for a sd band in an odd - even nucleus in the mass a = 190 region . by comparing the results with those of neighboring even - even hg nuclei ,
we have obtained information on pairing in the sd states .
the neutron separation energies in the nd and sd wells have been extracted by using data from @xmath55hg .
the separation energy in the sd well is 0.7 - 1.0 mev smaller than in the nd well , due partly to a reduction in the pair gap @xmath65 with deformation and partly to an change in @xmath68 .
we have compared the results with those from calculations based on hartree - fock - bogoliubov ( hfb ) theory with different skyrme interactions@xcite , and have found that the sly4 interaction , which yields 6.32 mev for the excitation energy of the i = 35/2 @xmath18 sd level , gives the best agreement .
similarly , the same interaction gives the best reproduction of the neutron separation energies in the nd and sd wells .
discussions with i. ragnarsson and a. afanasjev are gratefully acknowledged .
this research is supported in part by the u.s .
dept . of energy under contract nos . w-31 - 109-eng-38 and de - ac03 - 76sf00098 .
s. siem acknowledges a nato grant through the research council of norway .
+ t. l. khoo , m.p .
carpenter , t. lauritsen , d. ackermann , i. ahmad , d.j .
blumenthal , s.m .
fischer , r.v.f .
janssens , d. nisius , e.f .
moore , a. lopez - martens , t. dssing , r. krcken , s.j .
asztalos , j.a .
becker , l. bernstein , r.m .
clark , m.a .
deleplanque , r.m .
diamond , p. fallon , l.p .
farris , f. hannachi , e.a .
henry , a. korichi , i.y .
lee , a.o .
macchiavelli and f.s .
stephens , phys .
lett * 76 * , 1583 ( 1996 ) .
g. hackman , t. l. khoo , m.p .
carpenter , t. lauritsen , a. lopez - martens , i.j .
calderin , r.v.f .
janssens , d. ackermann , i. ahmad , s. agarwala , d.j .
blumenthal , s.m .
fischer , d. nisius , p. reiter , j. young , h. amro , e.f .
moore , f. hannachi , a. korichi , i.y .
lee , a.o .
macchiavelli , t. dssing , and t. nakatsukasa , phys .
lett * 79 * , 4100 ( 1997 ) .
a. lopez - martens , f. hannachi , a. korichi , c. schck , e. gueorguieva , ch .
vieu , b. haas , r. lucas , a. astier , , g. baldsiefen , m. carpenter , g. de france , r. duffait , l. ducroux , y. le coz , h. finck , a. grgen , h. hbel , t.l .
khoo , t. lauritsen , m. meyer , d. prevost , n. redon , c. rigollet , h. savajols , j.f .
sharpey - shafer , o. stezowski , ch .
theisen , u. van severen , j.p .
vivien , and a.n .
wilson , phys .
b380 * , 18 ( 1996 ) .
k. hauschild , l. a. bernstein , j. a. becker , d. e. archer , r. w. bauer , d. p. mcnabb , j.a.cizewski , k .- y .
ding , w. younes , r. krcken , r.m .
diamond , r.m .
clark , p. fallon , i.y .
lee , a.o .
macchiavelli , r. w. macleod , g. j. schmid , m.a .
deleplanque , f.s .
stephens , and w. h. kelly , phys . rev . * c55 * , 2819 ( 1997 ) .
d. p. mcnabb , j.a.cizewski , k .- y .
ding , n. fotiades , d. e. archer , j. a. becker , l. a. bernstein , k. hauschild , w. younes , r.m .
clark , p. fallon , i.y .
lee , a.o .
macchiavelli , and r. w. macleod , phys . rev .
* c56 * , 2474 ( 1997 ) .
t. lauritsen , m.p .
carpenter , t. dssing , p. fallon , b. herskind , r.v.f .
janssens , d.g .
jenkins , t.l .
khoo , f.g .
kondev , a. lopez - martens , a.o .
macchiavelli , d. ward , k.s .
abu saleem , i. ahmad , r. clark , m. cromaz , j.p .
greene , f. hannachi , a.m. heinz , a. korichi , g. lane , c.j .
lister , p. reiter , d. seweryniak , s. siem , r.c .
vondrasek , and i. wiedenhver , phys .
* 88 * , 042501 ( 2002 ) .
t. lauritsen , r.v.f .
janssens , m.p .
carpenter , p. fallon , b. herskind , d.g .
jenkins , t.l .
khoo , f.g .
kondev , a. lopez - martens , a.o .
macchiavelli , d. ward , k.s .
abu saleem , i. ahmad , r. m. clark , m. cromaz , t. dssing , a.m. heinz , a. korichi , g. lane , c.j .
lister , and d. seweryniak , phys .
lett . * 89 * , 282501 ( 2002 ) .
e. f. moore , r.v.f .
janssens , r. r. chasman , i. ahmad , t. l. khoo , f.l.h .
wolfs , d. ye , k. b. beard , u. garg , m. w. drigert , ph .
benet , z. w. grabowski , and j.a.cizewski , phys .
lett . * 63 * , 360 , ( 1989 ) .
r. g. henry , t. lauritsen , t.l .
khoo , i. ahmad , m.p .
carpenter , b. crowell , t. dssing , r.v.f .
janssens , f. hannachi , a. korichi , c. schck , f. azaiez , c.w .
beausang , r. beraud , c. bourgeois , r. m. clark , i. deloncle , j. duprat , b. gall , h. hbel , m. j. joyce , m. kaci , y. lecoz , m. meyer , e.s .
paul , n. perrin , n. poffe , m.g .
porquet , n. redon , h. sergolle , j.f .
sharpey - schafer , j. simpson , a.g .
smith , r. wadsworth , and p. willsau , phys .
lett . * 73 * , 777 ( 1994 ) .
t. lauritsen , t.l .
khoo , i. ahmad , m.p .
carpenter , r.v.f .
janssens , a. korichi , a. lopez - martens , h. amro , s. berger , l. calderin , t. dssing , s.m .
fischer , g. hackman , f. hannachi , c.j .
lister , e.f .
moore , d.t .
nisius , c. schck , and s. siem , phys . rev . *
c62 * , 044316 ( 2000 ) .
s. schwarz , f. ames , g. audi , d. beck , g. bollen , c. de coster , j. dilling , o. engels , r. fossion , j .- e .
garcia ramos , s. henry , f. herfurth , k. heyde , a. kellerbauer , h .- j .
kluge , a. kohl , e. lamour , d. lunney , i. martel , r.b .
moore , m. oinonen , h. raimbault - hartmann , c. scheidenberger , g. sikler , j. szerypo , c. weber , nucl.phys .
* a693 * , 533 ( 2001 ) .
m. p. carpenter , r.v.f .
janssens , b. cederwall , b. crowell , i. ahmad , j. a. becker , m. j. brinkman , m. a. deleplanque , r. m. diamond , p. fallon , l. p. farris , u. garg , d. gassmann , e. a. henry , r. g. henry , j. r. hughes , t. l. khoo , t. lauritsen , i. y. lee , a. o. machiavelli , e.f .
moore , d. nisius , and f. s. stephens , phys . rev .
* c51 * , 2400 ( 1995 ) .
t. lauritsen , ph .
benet , t.l .
khoo , k. b. beard , i. ahmad , m.p .
carpenter , p. j. daly , m. w. drigert , u. garg , p.b .
fernandez , r.v.f .
janssens , e.f .
moore , f.l.h .
wolfs , and d. ye , phys .
lett . * 69 * , 2479 ( 1992 ) .
t. l. khoo , t. lauritsen , i. ahmad , m.p .
carpenter , p.b .
fernandez , r.v.f .
janssens , e.f .
moore , f.l.h .
wolfs , ph .
benet , p. j. daly , k. b. beard , u. garg , d. ye , and m. w. drigert , nucl . phy . *
a557 * , 83c ( 1993 ) .
heenen and r.v.f .
janssens , phys . rev .
* c57 * 159 ( 1998 ) .
p. fallon , p .- h .
heenen , w. satua , r.m .
clark , f.s .
stephens , m.a .
deleplanque , r.m .
diamond , i.y .
lee , a.o .
macchiavelli , and k. vetter , phys . rev . * c60 * , 044301 ( 1999 ) .
t. radon , h. geissel , g. mnzenberg , b. franzke , th .
kerscher , f. nolden , yu . n. novikov , z. patyk , c. scheidenberger , f. attallah , k. beckert , t. beha , f. bosch , h. eickhoff , m. falch , y. fujita , m. hausmann , f. herfurth , h. irnich , h. c. jung , o. klepper , c. kozhuharov , yu .
a. litvinov , k. e. g. lbner , f. nickel , h. reich , w. schwab , b. schlitt , m. steck , k. smmerer , t. winkler , and h. wollnik , nucl . phys . *
a677 * , 75 ( 2000 ) .
.the different feeding and decay components of the spectrum in coincidence with the yrast sd band in @xmath0hg .
the total spin and energy removed , on average , by the different components are @xmath17i and @xmath17e ; @xmath29 and @xmath30 are , respectively , the average energy and spin removed per photon . from monte - carlo simulations , the quasi - continuum decay - out and statistical feeding components
have , respectively , @xmath87i@xmath88 = 0.5 and 0.66 @xmath18 per @xmath1 ray .
the unknown lines are defined as non - yrast discrete transitions . [ cols="<,^,^,^,^,^ " , ] state , which is the termination of the sd band decay , has been set at zero since ( a ) it facilitates the comparison of experimental and theoretical sd excitation energies with respect to this state and ( b ) it circumvents the uncertainty in its energy ( 128@xmath1622 kev , given in ref .
@xcite ) . ]
spectra in @xmath0hg for decays going through the yrast sd band ( upper figure ) and for decays through all ( mostly nd ) states ( lower figure ) .
the spectrum for feeding statistical transitions are from monte carlo simulations ; all other spectra are from experimental data . ] | the excitation energies and spins of the levels in the yrast superdeformed band of @xmath0hg have been determined from two single - step @xmath1 transitions and the quasi - continuum spectrum connecting the superdeformed and normal - deformed states .
the results are compared with those from theoretical mean - field calculations with different interactions .
a discussion of pairing in superdeformed states is also included . 2 |
although quantum gases are very dilute systems , most of their properties are governed by atomic interactions .
this allows to use them , for example , as _ quantum simulators _ to study the many - body physics of systems usually encountered in condensed matter physics @xcite .
however , in all usual quantum gases , the interactions can be described extremely well by a short range , isotropic _ contact _ potential , whose magnitude is proportional to the @xmath3-wave scattering length @xmath4 characterizing low energy collisions .
the dipole - dipole interaction taking place between particles having a permanent electric or magnetic dipole moment has radically different properties : it is long - range and anisotropic , as one readily sees on the expression @xmath5 giving the interaction energy @xmath6 between two polarized dipoles separated by @xmath7 ( @xmath8 is the angle between @xmath7 and the direction along which the dipoles are pointing ) .
these specific properties have attracted a lot of interest recently , and a large number of theoretical predictions have been made concerning dipolar quantum gases ( see e.g. ref . for a review ) : for instance , the stability of a dipolar bec depends crucially on the trap geometry ( see section [ sec : stab ] below ) ; in a quasi two - dimensional trap , the excitation spectrum can display a roton minimum instead of the usual bogoliubov shape ; finally , fascinating new quantum phases ( including supersolids ) are predicted to occur for dipolar bosons in an optical lattice . in practice one
always has a competition between contact and dipolar interactions ; it is therefore useful to define the following ( dimensionless ) ratio of the dipolar and contact coupling constants : @xmath9 the numerical factors are chosen in such a way that a _ homogeneous _ dipolar condensate is unstable against collapse for @xmath10 . for usual atomic magnetic moments @xmath11 ( e.g. for the alkalis ) , @xmath12 is very small ( typically a few @xmath13 ) and dipolar effects are extremely small . here
, we report on experiments with @xmath0cr , which has @xmath14 due to its large magnetic moment @xmath15 , and which also allows , _ via _ feshbach tuning of the scattering length @xmath4 , to even enhance @xmath12 .
this paper is organized as follows .
we first describe briefly in section [ sec : fr ] our experimental setup , with an emphasis on how we use a feshbach resonance in order to enhance dipolar effects and create a ` quantum ferrofluid ' .
section [ sec : stab ] is devoted to the study of the geometry dependence of the stability of a dipolar bec .
finally , we describe in section [ sec : collapse ] the dynamics following the collapse of an unstable dipolar condensate .
a bec of @xmath0cr containing about 50,000 atoms was obtained in 2005 by evaporative cooling of optically trapped chromium atoms @xcite . shortly after the achievement of condensation , a first effect of the dipole - dipole interaction could be observed in time - of flight experiments @xcite : the dipole - dipole interaction tends to _ elongate _ the bec along the magnetization direction . however , due to the small value of @xmath14 , the dipolar interaction was , in this experiment , only a small perturbation of the contact interaction , which essentially governed the expansion dynamics . the existence of several feshbach resonances @xcite in @xmath0cr opens the possibility to tune the scattering length @xmath4 using an external magnetic field @xmath16 , according to @xmath17 here , @xmath18 is the @xmath3-wave scattering length , @xmath19 the resonance position , and @xmath20 the resonance width . the broadest feshbach resonance in @xmath0cr is located at @xmath21 g and has a width @xmath20 of only @xmath22 g. this implies that the field control at the level of @xmath23 r.m.s . that we implemented allows us to tune @xmath4 close to 0 with a resolution of about one bohr radius .
this ` knob ' allowing one to change @xmath4 allowed us to perform time of flight experiments for two different orientations of the dipoles with respect to the trap axes @xcite , as in ref . , but now for increasing values of @xmath12 .
experimental results are shown in fig .
one clearly sees the dramatic effect of an increase of @xmath12 on the expansion dynamics .
in particular , for @xmath24 , the inversion of ellipticity of the condensate during time of flight ( the usual ` smoking - gun ' evidence for bec ) is inhibited by the strong dipole - dipole interaction .
the solid lines in fig .
[ fig1 ] are theoretical predictions ( without any adjustable parameters ) based on the gross - pitaevskii equation ( gpe ) generalized to take into account the non - local dipole - dipole interaction in the description of the macroscopic wavefunction @xmath25 of the bec : @xmath26 here @xmath27 is the contact interaction coupling constant . in this set of experiments ,
the trap geometry was not very far from spherical , which limited the study of dipolar condensates to values @xmath28 . to go beyond this value and reach the purely dipolar regime @xmath29
, we shall now see that one needs to tailor the confining potential , so that the attractive part of the dipole - dipole interaction does not destabilize the condensate .
it is well known that a bec with attractive contact interactions is unstable against long - wavelength fluctuations ( this _ phonon instability _ leads to a collapse of the bec having @xmath30 ) . as the dipole - dipole interaction has an attractive part for dipoles in a ` head - to - tail ' configuration
see equation ( [ eq : udd ] ) for @xmath31 , it is intuitively clear that in a prolate trap with the dipoles pointing along the weak direction of the trap [ see fig .
[ fig2]*(a ) * ] , the net effect of the dipolar interaction is attractive .
thus , in this configuration , one expects that when @xmath4 is reduced , the condensate becomes unstable , at a critical value @xmath32 which should be positive ( the small repulsive contact interaction being unable , at this point , to counteract the dipolar attraction ) .
conversely , in an oblate trap with the dipoles pointing along the strong confinement direction , the critical scattering length should be negative , and a _ purely dipolar quantum gas _ can be stabilized . in ref .
, this geometry - dependent stability of a dipolar condensate was studied experimentally .
a long period ( @xmath33 @xmath11 m ) optical lattice , obtained by interfering two laser beams at 1064 nm under a small angle of @xmath34 , was superimposed onto the optical dipole trap , allowing us to realize traps with cylindrical symmetry around the @xmath35-axis ( polarization direction ) and having an aspect ratio @xmath36 that could be varied over two orders of magnitude ( from @xmath37 to @xmath38 ) while keeping the average trapping frequency @xmath39 constant .
the experiment consists in creating a bec in a trap with a given aspect ratio @xmath40 , then ramping @xmath4 to a final value @xmath41 , and finally measuring the atom number @xmath42 in the condensate .
one observes that when @xmath41 is decreased below a critical value @xmath32 , @xmath42 suddenly drops to zero .
we stress that for all the traps we used , the condensate density was roughly the same .
figure [ fig2]*(b ) * shows the measured @xmath32 as a function of @xmath40 and clearly displays the expected behavior : for small @xmath40 ( prolate traps ) , @xmath32 is positive , and starts to decrease when the trap becomes more oblate . for @xmath43 ,
one has @xmath44 , meaning that a purely dipolar quantum gas ( @xmath45 ) can be stabilized by an appropriate trap geometry .
a simple way to go beyond the qualitative picture above and obtain an estimate for the instability threshold @xmath46 is to use a variational method .
inserting a gaussian ansatz ( with the axial and radial sizes @xmath47 and @xmath48 as variational parameters ) into the gross - pitaevskii energy functional whose minimization gives the gpe ( [ eq : gpe ] ) , one obtains the following energy to minimize : @xmath49 , \label{eq : ansatz}\end{aligned}\ ] ] where @xmath50 .
the first two terms are the kinetic and potential energies , while the third arises from contact and dipolar interactions .
the function @xmath51 is monotonically decreasing from 1 to @xmath52 as a result of the anisotropy of the dipolar interaction . for a given @xmath40
, one can find a ( possibly local ) minimum of @xmath53 at finite values of @xmath54 if and only if @xmath4 is larger than a critical value : this defines the stability threshold @xmath46 within this model .
the solid line in fig .
[ fig2]*(b ) * is the result obtained with this simple procedure , for the experimental parameters @xmath55 hz and @xmath56 .
one obtains a relatively good agreement with experimental data .
a numerical solution of the gpe ( [ eq : gpe ] ) gives even better agreement with measurements @xcite .
equation ( [ eq : ansatz ] ) allows one to understand easily , in the @xmath57 limit , the behavior of @xmath46 for @xmath58 and @xmath59 . indeed , in this limit
, it is the sign of the interaction term which determines the stability ; therefore one has @xmath60 where @xmath61 is the length defined in such a way that @xmath62 ( for @xmath0cr , one readily calculates , with help of ( [ eq : edd ] ) , that @xmath63 ) .
it is apparent on fig .
[ fig2]*(b ) * that for @xmath64 , the results are already close to the @xmath57 limit ( [ eq : limits ] ) .
it is natural to ask what happens if one drives the condensate into the unstable regime , e.g. by decreasing the scattering length below @xmath32 . in the case of pure contact interactions , a collapse of the condensate , followed by an explosion of a ` remnant ' bec ( _ bose - nova _ ) , has been observed in several systems @xcite .
more recently , the formation of soliton trains has also been reported @xcite .
we have studied the collapse dynamics of a dipolar condensate @xcite ( in a roughly spherical trap ) by ramping down rapidly the scattering length to a final value of @xmath65 , then waiting an adjustable holding time , and performing a time of flight of 8 ms before imaging the cloud .
figure [ fig3]*(a ) * presents the evolution of the condensate when the holding time is varied .
one observes that the cloud , initially elongated along the magnetization direction @xmath35 ( horizontal axis on the figure ) acquires rapidly a complicated structure with a four - fold symmetry , corresponding to a density distribution having a torus - like component close to the plane @xmath66 and two ` blobs ' close to the @xmath35-axis .
interestingly , this angular symmetry of the cloud is very close to the one of a @xmath2-wave @xmath67 , i.e. precisely the symmetry of the dipole - dipole interaction ( [ eq : udd ] ) . during the same time period
, the atom number in the condensate strongly decreases due to the three - body losses occurring because of the high densities transiently reached during the collapse .
the group of m. ueda in tokyo performed a three dimensional numerical simulation of the gpe ( [ eq : gpe ] ) , in which all input parameters were given their experimentally measured value @xcite .
three - body losses were accounted for by adding the imaginary term @xmath68 to eqn .
( [ eq : gpe ] ) , where @xmath69 m@xmath70s is the measured three - body loss coefficient . figure [ fig3]*(b ) * represents the results of the simulation .
the agreement is excellent , all the more if one keeps in mind that no adjustable parameter is introduced .
let s mention that the inclusion of a small delay ( also measured independently ) in the time variation @xmath71 of the scattering length , due to eddy currents in the vacuum chamber , had to be included to achieve a quantitative agreement !
the simulation also reproduces quantitatively the time dependence of the condensate atom number .
a fascinating prediction of the numerical simulation is the spontaneous formation , during the collapse , of two quantized vortex rings with opposite circulation ( and charge @xmath72 ) , as a result of the strongly anisotropic collapse : the collapse in the radial directions is fast and quickly followed by an outward flow , while axially the flow is still inward , thus giving rise to the circulation .
detecting experimentally the presence of vortex rings is very challenging , but might be done by using interferometric techniques ( e.g. matter wave heterodyning ) to reveal the winding of the phase of the bec wavefunction around the topological defects .
the results presented in this paper are the first dramatic manifestations of dipolar effects in quantum gases , and pave the way for future studies involving even more strongly interacting dipolar systems , especially the ones that may be obtained using the permanent electric dipole moments of heteronuclear molecules in their ground state . due to the large value of such dipole moments ( on the order of one debye )
, the long - range character of the dipolar interaction could then be used to achieve novel quantum phases in optical lattices @xcite , as well as to implement promising quantum information processing schemes @xcite .
however , already in the case of the comparatively weaker magnetic dipoles , extremely interesting theoretical proposals deserve experimental study ; to mention only one example , the generation of two - dimensional solitons @xcite ( whose stability arises from the long - range character of the dipole - dipole interaction ) is a very appealing experiment .
we thank h. saito , y. kawaguchi and m. ueda for their collaboration on ref . .
we acknowledge support by the german science foundation ( sfb / trr 21 , spp 1116 ) , the landesstiftung baden - wrttenberg and the eu ( marie - curie grant meif - ct-2006 - 038959 to t. l. ) . | we report on experiments exploring the physics of dipolar quantum gases using a @xmath0cr bose - einstein condensate ( bec ) . by means of a feshbach resonance ,
it is possible to reduce the effects of short range interactions and reach a regime where the physics is governed by the long - range , anisotropic dipole - dipole interaction between the large ( @xmath1 ) magnetic moments of chromium atoms .
several dramatic effects of the dipolar interaction are observed : the usual inversion of ellipticity of the condensate during time - of flight is inhibited , the stability of the dipolar gas depends strongly on the trap geometry , and the explosion following the collapse of an unstable dipolar condensate displays @xmath2-wave like features . |
robustness is a key feature in the analysis of complex systems , especially for complex biological systems .
many organisms have strong adaptability to environmental changes or failures in some of their components , and can live even if some of their genes are mutated . in particular , it is known that cancer cells are very robust @xcite .
therefore , understanding the origin of robustness of living cells has become an important research topic .
in particular , extensive studies have focused on the analysis of structural robustness of metabolic networks .
structural robustness refers to the tolerance of the system s behavior to changes in the structure of networks , and most existing studies focus on changes caused by knockout of gene(s ) or enzyme(s ) .
one of the reasons why extensive studies have been done on structural robustness of metabolic networks is that rather accurate data of metabolic networks are available via such databases as the kyoto encyclopedia of genes and genomes ( kegg ) @xcite and the encyclopedia of _ escherichia coli _ k-12 genes and metabolism ( ecocyc ) @xcite , and kinetic parameters , which are not necessarily available , are not required . in order to analyze the structural robustness of metabolic networks , the _ flux balance analysis _ ( fba ) methods have been widely used . in many of these approaches ,
_ elementary flux modes _ ( efms ) play a key role , where an efm is a minimal set of reactions that can operate at steady state @xcite . based on fba and/or efm ,
several studies have focused on finding a minimum reaction cut @xcite , that is , a minimum set of reaction ( or enzyme ) removals which prevent the production of a specified set of compounds .
other fba - based measures of robustness have also been proposed .
behre et al .
proposed a measure based on the number of remaining efms after knockout versus the number of efms in the unperturbed situation @xcite .
deutscher et al . proposed another measure using the shaply value from game theory @xcite .
other approaches have been proposed based on boolean models of metabolic networks in which reactions and compounds are modeled as and and or nodes , respectively .
handorf et al .
analyzed robustness of metabolic networks by introducing the concept of _ scope _ @xcite .
li et al . , sridhar et al . , and tamura et al .
developed integer programming - based methods for finding a minimum reaction cut in boolean models of metabolic networks @xcite .
smart et al . defined the _ topological flux balance _ ( tfb ) criterion based on a boolean model of metabolic networks and analyzed the _ damage _ ( number of reactions ) caused by knockout of a single reaction under tfb @xcite .
jiang et al .
defined and analyzed the _ impact degree _ , which is the number of reactions inactivated by knockout of a specific reaction @xcite . although there are some differences in the treatment of reversible reactions , the damage and the impact degree are very similar concepts .
cong et al .
extended the impact degree for knockout of multiple reactions @xcite . in this paper , we study the distribution of the impact degree caused by random knockout of a single reaction using the theory of branching process @xcite . in order to analyze earthquakes , saichev et al .
proposed a branching process with power - law distributions of offspring @xmath0 : @xmath1 , where @xmath2 is some constant , and approximately derived the distribution of the total number of offsprings @xcite .
we regard propagation of the impact of knockout of a reaction as a branching process , and apply their method to estimate the impact degree distribution , where the impact degree in our problem corresponds to the total number of offsprings in the branching process . in order to apply this method
, we develop a simple method for estimating the offspring distribution in a metabolic network .
although smart et al .
have already applied percolation theory and branching process to analysis of the size distribution of rigid clusters , defined as clusters of contagion nodes that do not contain any branched metabolite nodes ( see fig .
3a in @xcite ) , they did not explicitly estimate the damage distribution ( i.e. , the impact degree distribution ) .
we finally show an estimation method for the damage distributions .
the proposed method is applied to analysis of metabolic networks of four species : _ escherichia coli _ , _ bacillus subtilis _ , _ saccharomyces cerevisiae _ and _ homo sapiens_. the results show good agreement of impact degree distributions between empirical results and theoretical estimates .
jiang et al . proposed the _ impact degree _ as a measure of the importance of each reaction in a metabolic network @xcite .
the impact degree is defined as the number of inactivated reactions caused by knockout of a single reaction .
however , they did not consider the effect of cycles in metabolic networks . since cycles play an important role in metabolic networks , cong et al .
extended the impact degree so that the effect of cycles is taken into account @xcite by using a concept of the maximal valid assignment @xcite . here , we briefly review the definition of this extended impact degree @xcite . as in other works , we regard each metabolic network as a bipartite directed graph .
let @xmath3 and @xmath4 be a set of _ compound nodes _ and a set of _ reaction nodes _ respectively , where @xmath5 .
metabolic network _ is defined as a directed graph @xmath6 in which either @xmath7 or @xmath8 holds for each edge @xmath9 . means logical and of @xmath10 and @xmath11 . ]
each reaction and compound takes one of two states : 0 or 1 , where 0 and 1 correspond to inactive and active reactions ( compounds ) , respectively .
reverse reactions are treated as two irreversible reactions . in order to define the impact degree of a reaction
, we proceed as follows .
suppose that reaction @xmath12 is knocked out .
then , we start with the global state where all compounds are active ( i.e. , @xmath13 for all @xmath14 ) and all reactions but @xmath12 are active ( i.e. , @xmath15 for all @xmath16 and @xmath17 )
. then , we alternatively update the states of reactions and compounds by the following rules . 1 . for each reaction , there are three different compounds : consumed compounds ( i.e. , substrates ) , produced compounds ( i.e. , products ) , and directly unrelated compounds .
a reaction is inactivated if any of its consumed or produced compound is inactivated .
3 . for each compound
, there are three different reactions : consuming reactions , producing reactions , and directly unrelated reaction .
a compound is inactivated if all its consuming reactions or all its producing reactions are inactivated .
since no activation is possible in this process , the procedure converges to a stable state in a finite number of iterations .
the impact degree of reaction @xmath12 is defined as the number of inactivated reactions in the stable state .
this procedure simultaneously gives the definition of the impact degree and an algorithm to compute it .
let us illustrate the above process with the metabolic network shown in fig .
[ exnet ] .
suppose that reaction @xmath18 is knocked out .
then , the states of nodes change as shown in table [ exknock1 ] .
since four reactions ( including @xmath18 ) are inactivated in the stable state , the resulting impact degree is four .
next , suppose that reaction @xmath19 is knocked out . in this case , the states of nodes change as shown in table [ exknock2 ] and the resulting impact degree is two .
it is to be noted that @xmath20 is not inactivated because @xmath18 is still active in this case .
.impact degree calculation when @xmath18 in fig .
[ exnet ] is knocked out . [
cols="^,^,^,^,^,^,^,^,^,^,^,^,^,^",options="header " , ]
we proposed a model to estimate the impact degree distributions in metabolic networks , using a branching process approximation , and demonstrated its validity on real data .
the power - law model could more accurately estimate the impact degree distributions in real metabolic networks than the poisson model because the number of offsprings for each reaction node is assumed to follow the power - law distribution .
especially , the power - law model showed the significant agreements between the predicted distribution and the observed distribution although the case of _ h. sapiens _ represented the small @xmath21-value for the ks test ( i.e. , the low probability that the distribution is similar between models and observed data ) .
however , there is no great difference of the prediction accuracy for estimating the impact degree distributions between the power - law model and the poisson model ; thus , the poisson model may be useful for a rough estimate of the impact degree distributions .
intrinsically , the distribution of the total number of offsprings [ i.e. , @xmath22 is not significantly different between the power - law model and the poisson model in the case of smaller @xmath23 . as a simple example
, we here consider the case of @xmath24 ( i.e. , the critical case ) . in this case , the impact degree distributions @xmath25 of the power - law model and the poisson model correspond to @xmath26 and @xmath27 , respectively .
especially , @xmath25 of the power - law model is the power - law distribution with the exponent ranging between @xmath28 and @xmath29 because of @xmath30 ; thus it is not critically different from @xmath31 of the possion model for smaller @xmath23 .
since the impact degrees of real metabolic networks ( i.e. , @xmath23 ) were relatively small ( @xmath32 ) , there might be no great difference of the prediction accuracy between these models .
the similarity of predicted distributions of the total number of offsprings between the poisson model and the power - law model is also explained using the otter s theorem [ theorem 13.1 in @xcite ] .
this theorem indicates that the distribution @xmath25 of the total number of offsprings has the universal property of a power - law tail with the exponent @xmath33 as @xmath34 under mild conditions on offspring distributions , and it implies that the types of offspring distributions hardly influence the distribution of the total number of offsprings ( i.e. , impact degree distributions ) .
note the otter s theorem does not contradict with the analytical distribution @xmath25 of the power - law model because this theorem is not directly applicable to the power - law model due to the different assumptions in the derivation of @xmath25 between the power - law model and the otter s theorem .
the prediction performance is influenced by the assumption of offspring distributions and the fidelity of branching processes .
to purely evaluate the validity of branching process approximation , the empirical model is useful because of no assumption of offspring distributions .
it is expected that the empirical model show the best prediction accuracy because of using empirical offspring distributions . in the case of bacteria ( i.e. , _
e. coli _ and _ b. subtilis _ ) , this expectation is true , suggesting that the branching process approximation is useful for estimating the impact degree distributions . in the case of eukaryotes ( i.e. , _ s. cerevisiae _ and _ h. sapiens _ ) , on the other hand , we observed the unexpected results : the empirical model shows the relatively - low prediction accuracy . especially , the prediction accuracy of the empirical model is lowest in the case of _ h. sapiens _
( human ) .
this result implies limitations to the estimation of impact degree distributions based on the branching process approximation in the case of metabolic networks of eukaryotes ( i.e. , higher organisms ) .
a limitation of the branching process approximation is that we need to assume tree structures of networks ( i.e. , no cycles ) .
the presence of cycles may lead to an overestimation of the number of offsprings @xmath35 for each reaction , because some offsprings of a progenitor ( reaction node ) may have already been inactivated due to cycle structures .
from this reason , the models may overestimate the impact degree distributions . on the other hand , however , some reactions with more than one incoming edge may be inactivated in the presence of cycles , if all their parents are inactivated through different paths . in this case , the number of offsprings is underestimated .
the estimation of the number of offsprings depends on the relative importance of these two effects , and it may be not simple .
the difficulty in this estimation is also a limitation of the model .
the branching process model needs to be improved by considering additional assumptions other than metabolic network structures in order to obtain better predictions .
for example , the assumption of variable propagations in the branching process @xcite , in which the mean of offspring distributions differs at each propagation stage , may be useful because the degree of propagation may depend on metabolic dynamics such as gene expressions and metabolite concentrations . to apply this modified branching process model , time - series data on metabolic dynamics after gene disruptions , which are obtained by metabolomic analysis , are necessary for estimating the mean of offspring distribution at each propagation stage ( i.e. , time after gene disruptions ) .
since such data are unavailable at present , however , it is difficult to apply this modified branching process model .
similarly , other biologically - suitable assumptions are hardly determined because of few observed data on metabolic dynamics .
although there are above constraints on observed data on metabolic dynamics , we believe that the consideration of the additional information improves the prediction of impact degree distributions . in the future , the improvement of the prediction of the impact degree distributions using the branching process approximation may be possible with available data on metabolic dynamics .
the branching process approximation is useful for estimating the impact degree distributions in metabolic networks although it has the above limitations ; thus , it may be a powerful tool for evaluating a robustness of biological systems .
this work was partially supported by the jst presto program .
ta , tt and jpv were partially supported by a jsps / inserm grant .
ta was partially supported by grant - in - aid # 22650045 from mext , japan .
t. tamura , k. takemoto , t. akutsu , finding minimum reaction cuts of metabolic networks under a boolean model using integer programming and feedback vertex sets , international journal of knowledge discovery in bioinformatics 1 ( 2009 ) 1431 .
y. cong , t. tamura , t. akutsu , w - k .
ching , efficient computation of impact degrees for multiple reactions in metabolic networks with cycles , in : proc .
acm third international workshop on data and text mining in bioinformatics ( 2009 ) 67 .
i. dobson , b. a. carreras , number and propagation of line outages in cascading events in electric power transmission systems , in : proc .
48th annual allerton conference on communication , control & computing ( 2010 ) . | theoretical frameworks to estimate the tolerance of metabolic networks to various failures are important to evaluate the robustness of biological complex systems in systems biology . in this paper , we focus on a measure for robustness in metabolic networks , namely , the _ impact degree _ , and propose an approximation method to predict the probability distribution of impact degrees from metabolic network structures using the theory of branching process .
we demonstrate the relevance of this method by testing it on real - world metabolic networks .
although the approximation method possesses a few limitations , it may be a powerful tool for evaluating metabolic robustness .
metabolic network , branching process , power law , cascading failure 89.75.hc , 89.75.da |
systems with asymmetric transport characteristics are also known as ratchets and brownian motors since they can be used to generate dc currents from an ac or noisy voltage .
this mechanism has important applications in physics and biology @xcite . as specific examples , we mention diodes and photovoltaic current rectifiers . here
we address the possibility of realizing such effects in luttinger liquids , which is of interest in connection with nanostructured devices .
motivated by recent experiments on quantum dots @xcite , carbon nanotubes @xcite , and quantum hall systems @xcite , we examine ratchet effects caused by two unequal constrictions as point scatterers .
we consider a one - dimensional spinless electron liquid at zero temperature subject to an impurity potential @xmath0 and a pair interaction @xmath1 with the hamiltonian @xmath2 the impurity potential @xmath0 is chosen to describe two unequal point scatterers at a distance @xmath3 . in the absence of interactions , the application of a voltage @xmath4 leads to a current @xmath5 with a finite conductance @xmath6 .
the conductance is proportional to @xmath7 with the transmission probability @xmath8 for electrons with incident energy @xmath9 in the vicinity of the fermi energy @xmath10 . as a consequence of time reversal symmetry , @xmath8
does not depend on the direction of the incoming momentum and therefore noninteracting electrons have a symmetric transport characteristic with an odd function @xmath11 .
therefore , the inclusion of interactions is mandatory for the analysis of ratchet effects .
for the inclusion of interactions , the bosonized representation of the model is particularly convenient . the bosonization technique @xcite maps the quantum dynamics of the electron liquid onto a path integral for a bosonic field @xmath12 which essentially describes collective displacements of the electron liquid . in terms of this field , the particle density of the electrons reads @xmath13 } + \rm{h.c . }
\right ) \ .
\label{rho}\end{aligned}\ ] ] here , @xmath14 is the fermi wave vector and @xmath15 is a microscopic cutoff length scale of the order of @xmath16 .
the particle current is given by @xmath17 , and the charge current by @xmath18 . in terms of the field
@xmath12 , the relevant contributions to the action corresponding to the hamiltonian ( [ h ] ) read @xmath19 } + { \rm h.c . } ] \nonumber \\ & & - \frac{w_0}{\pi^{3/2 } \alpha } \partial_x \vartheta ( x ) [ e^{i [ 2 k_f x + 2 \sqrt \pi \vartheta(x ) ] } + { \rm h.c . } ] \bigg\ } \ . \label{s}\end{aligned}\ ] ] forward scattering by the interaction is included in the `` free '' bosonic theory via the luttinger parameter @xmath20^{-1/2}$ ] which is less than unity for repulsive interaction .
assuming that @xmath1 is short ranged due to screening effects , only its weight @xmath21 is effective .
the second and third lines of eq .
( [ s ] ) describe backscattering off the impurity potential and by the interaction , respectively . for a single point - like scatterer , @xmath22 , it was shown @xcite that repulsive electron interactions ( @xmath23 ) lead to a vanishing linear conductance .
a weak scatterer was found to suppress the conductance according to @xmath24 where @xmath25 is conductance change due to the presence of the scatterer . on the other hand , for a strong scatterer
, the conductance vanishes like @xmath26 with the amplitude @xmath27 for hopping across the scatterer .
we subsequently focus on the weak - barrier case , deferring the strong currugation limit to ref . @xcite .
the suppression of the conductance in the presence of interactions can be attributed to the backscattering of electrons from a hartree - type potential caused by friedel oscillation in the vicinity of the impurity . in the framework of the bosonic action ( [ s ] )
, this backscattering arises from the second - line contribution , and the third - line contribution is irrelevant for the asymptotics ( [ dg ] ) . to illustrate the mechanism at work ,
it is instructive to briefly recall the scattering features of single electrons by a double barrier @xmath28 . we assume the amplitudes @xmath29 and @xmath30 to be positive .
@xmath3 is the distance between the two barriers . comparing the probability density @xmath31 for a particle incident from @xmath32 with wave vector @xmath33 to the density @xmath34 for a particle incident from @xmath35 with wave vector @xmath36
, one observes that the densities are not only different but also _ not _ related by reflection symmetry , @xmath37 ) . in this term , the hartree approximation amounts to replacing @xmath38 which is proportional to the slowly varying part of the density , cf .
( [ rho ] ) by its average value @xmath39 .
this average is readily calculated from the action perturbatively to second order in @xmath40 .
then , the third - line contribution can be absorbed into the second - line contribution by replacing the bare potential @xmath0 with the effective potential @xmath41 including the correction @xmath42 .
in analogy to the analysis @xcite yielding eq .
( [ dg ] ) for a single scatterer , one then can calculate the corrections to conductance to second order in this effective potential . to estimate the strength of this effect , it is important to notice that , to leading order , this correction is proportional to the backscattering current off the impurities , i.e. , one expects @xmath43 corresponding to eq .
( [ dg ] ) .
the reinsertion of this correction into eq .
( [ dg ] ) suggests subleading ratchet corrections to the conductance of order @xmath44 .
performing the systematic calculation outline above , we obtain an asymmetric ratchet contribution to the conductance @xmath45 \label{res}\end{aligned}\ ] ] with the function @xmath46 invoking the bessel function @xmath47 .
equation ( [ res ] ) was obtained within the hartree approximation .
it becomes exact in a model of many bands @xmath48 interacting through the coupling @xmath49 .
however , the order - of - magnitude estimate of the current is valid in a more general case including our one - band model ( [ s ] ) .
this can be verified by analyzing the expression for the current in the forth order of the perturbation theory .
equation ( [ res ] ) is our main result .
its proportionality to @xmath50 reflects the fact that the ratchet effect vanishes in the absence of interactions .
it is valid provided the backscattering current can be obtained within perturbation theory .
this is the case for weak scattering , more specifically for @xmath51 within this limit , one can distinguish two regmies . * at lower voltages @xmath52 , @xmath53 and the effect is in agreement with the above estimate .
additional oscillating factors reflect resonances due to quantum interferences @xcite .
the absolute value of the ratchet current grows with decreasing voltage for @xmath54 .
* in the high - voltage regime @xmath55 , @xmath56
. then @xmath57 . in this regime
, the ratchet current grows with decreasing votalge for @xmath58 .
this result shows that the ratchet effect can be increasing with decreasing voltage . at low voltages beyond the point where the condition ( [ valid ] ) breaks down and the total current vanishes with decreasing voltage ,
also the ratchet current has to vanish .
nevertheless , it can give a sizeable contribution to the total current .
this explains the pronounced asymmetry observed experimentally in corrugated nanotubes @xcite .
although the hamiltonian ( [ h ] ) does not describe quantum hall systems , the bosonized form ( [ s ] ) captures tunneling of electron or quasiparticles between edges @xcite . in a fractional quantum hall state with filling factor
@xmath59 , the tunneling of quasiparticles corresponds to the case @xmath60 .
thus , the interesting regime with small @xmath61 is physically accessible . | we investigate a one - dimensional electron liquid with two point scatterers of different strength . in the presence of electron interactions ,
the nonlinear conductance is shown to depend on the current direction .
the resulting asymmetry of the transport characteristic gives rise to a ratchet effect , i.e. , the rectification of a dc current for an applied ac voltage . in the case of strong repulsive interactions , the ratchet current grows in a wide voltage interval with decreasing ac voltage . in the regime of weak
interaction the current - voltage curve exhibits oscillatory behavior .
our results apply to single - band quantum wires and to tunneling between quantum hall edges
. ratchets , luttinger liquids , conductance , current rectification , bosonization , impurity scattering , hartree approximation |
one of the most attractive properties of bilayer graphene ( which is made of two coupled atomic layers of carbon ) is the ability to induce a tunable spectral gap @xmath2 by applying a perpendicular electric field .
@xcite rapid advances in patterning graphene on the nanoscale now make it feasible to fabricate graphene ribbons with well defined edge termination , @xcite while experiments on narrow ribbons show that they can display a gap @xmath3 due to transverse size quantization . @xcite the ensuing facility to confine electrons in a controllable way via gate potentials and patterning makes bilayer nanoribbons a promising candidate for nanoelectronic applications .
systems suitable of being used as a basis for a transistor should exhibit a high resistance in the off state . in the case of nanoribbons with zigzag termination
, however , this requirement poses a problem , since the edges support current - carrying states with energies inside the bulk gap , and a much reduced hybridization gap @xmath4 due to exponentially small tunneling between the edges .@xcite on the other hand , several novel concepts ( notably , valleytronics @xcite ) exploit the existence and specific properties of such edge states .
( color online ) upper panel : sketch of a bilayer graphene nanoribbon of lattice constant @xmath5 and width @xmath6 ( inset : definition of hopping terms ) .
middle panel : dispersion relation of edge states for a nanoribbon with @xmath7 and layer - symmetry breaking potential @xmath8 , focussing on one of the two k points ( @xmath9 ) .
( dashed lines : two innermost bands for @xmath10 . ) here we use hopping matrix elements @xmath11 , but set @xmath12 .
the inset shows the transverse charge distribution of two edge states with energies indicated by the markers .
lower panel : the same but including hopping terms @xmath13 , @xmath14 .
, title="fig : " ] ( color online ) upper panel : sketch of a bilayer graphene nanoribbon of lattice constant @xmath5 and width @xmath6 ( inset : definition of hopping terms ) .
middle panel : dispersion relation of edge states for a nanoribbon with @xmath7 and layer - symmetry breaking potential @xmath8 , focussing on one of the two k points ( @xmath9 ) .
( dashed lines : two innermost bands for @xmath10 . ) here we use hopping matrix elements @xmath11 , but set @xmath12 .
the inset shows the transverse charge distribution of two edge states with energies indicated by the markers .
lower panel : the same but including hopping terms @xmath13 , @xmath14 .
, title="fig : " ] previous works on edge states in gated bilayer ribbons ignore two elements : ( i ) next - nearest - neighbor hopping , which is known to strongly affect the bulk properties @xcite and is of crucial importance for edge states in monolayer ribbons , @xcite and ( ii ) the effects of interactions , which are known to strongly influence the properties of quasi - one dimensional systems , including carbon - based systems .@xcite in this paper , we point out that next - nearest - neighbor hopping breaks the symmetry between the edges , resulting into a slow edge and a fast edge characterized by different values of the propagation velocity .
we then explore the consequences for the question whether many - body effects can help to open a gap in these channels . treating interactions in a hubbard model on the basis of bosonization and renormalization group methods at @xmath15
, we find that the edge channels display a mott transition at half filling . in presence of the higher - order hopping terms
, the associated charge gap @xmath16 on the slow edge can take on sizeable values , while on the fast edge the gap is negligibly small .
remarkably , the gap depends sensitively on the applied gate voltage @xmath1 ; the effects of interactions can therefore be controlled externally . furthermore , the dependence of @xmath16 on @xmath1 is non - monotonic , in striking contrast to the behavior of the field - induced gap @xmath2 in the bulk of the system .
this paper is organized as follows . in sec .
ii we present the hamiltonian for the graphene bilayer ribbon including the next - nearest - neighbor hopping and the hopping term between non - dimerised atoms in both layers .
we then describe the edge states for a zigzag ribbon of bilayer graphene and their different localization properties along the transverse direction of the ribbon by computing the inverse participation ratio . in sec .
iii we describe the effects of considering electron - electron interactions in the spectrum for edge states .
we find that because the different properties of the two edges the gap opened by the interaction is different for both edges .
we also briefly discuss a possible experimental verification of this different behavior .
we summarize our results in sec .
a graphene bilayer nanoribbons with zigzag edge termination is shown in fig.[fig1 ] ( left panel ) .
focussing on the low - energy bands which participate in electronic transport , such ribbons can be modelled by a tight - binding hamiltonian @xmath17 , where each carbon atom hosts one spin - degenerate electronic orbital with annihilation operator @xmath18 .
recent experiments specifically addressed the values of the parameters @xmath19 in bilayer graphene .
@xcite nearest - neighbor hopping in the same graphene layer is described by a hopping element @xmath20 , where @xmath21 .
the layers are bernal stacked , with hopping element @xmath22 between dimerised atoms that lie on top of each other .
the symmetry between both layers can be broken by top- or back - gates with a gate voltage @xmath0 , which induce a perpendicular electric field and give rise to an on - site potential @xmath23 on the two layers , which is obtained from @xmath0 by including the screening in the layers .
@xcite we assume that @xmath1 is uniform across the system , even near the boundaries , which in reality will be enforced to a good approximation by the proximity of the metallic gate close to the system . at the boundary ,
we impose standard hard - wall boundary conditions , which amount to setting the wavefunction to zero on lattice sites that lie outside the ribbon .
it is also conventional to include the direct coupling @xmath24 between the non - dimerised atoms in both layers .
@xcite ignoring for the moment other hopping terms , a typical dispersion relation of the edge states in a gated ribbon is shown in fig .
[ fig1 ] ( middle panel ) . of the four bands of edge states , two are massive , with a gap @xmath25 ( the same as for the bulk states ) , while the other two display a linear dispersion , corresponding to two counter - propagating states which are well localized at opposite edges ( see inset ) .
the same scenario is replicated at the other k point in the graphene brillouin zone , but with the edges interchanged .
therefore , each edge supports two counter propagating state , one from each k point .
notably , to this level of approximation , the propagation velocities on both edges are the same .
the middle and lower panels of fig .
[ fig1 ] shows how the dispersion changes when the two most prominent additional hopping elements are taken into account : the interlayer hopping @xmath26 between adjacent dimerized and non - dimerized carbon atoms , and the intralayer hopping @xmath27 between next - nearest neighbors .
both terms break the particle - hole symmetry , which then discriminates the different edges : the terminating atoms live on different layers and therefore posses a different on - site potential . in the dispersion relation , this takes the effect of an additional background velocity , which increases the velocity at one edge ( the fast edge ) while reducing the velocity at the other edge ( the slow edge ) .
the dependence of these velocities on the gate potential is shown in the top panel of fig .
. the additional hoppings also affect the transverse localization of the edge states .
this is shown in the bottom panel in terms of the inverse participation ( ipr ) @xmath28 , which is larger the better localized a state is ( here , the sum is over the transverse direction of the nanoribbon ) . at this point
it is important to mention that there are two types of zigzag edge termination in graphene bilayer,@xcite the so called @xmath29 and @xmath30 terminations .
when the on site potential @xmath1 is zero some qualitative differences in the bandstructure can be found in both cases .
however when @xmath0 leads to a nonvanishing @xmath1 in both cases two of the edge bands become gapped ( with a gap of the order of @xmath1 ) and the two other bands cross each other around the fermi points as discussed before for the two types of edge terminations . also in both cases the asymmetry between the velocities at different edges can be found , being the differences most prominent at the high energy bands .
thus we will not consider any distinction between the two terminations as long as we are only concerned in the physics at small energies around the fermi points . at small energies , the hybridization of the edge states leads to an avoided crossing , where the small residual gap @xmath31 vanishes exponentially with increasing width of the ribbon . a sizable hybridization gap @xmath31 exists only for small values of @xmath6 ( narrow ribbons ) .
for typical applied gate voltages and widths , each edge of the nanoribbon therefore behaves like a metallic one - dimensional system with two counter - propagating states , where each state is associated to one of the k points . in this scenario of low dimensionality
, one should expect that interactions play an important role . generally , however , interactions are felt most strongly when particles propagate slowly and are well confined .
thus , the concrete manifestations should depend on the distinct properties of the fast and slow edge states described above .
this is what we will explore in the remainder of this paper .
influence of @xmath32 and @xmath33 on the edge - state velocities ( top panel , @xmath34 ) and localization on each isolated edge ( @xmath35)(bottom panel , expressed in terms of the inverse participation ratio @xmath36 ) .
, title="fig : " ] influence of @xmath32 and @xmath33 on the edge - state velocities ( top panel , @xmath34 ) and localization on each isolated edge ( @xmath35)(bottom panel , expressed in terms of the inverse participation ratio @xmath36 ) .
, title="fig : " ]
within the tight - binding description , interactions can be incorporated via a hubbard term @xmath37 , where @xmath38 is the occupation operator on lattice site @xmath39 with electrons of spin @xmath40 .
assuming @xmath41 , each edge can be treated separately via a hamiltonian of non - chiral bosons : @xmath42 where @xmath43 labels the charge ( @xmath44 ) and spin ( @xmath45 ) sectors , while @xmath46 and @xmath47 are the velocity and the luttinger parameters , respectively .
interaction events that do not lead to backscattering renormalize these parameters in different ways , which yields the usual separation between the spin and charge degrees of freedom @xcite : @xmath48 @xmath49 where the upper sign applies to @xmath44 , while the lower sign applies to @xmath45 .
most of the interesting physics arises from backward scattering , which leads to a nonlinear sine - gordon term in the spin sector , @xmath50 at half filling , which we need to consider to assess the size of any interaction - induced gap , this is accompanied by umklapp processes , which contribute an analogous term to the charge sector : @xmath51 all these terms are controlled by the forward and backward coupling constants @xmath52 , which are obtained by writing the hubbard interaction term @xmath53 in the basis @xmath54 of the diagonalized one - particle hamiltonian around each k point ( @xmath9 ) : @xmath55 the left - right symmetry of the system results in the additional constraint @xmath56 , which involves the ipr of the transverse wavefunction . as seen above , as soon as @xmath32 and @xmath33 are taken into account the ipr strongly depends on the layer - symmetry breaking potential @xmath1 .
thus , the effective strength of interactions can be controlled by the applied gate voltage .
since the terms ( [ sinegordonspin ] ) and ( [ sinegordoncharge ] ) make the hamiltonian not exactly solvable , we assess their consequences using standard renormalization group ( rg ) arguments .
@xcite in first order in @xmath57 , the rg equations for @xmath58 and @xmath59 are @xmath60 where @xmath61 is a logarithmic renormalization scale .
the scaling behavior of @xmath62 depends on the value of @xmath59 . for @xmath63 ,
@xmath62 decreases when the scaling parameter @xmath64 increases , which renders the interaction term irrelevant . for @xmath65 , however , the interactions are relevant ; the intermediate case @xmath66 represents a quantum critical point .
charge gap @xmath67 as a function of @xmath1 , for selected values of @xmath68 , assuming @xmath12 ( upper panel ) , as well as @xmath13 , @xmath14 ( middle panel : slow edge ; lower panel : fast edge ) . note the different scales of the vertical axis.,title="fig : " ] charge gap @xmath67 as a function of @xmath1 , for selected values of @xmath68 , assuming @xmath12 ( upper panel ) , as well as @xmath13 , @xmath14 ( middle panel : slow edge ; lower panel : fast edge ) . note the different scales of the vertical axis.,title="fig : " ] charge gap @xmath67 as a function of @xmath1 , for selected values of @xmath68 , assuming @xmath12 ( upper panel ) , as well as @xmath13 , @xmath14 ( middle panel : slow edge ; lower panel : fast edge ) .
note the different scales of the vertical axis.,title="fig : " ] because of the constraint @xmath69 , the luttinger parameter ( [ kcharge ] ) can be written as @xmath70 for the case @xmath71 ( repulsive interactions ) , to which we restrict our attention , @xmath72 and the rg equations ( [ rgeqs ] ) imply that the spin sector will flow towards the noninteracting theory , with renormalized @xmath73 due to the emerging spin - rotation invariance . on the other hand , @xmath74 is smaller than one , and therefore the system may acquire a gap in the charge sector . for a more quantitative analysis , we introduce the new variable @xmath75 , so that eq .
( [ rgeqs ] ) takes the form @xmath76 in these new variables , the rg equations possess a first integral @xmath77 , whose value remains constant along the trajectories of the flow , and therefore can be calculated using the bare values @xmath78 and @xmath79 .
most importantly , none of the flow lines cross the line @xmath80 , where @xmath81 .
this line constitutes a separatrix between two regimes . for @xmath82 ,
@xmath83 scales to zero and @xmath84 approaches a definite renormalized value @xmath85 .
physically , this sector of the system is then well described by the noninteracting hamiltonian ( [ nonintbosh ] ) with renormalized @xmath86 . for @xmath87 , however , @xmath88 and @xmath89 flow towards strong coupling , where the rg equations are no longer valid .
the charge sector then acquires a gap @xmath16 . for bare value @xmath90
, this gap can be estimated using the selfconsistent harmonic approximation , @xcite which exploits that the gap scales with @xmath61 as @xmath91 , with @xmath92 .
this leads to @xmath93 , where we equate the real - space @xmath5 cutoff with the lattice constant .
the quantity @xmath94 is an ultraviolet cutoff in the momentum space .
since the edge dispersion relation is linear only for energies less than @xmath95 , @xmath96 , which delivers @xmath97 however , this estimate is only valid when @xmath98 . for @xmath99 ,
the gap must be obtained by terminating the flow at @xmath100 , where the first - order expansion breaks down , giving @xmath101 in order to determine which of these regimes applies to the bilayer graphene edge states , we have computed the ratio @xmath102 for some values of @xmath1 and @xmath68 .
consistently , @xmath103 , which means that the gap should be obtained from eq .
( [ alternativegap ] ) , where @xmath104 and @xmath105 follow from the results of fig.[fig1a ] .
the size of the resulting gap for different scenarios is shown in fig .
[ fig3 ] . in all cases , the dependence of @xmath16 on @xmath1 is non - monotonic . ignoring next - nearest - neighbor hopping ( left panel ) ,
the gap at both edges is identical , but takes a sizeable value only for unrealistically large values of the hubbard parameter .
when @xmath32 and @xmath33 are taken into account , the gap at the fast edge ( right panel ) is further suppressed .
however , the gap at the slow edge ( middle panel ) is dramatically increased , in particular in the region where the propagation velocity becomes small .
we have studied the effect of next - nearest - neighbor hopping and interactions on the edge states in gated bilayer graphene nanoribbons with zigzag termination .
the additional hopping results in the formation of a slow and a fast edge , where electrons propagate with different velocities .
the small velocity at the slow edge assists the formation of an interaction - induced gap @xmath16 .
whether a sizeable gap can be achieved at realistic interaction strengths would be best decided by an experiment which exploits that umklapp processes are suppressed when the temperature is raised .
the physics behind any experimentally observed gap can therefore be probed via the temperature dependence of the conductivity ( measured at the edge via side contacts ) , which for @xmath106 should scale as @xmath107 .
@xcite the experimental set up can be a four - terminal geometry which probes each edge separately ( @xmath35 and thus the two edges are decoupled for a sufficiently wide nanoribbon ) . because each edge state possesses different
a different fermi velocity @xmath74 is thus different as it is discussed in the text allowing to measure independently each conductivity .
this algebraic dependence is distinct from the exponential behavior of the conductivity when the gap originates in the single - particle spectrum ( such as the hybridization gap @xmath31 due to the finite width of the ribbon ) .
the simultaneous presence of the hybridization gap and interactions could be modelled by a bosonized interaction term @xmath108 , @xcite which couples the spin- and charge sectors .
moreover , in this situation , right- and left - moving states around the same k point become coupled because of their finite overlap .
the fact that these states propagate with different velocities adds interesting complications , which , however , go beyond the scope of the present work .
we gratefully acknowledge discussions with f. guinea , m. a. h. vozmediano and g. len suros and support by the ec via grant no .
mext - ct-2005 - 023778 . | we show that gated bilayer graphene zigzag ribbons possess a fast and a slow edge , characterized by edge state velocities that differ due to non - negligible next - nearest - neighbor hopping elements . by applying bosonization and renormalization group methods , we find that the slow edge can acquire a sizable interaction - induced gap , which is tunable via an external gate voltage @xmath0 .
in contrast to the gate - induced gap in the bulk , the interaction - induced gap depends non - monotonously on the on - site potential @xmath1 . |
contribution of long - range dispersive van der waals forces to the interaction between two molecules is quite significant .
these forces play an important role in the description of many physical and chemical phenomena .
some of these are adhesion , surface tension , physical adsorption @xcite , chemistry of rare gases @xcite , molecular chemistry in the interstellar medium @xcite etc .. the correlation between the electron density fluctuation at widely separated locations give rise to these forces . the van waals interaction between two molecules a and b in the lowest order
is represented by the orientationally averaged potential @xmath1 where r is the intermolecular distance and c@xmath2 is the orientationally averaged van der waals coefficient .
the coefficient c@xmath2 describes the dipole - dipole interactions between the two molecular species .
metal clusters , specifically those of alkali atoms na and k , played an important role in the development of cluster physics as a branch of modern physics and chemistry .
the knowledge of their van der waals coefficient @xmath0 is useful for the description of cluster - cluster collisions @xcite and also for characterizing the orientation of clusters in the bulk matter @xcite .
however , only very few papers devoted to the calculation of van der waals coefficients exist in the literature . in refs .
@xcite , time dependent kohn - sham ( tdks ) equation of time dependent density functional theory ( tddft ) within the spherical jellium background model ( sjbm ) was employed to calculate the van der waals coefficients . on the other hand , in ref .
@xcite , a purely particle- and current - density based hydrodynamic formalism of tddft within sjbm has been applied to calculate the coefficients .
this model is well suited for the description of the alkali metal clusters ( li , na , and k ) and correctly explains the greater stability of these clusters with magic number 2 , 8 , 20 , 34 , 40 , @xmath3 of atoms due to closing of the electronic shell @xcite .
although the hydrodynamic approach is less accurate than the tdks approach due to the approximate nature of the kinetic energy functional employed in it , nonetheless it is computationally advantageous for larger clusters in comparison to the orbital - based tdks approach .
the sjbm replaces the discrete ionic structure of clusters by a spherically symmetric uniform positive charge background and thus making it possible to carry out calculations for the optical response properties of reasonably large clusters of around 100 atoms within the tdks approach @xcite .
in contrast to this the hydrodynamic approach in conjunction with sjbm allows calculation of optical response properties of very large alkali atom clusters containing 10,000 atoms without much difficulty @xcite .
in past ten years or so , several all - electron _ ab initio _ calculations devoted to the ground state and the optical response properties of sodium clusters taking into account the actual geometrical arrangement of the sodium atoms have also been reported in the literature @xcite . however , these calculations are restricted to small sized clusters containing maximum up to 20 sodium atoms . moreover , these calculations are computationally more expensive and require substantially more resources than the jellium based calculations .
it is only very recently that the all - electron _ ab initio _ calculations of the van der waals coefficient c@xmath2 of small sized closed shell sodium clusters containing up to 20 atoms have been reported in the literature @xcite .
the main aim of this paper is to make a systematic comparison of the values of c@xmath2 obtained by employing the jellium based model and all - electron _ ab initio _ method within the tddft .
this comparison will enable us to assess the accuracy of jellium model in predicting the values of van der waals coefficient and such a study is important , as the jellium model often turns out to be much more efficient to handle particularly when dealing with larger cluster systems . before proceeding further
, it is important to note that density functional theory ( dft ) in principle should give the exact ground - state properties including the long range van der waals energies .
however , the widely used local density approximation ( lda ) and generalized gradient approximations ( gga ) @xcite exchange - correlation ( xc ) functionals fail to reproduce the van der waals energies .
this is because the lda and gga xc functionals can not completely simulate the correlated motion of electrons arising from coulomb interaction between distant non overlapping electronic systems .
it is only recently that attempts @xcite have been made to obtain van der waals energies directly from the ground - state energy functional by correcting the long range nature of the effective kohn - sham potential . on the other hand , it is possible to make reliable estimates of the van der waals coefficient @xmath0 by using expressions which relate this coefficient to the frequency dependent dipole polarizabilities at imaginary frequencies @xcite .
we follow the latter route for the calculation of these coefficients .
the paper is organized as follows : in section ii we discuss the theoretical method and the expressions employed to calculate the van der waals coefficient @xmath0 from the frequency dependent dipole polarizability .
results of our calculations are presented in section iii .
the paper is concluded in section iv .
in order to calculate the van der waals coefficient @xmath0 , we make use of the casimir - polder expression which relates @xmath0 to the frequency dependent dipole polarizability evaluated at imaginary frequency . in accordance with this expression
the orientation averaged dispersion coefficient between two molecules a and b is given by @xcite @xmath4 where @xmath5 is the isotropic average dipole polarizability of the j - th molecule and is given by @xmath6 in the above expression @xmath7 , @xmath8 and @xmath9 are diagonal elements of the dipole polarizability tensor .
therefore , the calculation of dispersion coefficient @xmath0 involves determining frequency dependent dipole polarizability tensor followed by the evaluation of the quadrature . for
the determination of the frequency dependent polarizability , we use two different methods ( i ) the hydrodynamic formalism of tddft in conjunction with the sjbm and ( ii ) all - electron _ ab initio _ method also in the realm of tddft . for details ,
we refer the reader to refs .
@xcite for the hydrodynamic formalism and ref .
@xcite for the _ ab initio _ method of calculation . in the following ,
we describe the choice of parameters made in this paper to perform the calculations of the dispersion coefficient c@xmath2 by two methods .
the basic idea of the sjbm is to replace the distribution of ionic cores by a constant spherically symmetric positive background of jellium density @xmath10 in a finite volume .
this positive background provides the attractive potential for the valence electrons contributed by each atom of the cluster .
the ground - state electronic distribution is then obtained by employing either orbital based kohn - sham or purely density based extended thomas - fermi ( etf ) formalisms of dft .
the calculation of the frequency dependent polarizabilty within the sjbm can be accomplished by tddft . like its time independent counterpart tddft can be formulated either in terms of the time dependent orbitals ( tdks formalism ) or in terms of the particle and the current densities ( hydrodynamic formalism ) . in this paper
, we employ the hydrodynamic approach formalism of tddft in conjunction with the sjbm to calculate the frequency dependent polarizabilities at imaginary frequencies .
the basic dynamical variables of the hydrodynamic theory are the time - dependent density @xmath11 and the velocity potential @xmath12 .
the velocity of the electron fluid is given by @xmath13 .
thus the total time - averaged energy can be expressed in terms of these two variables . for our purpose
we need to evaluate the second - order change in the time - averaged energy as this is related directly to the frequency dependent average dipole polarizability by the relation @xmath14 where @xmath15 is the frequency of the applied electromagnetic field .
it has been shown that the second - order change in the time - average energy @xmath16 is stationary with respect to the variations in the first - order induced density @xmath17 and the induced current - density @xmath18 @xcite .
consequently , @xmath16 can be determined by choosing appropriate variational forms for @xmath17 and @xmath19 and making @xmath20 stationary with respect to the parameters of @xmath17 and @xmath18 . for our calculations we use ten parameters
each for @xmath17 and @xmath18 and check the convergence by adding more parameters .
the variational method is also applicable for the imaginary frequencies @xcite as it requires replacing @xmath21 by @xmath22 in the expression of @xmath16 .
this allows us to determine dynamic multipolarizability at imaginary frequencies ( @xmath23 ) by exactly the same procedure as employed for getting @xmath24 .
it is well known that a tddft based response property calculation requires approximating the xc functional at two different levels .
the first one is the static xc potential needed to calculate the ground - state density of the system .
the second approximation is needed to represent the xc kernel @xmath25 which determines the xc contribution to the screening of an applied field . for the xc kernel
, we use reasonably accurate adiabatic local density approximation ( alda ) @xcite . on the other hand , for static xc energy
functional we use the dirac exchange energy functional @xcite and gunnarsson - lundqvist ( gl ) @xcite parametrized form for the correlation energy functional within the lda .
in addition to approximating the xc energy functional , the calculation of polarizability by the hydrodynamical approach of tddft also requires approximating the non - interacting kinetic energy functional @xmath26 $ ] also .
we choose the von weizsacker @xcite form for @xmath26 $ ] , which is given as @xmath27 = \frac{1}{8}\int\frac{{\bf\nabla}\rho\cdot{\bf\nabla}\rho } { \rho}d{\bf r}. \label{14}\ ] ] the von weiszacker functional is well suited for the description of response properties of closed shell metal clusters @xcite . in the present paper the ground - state densities @xmath28 of clusters
are obtained by employing purely density - based etf @xcite method within the sjbm of metal clusters . to carry out the all - electron _ ab initio _ calculation of the frequency dependent polarizabilities of the clusters , we employ the adf program package @xcite .
we refer the reader to ref .
@xcite for detailed description of the method adopted in this package for obtaining the frequency dependent polarizabilities . in this
package calculations of electronic and response properties of molecules are carried by using slater type orbital ( sto ) basis sets .
it is well known that for accurate calculations of response properties it is necessary to have large basis sets with both polarization and diffuse functions . for our purpose ,
we choose all electron even tempered basis set et - qz3p-2diffuse with two sets of diffuse functions consisting of ( 11s,9p,7d,3f ) functions for na atom .
the application of basis set with diffuse functions often leads to the problem of linear dependencies .
such problem have been circumvented by removing linear combinations of functions corresponding to small eigenvalues of the overlap matrix .
we expect that the size of the chosen basis set will make our results very close to the basis - set limit .
once the average polarizability at imaginary frequencies are obtained then the casimir - polder integral eq .
( [ casimirpolder ] ) is evaluated by employing thirty point gauss - chebyshev quadrature scheme as described in ref .
the convergence of the results has been checked by increasing number of frequency points .
like sjbm based calculations , all the _ ab initio _ calculations in the present paper are performed with the alda for the xc kernel . on the other hand , for the static xc potential needed to calculate the ground - state orbitals and their energies , we employ the lda potential as parametrized by vosko , wilk and nussair ( vwn ) @xcite .
we note here that both gl and vwn forms for the xc potential use the same dirac exchange energy functional @xcite but the parameterization for the correlation part is different .
we expect that this difference in the correlation energy functional will lead to a significantly smaller deviation in the results obtained via the _ ab initio _ and sjbm calculations than the discrepancy arising due to the consideration of the actual geometrical structure of the clusters in the _ ab initio _ calculations .
furthermore , in order to perform _ ab initio _ calculation of response properties we need to choose the ground - state geometries of clusters . for the dimer @xmath29
, we use the experimental bond length 3.0786 @xmath30 . on the other hand ,
for larger clusters ( 4- to 20-atom clusters ) we use the structures which are obtained via geometry optimization calculation with triple-@xmath31 plus two polarization functions ( tz2p ) basis set and becke - perdew ( bp86 ) xc potential @xcite .
all the optimizations are carried out with the convergence criteria for the norm of energy gradient and energy fixed at @xmath32a.u and @xmath33a.u , respectively .
the optimized structures obtained by us are in agreement with the corresponding results of refs .
@xcite . in case of a cluster having more than one isomers
we choose the one possessing the lowest energy for our calculations of the dipole polarizability . here
we note that our geometry optimization calculation for the cluster containing 20 sodium atoms yields lower energy for the structure with c@xmath34 symmetry than the one with t@xmath35 symmetry .
this is in contrast to the result of ref .
the next section is devoted to the discussion of results for the dispersion coefficient @xmath0 obtained by employing the above two methods .
.comparison of dispersion coefficient @xmath0 ( in atomic units ) for pairs of similar sodium clusters obtained with the all - electron _ ab initio _ and sjbm calculations . [ cols="^,^,^,^ " , ] having assessed the accuracy of results obtained within the sjbm , we now focus our attention on the study of relationship between c@xmath2 and @xmath36 . in the so - called london approximation
the van der waals coefficient c@xmath2 between two similar molecules can be represented in terms of the static polarizability @xmath36 as @xmath37 where @xmath38 is the effective or characteristic frequency of the system .
the above expression ( eq .
( [ londondispersion ] ) ) is obtained with the single pole approximation for the frequency dependent polarizability , which assumes that all the transitions can be replaced by only one effective transition .
this is also known as london dispersion formula and it provides a way to correlate the van der waals coefficient with the static polarizability . in doing
so the accuracy of c@xmath2 crucially depends on the precision with which the static polarizability is calculated . in order to study this correlation
, we calculate the average static polarizability of sodium clusters by the _ ab initio _ method and sjbm - hydrodynamic approach .
another reason for carrying out static polarizability calculations is that unlike c@xmath2 , a fairly large number of experimental results on the static polarizabilty exist in the literature @xcite , which allows us to compare our results with the corresponding experimental data .
the results of the polarizability calculations are shown in fig . 2 along with the experimental results of ref .
2 clearly shows that both _
ab initio _ and sjbm results are lower than the corresponding experimental data , however , the _ ab initio _ results are closer to the experimental data than the sjbm results . moreover , it is seen from this figure that like c@xmath2 , the deviation between the sjbm and _ ab initio _ results is minimum for magic number clusters and these deviations decrease with the larger clusters .
for example , the difference between the two results is 20@xmath39 ( _
ab initio _ number is 233.32 a.u . ) for the dimer , whereas the differnce reduces to around 11@xmath39 ( _ ab initio _ number is 1752.8 a.u . ) for the na@xmath40 cluster . for non - magic number clusters the differnce is more and but reduces with the increase in the size of the clusters .
therefore , we conclude that the static polarizability of large size clusters can also be obtained quite accurately by employing the sjbm - hydrodynamic approach with substantially less amount of computational effort .
the van der waals coefficient @xmath0 for sodium atom clusters containing even number of atoms ranging from 2 to 20 atoms have been calculated by employing all - electron _ ab initio _ method and the sjbm - hydrodynamic formalism within the realm of tddft .
the calculations are performed with the lda xc potentials .
the van der waals coefficient is obtained by using casimir - polder expression which requires frequency dependent dipole polarizabilties of the two interacting species at imaginary frequencies .
these are calculated by the all - electron _ ab initio _ and the sjbm - hydrodynamic methods of tddft .
the results of these two calculations are compared to ascertain the effect of ionic structure of the clusters on the van der waals coefficient .
we find that the differences between the results of all - electron _ ab initio _ and sjbm - hydrodynamic calculations are minimum for magic number clusters and these differences decrease with the increase in the number of atoms present in the clusters . even for non - magic number clusters
, we find that sjbm - hydrodynamic calculations lead to results which are very close to the corresponding _ ab initio _ numbers . from these results , we conclude that for clusters larger than na@xmath41 detailed ionic structure of the clusters does not have much effect on the results for the van der waals coefficient and also on their static polarizability .
this is an important results as carrying out _ ab initio _
calculations for the optical response properties of large clusters is very expensive .
this paper shows that for such systems sjbm based methods within tddft can yield quite accurate results for the van der waals coefficient .
25 d. langbein , _ theory of van der waals attraction _ , springer tracts in modern physics vol .
72 ( springer , berlin , 1974 ) .
k. o. christe , angew .
int . ed . *
40 * , 1419 ( 2004 ) . ,
astrophysics series * 17 * , ed .
d. r. flower , cambridge university press , cambridge ( 1990 ) .
a. s. amadon and w. h. marlow , phys .
a * 43 * , 5483 ( 1991 ) .
o. gunnarsson , s. satpathy , o. jepsen and o. k. anderson , phys .
67 * , 3002 ( 1991 ) .
lambin , a. a. lucas , j .-
vigneron , phys .
b * 46 * , 1794 ( 1992 ) . j. m. pacheco and w. ekardt , modern phys . lett .
b * 7 * , 573 ( 1993 ) j. m. pacheco and w. ekardt , phys .
* 68 * , 3694 ( 1992 ) .
a. banerjee and m. k. harbola , j. chem .
phys . * 117 * , 7845 ( 2002 ) ; a. banerjee and m. k. harbola , proc .
indian natn .
71 a * , 357 ( 2005 ) ; a. banerjee and m. k. harbola , pramana j. phys . * 66 * , 423 ( 2006 ) . m. brack , rev . mod
. phys . * 65 * , 677 ( 1993 ) and references their in ; w. a. de heer , rev .
* 65 * , 611 ( 1993 ) .
j. alonso and l. c. balbas _ topics in current chemistry _ * 182 * , ed .
r. f. nalewajski springer - verlag berlin ( 1996 ) and references their in .
m. madjet , c. guet , and w. r. johnson , phys .
a * 51 * , 1327 ( 1995 ) .
b * 39 * , 3533 ( 1989 ) . m. k. harbola and a. banerjee , j. theoretical comp
. chem . * 2 * , 301 ( 2003 ) .
a. banerjee and m. k. harbola , j. chem .
* 113 * , 5614 ( 2000 ) . j. l. martins , j. buttet , and c. car , phys
b * 31 * , 1804 ( 1985 ) .
i. moulett , j. l. martins , f. reuse , and j. buttet , phys . rev .
b * 42 * , 11598 ( 1990 ) .
u. r@xmath42thlisberger and w. andreoni , j. chem . phys . *
94 * , 8129 ( 1991 ) .
j. guan , m. e. casida , a. m. k@xmath42ster , and d. r. salahub , phys .
b. * 52 * , 2184 ( 1995 ) .
p. calmanici , k. jug , and a. m. k@xmath42ster , j. chem . phys .
* 111 * , 4613 ( 1999 ) .
s. k@xmath43mmel , t. berkus , p .- g.reinhard , m. brack , eur phys . j. d * 11 * , 239 ( 1999 ) .
l. kronik , i. vasiliev , and j. r. chelikowsky , phys .
b * 62 * , 9992 ( 2000 ) .
s. a. blundell , c. guet , and r. r. zope , phys .
. lett . * 84 * , 4826 ( 2000 ) .
s. j. a van gisbergen , j. m. pacheco , and e. j. baerends , phys . rev .
a * 63 * , 063201 ( 2001 ) .
i. a. solovyov , a. v. solovyov , w. greiner , phys .
a * 65 * , 053205 ( 2002 ) .
k. r. s. chandrakumar , t. k. ghanty , and s. k. ghosh , j. chem
. phys . * 120 * , 6487 ( 2004 ) .
a. jiemchooroj , p. norman and b. e. sernelius , j. chem .
phys * 125 * , 124306 ( 2006 ) .
a. banerjee , a. chakrabarti , and t. k. ghanty , accepted for publication in j. chem .
j. p. perdew , phys .
. lett . * 55 * , 1665 ( 1985 ) ; * 55 * , 2370(e ) ( 1985 ) .
a. d. becke , phys .
rev a * 38 * , 3098 ( 1988 ) .
j. p. perdew and y. wang , phys .
b , * 33 * , 8800 ( 1986 ) ; j. p. perdew , in _ electronic structure of solids _ , edited by p. ziesche and h. eschrig ( akademic - verlag , berlin , 1991 ) .
y. andersson , d. c. langreth , b. i. lundqvist , phys .
* 76 * , 102 ( 1996 ) .
j. f. dobson , b. p. dinte , phys .
, * 76 * , 1780 ( 1996 ) . w. kohn , d. makarov , phys .
, * 80 * , 4153 ( 1998 ) .
h. b. casimir and d. polder , phys . rev . * 73 * , 360 ( 1948 ) . a. j. stone , _ the theory of intermolecular forces _ , ( clarendon , oxford , 1996 ) . s. j. a. van gisbergen , j. g. snijders and e. j. baerends , j. chem .
* 103 * , 9347 ( 1995 ) .
e. k. u. gross , j. f. dobson and m. petersilka , in _ density functional theory , topics in current chemistry _ * 181 * , ed .
r. f. nalewajski ( springer , berlin , 1996 ) .
dirac , proc .
cambridge philos .
* 26 * , 376 ( 1930 ) .
o. gunnarson and b.i .
lundquist , phys .
b * 13 * , 4274 ( 1976 ) . c. f. von weizsacker , z. phys . * 96 * , 431 ( 1935 ) .
adf2006.01 , scm , theoretical chemistry , vrije universiteit , amsterdam , the netherlands , http://www.scm.com .
w. rijks and p. e. s. wormer , j. chem . phys . * 88 * , 5704 ( 1988 ) . s. h. vosko , l. wilk , and m. nussair , can . j. phys . * 58 * , 1200 ( 1980 ) .
a. d. becke , phys .
a * 38 * , 3098 ( 1988 ) .
j. p. perdew , phys .
b * 33 * , 8822 ( 1986 ) .
w.d . knight , k. clemenger , w. a. de heer , and w. a. saunders , phys .
b * 31 * , 2539 ( 1985 ) .
d. rayane , a. r. allouche , e. benichou , r. antoine , m. aubert - frecon , ph .
dugourd , m. broyer , c. ristori , f. chandezon , b. a. huber , and c. guet , eur .
j. d * 9 * , 243 ( 1999 ) ; g. tikhonov , v. kasperovich , k. wong , and v. kresin , phys .
a * 64 * , 063202 ( 2001 ) . | in this paper we employ two distinct approaches - all - electron _ ab initio _ method and the spherical jellium background model- within time dependent density functional theory to calculate the long range dipole - dipole dispersion coefficient ( van der waals coefficient ) @xmath0 of sodium atom clusters containing even number of atoms ranging from 2 to 20 atoms .
the dispersion coefficients are obtained via casimir - polder relation .
all the calculations are carried out with local density approximation for exchange - correlation energy functional .
these two sets of results are compared to assess the accuracy of jellium based results and to ascertain the effect of detail ionic structure of the clusters on the van der waals coefficient . |
due to a predicted high curie temperature ( above 300 k ) , diluted magnetic semiconductors ( dms ) are regarded as important materials for spintronics .
@xcite combining ferromagnetism with semiconductivity in oxides gives them an additional degree of freedom and functionality for fabricating unique devices with applications ranging from nonvolatile memory to quantum computing .
@xcite some theoretical @xcite and experimental @xcite reports have shown room temperature ferromagnetism ( rtfm ) in highly doped zno films .
the magnetic impurities are either implanted into the host matrix @xcite or introduced during the growth .
@xcite the experimental results regarding rtfm and its origin in zn@xmath0co@xmath1o samples in different forms including thin films , bulk , polycrystalline and nanocrystalline samples differ widely , and there is no consensus on either the nature of the magnetism ( whether the samples are para- , dia- or ferro - magnetic ) or on the origin of the ferromagnetism ( intrinsic , extrinsic , defect induced etc . ) . in the case of cobalt doped
zno films , the observed rtfm has been attributed either to co clustering , @xcite some external pollution @xcite or is thought to be intrinsic in nature .
@xcite cobalt and lithium co - doped zno films have revealed only co paramagnetism , and oxygen vacancies were considered as the origin for the rtfm .
@xcite in bulk samples the fm is attributed to co clustering , @xcite and in hydrogenated samples it is attributed either to the appearance of co clusters @xcite or oxygen vacancies .
@xcite in the case of polycrystalline and nanocrystalline samples , some groups have reported paramagnetism .
@xcite recently , nanocrystals of zn@xmath0co@xmath1o have been reported to be ferromagnetic @xcite or superparamagnetic , @xcite and nanowires showed a paramagnetic behavior with co concentrations up to @xmath5 .
@xcite very recently , thin films and nanoparticles of pure zno have been reported to be ferromagnetic , and their fm is attributed to defects on zn sites and surface point defects .
@xcite from a practical point of view , the ultimate goal of investigating materials such as cobalt doped zno is to fabricate semiconductor devices that have a spin - polarized nature to their electronic transport properties . to this end ,
explorations of the magnetotransport properties are an essential part of studies into cobalt doped zno .
the low temperature magnetoresistance ( mr ) of cobalt doped zno is typically tens of percent in magnetic fields of up to several tesla , @xcite and may show a crossover between positive and negative mr @xcite at different temperature and magnetic field values .
zno films containing magnetic nanoclusters within a non - magnetic insulating matrix have positive mr values as large as 811% at 5 k in fields of a few tesla , @xcite revealing the rich potential for magnetic - field influenced charge transport in dms .
the exploration of single crystals of dms is of great significance as this form is ideally suited for understanding the intrinsic properties of any material . in the present paper
we present the structural , magnetotransport and magnetic properties of both pristine zno and cobalt doped zno single crystals , which were grown by the molten salt solvent technique @xcite .
in the present case , the molten salt solvent technique has been employed to grow single crystals of both pristine zno and cobalt doped zno .
analytical reagent ( ar ) grade zno , koh@xmath66h@xmath7o and cocl@xmath82h@xmath7o were used as precursors .
calculated amounts @xcite of zno and cocl@xmath82h@xmath7o ( an appropriate amount was chosen corresponding to the desired concentration of co ) were slowly added to the molten koh .
the crucible was then covered with a specially designed silver lid which was fitted with a silver tube at its centre .
the other end of the silver tube was tapered into a cone . the crucible with the charge ( solvent + solute )
was then placed in the furnace such that the charge remained in the uniform temperature region of the furnace at a temperature of 480@xmath9c for @xmath1050 hours .
when immersed in the molten charge , the conical end of the tube provides suitable sites for the growth of the crystals , allows the heat of solidification to escape and also serves as a sheath for the thermocouple ( one end of which is placed in the tube ) .
these conditions result in the nucleation and isothermal growth of transparent , green - colored , needle - shaped cobalt doped single crystals of zno due to dissociation of zincate species unstable at 480@xmath9c , at the cylindrical walls of the silver crucible and the tapered end of the silver tube .
@xcite the crystals were extracted by dissolving the solvent in methyl alcohol .
dimensions of the needle - like crystals are typically 2.5 to 5 mm in length , 0.5 - 1.0 mm in width and 0.1 to 0.2 mm in thickness
. the single crystals of pristine and cobalt doped zno were powdered and their structural properties were studied using a philips ( xpert pro , model pw 3040 ) x - ray diffraction ( xrd ) system with cu k@xmath11 radiation at wavelength 1.54060 .
the semcf used for sem micrographs has rontecs edax ( energy dispersive analysis of x - rays ) system model quantax 200 , which is based on silicon drift detector ( sdd ) technology and provides an energy resolution of 127 ev at mn k@xmath11 .
the magnetic moment of a small amount of each kind of crystal was measured as a function of both applied magnetic field and temperature , using a quantum design mpms-7 squid magnetometer .
the temperature and magnetic field dependent electrical resistance of individual crystals was also measured .
the maximum applied magnetic field for these measurements was 4.5 t , and the range of temperatures employed was 2.5 - 350 k.
the x - ray diffractograms of both the pristine zno and zn@xmath0co@xmath1o crystals are shown in fig . [ fig1](a ) .
all the diffraction peaks correspond to zno in the wurtzite structure .
no secondary phases are detected in the xrd spectrum of zn@xmath0co@xmath1o crystals , which implies that the co@xmath4 ions are incorporated into the lattice substitutionally replacing zn@xmath4 .
it can also be seen that the xrd peaks of the zn@xmath0co@xmath1o are slightly shifted to lower angles as compared to the pure zno peaks , which implies that the lattice is slightly expanded relative to pure zno .
[ fig1](b ) shows the morphology of a zn@xmath0co@xmath1o single crystal as studied using scanning electron microscopy .
this depicts the characteristic hexagonal shape of a zno single crystal .
the exposed faces are hexagonal * m * 10@xmath120 and hexagonal * p * 10@xmath121 ( cone ) faces .
the precise elemental composition was calculated from the edax spectrum ; the concentration of co in the two samples has been estimated to be @xmath13 and @xmath2 .
[ cols="<,^,^,^,^,^,^,^,^,^,^,^ " , ] to study the electronic transport properties of the crystals , the temperature dependent resistance behavior was observed for zn@xmath0co@xmath1o single crystals .
electrical contacts were made to individual crystals using silver paste .
the 300 k resistivity of the pristine zno crystals is 3.6 @xmath14 cm , and the zn@xmath0co@xmath15o crystals 5.4 @xmath14 cm and 6.7 @xmath14 cm with @xmath16=0.10 and @xmath16=0.02 respectively .
the exponential decrease of resistance with increase in temperature as shown in fig .
[ fig2](a ) is a clear signature that the cobalt doped zno maintains the semiconducting behavior of undoped zno .
the two different slopes of the ln @xmath17 vs 1/t ( inset of fig .
[ fig2](a ) ) curve in the low ( 2.5 - 100 k ) and high temperature ( 100 - 300 k ) regions are signatures of two different conduction mechanisms being operative in these two temperature regions . at high temperatures ,
the conduction mechanism is thermally activated band conduction from donor levels near the conduction band , with an activation energy of 10 mev .
@xcite at low temperatures there is insufficient thermal energy to promote conduction through delocalized carriers . instead , below
we show that the variable range hopping ( vrh ) conduction mechanism is active at low temperatures , where the electrons hop between levels that are close to the fermi level irrespective of their spatial separation .
the relation between electrical conductivity and temperature for vrh is given by mott s equation @xcite @xmath18 the parameters @xmath19 and @xmath20 are given by the following expressions @xcite @xmath21 where @xmath22 is the phonon frequency at the debye temperature , @xmath23 is boltzmann s constant , @xmath24 is the density of localized electronic states at the fermi level and @xmath25 is the inverse localization length .
the exponent @xmath26 in eq .
[ mottvrh ] varies from @xmath27 when the density of states across the fermi level is constant to @xmath28 in the case of significant electron - electron interactions that lead to the formation of a coulomb gap . from eqs .
[ mottvrh ] and [ mottvrhcoeff1 ] we expect @xmath29@xmath30 . the nearly linear plot of @xmath29@xmath31 vs @xmath32 in fig .
[ fig2](b ) indicates that vrh is the dominant mechanism of conduction in zn@xmath33co@xmath34o below 100 k. we find similar behavior for both undoped zno and zn@xmath35co@xmath36o crystals .
the results of fitting the data for all three samples with the vrh model ( eqs .
[ mottvrh ] and [ mottvrhcoeff1 ] ) are reported in table [ table1 ] .
the mott parameters , i.e. average hopping distance ( @xmath37 ) and average hopping energy ( @xmath38 ) , were calculated from @xmath39 other conditions for vrh conduction are that the value of @xmath25@xmath37 must be greater than 1 , and that @xmath40 .
note that both of these conditions are satisfied in the zn@xmath0co@xmath15o crystals .
for the zno crystals the conditions are only satisifed above approximately 25 k , and at lower temperatures other conduction mechanisms may compete with vrh .
the values of @xmath41 , @xmath42 and @xmath37 determined by fitting to the vrh model are reasonable compared with results which exist for both doped and undoped zno .
@xcite thus , the analysis of the temperature dependent electronic transport data for both the zno and zn@xmath0co@xmath15o crystals provides convincing evidence for the vrh conduction process up to approximately 100 k , above which a thermally activated conduction mechanism becomes more appropriate .
measurements were also carried out to study the magnetoresistive behavior of the single crystals up to a magnetic field @xmath43 of 4.5 t and at temperatures down to 2.5 k. figs .
[ fig3](a ) and [ fig3](d ) show that for the zn@xmath0co@xmath15o crystals the magnetoresistance ( mr ) is weak and negative ( see inset fig .
[ fig3](a ) ) down to a certain temperature .
below that temperature the mr becomes positive ( i.e. changes its sign ) and becomes larger in magnitude .
the crossover in mr from negative to positive occurs at @xmath1020 k in the doped crystals .
it may be noted that the mr increases with the decrease in temperature in all cases , and also with increase in the concentration of cobalt .
the largest value attained , near saturation of the positive mr , is 20 % ( fig . [ fig3](d ) ) at 2.5 k for zn@xmath33co@xmath34o single crystals , which is at the upper end of mr values reported by other groups for zno - based dms @xcite .
a positive mr also arises in pristine zno crystals below 5 k ( not shown ) , but it is weaker than in the cobalt doped crystals , no more than 1% at 2.5 k , which is likely due to residual impurities .
there are several mechanisms which are typically cited to account for the mr in dms at low temperatures : * a negative mr at low fields owing to the magnetic field breaking the time - reversal symmetry of scattering paths involved in weak localization ; * a negative mr at large fields from a reduced scattering of the carriers as the magnetic field aligns the moments of the magnetic dopants ; * a positive mr at intermediate fields from a field - induced spin - splitting of the conduction band due to the s - d exchange interaction .
the absence of evidence for the first type , i.e. a sharp negative mr at @xmath440.1 t , as seen in a number of works @xcite , is consistent with the temperature dependent resistance measurements of fig .
[ fig2 ] , which indicate a hopping type conduction at low temperatures rather than weak localization of the carriers .
the second type of mr is seen in thin films of zno - based dms that exhibit ferromagnetism in magnetic measurements , and the absence of this component at the magnetic fields applied is consistent with the paramagnetic character of the single crystals as measured by squid magnetometry ( discussed below ) .
a phenomenological model for the third mechanism of mr has been developed by khosla and fischer @xcite , where the positive mr is proportional to @xmath45 .
this model has been used elsewhere to explain the mr behavior of doped zno thin films @xcite .
the predicted @xmath45 dependence appears to account for the data of figs .
[ fig3](b ) and [ fig3](e ) .
however , the khosla - fischer model considers the mr to arise from a field - induced splitting of bands of extended states . in the zn@xmath0co@xmath15o crystals of the present report , conduction is via localized states rather than through bands of extended states .
hence , none of the mechanisms of mr commonly applied to dms are applicable in our case , and we must consider an alternate mechanism to explain the low temperature positive mr , based on the vrh conduction process . at low temperatures , where hopping distances are large enough , a positive
mr arises due to the shrinking of the wavefunctions of the states available for an electron hop .
@xcite in low magnetic fields , the positive mr can then be described with @xmath46 where @xmath47 is the carrier localization length and @xmath26 is the vrh exponent of eq .
[ mottvrh ] .
this may account for the @xmath45 dependence at low fields as shown in figs .
[ fig3](b ) and [ fig3](e ) . at higher fields ,
there is a change to an approximate @xmath48 behavior , as seen before in thin films of cobalt doped zno in the low temperature hard gap regime of electronic transport @xcite .
note that at the lowest temperatures and highest fields , especially for the higher co concentration crystals , the mr is more weakly field dependent than in the low field region .
the data of fig .
[ fig4 ] show that , at 2.5 k in fields of @xmath491.5 t , the mr of the zn@xmath50co@xmath51o crystals can be approximately described by @xmath52@xmath53 , which is very close to the expected high - field mr in the vrh model of @xmath52@xmath54 @xcite .
we thus find that the mr properties of the zn@xmath0co@xmath15o crystals are completely explained as the effect of the magnetic field on the vrh conduction process .
the results of magnetic measurements on each crystal type are shown in fig .
the magnetization vs field ( @xmath55-@xmath56 ) measurements were carried out at room temperature as well as at 2.5 k. the @xmath55-@xmath56 curves of fig . [
fig5](a ) for cobalt doped samples at 2.5 k , normalized to their respective values at 4.5 t for each sample , exhibit neither remanence nor coercivity .
the maximum values of magnetization of the zn@xmath35co@xmath36o and zn@xmath33co@xmath34o crystals in 4.5 t applied field are 2.01 and 2.34 emu / g respectively , corresponding to absolute magnetic moment values of 5@xmath57 - 1@xmath57 emu .
in addition , the curves do not saturate , even at the highest applied field .
a representative sample of the temperature dependent magnetization ( @xmath55-@xmath58 ) measurements for the crystals , measured under both zero - field cooled ( @xmath59 ) and field - cooled ( @xmath60 ) conditions , is plotted in fig .
[ fig5](b ) . a ferromagnetic material , besides exhibiting distinct hysteresis in an @xmath55-@xmath56 loop , also demonstrates thermomagnetic irreversibility ( tmi ) , i.e. , @xmath61@xmath59 in @xmath55-@xmath58 curves .
the @xmath62 and @xmath63 curves of fig .
[ fig5](b ) are identical , showing no sign of tmi , further confirming the lack of ferromagnetic ordering in the sample ; instead the observed behavior is consistent with paramagnetism .
it is thus evident that the zn@xmath0co@xmath15o single crystals are paramagnetic , and doping of co does not result in any ferromagnetism .
to confirm that the magnetic behavior observed is due to the cobalt doping , the @xmath55-@xmath56 curves of the pristine zno sample recorded at both 2.5 k and room temperature ( 300 k ) are drawn in the inset of fig .
[ fig5](a ) .
it may be noted that the data recorded at room temperature exhibit a linear variation of magnetization @xmath55 with applied field , with a negative slope . this is a signature of a diamagnetic material .
however , at 2.5 k pristine zno crystals reveal a feeble paramagnetism with a magnetization 100@xmath64 weaker than that of the zn@xmath0co@xmath15o .
this weak low - temperature paramagnetism is possibly due to the presence of residual impurities as detected in the mr measurements . in paramagnets ,
a value of the effective magnetic moment per ion , @xmath65 , can be estimated from the slope ( @xmath66 ) of a @xmath67 vs @xmath58 plot ( curie law fit ) , where @xmath68 is the number of moles of co. the inset of fig .
[ fig5](b ) shows that the curie law applies over the full range of temperatures measured .
the value of @xmath65 was found to be 3.29 @xmath69 for zn@xmath35co@xmath36o , which compares favorably to values determined for other paramagnetic zno : co samples @xcite and is close to the theoretical free co@xmath4 ion spin - only value of 3.87 @xmath69 .
however , at the higher co concentration of @xmath2 , @xmath65 has a lower value of 2.16 @xmath69 . as the co concentration increases ,
the average co - co distance is reduced , which strengthens the direct antiferromagnetic exchange between co neighbors or superexchange via co - o - co bonds , and consequently the average moment per co is reduced from the case of isolated non - interacting co@xmath4 ions .
theoretically , the value of the total moment of paramagnetic co@xmath4 ions is @xmath70(=@xmath71+@xmath72)=3 + 3/2=9/2 .
however , for transition metals in semiconductors such as zno it is often found that the orbital angular momentum of the unpaired 3@xmath73 electrons is quenched , i.e. @xmath71=0 , and hence the total moment comes only from the spin momentum @xmath72=3/2 .
a brillouin function for the field dependence of the moment of isolated , non - interacting paramagnetic co@xmath4 ions with a spin - only moment of 3/2 is plotted in fig . [
fig5](a ) alongside the magnetic moment data measured at 2.5 k. the calculated field dependence for @xmath72=3/2 paramagnetic ions is extremely close to the experimental data from the cobalt doped samples , confirming that the field dependent magnetization is from a majority of co ions in a divalent , paramagnetic state .
the antiferromagnetically correlated fraction of co ions in the crystals is insensitive to the magnetic field .
finally , we summarize the state of the incorporated co ions in the single crystals of zno , and their effect on the magnetic and magnetotransport properties .
the resistivity and mr measurements point to the presence of localized electronic states , whereas the magnetic properties result from isolated co@xmath4 ions in the crystals . at higher co concentrations ,
a proportion of the co ions experience some antiferromagnetic exchange , likely with near neighboring co ions , which reduces the measured effective moment per incorporated ion . at low temperatures ,
charge carriers hop between the localized electronic states , and in the presence of a magnetic field the probability of hopping is reduced as the wavefunctions of the localized states shrink .
however , at higher magnetic fields , the paramagnetic co ions progressively align with the field , and a spin dependent vrh conduction process may be expected to lower the resistivity at sufficiently high fields @xcite .
no such negative mr is seen in fig .
[ fig3 ] , even at the lowest temperature and maximum field , where the @xmath55-@xmath56 data of fig .
[ fig5](a ) show that magnetic alignment of the co ions is significant .
furthermore , the mr is evidently decoupled from the magnetization of the crystals , as demonstrated by the different shapes of the mr in the two crystals as compared to the identical shape of their @xmath55-@xmath56 curves .
this independence of the magnetotransport from the magnetization in the zn@xmath0co@xmath15o crystals implies that the localized states involved in the conduction process may not be spin polarized through alignment of the co ion magnetic moments . although the formation of localized electronic states owing to the incorporation of co in zno is not controversial , our results support the idea that the electronic transport properties in this dms may be only weakly related to the magnetic properties .
some results on magnetic tunnel junctions fabricated from ferromagnetic cobalt doped zno have non - saturating positive magnetoresistance at low temperatures @xcite , rather than negative magnetoresistance as would be expected from tunneling of spin - polarized charge carriers between layers with their magnetic moments aligned in high fields .
additionally , those authors acknowledged that the behavior of the tunneling magnetoresistance does not correlate with the magnetization of these devices .
these results point to the possibility that the existence of ferromagnetism in cobalt doped zno is not a sufficient condition for generation of spin - polarized charge carriers .
pristine and cobalt doped zno single crystals were grown by a molten salt solvent technique at low temperatures ( 480 @xmath74c ) . the zn@xmath0co@xmath15o crystals are paramagnetic ; there is no sign of ferromagnetism at up to @xmath2 .
the paramagnetic properties of the crystals arise from isolated co@xmath4 ions . at temperatures above 100 k , conduction in the crystals
is by thermal activation , with a change to mott variable range hopping at lower temperatures .
doped and undoped zno crystals show almost negligible negative mr at room temperature , with a crossover to positive mr below 20 k. the positive mr reaches values as large as 20% in the zn@xmath0co@xmath15o crystals at 2.5 k , and is consistent with a field - induced shrinking of the localized state wavefunctions , which reduces the probability of a hop .
the magnetotransport is decoupled from the magnetization , and no spin dependent transport is observed at the highest level of magnetic alignment of the co@xmath4 ions .
our results indicate that the co incorporated in low temperature grown single crystals simultaneously introduces paramagnetic moments and localized electronic states into the zno electronic structure , however the bulk magnetic and magnetotransport properties are independent of each other
. one of the authors , n.s . , gratefully acknowledges support provided by the exchange agreement between the indian institute of technology delhi , india and the ecole polytechnique fdrale de lausanne , switzerland . | long needle - shaped single crystals of zn@xmath0co@xmath1o were grown at low temperatures using a molten salt solvent technique , up to @xmath2 .
the conduction process at low temperatures is determined to be by mott variable range hopping .
both pristine and cobalt doped crystals clearly exhibit a crossover from negative to positive magnetoresistance as the temperature is decreased .
the positive magnetoresistance of the zn@xmath0co@xmath1o single crystals increases with increased co concentration and reaches up to 20 % at low temperatures ( 2.5 k ) and high fields ( @xmath31 t ) .
squid magnetometry confirms that the zn@xmath0co@xmath1o crystals are predominantly paramagnetic in nature and the magnetic response is independent of co concentration .
the results indicate that cobalt doping of single crystalline zno introduces localized electronic states and isolated co@xmath4 ions into the host matrix , but that the magnetotransport and magnetic properties are decoupled . |
the study of pair - creation phenomena in the presence of an external field in a heat bath has attracted some attention in the last few years
. there have been attempts to find out the temperature dependence of the decay rate by various groups ( see for instance references @xcite to @xcite ) .
+ there exists many formalisms to calculate the pair - production rate for such a system .
one of them is to calculate the effective action , the imaginary part of which is related to the vacuum persistence probability and the other is to compute the density matrix using functional schrdinger representation and from thence calculate the production rate . in the past , there had been investigations where the imaginary part of the effective action was calculated using the real time formalism or thermo field dynamics ( tfd)@xcite and also the imaginary time formalism @xcite of finite temperature field theory . on the other hand j.hallin and per liljenberg had derived the vacuum persistence probability for such a system employing the elegant approach of functional schrdinger representation in reference @xcite .
unfortunately the results of these investigations do not see to agree with each othre . some of the studies have found a finite thermal correction to the production amplitude whereas other studies have claimed to have not found any . in references @xcite and @xcite a thermal correction to the decay rate have been found and the results of the two almost match except for a probable typographical error .
+ on the other hand the results of references @xcite , @xcite and @xcite are in contradiction to @xcite and @xcite . in this note
we shall concentrate on the findings of reference @xcite .
the authors of reference @xcite claim that at a fixed temperature , in the infinite time limit ( as is assumed in the effective action formalism ) , the thermal correction to pair production rate vanishes thereby contradicting the earlier results found using the finite temperature effective action formalism .
+ in this note , starting from eqn .
[ 208 ] of reference @xcite obtained by the functional schrdinger representation ( for the fermionic case ) , we will show that there is a _ finite correction _ to the pair production rate , even in the very large time limit at any nonzero temperature .
this is in direct contradiction to the results obtained in @xcite .
it must be noted here that though under some approximations this result reduces to the results obtained in reference @xcite and @xcite and we shall discuss the implications of that shortly .
+ to be little more precise , we find that in the limit of very small time and high temperature the results of reference @xcite matches with those of @xcite and @xcite found in the high temperature limit . on the other hand in the studies of @xcite and @xcite the production rate
is calculated in the infinite time limit as performed in the strict thermodynamic sense . in this paper
we will also point out that the production rate when properly computed , turns out to be a function of time and it tends to diverge once the infinite time limit is assumed .
in this section we start from the expression of probability @xmath0 , of finding the system with no fermion anti - fermion pairs in a state after a time @xmath1 , as is given in eqn . [
208 ] of hallin and liljenberg @xcite , @xmath2 \label{p}\ ] ]
the first integral here gives the zero temperature schwinger pair production probability and since the controversy in the literature is about the finite temperature piece we will concentrate only on the second term in eqn.[[p3 ] ] . in the second integral the quantity @xmath11 is given by @xmath12 and it also appears in the exponent thus making the evaluation with exact @xmath7 dependence more difficult .
therefore to circumvent this difficulty we propose to rewrite the exponential using an integral transform of the following form @xcite @xmath13 now if we identify @xmath14 with @xmath15 and the quantity @xmath16 as @xmath17 , we can write @xmath18 - \frac{{\bar{\beta}}^2}{4u } } \frac{d u}{u^{\frac{3}{2}}}. \label{p4}\end{aligned}\ ] ] with the aid of eqn.[[p4 ] ] , we obtain from the second integral ( eqn . [
[ p3 ] ] ) upon performing the @xmath19 integration @xmath20 - \frac{{\bar{\beta}}^2}{4u } } \nonumber \\ & = & \frac{\bar{\beta } } { \sqrt{\pi } } e^ { - { \bar{m}}^2 \pi } \int_{-\bar{t_f}}^{0 } d { \bar{p}}^3 \int_{0}^{\infty } \frac{d u}{u^{\frac{3}{2 } } } \frac { e^ { - u \left [ { \bar{m}}^2 + ( \bar{p}^3 + \bar{t_f } ) ^2 \right]- \frac{{\bar{\beta}}^2}{4u } } } { \left ( \pi + u \right ) } .
\label{p5}\end{aligned}\ ] ] hereafter one can interchange the order of the integrations and carry out the @xmath21 integration first . upon performing this integration
the resulting expression can be cast in the form of an error function with @xmath22 as its argument . rewriting the error function as a series we obtain @xmath23 in order to write the pair production rate in a more compact form , we make the following change of variables @xmath24 in eqn.[[sp5 ] ] and use the definition of pair production rate as given in eqn.[[s1 ] ] without taking the limit @xmath9 before hand .
and the resulting expression is , @xmath25 it can be seen from eqn.[[sp6 ] ] that for any nonzero value of @xmath7 the pair production rate , contrary to the claim made in @xcite , is always nonzero . at zero temperature ,
i.e , when @xmath26 , eqn.[[sp6 ] ] goes to zero because of the @xmath27 sitting in the exponent . and
it conforms to our general expectation that as temperature tends to zero the temperature dependent correction to the rate should also vanish . on the other hand at a nonzero temperature , in the thermodynamic sense , as @xmath9 the rate becomes divergent .
this is an unique feature that comes out of the treatment of reference @xcite .
this is the main result of this paper . in the remaining portion of the paper we shall try to evaluate this expression numerically for different values of the parameters and
write down the approximate analytic forms for each of these cases .
in this section we evaluate @xmath28 for three different values of the parameters namely ( a ) @xmath29 and @xmath30 or little more , ( b ) @xmath31 but @xmath32 and @xmath33 and lastly ( c ) both @xmath34 and @xmath35 are @xmath36 but @xmath31 . in the last case we would vary @xmath37 to show how @xmath28 changes with its variation .
+ we shall start with case ( a ) and then move over to the other two cases . in order to derive an approximate analytical form for @xmath28 for this case we start from eqn.[[p5 ] ] instead of eqn.[[sp6 ] ] as given in this text .
since in this case both @xmath7 and @xmath38 are assumed to be @xmath39 and the parameter @xmath40 is of the order of one or slightly more , the quantity @xmath41 } $ ] is very small .
since the integral there varies from @xmath42 to @xmath43 we would like to approximate the quantity @xmath44 in the denominator by u to arrive at @xmath45- \frac{{\bar{\beta}}^2}{4u } } .
\label{p6}\end{aligned}\ ] ]
having discussed the case for @xmath55 and @xmath56 , we will try to evaluate the pair production rate for @xmath31 but @xmath57 and @xmath33 . in order to do that
, it will be convenient if we proceed from eqn.[[sp6 ] ] .
the first important thing to notice here is , as @xmath58 , the dominant contribution to the quantity of our interest comes from the first term in the summation .
hence , for the dominant contribution , it is sufficient to retain the @xmath59 term where @xmath28 can be approximated by @xmath60 + since the dominant contribution to the exponential comes from @xmath61 , which for the parameter range of our interest turns out to be quite large , we will proceed as before by approximating the denominator by @xmath62 and then carrying out the @xmath62 integration using the modified bessel function of fractional order . on using the exact form of the bessel function eqn.[[app2 ] ] approximately comes out to be @xmath63 we have estimated @xmath28 numerically for @xmath64 and @xmath65 for various values of temperature ranging from @xmath66 to @xmath67 and have plotted them in figure [ [ expr ] ] .
it can be seen that the shape of the curve is basically due to the function @xmath68 as is given in eqn.[[app3 ] ] . +
lastly in figure [ [ time ] ] we have plotted @xmath28 for @xmath69 and @xmath70 as a function of @xmath7 , ranging from 100 to 101 in arbitrary units .
one can see from this curve that as time @xmath7 increases @xmath28 also keeps on increasing .
so in the strict thermodynamic limit as @xmath9 the @xmath28 would become divergent .
starting from eqn.[208 ] of ref.@xcite we have shown that for a fermionic system , contrary to the claim made there ( i.e. in @xcite ) , at any non - zero temperature there is a finite correction to schwinger s pair production rate for any value of @xmath7.it is shown that the final expression ( i.e. eqn.[[sp6 ] ] ) comes as a power series in @xmath71 .
though for small values of @xmath34 and @xmath37 , this expression seems to be finite , but in the proper thermodynamic sense it is not .
since for realistic situations @xmath34 can never be zero , on the otherhand @xmath37 in principle can approach infinity , hence in the infinite time limit , the quantity @xmath28 tends to diverge .
the reason behind this apparently contradictory results seems to depend on how one takes the limit @xmath72.in our view one should first carry out all the integrations and write down the final result as a function of @xmath7 , before taking the limit @xmath9 .
though we have carried out this detailed exercise for fermions only but in principle this can be extended for bosons too .
+ we would like to mention here that the same quantity obtained using the finite temperature effective action formalism ( where infinite time limit is implicitly assumed ) using both thermo field dynamics and imaginary time formalisms ( see e.g. @xcite and @xcite ) yields a nonzero finite result . in view of the previous results we find this corrected result of reference @xciteto
be quite amazing and in our view this deserves further study .
particularly one should try to understand why the different formalisms are giving different answers and which one of them is correct .
but in case the corrected result of reference @xcite which is obtained through the elegant formalism of functional schrdinger representation turns out to be right ( though we have some reservatio ns about it ) , it can certainly be used to estimate the thermal contribution to the soft processes dominated production mechanisms in the pre - equilibrium quark gluon plasma production phase of relativistic heavy ion collisions , at the sps and rhic .
its a pleasure to thank sanjay jain and apoorva patel for many interesting discussions and encouragements and very useful suggestions made during the course of this work .
the author would also like to thank s. konar , for the help rendered during the preparation of this manuscript .
cox , w.s .
hellman and a. yildiz , ann .
phys.(n.y . ) 154 , 211 , ( 1984 ) m. loewe and j.c .
rojas , phys .
rev . , 2689 , ( 1992 ) joakim hallin and per liljenberg , phys
1150,(1995 ) a. k. ganguly , p.k.kaw and j.c.parikh , phys . rev .
( 1995 ) 2091 .
p. elmfors and b .- s.skagerstam , phys .
lett . , 141 , ( 1995 ) .
j. schwinger , phys .
82.664(1951 ) .
asymptotic expansions : their derivation and interpretation _ academic press ( london and new york , 1973 ) | we show that contrary to the claim made by hallin and liljenberg in phys . rev .
1150,(1995 ) , ( hep - th/9412188 ) the thermal correction to the _ thermal decay or pair production rate _ for a system placed in a heat bath in the presence of an external electric field , is always nonzero in the finite as well as infinite time limit . using the formalism outlined there ,
we reestimate the decay rate for different values of temperature , mass and time.we also try to identify the parameter ranges where the quantity of interest agrees with that computed previously , at high temperature ( in the infinite time limit ) , from the imaginary part of the effective action using imaginary time and real time formalism of thermal field theory .
we also point out that in the strictly infinite time limit , the correct decay rate as obtained from the work of hallin et .
al . tends to diverge .
= 0.25 cm = 0.25 cm * comment on fermionic and bosonic pair - creation in an external electric field at finite temperature * + [ 16pt ] avijit kumar ganguly + centre for theoretical studies + indian institute of science , bangalore 560012 + |
the atlas and cms collaborations @xcite have narrowed down the allowed range of a light higgs mass to the region 115 - 131 gev .
in addition there are hints of the higgs mass being near @xmath1gev with sm like decay widths into @xmath2 and @xmath3 .
a light higgs is favored in supersymmetry although the mssm predicts a tree level upper bound on the lightest cp - even higgs mass as @xmath4 . within mssm , loop corrections can give required large corrections to higgs mass provided the stop is heavier than 1 tev or there is near maximal stop mixing .
implications of the 125 gev higgs for the mssm and constrained - mssm parameter space have been extensively studied @xcite .
going beyond mssm , in order to get a larger tree - level higgs mass , the simplest extension is a singlet superfield in the nmssm model @xcite .
the singlet interaction with the two higgs doublet of mssm is via the @xmath5 term .
the higgs mass is now given by the relation @xmath6 , where @xmath7 is due to radiative - correction .
taking @xmath8 ( larger values would make it flow to the non - perturbative regime much below the gut scale ) and @xmath9 the radiative correction needed to get a 125 gev higgs mass is @xmath10gev which is an improvement over the @xmath11gev needed in the mssm
. however fine - tuning of the stop mass is still required in nmssm to get the required higgs mass @xcite . also by extending the mssm gauge group in a suitable way
, the new higgs sector dynamics can push the tree - level mass well above the tree - level mssm limit if it couples to the new gauge sector @xcite . in most of the cases
the non - decoupling d - terms contribute non - trivially to increase the tree - level mass of the sm - like higgs boson .
recent anaysis of the susy model based on @xmath12 gauge group @xcite has shown that the tree level physical higgs mass can be atmost 110 gev and through the one - loop correction it can be raised considerably . another recent work on mssm extended by a u(1 )
gauged peccei - quinn symmetry @xcite where the new d - terms can raise the tree - level mass well enough to accommodate the 125 gev higgs boson without significant radiative correction and hence requires less fine - tuning .
an important aspect of the 125 gev higgs mass is that the parameter space of thermal relic for dark matter is severely restricted . in mssm ,
the lsp is a higgsino at the tev scale @xcite . in nmssm , susy partner of the singlet scalar -the singlino mixes with the neutralinos to provide a light dark matter
recent analysis @xcite has shown that the benchmark parameters which give a 125 gev higgs also provide a neutralino dark matter candidate with mass in the range of 68 - 85 gev . to our knowledge ,
the dark matter in triplet - extended mssm has not been studied so far .
the extension of mssm by extending it with a @xmath13 and @xmath14 su(2 ) triplet superfields has been studied @xcite where the tree level contribution to the higgs mass from the triplet higgs sector has been calculated .
it has been shown in @xcite that with the @xmath13 triplet superfield the tree - level higgs mass can be raised to 113 gev which would still require substantial loop corrections from stops .
recently , the mssm extended by two real triplets ( @xmath15 ) and one singlet@xcite has been studied with a motivation to solve the @xmath16-problem as well as to obtain a large correction to the lightest higgs mass .
the analysis of the dark matter sector of this model will be complicated as the lsp will be the lightest eigenstate of the @xmath17 neutralino mass matrix which has not yet been done . in this paper
we explore the minimal extension of the mssm which can give a tree - level higgs mass of 119 - 120 gev .
we find that by extending the mssm by adding a singlet and a @xmath13 su(2 ) triplet superfields , this aim can be achieved .
the upper bound on the tree - level mass of the lightest cp - even higgs is given in equation ( [ bound ] ) . with this tree level higgs
mass the stop mass need not be very heavy and this solves the fine - tuning problem of the higgs mass in mssm and nmssm @xcite .
we also study the dark matter candidates in this model which is obtained by diagonalizing the @xmath18 neutralino mass matrix .
we find a viable dark matter with mass 100 gev , which is a mixture of the higgsino and triplino ( the fermionic partner of the neutral component of the triplet higgs ) .
we fix two sets of benchmark parameters at the electroweak scale which would give a the 125 gev and dark matter relic density @xmath19 compatible with wmap-7 measurements @xcite .
we find that the direct detection cross section of the dark matter is @xmath20 , which is compatible with the direct detection experiments like xenon100 @xcite . in section([model ] ) we display the superpotential of our model and we derive the scalar potential from the d - terms and f - terms and from the various soft - breaking terms . in section([higgs ] ) we give a detailed analysis of the higgs sector and we calculate the cp - even , cp - odd and charged higgs mass matrices . in section([neut ] ) , the neutralino and the chargino mass matrices are discussed .
the numerical results based on this model are discussed in detail in section([results ] ) .
we show the results for two sets of benchmark points which include the parameters like couplings , tri - linear soft breaking terms , soft masses and the fermionic and scalar mass spectrum .
we have also taken into account the one - loop corrections to the lightest physical higgs mass and shown a quantitative improvement of the level of fine - tuning compared to other models . in section([dm ] ) we discuss the dark matter from the neutralino sector of this model and its phenomenology , which is one of the main results of this paper . in the concluding section we summarize the results and we point out some directions for further study of the triplet - singlet model which will enable the model to be tested at the lhc .
in this model , we have extended the superpotential of the minimal supersymmetric standard model by adding one singlet chiral superfield @xmath21 and one su(2 ) triplet chiral superfields @xmath22 with hypercharge @xmath23 .
the most general form of the superpotential for this singlet - triplet extended model can be written as , where , @xmath24 are the higgs doublets of the mssm and the yukawa superpotential @xmath25 is given as , in terms of the components , we have @xmath26 , @xmath27 and @xmath28 here , @xmath29 , which would not have been true for real higgs triplet in non - supersymmetric models .
we can solve the @xmath16-problem by starting with a scale invariant superpotential , given as where the su(2 ) invariant dot product is defined as , this superpotential([w_s ] ) also has an accidental @xmath30-symmetry , i.e. invariance of the superpotential on multiplication of the chiral superfields by the factor of @xmath31 . by ,
this choice we are eliminating the @xmath16-parameter but an effective @xmath16-term is generated when the neutral components of @xmath21 and @xmath32 acquire vev s @xmath33 and @xmath34 respectively , @xmath35 therefore , in terms of the neutral components of the super - fields the equation ( [ w_s ] ) sans @xmath36 can be re - written as , the scalar potential involving only higgs field can be written as , @xmath37 in the above equation , @xmath38 consists of the soft - supersymmetry breaking term associated with the superpotential in equation([w_s ] ) , is given by in equation([scalarv ] ) @xmath39 is the supersymmetric potential from f - terms , given by whereas the f - term for the neutral scalar potential can be derived from equation([w - neu ] ) as , where , @xmath40 stands for @xmath41 and @xmath42 is the scalar counter - part of the neutral superpotential @xmath43 .
finally , @xmath44 is supersymmetric potential from d - terms in equation ( [ scalarv ] ) , given by after electroweak symmetry breaking , only the neutral components of the scalars fields acquire vev s , i.e , @xmath45 , @xmath46 , @xmath47 and @xmath48 the neutral - scalar part of the chiral superfields can be decomposed into real and imaginary parts , the minimization conditions are derived from the fact that , @xmath49 we can determine the soft breaking mass parameters like @xmath50 , @xmath51 , @xmath52 and @xmath53 using the following minimization conditions , where , @xmath54 \label{aeff}\ ] ] and @xmath55 , @xmath56 . due to the addition of the triplets ,
the gauge bosons receive additional contribution in their masses like , @xmath57 the @xmath58-parameter at the tree - level is defined as , @xmath59 clearly , the @xmath60-parameter deviates from unity by a factor of @xmath61 .
using the recent bound on @xmath60-parameter at 95@xmath62 c.l .
we can determine the bound on the triplet higgs vev @xmath34 .
@xmath60 can be confined in the range 0.9799 - 1.0066 @xcite and hence @xmath63 gev at 95@xmath62 c.l .
the symmetric cp - even higgs mass matrix is written in the basis of ( @xmath64 , @xmath65 , @xmath66 , @xmath67 ) with 10 independent components .
after electroweak symmetry breaking ( ewsb ) the entries of the squared mass - matrix are , where @xmath68 s are defined as , and @xmath69 is defined in equation([aeff ] ) .
* _ bound on the lightest higgs mass : _ the bound on the lightest higgs mass is derived from the fact that , the smallest eigenvalue of a real , symmetric @xmath70 matrix is smaller than the smallest eigenvalue of the upper left @xmath71 sub - matrix@xcite . using this
we obtain an upper bound on the lightest cp - even higgs mass , @xmath72 \label{bound}\end{aligned}\ ] ] the bound on lightest higgs mass has been considerably improved over the mssm due to the additional contribution from the singlet and triplet gauge fields . using equation([bound ] )
we can put constraints on the parameters like @xmath73 , @xmath74 and @xmath75 satisfying the recent bound on higgs mass from atlas and cms .
the elements of the @xmath76 cp - odd higgs squared mass matrix , after ewsb , in the basis of ( @xmath77 , @xmath78 , @xmath79 , @xmath80 ) are , where , @xmath81 this matrix always contains a goldstone mode @xmath82 ( gives mass to z - boson ) , which can be written as , @xmath83 and we rotate the mass matrix in the basis ( @xmath82 , @xmath84 , @xmath85 , @xmath86 ) where , @xmath87 after removing the goldstone mode , we again rotate the remaining 3@xmath883 mass matrix and finally obtain , where @xmath89 , @xmath90 , @xmath91 are the massive modes .
the charged higgs sector comprises of a @xmath92 symmetric matrix , written in the basis ( @xmath93 , @xmath94 , @xmath95 , @xmath96 ) , which has 10 independent components , ( after ewsb ) given by where @xmath97 s are defined as , @xmath98 after diagonalization , we obtain one massless goldstone state @xmath99 ( gives mass to @xmath100-boson , since @xmath101 ) , @xmath102 and three other massive modes like @xmath103
the neutralino mass matrix extended by the singlet and triplet sector , in the basis ( @xmath104 ) is given by , @xmath105 where , @xmath106 , @xmath107 are the soft breaking mass parameters for bino and wino respectively and @xmath108 , @xmath109 , @xmath110 and @xmath111 the left - most 4@xmath884 entries are exactly identical with that in mssm , except the @xmath112-term which is defined in equation([mu ] ) . as the triplet and
the singlet fermion does not have any interaction with the neutral gauginos the right - most 2@xmath882 entries are zero .
+ the chargino mass terms in the lagrangian can be written as , @xmath113\ ] ] where , the basis @xmath114 and @xmath115 are specified as , @xmath116 , @xmath117 and the chargino matrix in the gauge basis is given by , @xmath118
[ cols="^,^ " , ] in both the cases we do not require large contribution from the radiative corrections to raise the lightest physical higgs mass so as to satisfy the value of 125 gev .
this in turn implies that the contribution from the stop - top sector is not significant as in the case of mssm .
in fact in absence of fine tuning the correction to lightest physical higgs mass from the stop - top sector is given by , @xmath119 for , @xmath120 and @xmath121 being @xmath122 ( 200 gev ) , this amounts to a correction of only a few gev .
in this model , the lightest physical higgs mass at the tree - level is boosted compared to nmssm , and other triplet extended model @xcite as it gets contribution from both singlet and triplet sector([bound ] ) .
therefore , we can obtain a higgs boson close to 125 gev even at the tree level . after including the leading order radiative corrections from the stop - top and triplet sector , we get @xmath123 where , @xmath120 and @xmath121 are the soft masses of the stops
, @xmath124 is the soft trilinear coupling , @xmath125 is the yukawa coupling and q is the fundamental scale of susy - breaking . + the fine - tuning parameter can be quantified @xciteas , @xmath126 in case of mssm ( only first term in eqn.[fine ] is present ) , we have @xmath127 but the tree - level bound on higgs mass is @xmath128 .
therefore , one is forced to consider large values for @xmath120 , @xmath121 and @xmath124 , say 1 tev in order to raise the lightest physical higgs boson mass upto 125 gev . in this case ,
@xmath129 and thus it leads to maximal stop mixing .
+ in nmssm , the radiative correction needed to get a 125 gev higgs mass is @xmath130gev .
there is no doubt an improvement over mssm , but still fine tuning is required in the stop - top sector@xcite . in the model with one triplet @xcite , the lightest physical higgs mass can be raised to 113 gev . here , the required value of radiative correction is @xmath131 gev .
now , the fine tuning due to the triplet sector is , @xmath132 where @xmath133 . the value of @xmath134 can be as large as 40 .
therefore , this model can no longer be considered as a zero fine - tuning model .
+ now coming to our model , we require @xmath135 gev only- here we see a distinct improvement of 20 - 50 gev compared to other models discussed so far . also , @xmath136 being comparable to @xmath125 , we do not need heavy stops or large stop - top mixing to get the required higgs mass .
for example , @xmath137 gev , @xmath138 gev and @xmath139 tev , we obtain @xmath140 .
thus , we can achieve little fine - tuning compared to other models , since the lightest physical higgs mass can be large at tree level and does not require large contribution from the radiative corrections . here
we note that , the higgs - triplet - higgs coupling @xmath74 ( 0.75 and 1.02 ) becomes non - perturbative at gut scale .
but , these choices of @xmath74 actually helps to raise the higgs mass close to 125 gev at the tree - level .
other alternative could be of course having small @xmath74 , but then we would require large radiative corrections .
therefore , we improve the level of fine - tuning at the cost of giving up perturbativity of @xmath74 at gut scale .
we have analyzed the neutralino sector where the lightest neutralino ( lsp ) , is a mixture of higgsino - triplino and turns out to be a viable dark matter candidate .
the components of @xmath141 ( for @xmath142 ) , i.e. the lsp are , @xmath143 since the lsp has mass @xmath144(100 gev ) , there are two possibilities of final states into which it can annihilate , i.e. ( i)fermion final states and ( ii)gauge boson final states . for annihilation into fermions , except
@xmath145 it can go to any other @xmath146 pairs via pseudo - scalar higgs , z - boson exchange and sfermion exchange .
but , if we consider the neutralino to be more like triplino , then its coupling with z - boson is forbidden .
generally , it can annihilate into gauge boson pairs via several processes like chargino exchange , scalar higgs exchange and z - boson exchange .
but the dominant contribution comes from annihilation into @xmath100 via chargino exchange , which finally leads to the relic density of 0.117 , consistent with wmap @xcite . the scalar interaction between the dark matter ( i.e neutralino lsp ) and the quark is given by , @xmath147 where @xmath148 is the coupling between the quark and the neutralino .
the scalar cross section for the neutralino scattering off a target nucleus ( one has to sum over the proton and neutrons in the target ) is given by , @xmath149 where , @xmath150 is the reduced mass of the nucleon and @xmath151 is the neutralino coupling to proton or neutron@xcite , given by @xmath152 where @xmath153 and @xmath154 @xcite .
@xmath155 is related to these values by @xmath156 the term in eq .
[ scalarterms ] which includes @xmath155 results from the coupling of the wimp to gluons in the target nuclei through a heavy quark loop
. we can approximate @xmath157 where , @xmath158 is the product of different coupling and mixings , @xmath159 is the mass of the quark and @xmath160 ( @xmath161 being the dark matter mass )
. the parameter @xmath158 plays a crucial role in determining the spin - independent cross - section and is highly model dependent .
using this we estimate @xmath162 and the value of the spin - independent cross - section is @xmath0 , which is below the exclusion limits of xenon100@xcite and other direct detection experiments .
in this paper we have explored an extension of mssm where the higgs sector is extended by a singlet and a @xmath13 triplet superfield .
this is the minimal model which gives a tree level higgs mass of @xmath122(119 - 120 gev ) and the one - loop correction can easily raise it to 125 gev without significant contribution from the stop - top sector .
however , @xmath136 ( at electroweak scale ) becomes non - perturbative at the gut scale , while all other couplings remain perturbative upto gut scale - on the other hand this is the price we pay to retain small fine - tuning . in addition , we see that the triplino and singlino contributions to the neutralino mass matrix gives a viable dark matter candidate with mass around @xmath163 gev which may be seen at the lhc from the missing transverse energy signals @xcite . in mssm and
nmssm the problem for getting the correct relic density of dark matter is related to the necessity of choosing chargino and scalar masses to be in the multi tev scale to fit the higgs mass from radiative corrections .
the dm mass in mssm is around 700 gev while in nmssm it is possible to obtain viable dm in the 100 gev range .
the main advantage of our model for the dark matter is that since the sparticle masses need not be very large compared to the electroweak scale the wimp miracle is restored and we are able to get dm mass in the 100 gev range over a large parameter space of our model .
the data from lhc with integrated luminosity of @xmath164 has not only given an indication the higgs mass but there is also a measurement of the higgs decay branchings into different channels .
detailed analysis @xcite of the 125 gev higgs branching fractions seen at the lhc indicates that the signal ratio for higgs decay into two photons is larger than sm prediction by a factor of @xmath165 , decay into @xmath166 and @xmath167 channels is smaller than sm by a factor of @xmath168 and into @xmath169 and @xmath170 channels it is factor @xmath171 consistent with sm .
the lightest cp - even higgs ( [ higgs - frac ] ) has a sizable fraction of the singlet and the higgs decay phenomenology will be distinguishable from the mssm @xcite and likely to be similar to the nmssm scenario @xcite .
but , there will be some contribution from the triplet sector too .
the phenomenological aspects of the real - triplet extended sm has been studied in @xcite .
more data from lhc will pinpoint or rule out the extended higgs sector models and it would be useful to study the singlet - triplet extended mssm model in greater detail with emphasis on the lhc signal in the future .
12 g. aad _ et al . _ [ atlas collaboration ] , phys .
b * 710 * , 49 ( 2012 ) [ arxiv:1202.1408 [ hep - ex ] ] .
s. chatrchyan _ et al . _
[ cms collaboration ] , phys .
b * 710 * , 26 ( 2012 ) [ arxiv:1202.1488 [ hep - ex ] ] .
h. baer , v. barger and a. mustafayev , arxiv:1112.3017 [ hep - ph ] ; + s. heinemeyer , o. stal and g. weiglein , phys .
b * 710 * , 201 ( 2012 ) [ arxiv:1112.3026 [ hep - ph ] ] ; + a. arbey , m. battaglia , a. djouadi , f. mahmoudi and j. quevillon , phys . lett .
b * 708 * , 162 ( 2012 ) [ arxiv:1112.3028 [ hep - ph ] ] ; + p. draper , p. meade , m. reece and d. shih , arxiv:1112.3068 [ hep - ph ] ; + a. arbey , m. battaglia and f. mahmoudi , eur .
j. c * 72 * , 1906 ( 2012 ) [ arxiv:1112.3032 [ hep - ph ] ] ; + m. kadastik , k. kannike , a. racioppi and m. raidal , arxiv:1112.3647 [ hep - ph ] ; + j. cao , z. heng , d. li and j. m. yang , arxiv:1112.4391 [ hep - ph ] ; + j. ellis and k. a. olive , arxiv:1202.3262 [ hep - ph ] ; + m. a. ajaib , i. gogoladze , f. nasir and q. shafi , arxiv:1204.2856 [ hep - ph ] ; + f. brummer , s. kraml and s. kulkarni , arxiv:1204.5977 [ hep - ph ] .
m. drees , int .
j. mod .
phys . a * 4 * , 3635 ( 1989 ) ; + u. ellwanger , c. hugonie and a. m. teixeira , phys
. rept .
* 496 * , 1 ( 2010 ) ; + g. g. ross and k. schmidt - hoberg , arxiv:1108.1284 [ hep - ph ] ; + l. j. hall , d. pinner and j. t. ruderman , arxiv:1112.2703 [ hep - ph ] ; + s. f. king , m. muhlleitner and r. nevzorov , nucl .
b * 860 * , 207 ( 2012 ) [ arxiv:1201.2671 [ hep - ph ] ] ; + z. kang , j. li and t. li , arxiv:1201.5305 [ hep - ph ] ; + j. cao , z. heng , j. m. yang , y. zhang and j. zhu , jhep * 1203 * , 086 ( 2012 ) [ arxiv:1202.5821 [ hep - ph ] ] ; + d. a. vasquez , g. belanger , c. boehm , j. da silva , p. richardson and c. wymant , arxiv:1203.3446 [ hep - ph ] ; + u. ellwanger and c. hugonie , arxiv:1203.5048 [ hep - ph ] .
s. f. king , m. muhlleitner and r. nevzorov , nucl .
b * 860 * , 207 ( 2012 ) [ arxiv:1201.2671 [ hep - ph ] ] .
h. e. haber and m. sher , phys .
d * 35 * , 2206 ( 1987 ) ; + m. drees , phys .
d * 35 * , 2910 ( 1987 ) ; + k. s. babu , x. -g .
he and e. ma , phys .
d * 36 * , 878 ( 1987 ) ; + m. cvetic , d. a. demir , j. r. espinosa , l. l. everett and p. langacker , phys .
d * 56 * , 2861 ( 1997 ) [ erratum - ibid .
d * 58 * , 119905 ( 1998 ) ] [ hep - ph/9703317 ] ; + p. batra , a. delgado , d. e. kaplan and t. m. p. tait , jhep * 0402 * , 043 ( 2004 ) [ hep - ph/0309149 ] ; + a. maloney , a. pierce and j. g. wacker , jhep * 0606 * , 034 ( 2006 ) [ hep - ph/0409127 ] ; + y. zhang , h. an , x. -d .
ji and r. n. mohapatra , phys .
d * 78 * , 011302 ( 2008 ) [ arxiv:0804.0268 [ hep - ph ] ] .
m. hirsch , m. malinsky , w. porod , l. reichert and f. staub , jhep * 1202 * , 084 ( 2012 ) [ arxiv:1110.3037 [ hep - ph ] ] . h. an , t. liu and l. -t .
wang , arxiv:1207.2473 [ hep - ph ] .
h. baer , v. barger and a. mustafayev , arxiv:1202.4038 [ hep - ph ] .
p. draper , t. liu , c. e. m. wagner , l. -t .
wang and h. zhang , phys .
lett . *
106 * , 121805 ( 2011 ) [ arxiv:1009.3963 [ hep - ph ] ] .
j. -j .
cao , k. -i .
hikasa , w. wang , j. m. yang , phys .
b * 703 * , 292 ( 2011 ) [ arxiv:1104.1754 [ hep - ph ] ] .
j. r. espinosa and m. quiros , phys .
b * 279 * , 92 ( 1992 ) .
j. r. espinosa and m. quiros , nucl .
b * 384 * , 113 ( 1992 ) .
s. di chiara and k. hsieh , phys .
d * 78 * , 055016 ( 2008 ) [ arxiv:0805.2623 [ hep - ph ] ] .
k. agashe , a. azatov , a. katz and d. kim , phys .
d * 84 * , 115024 ( 2011 ) [ arxiv:1109.2842 [ hep - ph ] ] .
s. r. coleman and e. j. weinberg , phys .
d * 7 * , 1888 ( 1973 ) .
r. barbieri and g. f. giudice , nucl .
b * 306 * , 63 ( 1988 ) .
u. ellwanger , g. espitalier - noel and c. hugonie , jhep * 1109 * , 105 ( 2011 ) [ arxiv:1107.2472 [ hep - ph ] ] .
d. larson _
et al . _ ,
astrophys .
j. suppl .
* 192 * , 16 ( 2011 ) [ arxiv:1001.4635 [ astro-ph.co ] ] . l. baudis and f. t. x. collaboration , arxiv:1203.1589 [ astro-ph.im ] . g. jungman , m. kamionkowski and k. griest , phys .
rept . * 267 * , 195 ( 1996 ) [ hep - ph/9506380 ] .
g. bertone , d. hooper and j. silk , phys .
rept . * 405 * , 279 ( 2005 ) [ hep - ph/0404175 ] .
j. r. ellis , a. ferstl and k. a. olive , phys .
b 481 , ( 2000 ) 304 , [ arxiv : hep - ph/0001005 ] .
m. m. nojiri , g. polesello and d. r. tovey , jhep * 0603 * , 063 ( 2006 ) [ hep - ph/0512204 ] ; + p. nath , b. d. nelson , h. davoudiasl , b. dutta , d. feldman , z. liu , t. han and p. langacker _ et al .
_ , nucl .
* 200 - 202 * , 185 ( 2010 ) [ arxiv:1001.2693 [ hep - ph ] ] .
giardino , k. kannike , m. raidal and a. strumia , arxiv:1203.4254 [ hep - ph ] . | we study the extension of minimal supersymmetric standard model by adding one singlet and one hypercharge zero su(2 ) triplet chiral superfield .
the triplet sector gives an additional contributions to the scalar masses and we find that the lightest cp - even higgs boson can have a mass of 119 - 120 gev at the tree level , and radiative correction raises the value to 125 gev . in this model
no significant contributions from stop loops is needed to get the required higgs mass which alleviates the fine tuning problem of fixing the stop mass to a high precision at the gut scale .
in addition this model gives a neutralino dark matter of mass around 100 gev which is a mixture of higgsino and triplino with a dark matter density consistent with wmap observations .
the spin - independent scattering cross - section with nucleons is @xmath0 , which makes it consistent with the bounds from direct detection experiments like xenon100 and others . |
in gauge theories , the coulomb gauge has special position .
the number of dynamical variables is the same as the number of physical degrees of freedom . moreover ,
if we go to the hamiltonian , phase - space , first - order formalism , there are no ghosts ; and , because of the existence of a hamiltonian , unitarity should be manifest .
nevertheless , there are complications , if not problems , with the coulomb gauge . in the lagrangian , second - order , formalism , there are ` energy - divergences ' in individual feynman graphs
. these are divergences over the energy - integration , @xmath0 , in a loop , for fixed values of the 3-momentum @xmath1 .
these are difficult to regularize : dimensional regularization does not touch them , and any other regularization risks doing violence to gauge - invariance ( but see leibbrandt et al for a modified form of dim .
these energy - divergences do cancel when all graphs are combined @xcite ; but it makes one uneasy to be manipulating divergent and unregulated integrals .
the problem of energy - divergences is eased by going to the hamiltonian , phase - space , first - order formalism , in which time derivatives of the gluon field @xmath2 are eliminated in favour of the conjugate momentum field @xmath3 .
this has the advantage of being a true hamiltonian formalism , and unitarity should be manifestly obeyed . also , there are no ghosts ( ghost loops cancel part of the closed coulomb loops ) . for a sample calculation in this formalism
see @xcite , and for possible connection to confinement see @xcite .
but there are still problems .
there is a question of operator - ordering in the hamiltonian @xcite ( see also @xcite , @xcite ) , which may require higher - order terms .
it has been shown @xcite that these operator - ordering problems are connected with ambiguous multiple energy - integrals in higher orders . in this paper , we are concerned with a simpler problem , which arises at 2-loop order . in general , in the hamiltonian formalism to this order , the two integrals over the internal energies converge with the two internal spatial momenta held fixed . however , renormalization demands that we first perform the energy and momentum integrals for each subgraph , then make subtractions for ultraviolet divergences , and then perform the remaining energy and momentum integrals . with this sequence of operations ,
one does in general find an energy - divergence in the final energy integral .
we illustrate this with a simple example in which quark - loop subgraphs are inserted into the second - order gluon self - energy graphs .
we perform the energy - momentum integral in the subgraph first , then do the renormalization subtraction .
individual graphs now have energy - divergences in the final energy integral , but these cancel when graphs are combined .
we show that the cancellation is a consequence of the ward identities obeyed by the quark - loop sub - diagrams .
of course , it is reassuring to check that the energy - divergences do cancel .
but we are back in the uncomfortable position of having to handle divergent ( and unregularized ) integrals at intermediate stages of the cancellation .
this contrasts with the feynman gauge , where all integrals to all orders are unambiguously regularized by dimensional regularization .
*p * bf
p lorentz indices are denoted by greek letters , spatial indices by @xmath4 , and colour indices by @xmath5 .
we use the metric tensor @xmath6 , where @xmath7 lorentz vectors are written @xmath8 we define the lorentz tensor @xmath9 by @xmath10 and for any vector @xmath11 we define the vector @xmath12 by @xmath13 ( of course , these definitions refer to the particular time - like vector @xmath14 with respect to which we have chosen to define the coulomb gauge . )
we define a spatial transverse tensor @xmath15 by @xmath16 so that @xmath17 in terms of colour matrices @xmath18 , @xmath19 is defined by @xmath20 and in terms of the structure constants @xmath21 , @xmath22 is defined by @xmath23 the renormalized coupling constant is @xmath24 .
quarks have mass @xmath25 .
we use dimensional regularization with spacetime dimension @xmath26 .
the lagrangian density in the phase - space formalism is @xmath27 where @xmath28.\ ] ] the hamiltonian form of the coulomb gauge has dynamical , conjugate fields @xmath29 .
the coulomb potential @xmath30 is not a dynamical variable
. it could be eliminated , but we find it convenient to leave it in the feynman rules .
we use a continuous line to represent @xmath3 , a dashed line for @xmath2 and a dotted line for @xmath30 .
the propagators are not diagonal in @xmath31 .
the propagators are shown in fig.1 ( the arrows on the lines show the direction of the momentum @xmath32 ) .
only one vertex will be relevant to our calculation , that shown in fig.2 .
, dashed lines @xmath33 and dotted lines @xmath34 .
the arrow indicates the sense of momentum flow.,width=234 ] its value is @xmath35 in the hamiltonian formalism , this is the only vertex involving the coulomb field .
it is this feature which implies there are no energy - divergences to 1-loop order .
we are going to insert quark loops into a gluon diagram , so we need a notation for quark 1-loop effective action .
let the 2-gluon term in this effective action be ( in momentum space ) @xmath36 we will require to know @xmath37 only for @xmath38 in this region , we have @xmath39,\ ] ] ( using minimal subtraction with a mass unit @xmath40 ) . the 3-gluon term in the effective action will be denoted by @xmath41 where @xmath42 , and the quantum numbers of the three gluons are @xmath43 ( all momenta are directed into the vertex ) .
we will not need to know the value of @xmath44 in general .
finally the 4-gluon term in the effective action will be denoted as @xmath45 where the quantum numbers are @xmath46 .
again , we do not need to know the value of @xmath47 in general .
both @xmath48 and @xmath47 have the symmetries required for bose symmetry .
note that the effective action due to quark loops is a functional of @xmath49 , and does not depend upon @xmath3 .
there are terms in the effective action which involve the coulomb field @xmath30 , and this is the reason that energy - divergences re - appear .
renormalization requires also the presence of counter - terms with a different structure from the interactions in the original formalism .
the effective action obeys the following ward identities : @xmath50,\ ] ] @xmath51 @xmath52 these identities express the gauge invariance of the quark loop contribution to the effective action .
they are a special case of the brs identities ( see for example itzykson and zuber equation ( 12 - 144 ) ) when there are no ghost contributions @xcite .
we shall show that these identities are sufficient to ensure the cancellation of energy - divergences between graphs .
the simplest example of the energy - divergences occurs in the gluon 2-point function , to 2-loop order .
the relevant graphs are shown in fig.3 , where the thick circles represent terms from the quark - loop effective action ( in ( 12 ) , ( 14 ) and ( 15 ) ) . in fig.3(vi ) , the sum of the three subgraphs shown corresponds to ( 15 ) .
we then see that ( 19 ) , ( 20 ) and ( 21 ) are each divergent as integrals over @xmath62 for fixed @xmath1 .
( actually , for @xmath63 this is true only of the contributon from the subtraction term in ( 13 ) . )
if we take the limit @xmath64 first , then we get a double log energy - divergence . to find the behaviour of the integrals in ( 22 ) , ( 23 ) and ( 24 ) , it is sufficient to use the ward identities ( 16 ) and ( 17 ) in the large @xmath62 limit
. then ( 16 ) gives @xmath65 also , @xmath66 @xmath67 similarly , ( 17 ) implies that @xmath68\ ] ] @xmath69 using ( 27 ) again . from ( 27 ) , ( 28 ) and ( 29 ) , we see that all the integrals in ( 18 ) have energy - divergences of the same kind , and the divergent part has the form @xmath70,\ ] ] where @xmath71 @xmath72 @xmath73 @xmath74@xmath75 @xmath76 these last six expressions cancel , so the energy - divergences in the separate terms in ( 18 ) cancel out in the sum .
probably similar cancellations occur in two - loop graphs made entirely of gluon lines .
but in this case there is the extra complication of the ambiguous integrals connected to the christ - lee terms @xcite @xcite .
+ acknowledgement .
this work was supported by mzos of the republic of croatia under contract no . 0098003
( a.a . ) .
we are grateful to dr .
g. duplani for drawing the figures .
99 g. heinrich and g. leibbrandt , nucl .
b * 575 * , 359 ( 2000 ) r.n .
mohapatra , phys .
rev . d4 * 22*,378 , 1007 ( 1971 ) a. andrai , eur .
j. c * 37 * , 307 - 313 ( 2004 ) a. cucchieri , d. zwanziger , phys .
d * 65 * , 014002 ( 2002 ) n. christ , t. d. lee , phys .
rev . d*22 * , 939 ( 1980 ) p. doust , ann . of phys . *
177 * , 169 ( 1987 ) ; p. doust , j. c. taylor , phys .
lett . * 197 * , 232 ( 1987 ) j. c. taylor , in physical and nonstandard gauges , proceedings , vienna , austria 1989 , edited by p. gaigg , w. kummer , m. schweda j. schwinger , phys . rev . *
127 * , 234 ( 1962 ) ; phys . rev . * 130 * 406 ( 1963 ) h. cheng , e. c. tsai , phys
b * 176 * 130 ( 1986 ) ; phys .
* 57 * 511 ( 1986 ) c. itzykson and j. b. zuber , quantum field theory , international series in pure and applied physics , mcgraw - hill inc . | in the coulomb gauge of nonabelian gauge theories there are in general , in individual graphs , ` energy - divergences ' on integrating over the loop energy variable for fixed loop momentum .
these divergences are avoided in the hamiltonian , phase - space formulation .
but , even in this formulation , energy - divergences re - appear at 2-loop order .
we show in an example how these cancel between graphs as a consequence of ward identities .
+ pacs numbers : 11.15.bt ; 11.10.gh + keywords : coulomb gauge ; renormalization ; qcd |
twin - screw devices are widely applied in various industries , including polymer processing , rubber compounding , food processing , and pharmaceutical development , because a high mixing performance can be achieved by the inter - meshing configuration of twin screws @xcite . in the plastics industry ,
twin - screw extruders are used to develop polymer materials with certain desired properties by mixing different fillers and additives with polymer melts . for this melt - mixing process
, several types of screw elements have been developed for different material processabilities and different mixing qualities , since the quality of the mixing process is primarily determined by the geometric structure of the mixing element . in selecting and/or developing a mixing element ,
one fundamental issue is a systematic understanding of the relation between the melt flow driven by the geometric structure of the mixing element and the mixing characteristics . among various types of mixing elements @xcite ,
kneading disks ( kd ) are the most commonly used due to their mixing efficiency . in order to tune the mixing characteristics of conventional kd for application to various purposes ,
a modified type of kd has been proposed , called `` pitched - tip kneading disks '' ( ptkd ) @xcite . in ptkd ,
two disk tips are pitched to the screw axes , whereas the tips in the conventional kd are parallel to the screw axes .
the combined choice of the tip angle and the disk - stagger angle can make it possible to control both the distributive and dispersive mixing abilities better than in conventional kd . in the previous study @xcite
, it has been found that different combinations of tip angle and disk - stagger angle actually lead to different mixing characteristics . nonetheless , the essential role of the tip angle , specifically its direction and size , on the modification of the flow pattern and the resulting mixing characteristics are still unclear .
in this article , we investigate how pitched tips modify the mixing characteristics of conventional kd in twin - screw extrusion . for this purpose
, we consider different directions and sizes of the tip angle on a neutrally staggered kd , and discuss their mixing characteristics using numerical simulation of the polymer - melt flow in the kneading zone .
conventional kneading disks ( kd ) are composed of several oval blocks combined with a stagger angle between two adjacent blocks ( fig . [ fig1](c ) ) .
the tips of the blocks used in conventional kd are parallel to the screw axes .
in contrast to this , in pitched - tip kd , the tips of the blocks are twisted to the screw axes @xcite . pitched tips
modify the channel geometry of the conventional kd . furthermore , pitched tips add additional drag ability along with screw rotation to conventional kd .
the tip angles associated with forward pump will be called `` forward tip '' ( ft ) and define its tip angle as positive .
conversely , for negative tip angles , pitched tips add backward pump to a block , so that we will call it `` backward tip '' ( bt ) .
along with this terminology , we will call the conventional block with zero tip angle the `` neutral tip '' ( nt ) .
the typical ptkd has been designed by adjusting both the tip angle and the disk - stagger angle , and therefore its mixing characteristics are determined by the combination of the tip angle and the disk - stagger angle @xcite .
effects solely due to the pitched tip have not been discussed systematically in previous works . in this paper , we investigate the effects of the pitched tip direction and its angle on the mixing characteristics . for this purpose ,
ptkds with a disk - stagger angle and different tip angles , shown in fig .
[ fig1 ] , are chosen .
based on five - block kneading elements of @xmath0 with a disk - stagger angle of 90@xmath1 , pitched tips with different tip angles from -30@xmath1 to 30@xmath1 are arranged .
+ + + +
the flow of a polymer melt in the melt - mixing zone of a twin - screw extruder has been numerically solved to allow understanding the relation between the melt - mixing characteristics and the geometry of the kneading elements .
the numerical method to study the melt - mixing zone in this paper follows that of the previous paper @xcite .
here we give a summary of the numerical simulation .
we focus on the situation where the material fully fills the channel .
the reynolds number is assumed to be much less than unity , so that inertial effects are neglected .
the flow is assumed to be incompressible , and in a pseudo - steady state to screw rotation , as has often been assumed in polymer flow in twin - screw extruders @xcite . with these assumptions ,
the governing equations become @xmath2 where @xmath3 is the velocity , @xmath4 is the pressure , @xmath5 is the deviatoric stress , @xmath6 is the mass density , @xmath7 is the specific heat capacity , @xmath8 is the temperature , @xmath9 is the thermal conductivity , and @xmath10/2 $ ] is the strain - rate tensor , where @xmath11 indicates the transpose .
the fluid is assumed to be a viscous shear - thinning fluid that follows cross exponential viscosity @xcite , @xmath12 , \label{eq : exponential_model } \\ \dot{\gamma } & = \sqrt{2{\mbox{\boldmath $ d$}}:{\mbox{\boldmath $ d$ } } } , \ ] ] whose parameters are obtained by fitting the shear viscosity of a polypropylene melt taken from @xcite , and the values are @xmath13k , @xmath14pa@xmath15s , @xmath16 , @xmath17k@xmath18 , and @xmath19s .
the mass density , specific heat capacity , and thermal conductivity are taken from @xcite as well , and the values are @xmath20kg / m@xmath21 , @xmath22j/(kg@xmath15k ) , and @xmath23w/(m@xmath15k ) . as operational conditions ,
the screw rotation speed is set to 200rpm , while the volume flow rate varies in the range of 5120cm@xmath21/s ( @xmath2513318kg / h ) , so that the value of @xmath26 lies in the range of 1.536cm@xmath21 .
the no - slip condition on the velocity at the barrel and screw surfaces is assumed .
the velocity at the inlet and outlet boundaries was set to be uniform under the given volumetric flow rate .
the pressure at the outlet boundary was fixed to be a constant value .
the temperatures on the barrel surface and at the inlet boundary were set to 473.15 k and 453.15 k , respectively .
the natural boundary conditions for the temperature equation in the exit boundary plane and the screw surface were assigned .
the time evolution of the velocity and temperature fields was constructed with the converged fields for every three degrees of screw rotation .
the trajectories of the passive tracers were determined based on the solved velocity field .
the set of tracer trajectories was utilized to compute the residence time distribution @xcite and to investigate the flow history and mixing process .
the lagrangian - history average of a quantity @xmath27 over the trajectory of @xmath28th tracer is defined as @xmath29 where @xmath30 and @xmath31 are the residence time and position of the @xmath28th tracer , respectively , and @xmath32 is the dirac delta distribution .
the statistical distribution of @xmath33 characterizes the global flow characteristics in the melt - mixing zone .
initially , 2000 points were uniformly distributed in a certain section relative to the axial position , which was arbitrarily set in the second disk .
they were advected until they reached the outlet section .
when computing the tracer advection , some tracers went out of bounds because the time resolution of the velocity field was limited . to circumvent the effect of the lost tracers on the statistics , we set the number of initial points to 2000 , which ensured that a sufficient number of points reached the outlet .
in order to see the effect of the tip angle on the inhomogeneity of cross - sectional mixing , we computed the fluctuation of the residence time . figure [ fig2 ] shows the probability density of the residence time normalized by the mean residence time for five different ptkds under @xmath34cm@xmath21 . from this figure
, we observe that the residence time distribution is almost unaffected by the forward tip angle while the fraction of the longer residence increases with a backward tip angle . in fig .
[ fig3 ] , there is drawn the relative standard deviation of the residence time to its mean value as a function of @xmath26 .
it turns out that the fluctuation of the residence time in ptkds hardly depends on @xmath26 , and takes the value of 0.50.6 . at higher values of @xmath26 ,
the residence time fluctuation is approximately the same level as that of the conventional kd .
this fact suggests that ptkds have as good mixing ability as the conventional kd with the same disk - stagger angle at sufficiently high @xmath26 .
the effect of the tip angle on the residence time distribution depends on the tip direction .
while for ft type the residence time fluctuation does not show any angle - dependence , for bt type the residence time fluctuation decreases slightly with the backward tip angle .
although this slight change of residence time fluctuation reflects the effect of the tip angle , this change does not seem to be substantial in terms of mixing ability . in short ,
the residence time distribution is not substantially modified by the addition of a tip angle to a conventional kd .
next , we discuss the effects of a pitched tip on dispersive mixing . the mean applied stress during residence , @xmath35 , is calculated by inserting the stress magnitude @xmath36 into eq .
( [ eq : lagrangian_average_of_f ] ) .
figure [ fig4 ] shows the probability density function of @xmath35 for ft15 , ft30 , and nt under @xmath34cm@xmath21 . as the forward tip angle increases , the fluctuation of @xmath35 increases , whereas the average of @xmath35 stays at a similar level as that of nt .
the average value and the relative standard deviation of @xmath35 as a function of @xmath26 are drawn in figs .
[ fig5](a ) and ( b ) , respectively .
the average value of @xmath35 is insensitive to @xmath26 , and takes a similar value as that of nt .
in contrast , the fluctuation of @xmath35 increases with @xmath26 , and is larger for a larger forward tip angle .
these results clearly indicate that the forward tip increases the inhomogeneity of dispersive mixing quality especially at high throughput operation .
figure [ fig6 ] shows the probability density function of @xmath35 for bt15 , bt30 , and nt under @xmath34cm@xmath21 . in fig .
[ fig6 ] , we observe an increase both in the average level and the fluctuation of @xmath35 with an increase in the backward tip angle . the average value and the relative standard deviation of @xmath35 as a function of @xmath26 are drawn in figs .
[ fig7](a ) and ( b ) , respectively . in fig .
[ fig7 ] , the average value of @xmath35 becomes larger for a larger backward tip angle , suggesting that the bt type enhances the dispersive mixing ability compared to the nt type .
in contrast , the fluctuation of @xmath35 is rather insensitive to the backward tip angle .
that is , irrespective of bt angle , the inhomogeneity of the dispersive mixing is at the same level as that of the nt type with an increase of @xmath26 , the fluctuation of @xmath35 for bt slightly increases , suggesting that inhomogeneity in the dispersive mixing is smaller for lower @xmath26 . in short , for bt type , both high level and small inhomogeneity of @xmath35 is achieved at low @xmath26 conditions , indicating that that dispersive mixing in bt type is most effective when @xmath26 is low and the bt angle is large . in order to further study the effects of the tip - angle on dispersive mixing
, the passage of fluid at high - stress regions has been investigated .
irrespective of the tip angle , the highest shear stress is achieved in the small gap regions like the tip barrel clearance and the inter - meshing region .
the substantial dispersion process mainly takes place when fluid elements pass through such high - stress regions .
however , since the high - stress region occupies only a small fraction of the whole channel , the fraction of the fluid passing through these regions and the number of passages are essentially determined by the flow pattern at the low - stress ( low - shear - rate ) regions .
thus , the geometric structures of the mixing elements are directly responsible for the passage of fluid through the high - stress regions , eventually leading to a better dispersive mixing performance . by setting a threshold value for the magnitude of the shear stress tensor ,
the regions of tip barrel clearance and the screw inter - meshing region are defined as the high - stress regions .
for instance , the region where the shear stress magnitude is larger than 0.35mpa for nt type under @xmath37cm@xmath21 is shown in fig
. [ figa ] , which clearly shows that the high - stress region is localized at around tip - clearance and inter - meshing regions .
we calculated the fraction of the tracers passing through the high - stress regions as well as the residence time in those regions .
for each tracer , the residence time in the high - stress regions , @xmath38 , as well as the residence time in the whole mixing zone , @xmath30 , has been studied . figures [ fig8](a ) and ( b ) show the probability density function of the ratio @xmath39 under @xmath34cm@xmath21 . at a first look at figs .
[ fig8](a ) and ( b ) , the residence time in the high - stress regions is at most 0.1 during the residence in whole zone , irrespective of the tip angle . both for ft and bt types ,
the residence in the high - stress regions becomes longer compared to the nt type , indicating that the pitched - tip geometry can enhance the efficiency of conveying the fluid to the high - stress regions .
the increase of the relative residence in the high - stress regions is more pronounced with the bt type .
this observation partly explains the enhancement of the dispersive mixing by the pitched tip . from figs .
[ fig8](a ) and ( b ) , we also observe that the probability of vanishing @xmath38 takes a finite value , indicating that not all the fluid elements pass through the high - stress regions . in order to study this aspect
, we discuss the fraction of the tracers passing the high - stress regions , denoted by @xmath40 . in figs .
[ fig9](a ) and ( b ) , @xmath40 is plotted to the mean residence time of high - stress regions , @xmath41 , for different @xmath26 values , in which the larger symbol is for the larger value of @xmath26 . in the plot of @xmath42 and @xmath43 ,
the upper locations means that a large fraction of tracers pass through the high - stress regions , while the right locations means that the tracers stay long time in the high - stress regions .
hence , the upper and right locations indicate a high potential for dispersive mixing . for conventional nt type , a fraction , 0.40.5 , of the fluid passes through the high - stress region for the @xmath26 value studied .
furthermore , @xmath42 and @xmath43 are rather insensitive to the value of @xmath26 .
in contrast , for ft and bt types , @xmath42 and @xmath43 show a relatively large dependence on @xmath26 .
we observe a general trend such that , as @xmath26 decreases , both @xmath42 and @xmath43 increase .
in addition , the characteristics of the passage through the high - stress region have different tip - angle dependencies with different directions of the tip angle . while for ft type , @xmath42 and @xmath43 does not depend on the tip angle , for bt type , @xmath42 and @xmath43 take larger values for larger backward tip angles .
this fact is consistent with the tip - angle dependence of the mean applied stress during residence in fig .
[ fig5](a ) and [ fig7](a ) .
the characteristics of the passage through the high - stress region reveal that the backward tip is effective at enhancing the dispersive mixing ability . from the data
above , it turns out that the pitched tip modifies the dispersive mixing ability of kneading disks , and the backward tip enhanced the potential of dispersive mixing while keeping the distributive mixing ability .
we discuss the relation between the flow pattern induced by the geometry of the pitched - tip kd and the dispersive mixing ability .
pitched tips add an additional drag ability along the screw rotation to conventional kd . for backward tips
increase the drag in the backward extrusion direction , for @xmath44 condition , the pressure drop in bt type becomes larger than that in nt , leading to the pressure flow in the extrusion direction being more pronounced in bt .
the enhanced pressure flow in bt promotes the passage of fluid across the small gap at the backward tips , resulting in an increase of the passage through the high - stress regions and the mean applied stress . in other words ,
the geometry of backward tips cause a counteracting effect by blocking against the forward flow and the additional forward pressure flow while keeping the leakage flow in the inter - disk spaces .
since this effect is driven by the screw rotation , it is maximized with the small @xmath26 condition , which is clearly demonstrated by the characteristics of the passage through the high - stress region in fig .
[ fig9 ] . in the case of ft type , since the forward pitched tips increase the drag in the forward direction , for @xmath44 conditions the pressure drop become smaller than for nt , or even becomes negative , depending on @xmath26 .
thus , the counteracting effect of the pressure flow and the blockage by the tips is weaker in the ft type than in the bt type . in short ,
the geometric modification of conventional kd by pitched tips induces both a change in the channel geometry and a modification of the pressure flow , causing a change in dispersive mixing ability .
in pitched - tip kneading disks ( ptkd ) , a novel melt - mixing element used in twin - screw extrusion , the effects of the direction and size of the tip angle on the mixing characteristics were investigated based on the numerical simulation of a three - dimensional flow . under a material feed rate and a screw rotation speed , the tip angle on a conventional kneading - disk ( kd )
element modifies the following two properties of the flow : ( i ) additional drag ability caused by screw rotation modifies the pressure flow , ( ii ) the pitched - tip acts as a blockage to the pressure flow induced by the pitched - tip .
these counteracting effects result in an increase of the fraction of the fluid elements passing through the high - stress regions , leading to a modification of the dispersive mixing ability .
forward tips on neutrally staggered kd increase the inhomogeneity of the applied stress during residence compared to the conventional kd , while keeping the residence time fluctuation .
in contrast , backward tips on neutrally staggered kd increase the mean applied stress during residence , suggesting an enhancement of the dispersive mixing ability . these effects are more pronounced for larger tip angles , and are dependent on the ratio of the material feed rate to the screw rotation speed .
these findings about the role of the tip angle can be useful for a better understanding of the mixing characteristics of a general class of ptkd elements and for choosing the optimal combination of the disk - stagger angle and tip angle .
the numerical calculations have been partly carried out using the computer facilities at the research institute for information technology at kyushu university . this work has been supported by grants - in - aid for scientific research ( jsps kakenhi ) under grant nos . 26400433 , 24656473 , and 15h04175 . | in mixing highly viscous materials , like polymers , foods , and rubbers , the geometric structure of the mixing device is a determining factor for the quality of the mixing process . in pitched - tip kneading disks ( ptkd ) , a novel type of mixing element , based on conventional kneading disks ( kd ) , the tip angle is modified to change the channel geometry as well as the drag ability of kd .
we discuss the effects of the tip angle in ptkd on mixing characteristics based on numerical simulation of the flow in the melt - mixing zone under different feed rates and a screw rotation speed .
it turns out that the passage of fluid through the high - stress regions increases in ptkd compared to conventional kd regardless of the directions and sizes of the tip angle , while the fluctuation in residence time stays at the same level as the conventional kd .
furthermore , pitched tips of backward direction increase the mean applied stress on the fluid elements during its residence in the melt - mixing zone , suggesting the enhancement of dispersive mixing quality in ptkd .
these understandings of the role of the tip angle on kd can give a basic guide in selecting and designing suitable angle parameters of ptkd for different mixing purposes .
polymer processing , mixing , twin - screw extrusion , numerical simulation |
the simple `` closed - box '' model of galaxy evolution @xcite predicts a correlation between the stellar mass and the gas - phase metallicity of a galaxy ( the @xmath0-@xmath1 relation ) .
such a relation is indeed seen in the nearby universe @xcite . and is found to be in place out to @xmath18 @xcite , progressively evolving to lower metallicities . tracing the @xmath0-@xmath1 relation and its evolution to even earlier times will provide insight into the earliest epochs of galaxy evolution , while allowing us to probe the relative importance of various galactic - scale phenomena proposed at @xmath13 .
however , studying the @xmath0-@xmath1 relation beyond @xmath16 is challenging because the nebular emission lines required for robust metallicity measurements ( e.g. , h@xmath19 , h@xmath20 , n@xmath21ii @xmath22 , o@xmath21iii @xmath23 , o@xmath21ii @xmath24 ) shift into different near- and mid - ir wavebands , where existing spectrographs have reduced sensitivity compared to the optical band .
this is aggravated by the faintness of the galaxies , making individual abundance measurements nearly impossible . an alternative way to determine metallicities at @xmath25 ( and in principle at @xmath26 and beyond @xcite )
is absorption spectroscopy of gamma - ray burst ( grb ) optical / near - ir afterglows .
long - duration grbs are known to be associated with the deaths of massive stars ( e.g. @xcite ) , and therefore with sites of active star formation .
the large optical luminosities of grb afterglows and their intrinsic featureless spectra provide a unique way to measure interstellar medium ( ism ) metallicities for galaxies at @xmath27 from rest - frame ultraviolet metal absorption lines and ly@xmath19 absorption . since the afterglows are significantly brighter than the underlying host galaxies , this technique allows us to measure metallicities independent of the galaxy brightness .
this approach has now been exploited at least to @xmath28 using optical spectra ( e.g. , @xcite ) , and with near - ir spectrographs it can be implemented to @xmath29 .
naturally , to explore the @xmath0-@xmath1 relation at @xmath25 we also require a determination of the grb host galaxy stellar masses , and hence follow - up infrared observations with the _ spitzer _
space telescope to probe the rest - frame optical luminosity .
we present the first large set of _ spitzer _
observations for grb host galaxies at @xmath2 , and combine the inferred masses with measured metallicities to explore the @xmath0-@xmath1 relation beyond @xmath30 .
a full analysis is presented in laskar et al .
( 2011 ) , to be submitted to apj .
we obtained deep observations ( about 2 hr per target ) of 20 long - duration grb host galaxies in the range @xmath31 using the 3.6 @xmath32 m band of the infra - red array camera ( irac ; @xcite ) on - board the _ spitzer _ space telescope .
the effective wavelength of the irac 3.6 @xmath32 m band probes the rest - frame spectral energy distribution ( sed ) redward of about 5500 and therefore provides a robust measure of the stellar mass .
we used optical afterglow images to perform relative astrometry on the _ spitzer _ mosaics and to locate the grb hosts . at the depth of our observations , _
spitzer _
images are confusion - limited for faint sources , requiring us to model and subtract nearby contaminating sources in 3 cases .
we detect 5 grb hosts at 3.6 @xmath32 m ( grbs 050319 , 050814 , 060707 , 060210 , and 060926 ) , with flux densities ranging from about 0.55 to 1.65 @xmath5jy , with a typical upper limit of about 0.25 @xmath5jy ( 3@xmath33 ) . to assess the typical flux density of the non - detected hosts we carry out a stacking analysis of 11 non - detections with accurate relative astrometry .
the stacked image does not show a detection down to a 3@xmath33 limit of @xmath34 njy .
+ computing stellar masses from observed luminosities in a given wave - band requires knowledge of the mass - to - light ratio and hence the stellar population age and metallicity . when multi - band photometry is available , modeling of the spectral energy distribution ( sed ) using stellar population synthesis models can be used to determine stellar masses , provided that a single stellar population is assumed . when multi - band photometry is not available , the resulting uncertainty in the mass - to - light ratio ( e.g. , at @xmath35 @xmath32 m ) is about an order of magnitude ( e.g. , @xcite ) . here
, since we lack broad - band photometry , we determine a range of mass - to - light ratios for each galaxy in the observed 3.6 @xmath32 m band using a wide range of population ages and the single stellar population models of @xcite with a salpeter imf and assuming an instantaneous burst of star formation ( e - folding time , @xmath36 ) .
as expected , the 3.6 @xmath32 m mass - to - light ratio for these models increases with stellar population age beyond @xmath37 myr .
an upper bound on the mass - to - light ratio is achieved by setting the stellar population age to the age of the universe at each host redshift ( @xmath38 gyr at the median redshift of our sample ) .
we stress that this leads to a very conservative maximum mass for each host galaxy since studies of lyman break galaxies ( lbgs ) and ly@xmath19 emitters ( laes ) at @xmath25 indicate typical population ages of @xmath39 gyr @xcite . the median age for long - duration grb hosts at @xmath40 of about 70 myr @xcite is probably more typical .
the variation in mass - to - light ratio between these age values is about an order of magnitude .
the maximal masses inferred for our sample are @xmath41 m@xmath4 , while the typical ( maximal ) upper limits are @xmath42 m@xmath4 .
the masses inferred for a 70 myr old population are about @xmath43 m@xmath4 , with typical upper limits of @xmath44 m@xmath4 .
the mass limit from the stack of 11 grb hosts is @xmath45 m@xmath4 for a 70 myr old population , and @xmath46 m@xmath4 for the maximal age .
we also include a literature sample of 5 long - duration grb hosts at @xmath25 , incorporating two detections with maximal masses of @xmath47 m@xmath4 ( grb060510b ; @xcite ) and @xmath48 m@xmath4 ( grb080607 ; @xcite ) .
a typical optical afterglow spectrum exhibits a wide range of ism absorption features due to rest - frame uv transitions of low- and high - ionization metal species , which allow a direct determination of the column density of these elements along the grb line of sight through the host galaxy .
combined with a determination of the neutral hydrogen column density via the ly@xmath19 line , it is possible to determine the ism abundances ( e.g. , @xcite ) .
we use s@xmath21ii , when available , as a measure of the metallicity , primarily since sulfur is not strongly depleted onto dust .
we also place lower limits on the metallicity using si@xmath21ii , si@xmath21iv and c@xmath21ii detections reported by @xcite . of the 18 grbs in our sample , six have determined [ s / h ] values ,
five have no metallicity information , and six have upper limits on their metallicity from non - detections of s@xmath21ii as well as lower limits based on si@xmath21ii or si@xmath21iv detections . for grb050908 , the metallicity upper limit ( @xmath49 z@xmath50 ) is not meaningful and we only report a lower limit based on a c@xmath21ii detection . using these values we find a wide range of metallicities ( using the solar abundance values from @xcite ) : @xmath51 z@xmath4 for the _ spitzer_-detected grb hosts , which have stellar masses of @xmath52 m@xmath4 .
we divide the grbs with available metallicity information ( either a metallicity detection or a bounded range ) into two mass bins the 3.6 @xmath32 m detections with @xmath53 m@xmath50 ( group 1 ) and the objects included in the stack ( group 2 ) . for each group
, we perform a monte carlo simulation to estimate the mean metallicity , yielding @xmath54 and @xmath55 ( @xmath56 intervals ) . for group 1 ,
the mean of the maximum inferred stellar masses is @xmath57 m@xmath50 , while that of the masses inferred from the 70 myr populations is @xmath58 m@xmath50 . to obtain mass estimates for group 2 , we scale our stack limit obtained for 11 non - detections by @xmath59 . using the mean maximum mass - to - light ratio of the objects in group 2 yields an upper limit on the mean stellar mass of these seven objects of @xmath60 , while using the mean mass - to - light ratio at 70 myr yields a mass limit of @xmath61 . in figure 2
, we present the absorption line metallicities plotted versus the stellar masses inferred from our _ spitzer _
observations and the literature studies .
the 6 grb hosts that do not have any constraints on their ism metallicity are excluded from this plot .
we find that our two points fall below the observed relations at @xmath62 , providing evidence that the galaxy @xmath0-@xmath1 relation continues to evolve at @xmath2 , with our stack range probing a somewhat lower mass scale than the lbg studies at @xmath63 @xcite .
and @xmath64 intervals for estimates of the mean metallicity at two mass bins of @xmath65 m@xmath4 ( 3.6 @xmath5 m detections ) and @xmath66 m@xmath4 ( scaled stack limit - see text ) .
the dashed vertical line indicates the upper limit on the mean mass of the stack for a 70 myr population .
these data are consistent with a decline in metallicity with lower stellar mass an @xmath0-@xmath1 relation .
also shown are the relations for @xmath67 @xcite , @xmath68 @xcite , @xmath69 @xcite , and @xmath70 ( * ? ? ?
* ; * ? ? ?
* filled squares ) ; the curves at @xmath71 are @xmath0-@xmath1 relations re - calibrated by @xcite .
our two points at @xmath2 fall below these relations , suggesting that the @xmath0-@xmath1 relation continues to evolve to @xmath72 .
[ fig : mz ] ]
we present the first study of the galaxy mass - metallicity relation at redshifts of @xmath2 using grb afterglow absorption metallicities and _ spitzer _ follow - up observations .
five of the 20 grb hosts in our sample are detected above a 3@xmath33 flux density threshold of 0.25 @xmath5jy , corresponding to a typical stellar mass of @xmath73 m@xmath4 .
we further place a limit of @xmath74 m@xmath4 on the non - detected hosts based on a stacking analysis . using the metallicities of the host galaxies ( inferred mainly from s@xmath21ii ) we find a range of @xmath75 at a fixed stellar mass of @xmath65 m@xmath4 .
the mean metallicity at this mass scale is about @xmath76 z@xmath4 .
the mean metallicity associated with 7 of the 11 the non - detected hosts , which have an upper limit of @xmath66 m@xmath4 , is @xmath15 z@xmath4 .
thus , there appears to be an overall decline in metallicity with decreasing mass , a hint of an @xmath0-@xmath1 relation .
furthermore , our two points on the @xmath0-@xmath1 relation lie below the relations at lower redshifts , suggesting that the relation continues to evolve at least to @xmath77 .
clearly , additional observations are required to confirm and increase the statistical significance of this result .
a sample of 20 additional grbs at @xmath25 from 2007 through the present is available for study .
this will allow us to double the existing sample . the _ james webb space telescope _
( jwst ) will provide a much deeper view of the @xmath0-@xmath1 relation at high redshift .
for instance , the nircam instrument on jwst will allow us to detect grb hosts at @xmath30 down to a mass of @xmath78 m@xmath4 , and at @xmath79 to @xmath80 m@xmath4 with similar integration times as our observations .
equally important , the nirspec instrument ( @xmath81 @xmath32 m ) will allow us to determine emission - line metallicities for some of the hosts , and hence to cross - calibrate the afterglow absorption metallicities . with on - going afterglow absorption metallicity measurements
, the grb sample will continue to play a key role in our study of high redshift galaxies .
m. asplund , n. grevesse , and a. j. sauval , `` the solar chemical composition , '' in _ cosmic abundances as records of stellar evolution and nucleosynthesis _ , edited by t. g. barnes iii & f. n. bash , 2005 , vol .
336 of _ astronomical society of the pacific conference series _ , pp . 25+ . | long - duration gamma - ray bursts ( grbs ) provide a premier tool for studying high - redshift star - forming galaxies thanks to their extreme brightness and association with massive stars . here
we use grbs to study the galaxy mass - metallicity ( @xmath0-@xmath1 ) relation at @xmath2 , where conventional direct metallicity measurements are extremely challenging .
we use the interstellar medium metallicities of long - duration grb hosts derived from afterglow absorption spectroscopy ( @xmath3 z@xmath4 ) , in conjunction with host galaxy stellar masses determined from deep _ spitzer _ 3.6 @xmath5 m observations of 20 grb hosts .
we detect about 1/4 of the hosts with @xmath6 to @xmath7 mag , and place a limit of @xmath8 mag on the remaining hosts from a stacking analysis . using a conservative range of mass - to - light ratios for simple stellar populations ( with ages of 70 myr to @xmath9 gyr ) ,
we infer the host stellar masses and present the galaxy mass - metallicity measurements at @xmath2 ( @xmath10 ) .
we find that the detected grb hosts , with @xmath11 m@xmath4 , display a wide range of metallicities , but that the mean metallicity at this mass scale , @xmath12 z@xmath4 , is lower than measurements at @xmath13 . combined with stacking of the non - detected hosts ( with @xmath14 m@xmath4 and @xmath15 z@xmath4 )
, we find evidence for the existence of an @xmath0-@xmath1 relation at @xmath16 and continued evolution of this relation to systematically lower metallicities from @xmath17 .
address = harvard - smithsonian center for astrophysics , 60 garden street , cambridge , ma 02138 address = harvard - smithsonian center for astrophysics , 60 garden street , cambridge , ma 02138 address = spitzer science center , california institute of technology , pasadena , ca 91125 |
* antonio : * you received your degree in economics from `` l. bocconi '' university in milano
. why did you decide to study economics ?
* eugenio : * i enrolled in an economics degree essentially because it was the only choice i had .
having attended a technical high school for accountants , at the time the law did not allow students from this kind of secondary school to study mathematics at university , which would have been my favorite option .
you needed to attend college preparatory schools to enroll in subjects like mathematics .
my family could not afford my university studies and i was expected to get a job right after completing high school .
the choice i made at the age of 14 was coherent with this scenario . by the time i got my diploma from high school ,
the situation had improved a little bit and i could afford going to university .
this was also due to a financial aid program , adopted by the farseeing italian center left governments of the time , for students belonging to economically disadvantaged families .
the money from this program ( it was around 200 euros , per year , in 1965 ) and the earnings deriving from private lessons i used to teach to other students allowed me to obtain a degree at bocconi university .
* igor : * why did you choose bocconi university and not another university ?
* e : * in the 1960s bocconi was considered a prestigious university : a degree from it represented the key for obtaining a good and rewarding job on the market .
these elements exerted an influence on me and my family .
however , there was also a practical reason : milano is just one hour by train from my hometown , cremona , and i could go back home every weekend .
yet another reason is the influence of my friend lorenzo peccati , a mathematical economist from cremona , who was a student at bocconi while i was still at high school .
he was well aware of my bent for mathematics and suggested a few advanced textbooks where i started reading the mathematical tools used in economic modeling .
in particular , i was excited at reading the italian translation of the monograph allen ( @xcite ) on mathematical analysis for economists and this convinced me that bocconi would still have allowed me to study maths .
* i : * this is a funny coincidence since i was convinced by lorenzo peccati s son , giovanni , by now a well - known probabilist , to enroll at bocconi and for precisely the same reasons . where did your passion for mathematics come from ?
* e : * i was very lucky at high school because i had a brilliant maths teacher , sidomo vailati .
he had a variety of scientific and cultural interests and also did consulting for a few private companies , thanks to his unusual , at least in that period , knowledge of statistics , probability and operations research .
he certainly was a self - taught man in the area of stochastics . at that time ,
probability and statistics , unlike analysis , geometry and algebra , were not perceived as relevant topics within mathematics degrees : they were only present in a few optional courses . to my knowledge , the only exception was the university of roma due to the presence of bruno de finetti .
in fact , this unfortunate situation lasted until the 1970s when the first full professors in probability , apart from de finetti , were recruited after national competitions . turning back to vailati
, it is worth recalling that , among some courses for high school teachers organized by the italian ministry of education , he also took a course in probability , which was delivered by de finetti . as a consequence , at the age of 16
i was introduced to the realm of subjectivism and learned the first elements of probability and its applications .
these first years of exposure to de finetti s approach have stimulated an intellectual and scientific interest that has certainly influenced my later research . *
a : * how was the environment at bocconi university in the years you have been there ? * e : * bocconi had very few professors among its own faculty and heavily relied upon adjunct faculty holding positions in other universities .
these few professors were all influential personalities of the time , playing significant roles in the italian social , political and economic life of the 1960s .
for instance , giovanni demaria was a paretian economist who acted as economic consultant for the constituent assembly that created the constitution that lies at the foundation of the modern italian republic after world war ii .
there was also a special feeling between bocconi and milano , a city that had been able to overcome the disasters of world war ii and was experiencing dramatic economic growth led by the manufacturing sector .
bocconi looked to me , and many others , as a vital part of milano and contributed to consolidate this process .
then , during the last couple of years , the student protests of 1968 started .
despite being a private university , bocconi experienced serious clashes and some of its students played an active role in the movement . * i : * did you like studying economics ? *
e : * i was very fond of economics .
the professors i was interacting with were quite enthusiastic about my inclination toward developing mathematical tools useful for economic modeling .
there is an episode that occurred during my third year that i like to recall .
i was attending a course in political economy which included a series of seminars and one of them concerned the relationship between the italian central bank and the department of treasury , which at that time was the subject of a lively debate .
for an economic interpretation of the relationship between the two institutions , we were suggested to refer to an article by giorgio la malfa and franco modigliani ; the latter was later awarded the nobel prize in economics in 1985 .
the main contribution of the paper was the proposal of a static model .
playing a bit around with that model , i was able to derive a dynamic version of it , which seemed in line with the real situation in italy .
this was appreciated by the other students and the teaching assistants .
also in connection with the course on public finance , i devised a model describing the evolution of certain taxing decisions .
overall , i think i had quite a good economic intuition .
* a : * what did lead you to study probability and statistics ?
* e : * i was both impressed and fascinated by the first year course in mathematics that was taught by giovanni ricci .
it was more advanced than a traditional calculus course .
the second year maths course , delivered by giuseppe avondo - bodino , included also a part devoted to probability , which actually covered essentially the same material nowadays taught in first probability courses in maths degrees . during the second year of my degree
i also attended a course on statistical inference held by francesco brambilla , which was important for my education .
finally , my third maths course by eugenio levi contained some probabilistic applications .
this experience revived my curiosity , dating back to high school , for foundational aspects of probability .
moreover , i perceived probability as a tough subject and therefore more challenging and stimulating than others i was studying .
* i : * what was the topic of your degree thesis ?
* e : * i asked avondo - bodino to be my thesis supervisor .
he was a passionate fisherian and hostile toward the bayesian paradigm : it might , thus , seem curious that the title of my thesis was `` the bayesian approach to hypothesis testing . ''
indeed , he chose that topic with the aim of proving the fallacy of the bayesian paradigm : this is revealed by the fact that the potential of the bayesian approach was going to be assessed with respect to hypothesis testing problems that had already received well - established answers within the frequentist framework . to be honest , he did not even like the neyman
pearson approach : according to him , it introduced subjective elements since it relied on decision theory .
my task was essentially to : ( i ) collect as much material as possible on hypothesis testing , ( ii ) evaluate the possible impact of the bayesian approach and ( iii ) establish whether it could be a sensibile alternative to the frequentist approach .
and my supervisor obviously expected a negative answer to the last question . *
a : * and how did it work out ? what were your first impressions on the bayesian approach ?
* e : * while working on the thesis , i developed some skepticism about the automatic implementation of bayes theorem , which was a legacy from laplace and his followers .
however , my viewpoint was limited .
in fact , writing the thesis was not an easy job , especially because i could not rely on many systematic and exhaustive treatments .
there were , of course , de finetti s papers , but , given the unorthodox way they were written , i was not able to understand the connection between his theory and the anglo - american neo - bayesian approach typicallyadopted in papers appearing in statistics journals at that time
. de finetti s work did not follow the standard bayes
laplace paradigm : in contrast , he re - constructed it and recast it in a way to be coherent with his approach to prediction . the books i referred to were lindley ( @xcite ) , raiffa and schlaifer ( @xcite ) and , mostly , ferguson ( @xcite ) , which contained a beautiful part on the bayesian approach from the viewpoint of wald s decision theory .
the 1959 lecture notes of de finetti s course at a summer school in varenna [ later translated in de finetti ( @xcite ) ] and savage ( @xcite ) were also helpful .
i obtained a few minor results in terms of interpretation and comparison and also derived a `` rule '' for the choice of type - i error probability @xmath0 .
as soon as i completed the thesis , the monograph degroot ( @xcite ) appeared : i found it very interesting and it proved to be very useful for my statistical education . *
a : * one of your best friends and main coauthors is certainly donato michele cifarelli .
did you meet him while studying at bocconi ?
* e : * yes , he is actually 10 years older than me and was my teaching assistant while i was attending the statistics course .
he delivered insightful lectures where it was apparent that he had remarkable mathematical skills and also a deep knowledge of the book by e. lehmann .
therefore , it was quite natural to seek his help when i started working on the thesis , which required me to study also the frequentist approaches .
his advice was very important , although at the time he looked , at least to me , not interested into the frequentist versus bayes debate .
i was impressed by his vast knowledge of frequentist methods , both parametric and nonparametric , as well as of probability theory and stochastic processes .
i very much liked the fact that he preferred fundamental , though maybe difficult , at least for us , books to much more immediate cookbooks .
for instance , he knew in great detail the celebrated monograph by j. l. doob on stochastic processes .
we have been bound by a deep friendship and reciprocal esteem ever since .
* i : * who were other important scholars you met at the time and who influenced your early approach to mathematics ?
* e : * overall bocconi was an intriguing place , at least in italy , for probability and statistics , given these subjects were , as i said , almost absent from most maths degrees . at the time
italian statistics , and academia in general , did not have systematic contacts with the international community and the head of the institute of statistics , brambilla , had the merit of introducing and spreading the great developments in statistics and operations research , which took place in the uk and the us .
he had contacts with foreign scholars : for example , he is also cited by leonard j. savage in his 1954 book . at the heart of these scientific activities was the centre for operations research , which , among its sponsors , could count on adriano olivetti , an enlightened and revolutionary entrepreneur of the time , who ran the company producing the celebrated typewriter `` lettera 22 '' now displayed at the museum of modern arts in new york city .
his company is also well - known for its pioneering contributions to the development of personal computers . among the various cultural and scientific activities created and supported by adriano olivetti
, it is worth mentioning the journal `` tecnica e organizzazione '' : brambilla was a co - editor of the journal when de finetti published an important paper on the essentials of computational techniques based on monte carlo methods , `` macchine che pensano e che fanno pensare '' ( `` machines that think and that make you think '' ) .
brambilla was a remarkable figure : he had been the assistant of ferruccio parri who , besides being one of the first italian prime ministers after the war , was also imprisoned by the germans during world war ii since he had been one of the antifascist opposition leaders . *
a : * what happened after you graduated ?
* e : * i really enjoyed working on my thesis and i was eager to continue , at least for some time , with research .
programs did not exist in italy since they were only introduced in the mid-1980s .
so i was doomed to the military service which was compulsory and would have lasted for 15 months .
i tried to postpone my entry for as long as possible , since i wanted to compete for a scholarship from the italian ministry of university .
had i obtained it , i could have freezed it until the end of the military service .
thankfully my strategy was successful and , when i was discharged in january 1972 , i was able to go back to university .
after graduation and before starting the military service , i shared the office at bocconi with cifarelli , and together we attended various maths courses at the state university in milano .
i then sat the exams during my military service , but in the end i did not complete a maths degree , since i was already involved in developing my own research and i was willing to publish !
nonetheless , those studies turned out to be very useful for me .
* i : * unlike many italian academics , and more in line with what happens abroad , you have been working in many different universities . was this important for your professional development ?
* e : * definitely .
in addition to working in various universities , i also experienced very different environments .
it has been very helpful from both a scientific and personal point of view .
i met many statisticians and mathematicians with very different backgrounds .
my first experience outside bocconi was in torino : it was a small department , most colleagues were of my age and so it was pretty easy to settle in .
afterward i moved to bologna in a much larger department more in line with the italian statistics tradition .
then i got back to milano : first at the mathematics department of the state university and , then , to bocconi .
finally pavia , which is one of the oldest universities founded in 1361 and in a very prestigious mathematics department of which i am a proud member . *
a : * tell us a bit about your first steps in the italian academia in torino .
* e : * my supervisor , avondo - bodino , was full professor in torino and a lecturer at his department resigned and decided to leave the academia . since they needed a replacement , in december 1973 i moved there with the concrete opportunity of obtaining , later , a permanent position .
due to absurd bureaucratic reasons , i finally obtained an assistant professorship only in 1978 .
* i : * you obtained a full professorship position in a national competition at a young age in 1980 and , therefore , moved to bologna which hosted one of the few faculties of statistics in italy .
then back to milano . *
e : * that was a time italy was investing in universities , unlike now .
therefore , the career perspectives were quite good also in the academia if you worked hard . incidentally ,
the head of the selection committee was de finetti , i must say a recurrent figure in my life . in 1980
the only autonomous statistics faculties were roma and padova , whereas bologna and palermo were offering statistics degrees but within economics faculties .
statistics became a faculty in bologna only toward the end of the 1980s .
in fact , and in contrast to what happens outside italy , faculties are pivotal players in italian academia , mainly because they manage the recruitment .
now it seems that things will change but , as we say in italy , everything changes so that nothing changes .
anyhow , in bologna i mostly taught probability courses , but my ties to milano were still strong , especially because of my collaboration with cifarelli .
therefore , i accepted the offer from the mathematics department of the university of milano in 1984 , where they did not have a faculty member doing research in probability and mathematical statistics until my arrival .
a key role for my transfer was played by an analyst , marco cugiani , whom i also replaced as director of a research institute of the c.n.r .
, nowadays the milano branch of the institute of applied mathematics and information technology . *
a : * in 1989 you moved back to your beloved bocconi university .
did you find any substantial changes since the last time you had been there ?
* e : * at bocconi things had changed a lot , the most apparent being that it had turned from an elite to a larger and more open university with something like 10,000 students .
hence , i think that some changes were necessary . as for myself , i was in a somehow privileged position since i was mostly teaching advanced and not compulsory courses , which were much more challenging than most other courses . therefore , starting from a yearly basin of more than 2,000 students of very good quality , by self - selection i had small numbers of students , who were highly motivated and of the highest quality .
i guess i have to mention you two , will not i ? but let me also mention chiara sabatti and giovanni peccati , among many others . * i : * what convinced you to move to another university in 1998 ?
* e : * during the 1990s , while i was there , an even more radical reorganization was occurring : courses of the type i was teaching were perceived as too `` aristocratic '' and had too few students so that they were doomed to be shut down . and
the same destiny was foreseen for the most challenging degrees .
in fact , when i moved back in 1989 , i did it with the aim of setting up a statistics degree : i was very disappointed when the project was officially turned down in 1997 .
therefore , i decided that my experience at bocconi was concluded and that i wanted to move back to a maths department
. nonetheless , i still have a special affective relation with bocconi.=1 * i : * then you moved to pavia .
and i followed you , given i had just graduated from bocconi and started my ph.d . in pavia .
it was a new challenge for you at a mature age , was it not ?
* e : * the only science faculty members and future colleagues of mine in pavia i knew in person , prior to moving , were three brilliant mathematicians : maurizio cornalba ( we were both members of the scientific committee of the italian mathematical union ) , franco brezzi ( we got to know each other at the meetings of the italian national research council ) and enrico magenes .
enrico magenes , who passed away last november , shaped the department in its current form and significantly contributed to its international standing . when i moved , i was one of only two probabilists and since then we have been able to hire two additional assistant professors
. once arrived , with great enthusiasm i immediately got involved in the ph.d .
program and the outcome has been rewarding , as also witnessed by the achievements of some former ph.d .
students .
* a : * throughout your career you have been working at faculties of science and economics , at public and private universities .
what environment do you think better fits the needs of a researcher in probability and statistical science ?
* i think that the ideal environment , relatively to the italian experience which is the only one i am aware of , is a maths department which is open toward the modern trends and research directions of mathematics and therefore not too much bound to traditional subjects like analysis , geometry and mathematical physics . and thankfully there are various departments that comply with these criteria .
however , i must admit i am a bit pessimistic about the future : the italian university system in general , and departments involved in basic research in particular , are struggling and suffering due to the indiscriminate financial cuts in recent years .
these have been implemented somehow light - heartedly by the government since cuts in basic research funding are unlikely , at least in italy , to cause immediate social upheaval .
in fact , it would be important to abandon the habit of uniformly distributed cuts and aim at creating , or consolidating , niches of excellence .
and i have seen many such niches during my career .
* i : * at the beginning of your academic career you started working on inferential problems according to the frequentist approach . * e : * my interest in frequentist inference started soon after completing my degree thesis and heavily benefited from the collaboration with cifarelli . and one of the first topics we started working on was hypothesis testing .
corrado gini and other italian statisticians had introduced a considerable number of summary statistics that were originally used only for exploratory data analysis to measure , for instance , concentration , variability , dependence and similarity between sets of data , and so on .
our idea was to use such summary measures for inferential purposes and , specifically , as test statistics for studying dependence in nonparametric problems .
an early contribution in this direction was achieved by cifarelli ( @xcite ) who studied the asymptotic distribution of a statistic arising in a test of homogeneity for two - sample problems .
the paper contained a remarkable result on the distribution of the integral of the absolute value of the brownian bridge .
our initial efforts led to a paper ( cifarelli and regazzini , @xcite ) that we are very proud of : there we determine the limiting distribution of a measure of monotone dependence introduced by gini .
the program we set was very appealing and consisted in checking whether these statistics , when used for hypothesis testing , yielded tests that were more efficient than those commonly used at the time .
for example , the index of monotone dependence i was mentioning was compared with spearman s @xmath1 and with kendall s @xmath2 and in some cases it featured better performances . *
a : * around the mid-1970s you turned back to bayesianism .
what about your skepticism ?
* e : * yes , and my experience at university of torino was fundamental in this respect .
the department library held the collection of all de finetti s papers , well kept and easily accessible .
i started looking at contributions cited by savage as decisive for the foundations of the bayesian paradigm .
my curiosity was fueled by the fact that , as i said , de finetti s work appeared to me completely disconnected from the bayesianism i had studied on books and journal articles .
it was a challenging task since de finetti s writing style , which was actually one of the main aspects he was criticized for , was unorthodox and sometimes seemingly cryptic . nonetheless , hard work and stubbornness finally allowed me to understand why de finetti introduced exchangeability and the role such a form of symmetry plays in the reconstruction he gave of the bayes
laplace paradigm .
this really opened my eyes on a new world providing a coherent and unified view of statistical inference , where subjective probabilities play an important role .
in fact , suddenly the subjective interpretation of probability was the only one that made sense to me from both a philosophical and a mathematical point of view .
* i : * you were then able to convince cifarelli to enter the realm of bayesian statistics .
* e : * i have to say that cifarelli shared my same doubts on the foundations of the bayesian approach to statistical inference .
however , after completing my study program in torino i pointed him to the references where de finetti was answering our questions and solving our doubts . besides de finetti s well - known papers , i had discovered many other `` minor '' contributions that were important for understanding the unified framework he had in mind . and , after struggling to understand
, i started to love his style : entering his world had been very demanding , but once i succeeded the reward was incomparable . in his work one could find ideas , hints and concepts whose expressive force was much more powerful than a standard presentation of definitions , theorems and cool mathematical technicalities .
spurred by the enthusiasm , i had been able to convince cifarelli and we started working together in this direction . *
i : * is this when you started your research on bayesian nonparametrics ? *
e : * in some sense , yes . on the one hand , we were hoping to be able to tackle in a bayesian setting the same issues we had addressed within classical nonparametric inference . on the other hand , we guessed that our starting point should have been de finetti s representation theorem as stated in de finetti ( @xcite ) which we could consider as being nonparametric . in this fundamental paper ,
the law of an exchangeable sequence is described as a mixture on a space of probability measures and the prior is the almost sure limit , in a weak sense , of the empirical measure generated by the data .
this motivated the investigation of random probability measures ( rpm s ) for statistical inference and might have led to extend the bayes
laplace paradigm .
we planned to consider estimation of functionals of rpm s such as the mean , the variance or other characteristic parameters of the unknown distribution .
the necessary preliminary step was to determine the posterior distribution of these functionals .
a helpful reference was a short paper ( de finetti , @xcite ) , where de finetti provides a reformulation in bayesian terms of methods used in exploratory data analysis for smoothing the empirical distribution . moving from this , he basically addressed in a nonparametric framework both the issue of prediction and of evaluation of the posterior distribution on a set of probability measures . unfortunately , we had no clue on how to define a probability distribution on a space of probability measures that would be analytically tractable . of course , we were not aware of t. s. ferguson s paper ( ferguson , @xcite ) .
we were stuck and all the attempts we made led us nowhere . * a : * was there any decisive event that helped you overcoming these difficulties ?
* e : * in 1976 i met andrew l. rukhin who had left the soviet union and was in italy just before migrating to the us .
we discussed our research activities and i described to him the technical problems cifarelli and i were dealing with .
he suggested we go through ferguson s paper in order to find an answer to our questions . and , indeed , that was the case : that paper allowed us to resume our project .
so we started considering linear functionals of the dirichlet process with the aim of determining their probability distributions analytically . *
a and i : * let us also recall that the study of the dirichlet process suited your passion for classical music very well
! * e : * gustav dirichlet is associated with the distribution because he evaluated the integral on the simplex .
the musical connection is that he married rebecka henriette mendelssohn , younger sister of felix mendelssohn , the famous german composer .
* i : * were there other bayesians in italy at the time ?
* e : * a few years after its re - flourishing at an international level , due to the work of leonard j. savage , the bayesian approach was sort of rediscovered in italy as well .
this may sound surprising given de finetti is italian : however , one has to consider that de finetti only entered academia in 1946 , at the age of 40 , when his research was already focused on different topics .
interestingly , he had obtained the position already in 1939 , but could only start his job in 1946 after the fall of the fascist regime due to a law forbidding the appointment of unmarried professors , as was de finetti s case . anyhow ,
in those years there was a large group led by giuseppe pompilj in roma and some scholars started to work on bayesian statistics , like ludovico piccinato . in roma
there were also some of de finetti s students like , for instance , fabio spizzichino .
i should also mention a group based in trieste and coordinated by luciano daboni , who started working under de finetti s supervision soon after gaining his university degree . besides actuarial mathematics , they focused mainly on exchangeable processes and foundational issues of bayesian inference and , during the years , i had many fruitful interactions with them . *
a : * even if more interested in the bayesian paradigm , you did not avoid doing research based on a frequentist approach .
it seems you did not , and still do not , see any ideological contraposition between bayesianism and frequentism .
* e : * i have never seen this as an ideological contraposition .
i think that ideological positions make sense only outside the realm of mathematics . anyhow , even when i was working on statistical problems according to the frequentist approach , i always had the feeling that the bayesian framework was far more complete and logically sound .
i was not enthusiastic about the automatic use of priors on unobservable parameters : the subjective views i had on probability were in conflict with such a treatment of the bayes
laplace paradigm , as i believe that inference must concern quantities that can be empirically observed .
but , on the other hand , the fisherian attitude appeared to me as too drastic , because prior beliefs should play a role in statistical inference .
once able to fully understand the consequences of de finetti s results , i became convinced that bayesianism was the only acceptable way of inductive reasoning . *
a : * current developments in bayesian inference involve a heavy use of simulation algorithms .
do you still think there is a need for putting a strong effort in determining exact forms of bayesian inferences ( or , at least , error evaluation when approximations are used ) , even when these are difficult to use in practice ?
* e : * computational techniques have been decisive in making bayesian models applicable to real world problems and some recent applications i saw are simply amazing .
i definitely think that the advantages they yield largely surpass some drawbacks associated with their uses .
that said , i would still like to make a point , which i think is important since it has to do with how statistical modeling is conceived .
indeed , models should be devised as simple as possible , while still preserving the capability of capturing the essential features of the phenomenon under study .
such a simplification could be achieved by first detecting inessential elements and , then , dropping them when it comes to the point of specifying the model .
this attitude is natural when one aims at achieving exact estimates of the quantities of interest .
however , if the need for pushing analytic results as far as possible disappears , it is likely that the models become more loose and unnecessarily complex .
both parsimony and extreme care in the formalization of models are still important guidelines for research : the only difference is that they now need to be spelled out clearly , while they were implicitly followed in the past .
another related and important point concerns approximation .
when exact inferences are not possible , one should put some effort in providing an upper bound to the error of approximation yielded by the numerical techniques that are used .
i have tried myself to work in this direction , for instance , in relation to approximating the probability distribution of the mean of a dirichlet process .
i know this is a challenging task , but it can not be avoided .
* i : * there is no doubt your research has been deeply influenced by de finetti s work . which was the first paper of de
finetti you read through ?
* e : * while i was completing my thesis at bocconi i came across his joint paper with savage ( de finetti and savage , @xcite ) .
it contained a discussion on the choice of the prior distribution and was mainly illustrative with no deep mathematics involved but still evocative for a novice . *
a : * his most renowned piece of work certainly is the two - volume book on probability theory , de finetti ( @xcite ) .
what else would you suggest to a student who is willing to study and understand de finetti s stance in probability and statistics ?
* e : * i would certainly suggest de finetti ( @xcite ) , two volumes containing selected papers by de finetti , which have been published by the italian mathematical union in 2006 in occasion of the centenary of his birth .
the first volume is on probability and statistics , whereas the second is on applied maths and on the teaching of maths . as for his subjective views on probability
, one should refer to de finetti ( @xcite ) .
one should also read de finetti ( @xcite ) .
another important piece of work is de finetti ( @xcite ) .
finally , de finetti ( @xcite ) contains a selection of some of de finetti s papers with english translation .
unfortunately , some significant contributions , at least to my knowledge , have been only published in italian , such as those related to independent increments processes and some others on the subjectivistic definition and interpretation of probability .
* i : * as you just mentioned , the fact that he was not writing in english hindered the circulation of his ideas and results in the scientific community .
* e : * this is definitely true .
for example , it is probably unknown to many that de finetti introduced the celebrated @xmath2 index a few years before kendall ( de finetti , @xcite ) . in 1939
he obtained some important results on optional stopping : indeed , de finetti ( @xcite ) deals with the gambler s ruin problem , where one can also find an embryonic version of the girsanov theorem .
another important contribution was the continuity theorem for characteristic functions : he proved it in the appendix of de finetti ( @xcite ) .
besides these , it is worth listing a few other contributions for which a priority to de finetti should be acknowledged : he completed what is now known as the glivenko
cantelli theorem before francesco p. cantelli in de finetti ( @xcite ) ; in de finetti ( @xcite ) he devised a model that anticipated the portfolio theory for which markowitz was awarded the nobel prize ; he proved the theorem on almost everywhere nondifferentiability of the trajectories of the brownian motion in de finetti ( @xcite ) . *
a : * with reference to the de finetti ( @xcite ) paper , which is actually our favorite , should we not as italians propose _ lvy processes _ be called _ de finetti lvy processes _ instead ?
* e : * as i mentioned before , the answer is affirmative .
indeed , de finetti started from a more general problem of providing the random counterparts of a volterra classification for the ordinary laws of physics . in this context
he identified processes with independent and homogeneous increments as those whose characteristic function satisfies the first of the equations in volterra s classification , namely , @xmath3 .
as a by - product , he also introduced implicitly the notion of infinite divisibility . in a subsequent paper ,
de finetti ( @xcite ) , he further characterized the class of infinitely divisible laws as the class of distribution limits of compound poisson processes , thus providing a representation theorem for infinitely divisible distributions .
lvy was not aware of de finetti ( @xcite ) and resorted to a different approach to obtain more general and deep results .
the contribution by khintchine to the well - known lvy
khintchine representation originates from a paper published in 1937 ( see khintchine , @xcite ) : khintchine s paper builds upon kolmogorov ( @xcite ) , where kolmogorov explicitly mentioned ( even in the title of the article ) that he was resorting to the approach set forth by de finetti .
so , yes , it should definitely be _ de finetti
lvy processes_.
* i : * and what were his connections with the broader international scientific community ?
* e : * his first international contacts , before graduating in mathematics at the university of milano , are related to a paper , de finetti ( @xcite ) , he wrote on mendelian inheritance , which had quite an impact in biology .
it was his first paper and appeared on the italian journal _ metron_. his results also attracted the attention of alfred j. lotka and jacques s. hadamard .
the latter was so impressed by de finetti s achievements that he suggested georges darmois to study the paper , as witnessed by one of the letters that de finetti wrote to his mother in 1929 and that have recently been published by his daughter fulvia .
this research also originated the so - called de finetti diagrams that are extensively used in population genetics . *
a : * an important event at which de finetti drew attention on his research in probability was the international congress of mathematicians ( icm),which was held in bologna in 1928 .
* e : * that conference was definitely important for the development of de finetti s interactions with foreign scholars .
on that occasion he presented his first results on exchangeability and made contact with maurice r. frchet , who later invited him to the institut henri poincar in 1935 and to the colloque de gnve in 1937 where he then also met jerzy neyman and others .
he had frequent interactions with paul lvy and aleksandr khintchine , respectively , on independent increment processes and on the proof of the representation theorem for exchangeable sequences .
he was also in contact with andrey n. kolmogorov , as witnessed by the kolmogorov ( @xcite ) paper on infinite divisibility whose title contains an explicit reference to `` a problem of de finetti . ''
both kolmogorov and de finetti also worked at the same time on the derivation of a representation theorem for associative means , now known as the de finetti kolmogorov nagumo theorem .
he also got in contact with many eminent mathematicians via mail .
in fact , he used to have a notebook in which he recorded to whom he had sent which of his papers : de finetti s daughter , fulvia , once showed it to me and the names are impressive . after world war ii he had significant scientific collaborations with leonard j. savage and lester dubins and he interacted also with william feller and abraham wald .
* i : * were his views on the subjective approach to probability theory held in high regard ?
* e : * in mathematics his work has been largelyignored , and not only because of the subjective interpretation .
indeed , the mathematical approachyielded by such interpretation does not require @xmath4-additivity .
in fact , finitely additive laws also become admissible and the traditional measure
theoretic approach to probability theory represents obviously a particular case .
countably additive probabilities are coherent in de finetti s sense but are just a subclass of coherent laws .
and de finetti himself was well aware that many results could have been neater by assuming countable additivity .
we may reasonably conjecture that his position in favor of including finitely additive probabilities somehow put him off from focusing on the particular countably additive setup .
this could explain , for example , why he did not further investigate processes with independent increments .
it is to be noted that the framework for his subjective approach had been settled by 1931 and , as evident from his published mail exchange with m. frchet , he fought for it for a while .
* i : * and what about the impact on statistical practice ?
* e : * in bayesian statistics references to subjectivism are quite frequent , but i actually see little of de finetti behind them . first , in the subjective approach also finitely additive laws are allowed and , therefore , a proper subjectivist should try to analyze statistical problems in this setup .
this point is very important in the case where `` transcendent '' conditions such as convergence of sequences of random elements , forms of the corresponding limits , etc.are involved : one should , then , establish the extent to which the conclusions depend on the specific @xmath4-additive extension ( usually unique ) of the original finite - dimensional distributions .
second , from an interpretation point of view , subjectivism and objectivism are often mixed up and bayes theorem is applied in an automatic way , whereas subjectivism would require probabilistic statements to be made on verifiable events .
subjectivism seems more a kind of catch - phrase than a real commitment . in my opinion , the papers of l. j. savage , l. dubins , j. pitman , p. diaconis and d. freedman are the ones that adhere most closely to de finetti s views . * a : * did your convinced support of subjective probability affect the way you teach probability courses ?
* e : * this represented a sort of dilemma throughout my career .
focusing solely on de finetti s mathematical theory of probability would have implied providing students with an unorthodox background in probability : it could have been an enrichment for some of them but also a drawback for some others , especially for those who needed to use probability as a mere tool in other disciplines . therefore , most of the courses i taught were within the @xmath4-additivity framework .
nonetheless , i have always tried to illustrate extensively some distinctive features of the subjective viewpoint in one of my first lectures .
this was useful since it provided students with a more complete picture of the subject and allowed them to understand that the results i was going to state and prove were valid on a special class of probabilities sharing the property of countable additivity .
students were , then , aware that it was somehow like teaching them a course in analysis that was just about analytic functions !
the connection with conditional properties was far more difficult to point out .
as for the subjective interpretation , it is still possible to preserve it even when confining to @xmath4-additive probabilities . * a : * can you provide some further insight on this last issue ?
* e : * the difficulty i am referring to arises due to the fact that kolmogorov s definition can not be seen as a special case of coherent conditional probabilities .
in fact , the kolmogorov approach lacks an appropriate axiomatization and interpretation of conditional probability : the definition is by means of a limiting procedure .
the perspective is then completely different .
for example , de finetti s approach necessarily leads to conditional probabilities that are regular and proper , whereas it is well known that kolmogorov s definition does not . in order to grasp these mathematical and conceptual differences on conditional expectations and probabilities , one can refer to the works by l. dubins , david blackwell , czeslaw ryll - nardzewski , william d. sudderth , roger a. purves , pietro rigo , patrizia berti and also myself .
* i : * tell us about your meetings with de finetti . *
e : * i first met him in 1969 at a summer course on mathematical economics in urbino .
since i was working on my thesis , i took the opportunity to ask him a few questions about his paper with savage ( de finetti and savage , @xcite ) i had read .
he was not very talkative and probably thought i was not understanding anything .
he was right , but i still went away with the impression that it was not simple at all to interact with him .
afterward i met him at some conferences during the 1970s , but at that time he was not working on statistics and probability with the same intensity and creativity of the early days : he was more inclined to elaborate on general philosophical and foundational aspects .
the only thing i can say about our meetings is that i had the impression he was interested in nontrivial and original approaches or attitudes that to some other people might have appeared as singularities . for example
, in bologna he once told me he had been fascinated by the mathematical physics lectures held at the polytechnic in milano by a lecturer , bruno finzi , whose assignments were notoriously challenging and contained exercises that finzi himself could not solve .
he recalled the solutions he had been able to give were very original and much appreciated by finzi .
he also told me he had appreciated lectures on economics of insurance companies delivered by ulisse gobbi at the polytechnic in milano because they had been the source of inspiration for the mathematical modeling of many aspects of economics he had later investigated .
i am surprised by this , since in gobbi s work i did not find any mathematical formalism.=-1 * a : * he was also engaged in public life and gained some popularity because of his political experience .
* e : * his political experiences can be well understood if one refers to the environment where he grew up .
de finetti s family was wealthy and highly educated .
they were part of the italian community in territories of the habsburg empire , and his father was an engineer working for the austro - hungarian railway . during his childhood
he had learned about the irredentist ideas of the italian minority that was aiming at unification with italy .
such aspirations quite naturally developed into strong nationalist feelings once the area became part of italy .
moreover , having been part of a minority , he developed a strong sensitivity toward injustice in all respects and , therefore , also a strong criticism toward some social implications of capitalism of the time .
this blend of ideas somehow naturally led him to support the rising fascist party : its initial political and social program included a series of reforms whose goal was the complete state control of the economy .
as de finetti himself wrote a few years before dying , the direction of the whole economy , once freed from the terrible tangle of individual and interest group selfishness , should lean toward the collective achievement of a paretian `` optimum '' and should be further inspired by `` fairness '' criteria . *
a : * hence , his support to fascism was mainly the result of ideal feelings that were fueled by strong social and economic views . *
e : * this is further witnessed by the fact that after the fall of fascism , he sympathized with left - wing movements without adhering to a large political party . finally , during the 1970s he started being involved in important campaigns for civil rights and for social justice .
the italian party that better fitted his political thoughts of the time was the radical party .
* i : * can you tell us something about it ?
it seems that , while being involved in political activities set forth by the radical party , he spent one night in jail !
* e : * in fact , he did not end up in jail because the order to release him arrived before being imprisoned . to make a long story short , he was editor of a newspaper of the radical party , which was publishing letters of conscientious objectors who refused to perform the compulsory military service .
this was illegal at the time .
the day he learned he was going to be arrested , he asked the police whether it was possible to arrest him at the _ accademia dei lincei _ , the most prestigious italian science academy , where he was going to have an official meeting the day after .
he motivated such a seemingly bizarre request with the fact that the police could have saved some money by not picking him up by car at home : the _ accademia dei lincei _
building was , indeed , just a few steps away from the prison he was supposed to go to .
however , the order to release him arrived as soon as he got to jail .
this episode had a huge echo in the press . *
a : * we also recall a story you told us about kolmogorov visiting roma and wanting to meet de finetti .
* e : * in 1962 kolmogorov was awarded the balzan prize for mathematics , the other awardees beingpope giovanni xxiii for peace , paul hindemith for arts , samuel e. morison for humanities and karl von frisch for biology .
two well - known mathematicians , gaetano fichera and olga a. oleinik , went to collect him at the roma airport and asked him what they could do for him . and , as fichera reported , his answer was , `` if you know him , then you should organize a meeting with de finetti . '' * a and i : * de finetti s papers are scattered with brilliant ideas , sometimes only sketched .
what are the aspects of de finetti s work which still need to be developed ?
* e : * as for some specific topics , such as exchangeability and processes with independent increments , in my opinion most of his ideas have already been extensively developed and not much is left to investigate in the precise direction he had originally thought of . on the other hand , i believe that much is still left to investigate on the general foundations of probability theory that emerge from his work and that he strongly supported
. these studies might have a relevant impact in statistics , in physics and in other research areas .
the advances i am thinking of concern both the interpretation of probability and the enlargement along with its mathematical implications of the class of admissible probability laws to include also the finitely additive ones .
* a : * you investigated quite extensively the development of statistics and probability in italy in the first half of the 20th century ( e.g. , regazzini , @xcite ) .
can you tell us about it ?
* e : * in contrast to what happened in the anglo american world or in russia , in italy probability and statistics developed along almost separate paths .
probability started growing in mathematical environments .
as far as i know , the first to deal with the topic in a comprehensive way was guido castelnuovo , a famous mathematician who was mainly doing research in algebra and geometry .
his 1919 book on probability ( castelnuovo , @xcite ) was used as a textbook for quite some time in those few mathematics degrees where probability was taught .
the interpretation of probability was frequentist , in line with a view that would have been later shared also by frchet , lvy and kolmogorov , and covered results of the russian school up to andrey markov and aleksandr m. lyapunov . already , back in 1915 he had the idea of setting up a school of statistics and actuarial sciences at the university of roma , which was then created in 1927 .
it had considerable success with many enrolled foreign students and then became a proper faculty in 1936 with gini . in the preparation of his book castelnuovo
was helped by cantelli , who is considered , also at an international level , one of the first modern probabilists .
he derived , among other contributions , versions of the laws of large numbers , the borel
cantelli lemma , a mathematical theory of risk that was named after him , and developed an autonomous abstract measure
theoretic theory of probability , which appeared before kolmogorov s .
it is interesting to recall that in this last development a crucial point was the proof of the existence of measurable maps defined on @xmath5 $ ] , endowed with the uniform distribution , in such a way they have prescribed probability laws : such an approach also reflects the idea of adhering to the classical definition of probability due to laplace .
anyhow , this problem led him and his students to anticipate at least part of what is nowadays known as the skorokhod representation . a distinguished scholar who obtained important results along the lines of research undertaken by cantelli was giuseppe ottaviani , who is also known for his inequalities that are related to cantelli s theory of risk .
francesco g. tricomi , eminent analyst andfriend of cantelli , also gave some contributions to probability as did carlo e. bonferroni , who is well known for his inequalities .
* a : * given such a glorious tradition , it is quite surprising , as you said earlier , that the first full professors in probability were appointed by italian universities only in the 1970s , with the notable exception of de finetti .
* e : * actually , at the beginning of the 1970s only two professors in probability were recruited , namely , giorgio dallaglio and giorgio letta .
dallaglio was at the faculty of statistics in rome and was a member of the before mentioned group led by pompilj .
letta is from pisa and spent several research periods in germany and france .
the latter experience stimulated collaborations between italian probabilists some of whom were letta s students and french probabilists in paris and strasbourg , a fruitful trend which is still ongoing .
then a larger group of people were appointed at the end of the 1970s in various italian universities .
* i : * and what about statistics ?
* e : * in the last three decades of the 19th century , topics that are today ascribed to mathematical statistics were taught in geodesy or astronomy courses .
lectures by a not well - known italian mathematician , paolo pizzetti , were very interesting and contained some innovative ideas on significance tests .
more conventional , at least according to the italian framework , statistics courses were in law faculties : many academic statisticians actually had a degree in law .
most of them were involved in official statistics and it was therefore natural that the interactions between statisticians and probabilists were rather limited .
the first modern italian statistician was rodolfo benini , who had a law degree from the university of pavia and developed statistical methods for demographic , sociological and economic problems around the end of the 19th and the beginning of the 20th century .
i recall once i came across historical documents presented in noteworthy conferences of the american and british economic societies where benini is referred to as one of the founders of econometrics .
i think this due to his analysis of income and wealth distributions and to the pioneering use of multiple regression methods to estimate , for example , demand curves .
he also had the idea of studying contingency tables with fixed marginals . among his successors ,
the main figure is certainly gini , also a graduate in law .
his methodological contributions to statistics were praiseworthy and were later studied not only in relation to mere data analysis .
gini dominated italian statistics until his death in 1965 and created a school of faithful followers .
a prominent group of scholars was led by pompilj at the faculty of statistics in roma . as i recalled earlier , dallaglio was one of its members and he obtained noteworthy mathematical results that can be traced back to the ginian analysis of statistical relationships .
his results , however , have a remarkable independent interest : for example , he provided a relevant contribution to the definition and to the properties of what is today known as the wasserstein distance .
see dallaglio ( @xcite ) . *
a : * you mentioned paolo pizzetti who seems to be a neglected figure within the italian statistics community , was he not ? we have never heard of him in our statistics courses .
* e : * yes , he was unfairly neglected .
his contributions , which appeared in the 1880 s , were very innovative and relied on an original approach that somehow anticipated a few distinguishing ideas lying at the foundations of statistics as set forth by karl pearson and by ronald a. fisher .
as an example , he proposed procedures that were very similar to the significance tests fisher would have later adopted as a distinctive feature of his methods .
pizzetti also had remarkable mathematical skills that allowed him to determine the exact distribution of certain statistics used for data analysis .
and he was well aware of the results achieved , in this direction , by a german geodesist friedrich r. helmert .
he reproved helmert s results with the aim of extending them and relied on innovative methods and techniques that fisher himself would have later proposed independently .
this is very well documented in a recent historical monograph by anders hald .
as you can easily guess , pizzetti s ideas were totally different from those that gini would have later expressed apropos of the fisherian tests .
indeed , gini was very critical about the use of significance tests and his criticisms were shared by de finetti .
this may partly explain why pizzetti is not known by many statisticians .
his work was somehow considered as heterodox for quite some time , as demonstrated by the 1960s reprint of pizzetti s 1892 book ( pizzetti , @xcite ) . in the preface , written by v. castellano , p. fortunati and g. pompilj
, it is claimed that parts of pizzetti s work were `` misleading and contained errors that had been masterfully pointed out by gini in gini ( @xcite ) . '' and the `` misleading parts '' they were referring to are exactly those where pizzetti uses his results for devising statistical tests . *
a and i : * the excerpt you read can partly explain the isolation of the italian statistics community in those years .
* e : * it partially does .
indeed , i think that gini s critical remarks make sense .
the point is that they were not complemented by alternative proposals that could take his concerns into account .
hence , it was almost inevitable that gini s position would have become marginal and isolated within the broader international community .
it should be recalled that isolation fitted very well with the political climate favoring autarkic tendencies during the fascist regime and it unfortunately further consolidated over the years in the italian statistics community , at least in academia .
this obviously had a long - lasting negative impact , from which italian statistics started recovering only in the 1970s . *
a : * gini was appreciated both for his scientific achievements and for his praiseworthy services as a scientific expert within various important italian institutions . *
e : * definitely .
he was founding president of the italian central institute of statistics in 1926 and set up first the school and then the faculty of statistics , demographic and actuarial sciences in roma in 1936 .
he was in constant contact , also meeting him in person , with mussolini , who used to pay attention to statistical analyses for taking decisions on policy issues .
for example , he acted as a technical advisor within the programs of demographic and eugenics policies pursued by the fascist regime .
later he also founded the italian statistical society , of which he has been president for 20 years .
besides the scientific and institutional authoritativeness he gained in italy , it should be recalled that he obtained countless recognitions abroad as well . among them
i could mention that he became honorary fellow of the royal statistical society , vice president of the international sociological institute , and honorary member of the international statistical institute . in 1920
he was the founding editor of the journal _ metron _ , which published papers by many eminent statisticians of the time , such as r. a. fisher , a. a. chuprov , a. j. lotka , s. s. wilks , e. e. slutsky , s. kullback , h. wold and a. l. bowley .
* i : * we have also heard of some funny stories about gini bearing ill - luck .
can you tell us something more ?
* e : * yes , this is somehow true , but it is to be considered within the typical italian attitude of making fun of powerful people , as gini certainly was .
there are various minor anecdotes and a dramatic episode that would allow to conjecture a `` correlation '' of the type you are referring to . as for the latter , something incredible happened in 1927 : he was on the steamboat `` princess mafalda , '' which shipwrecked off the brazilian coast between salvador de bahia and rio de janeiro , and he was among the few survivors , the `` legend '' says thanks to his rowing skills , a sport he had practiced in youth .
a less dramatic and funnier story i have heard of concerns an episode where , chatting with a colleague of his , he paid a compliment to a young female student s legs whom they met on the stairs : after a few steps she fell down and broke her leg .
i remember that ottaviani did not mention his name , he referred to him as _ the unnamed _
, since mentioning his name could have led to something bad happening .
all kidding aside , after the shipwreck in brazil , he criticized the italian authorities for the poor assistance from the italian navy and , more in general , from the italian government .
these complaints caused him a lot of troubles with the fascist regime in italy .
he had , indeed , a strong and straight attitude that helped him to protect scientific matters and appointments from political influence .
of course , this position attracted the aversion of many fascist party officials who strove for mussolini to remove him as president of the italian central institute of statistics . and
his criticisms on the occasion of the shipwreck were added to the list of gini s `` offences '' to the regime that led to his resignation in 1932 . however , as i said before , he kept collaborating with the regime as a scientific expert in demography , statistics and eugenics . *
a : * cantelli , de finetti and gini were the towering figures in probability and statistics before world war ii in italy .
they were also completely different characters .
how did they get along ?
* e : * gini published de finetti s work on mendelian inheritance in _ metron _ and offered de finetti a job at the italian central office of statistics before he graduated . while at the italian central office , de finetti was involved in a project for predicting the evolution of the italian population and crucially designed all modeling aspects of the project .
he then wanted this to be credited as his contribution , but gini was reluctant to do so .
this episode is well documented in one of the letters de finetti wrote to his mother and contained in the collection published by his daughter that i have already mentioned . in any case , at the end of his four year contract in 1931 , de finetti moved back to trieste and started to work for the insurance company _
generali_. the relationship between de finetti and cantelli was quite a difficult one , since they were in strong disagreement on the interpretation of probability .
cantelli did not want to hear anything about finite additivity and he also tried to prove that @xmath4-additivity was a necessary property . *
i : * in addition to _ metron _ , there was also the _ giornale dellistituto italiano degli attuari _
( _ giia _ ) , which was a top journal in statistics and probability during the 1930s .
why have they both lost their international reputation since then ?
* e : * the _ giia _ was established in 1930 , the same year _ the annals of mathematical statistics _ published their first issue .
it was edited by cantelli and the most distinguished scholars of the time , such as cramr , frchet , kolmogorov , khintchine , lvy , neyman and von mises , published fundamental contributions on it . world war ii ruined everything , since its publication was suspended and the _ giia _ lost its elite status among the top probability and statistics journals which , during and soon after the war , included _
the annals of mathematical statistics _ , along with _
biometrika _ and the _ journal of the royal statistical society_. the other italian prestigious journal , _ metron _ , which was established in 1920 , paid a high price for the line of development of italian methodological statistics and actually already declined before the war . *
a : * in 1978 you actually published a very interesting paper characterizing the dirichlet process in terms of linear predictive distributions ( regazzini , @xcite ) on the _
giia_. why did you decide this was a suitable outlet for your paper ?
* e : * after my discussions with rukhin , the dirichlet process became a main ingredient of my research agenda .
in fact , i was dealing with risk premium models , for insurance companies , which were linear combinations of an empirical part and an expected value related to some prior guess
they identify the so - called credibility premium .
i thought to revisit the problem coherently with the predictive distributions generated by an exchangeable sequence and asked myself what the underlying de finetti measure was : it turned out to be the law of a dirichlet process .
i wrote this paper while i was working with cifarelli on the distribution of linear functionals of the dirichlet process .
i then presented it at a conference , where luciano daboni , an editorial board member of _ giia _ , was present : he liked the paper a lot , invited me to give a seminar in trieste and proposed for me to publish it in _
giia_. some years later the same result was independently obtained by albert y. lo ( lo , @xcite ) .
* i : * during the 1980s you were probably one of the few statisticians in italy who published their papers in international journals .
do you have any idea why this happened at the time ?
* e : * well , first of all , most people , both in statistics and probability , did not even try to submit their work abroad . it was simply not necessary for the progress in academic careers .
even many mathematicians only published in italian journals .
overall , the need for trying to spread one s own work at an international level was not felt yet .
actually , it was probably not even felt in the anglo - american world : it just happened that their journals then became the `` international '' ones . by the way ,
papers that were published in italian journals with a very limited spreading were not all necessarily of bad quality . on the contrary ,
some of them are very well known even abroad and contain innovative ideas .
anyhow , in recent years things have changed substantially and young researchers submit their work to the best international journals . *
a : * we , as students , have nice memories of summer schools organized by the italian scientific community to support the spreading of probability and statistics .
you have been an active part of this initiatives . *
e : * after attending some of them as a student , i have been involved several times in organizing and teaching at summer schools that took place in various beautiful locations in italy , such as cortona , perugia , livigno and rhme - notre dame .
in addition to being an opportunity to meet talented students , summer schools also allowed me to get in contact , and actually build up friendships , with some authoritative scholars such as alan agresti , patrick billingsley , albert y. lo , slava sazonov , henry teicher and jon wellner .
unfortunately , the cuts operated through the years by the italian governments have made it more difficult to sustain the organization of such praiseworthy initiatives . *
a : * you then also started to collaborate with sazonov .
in fact , one of your articles we really enjoyed reading , for the wealth of results and techniques it offers , is regazzini and sazonov ( @xcite ) .
how did you convince him to do research on bayesian statistics ?
* e : * slava was a wonderful person i really miss .
he was a loyal friend to me and an extraordinary mathematician .
i first met him at a conference at the beginning of the 1990s and , then invited him to deliver a course , jointly with albert lo , at a summer school organized by bocconi university in 1992 .
he then taught also in the 1993 and 1994 editions .
we started collaborating in 1996 , while he was teaching a course on `` probability theory in hilbert spaces '' at the italian national research council in milano .
our first joint work concerned central limit theorems for partially exchangeable arrays of random elements taking values in a hilbert space . at the time i was also preparing my lectures for a ph.d .
course on bayesian nonparametrics to be taught in roma and i was dealing with the problem of estimating a statistical model by means of a mixture of dirichlet processes .
such a problem was suggested by diaconis and ylvisaker ( @xcite ) : with slava we showed that it is possible to construct a mixture of laws of dirichlet processes that approximates the distribution of any random probability measure , with respect to the topology of weak convergence .
and we have been able to obtain , under suitable assumptions , the corresponding approximation bounds for the posterior measures .
these results were presented at the 1st workshop on bayesian nonparametrics that took place in belgirate ( italy ) in 1997 .
while we were working on this paper , my mother became seriously ill and slava has been very important in supporting me in such a difficult period .
* i : * you have always had good relationships with probabilists and statisticians from russia .
for example , ildar ibragimov is another good friend of yours who has been several times in pavia contributing to the ph.d .
i had the pleasure to attend his lectures and really enjoyed them .
* e : * it was actually slava who suggested i contact ildar ibragimov .
in fact , i had asked slava indications for possible instructors for ph.d . courses . and
slava mentioned about ildar and told me that in addition to being a great scientist he was an excellent teacher .
of course , i knew ildar by fame and i feared he would have not accepted my invitation but he did .
i got the chance to meet in person not only a brilliant mathematician but also a wonderful person .
his courses in pavia were greatly appreciated and i liked the fact that he , and also slava , was trying to adapt his lectures to the students background .
classes are quite composite , with most students having either mathematics or economics degrees .
the former typically have good backgrounds in pure maths but not in statistics and probability , whereas for the latter it is the opposite .
i remember ildar asking me , `` how come i have students in my class who know about radon
nikodm derivatives of stochastic processes but struggle with fourier transform coefficients ? ''
* a : * your papers with cifarelli on functionals of the dirichlet process ( cifarelli and regazzini , @xcite , @xcite ) are probably your most well - known contributions to bayesian nonparametrics . and
it is amazing how many connections your results have with a variety of research areas such as combinatorics , mathematical physics , theory of stochastic processes , the moments problem , and so on . were you aware of these ? *
e : * as i said , our original problem was merely of a statistical nature .
from an analytical point of view , the task we were facing was very challenging , but we were not aware of the connections with seemingly unrelated areas of mathematics .
we learned about some of these relations thanks to the paper by persi diaconis and johannes kemperman that was presented at the valencia meeting in 1994 ; see diaconis and kemperman ( @xcite ) .
in addition to embedding the whole problem in a wider mathematical context , it is also very well written and sketches a few open problems ; i strongly recommend reading it .
it is also thanks to this very same paper that my work with cifarelli gained some popularity .
* i : * the basic trick you resorted to was the inversion of a cauchy
stieltjes transform for the mean of the dirichlet process .
how did you arrive to this intuition ?
* e : * the procedure actually relied on the determination of recursive relations for the moments of the linear functional .
such a strategy was inspired by the work of m. kac who used it to obtain the well - known feynman
kac formula ; see , for example , kac ( @xcite ) .
this closeness is further revealed by the adoption , in our paper , cifarelli and regazzini ( @xcite ) , of the same symbols used by kac !
cifarelli had successfully used it to establish a closed form expression for the probability distribution of the integral of the absolute value of the brownian bridge in cifarelli ( @xcite ) .
these recursive relations we obtained allowed us to determine the laplace transform whose iteration yields the cauchy stieltjestransform .
we then resorted to the inversion formulae of the cauchy
stieltjes transform to deduce an exact form for the probability distribution of a linear functional of the dirichlet process .
most of these ideas were already contained in cifarelli and regazzini ( @xcite ) . in cifarelli and regazzini ( @xcite )
we basically completed that paper and provided some further insight . * a
: * more recently you developed an alternative method based on an inversion formula for the characteristic function .
* e : * the approach you are referring to was inspired by the representation of the dirichlet process as the normalization of a gamma process that was first pointed out by ferguson himself in his 1973 paper .
this representation combined with a suitable inversion formula led to new forms for the probability distribution of the mean of a dirichlet process , which are recorded in a paper with alessandra guglielmi and giulia di nunno .
i have then extended , with the two of you , the approach to deal with means of random probability measures induced by the normalization of a generic process with independent increments .
* i : * at the moment , bayesian nonparametric regression is a hot topic . in this respect ,
a paper of cifarelli and yourself has been recently `` rediscovered '' ( cifarelli and regazzini , @xcite ) .
can you talk to us about its origin and contents ?
* e : * the original goal of our research was to determine a probability distribution for partially exchangeable arrays of random elements .
in particular , we were looking for a solution that could be treated analytically , while avoiding the independence assumption among rows .
these were the two reasons which led us to the idea of resorting to the mixture of products of dirichlet processes .
we have been able to determine the associated system of predictive laws and the distribution of vectors of functionals . in a parametric
setting , partial exchangeability had been incorporated in a paper by lindley and smith ( lindley and smith , @xcite ) .
i then used our model to study credibility formulae with collateral data .
cifarelli had also developed the model for applications to anova and linear models , the latter in collaboration with marco scarsini and pietro muliere .
we did not even submit the paper to a journal , since , as i said , at the time a technical report or a journal publication counted the same for us .
i am really pleased to see the recent explosion of proposals on dependent nonparametric models , somehow in the spirit of our 1978 paper , developed by s. maceachern , p. mller , d. dunson and many others.=-1 * i : * in our opinion , the work of two probabilists , john f. c. kingman and jim pitman , has to be listed among the main and most far reaching contributions to bayesian nonparametrics , even if not directly focused on it .
do you share this view ?
* e : * i am strongly in favor of a bayesian approach that solely relies on the specification of distributions for observable random elements .
therefore , in general , i like all those contributions and tools that aim at providing systems of predictive distributions related to modeling and applications .
these do not resort to conditional distributions , given parameters ( either finite or infinite - dimensional ) that in some applications would be devoid of any empirical meaning . and , the works by kingman and pitman ,
although originated in different research areas , have an important impact on bayesian statistics .
even though i read their papers only recently , i have appreciated them very much since they open up the possibility of implementing the bayesian paradigm in the direction i lean toward . *
a : * the contributions of kingman and pitman you just mentioned are closely related to exchangeability , a topic you extensively worked on both from a statistical and probabilistic point of view .
* e : * my interest in exchangeability was stimulated by reading de finetti s papers .
the first place where i came across the statement of de finetti s representation theorem was the monograph by love .
but i could not understand its statistical implications .
i could appreciate its relevance for inductive reasoning only through a careful study of de finetti ( @xcite , @xcite ) : in my opinion , these papers really stand out in terms of conceptual and mathematical rigor and effectiveness in highlighting the role of exchangeability for induction , and remain unbeaten to date .
of course , the modern uses of exchangeability and the key role it plays in modeling a variety of phenomena are probably beyond what de finetti could have expected .
* i : * you have also been working on characterization theorems in this context . *
e : * you are probably referring to results i have obtained with sandra fortini and lucia ladelli and that characterize systems of predictive distributions associated with exchangeable sequences of random elements .
i have also noted that these kinds of results have recently attracted more and more interest in bayesian nonparametrics practice .
another interesting characterization was obtained in a paper i coauthored with giovanni petris where we dealt with exchangeability in the presence of finitely additive probabilities : we stated and proved a weak version of the representation theorem that reduces to the celebrated de finetti theorem ( strong version ) if one specializes to the case of @xmath4-additive probabilities . in this situation , we were also able to use the representation theorem to show existence of a random probability measure defined by means of a system of finite - dimensional distributions agreeing with ferguson s framework . *
i : * you have also provided nice contributions to the investigation of properties of partially exchangeability . *
e : * indeed , i have been , and i still am , interested in forms of dependence more general than exchangeability , as witnessed by some contributions i have already mentioned before , such as the paper on mixtures of products of dirichlet processes or the formulation of a central limit theorem for partially exchangeable arrays .
besides these , i wish to mention a nice characterization of partially exchangeable arrays that has been established in a paper i wrote with fortini , ladelli and petris .
indeed , we proved a conjecture formulated in de finetti ( @xcite ) , according to which a suitable random matrix related to the transitions of a recurrent process is partially exchangeable if and only if the law of the process can be represented as a mixture of laws of markov chains .
moreover , we have been able to show that de finetti s definition of partial exchangeability is equivalent to the one provided by diaconis and freedman in a couple of papers they wrote in 1980 .
* i : * in some of your work you have also provided some insight into an approach to bayesian statistical inference based on finitely additive conditional probabilities . *
e : * i started getting involved into research on finitely additive conditional probabilities after reading some papers by r. scozzafava in the first half of the 1980s .
in fact , i grew convinced that countable additivity was not justifiable as a necessary condition unlike finite additivity which is necessary for the validity of de finetti s coherence principle
. therefore , finitely additive probabilities have to be considered as admissible and i became interested in revisiting known results in probability as particular cases of the finitely additive framework .
in particular , i found the interpretation of the definition of conditional probability , as given by kolmogorov , unsatisfactory .
conditioning is based on classes of events that partition the whole sample space and that become finer and finer : conditional probability is then obtained through a limiting process in terms of a radon nikodm derivative , and depends on the class of events one conditions on . in de finetti s approach , a conditional probability , given an event , is defined through a natural , and unavoidable , strengthening of the coherence principle .
de finetti himself had hinted at such a possibility , without developing his idea in general mathematical terms .
i tried to make this more explicit in some papers i wrote during the 1980s in regazzini ( @xcite , @xcite ) .
these topics have been object of further investigation by my friends p. berti and p. rigo .
an important point is that many situations that appear as paradoxical if one refers to kolmogorov s conditional probabilities can be justified within the finitely additive framework . *
a : * can you provide us with an example ?
* e : * the most well - known is probably borel s paradox .
indeed , if a uniform distribution on the surface of a sphere is defined , with respect to a specific choice of geographic coordinates ( namely , latitude and longitude ) , one would expect that the conditional distribution for latitude , given a fixed longitude , is uniform . however , this does not happen in kolmogorov s framework .
in de finetti s approach , instead , one can adopt the more intuitive probability assessment even if it would be nondisintegrable .
the reason for such a behavior can be traced back to the specific notion of conditional probability according to kolmogorov s approach , since it does not admit the evaluation of the probability of an `` isolated event '' with probability zero . on the contrary , de finetti s setup is open to different solutions : indeed , disintegrability turns out to be _ not necessary _ for coherence .
* i : * another amusing aspect of finitely additive conditional probabilities emerges from your work on well - calibration of systems of predictive distributions . * e : * loosely speaking , well - calibration corresponds to situations where the distance between weighted averages of forecast probabilites and empirical observations converges to zero as the number of observations , and forecasts , increases .
kolmogorov s theory always yields well - calibrated predictions or forecasts .
this corresponds to a somehow unrealistic situation in practice , since one would also expect cases of not well - calibration . with p. berti and p.
rigo we were interested in checking whether the same was true within de finetti s theory as well .
our curiosity to this problem was stimulated by a paper of phil dawid ( dawid , @xcite ) .
the answer we got was naturally affirmative for _ strategic _ conditional probabilities .
the term strategic was coined by dubins and savage in their well - known monograph where they resorted to de finetti s theory to solve quite complicated measurability problems .
strategic conditional probabilities do indeed preserve , in a finitely additive setting , the disintegrability property that characterizes kolmogorov s definition .
as for well - calibration , we were able to show that , beyond strategic evaluations , there exist not well - calibrated coherent bayesian predictors with positive probability.=1 * a : * many critics of de finetti s subjectivistic standpoint in probability theory use , as an argument for supporting their position , de finetti s sentence `` probability does not exist . ''
what can be replied to such objections ?
* e : * first of all , one should consider the provocative nature of de finetti s sentence .
moreover , its meaning should not be decontextualized .
according to de finetti , if one wants to give probability an objective meaning , one should prove its existence .
in other words , there should be an existence theorem , a clear proof of the existence of an object termed `` probability . '' for example , the interpretation of probability as a limiting frequency can not be considered as a proof , even if just empirical , of its existence .
hence , he used the expression `` probability does not exist '' just to make the point that probability has simply a subjective meaning .
it is also to be said that most of the criticism raised against subjectivism basically refers to the contents of his two - volume monograph , de finetti ( @xcite ) , which is , according to its subtitle , `` a critical introductory treatment . '' in my opinion , de finetti s position can be better discussed by relying on his early works , which are more concise , go straight to the point and display more mathematical and formal details .
* i : * a noteworthy scholar who contributed to the theory of finitely additive probabilities was lester dubins .
you were also a good friend of his and had the chance to host him in milano .
* e : * dubins had been in italy several times and he delivered courses at summer schools .
he was very fond of italy and , in the second half of the 1980s , i invited him once to stay for a month in milano .
we had discussions on various research topics .
he was the source of many ideas that i later developed in my research . in those years
i was mainly working on technical aspects of nonparametric inference , whereas he could provide me with many insights into theoretical issues related to finite additivity that turned out to be of great importance to me . *
a and i : * you have lately become interested in some problems in mathematical physics .
how did it happen ?
* e : * moving to pavia in 1998 , i joined a mathematics department with a few internationally well - known mathematical physics scholars .
i started interacting with them and at some point a colleague of mine , ester gabetta , showed me some papers where the central limit theorem was used to describe the convergence to equilibrium of the solution of certain kinetic equations . in particular , i read two papers , mckean ( @xcite ) , that spurred my enthusiasm for the topic .
i tried to understand and extend the connections with probability , and could count on the collaboration of my colleagues to help me understand the problem from the perspective of physics .
furthermore , the encouragement from eric carlen and maria carvalho has been important for pursuing my research in this direction .
in fact , they liked our first results and suggested us to publish them ( see gabetta and regazzini , @xcite ) : there we obtained some identities that came in handy for later developments of the work in this area . *
a and i : * was this line of research as rewarding as others you have pursued in your career ?
* e : * i would say i am happy about what i have achieved so far with my coauthors . starting from the kac model , which is generally considered as a toy model , we obtained some interesting results concerning the characterization of the initial data in order to gain convergence to equilibrium .
we have also considered situations where the energy , interpreted as the variance of the initial datum , is infinite and we performed an analysis of the speed of convergence . in these studies ,
i have also collaborated with lucia ladelli and federico bassetti .
later , i have supervised the thesis of emanuele dolera , a ph.d .
student in pavia .
this work has required a strong effort that was rewarded by the achievement of a noteworthy result proving the validity of a conjecture formulated in the 1966 mckean paper . in the last 40 years
many scholars have worked hard with the aim of proving it .
* i : * in some of the previous questions we have lingered on the subjectivistic interpretation of probability .
what is the most relevant impact this has on statistics ?
* e : * a crucial point to understand is whether it is worth preserving an axiomatization based on countable additivity . of course , i think it does not generally have a statistical justification that makes its use necessary .
if finitely additive probabilities are also admissible , then a considerable number of results in the literature should be revisited .
i have already mentioned that one should reconsider the definition of conditional expectation . moreover , a number of limiting theorems should be reformulated in order to account for this more general framework .
these issues are also of great relevance in statistics regardless of the approach , either frequentist or bayesian , one adopts . *
a : * does this lead , among others , to a rethinking of bayesian procedures ? *
e : * indeed , bayesian procedures are typically implemented by assuming complete additivity and this leads to assume some of its implications as necessary .
let us consider , as an example , the dirichlet process .
a well - known result is that the dirichlet process selects , almost surely , discrete probability measures .
however , such a property holds true for the countably additive extension of the collections of finite - dimensional probability distributions of the process .
there are other non @xmath4-additive extensions for which the dirichlet process selects nondiscrete distributions with positive probability .
this points to the fact that in statistical practice one should avoid assessing a probability for objects devoid of empirical evidence .
for example , take the proposition stating that de finetti s measure is the law of the ( almost sure ) weak limit of the empirical distribution : thus , it depends on infinitely many observations and concerns `` transcendent''in de finetti s words conditions not directly verifiable .
the conclusion of such a proposition could be obviously false with non @xmath4-additive extensions .
on the other hand , the fact that de finetti s measure is the weak limit of the low of the empirical distribution , as the sample size increases , is , in any case , true : in my opinion this suffices with respect to sound statistical goals .
i think this is an important foundational aspect , which is often neglected and should be further investigated .
* i : * are you saying that one should have clear in mind the different levels at which mathematics and statistical applications operate ?
* e : * more or less , that is what i mean .
indeed , it is true that mathematics makes parameters interpretable as limits of ( or of functionals of ) empirical processes , but it does not automatically grant that inference on them are legitimate . *
a : * does this position contrast with the usual way of presenting a bayesian model as the combination of a likelihood and a prior ?
* e : * let me start by making an important point that reflects my view on statistics : if inference is seen as a decision problem to be solved under uncertainty and if one agrees that probability is a tool to resort to , then there is no other choice but the bayesian approach .
nonetheless , i agree with what diaconis and ylvisaker say at the beginning of their paper diaconis and ylvisaker ( @xcite ) : bayesian statistics can not be reduced to the elicitation of a prior and the automatic application of bayes theorem .
hence , i would give an affirmative answer to your question if one conditions on unobservable quantities .
but this is not limiting the scope of bayesian inference at all .
indeed , one can think of inferential procedures that can still be implemented in this more general framework , even when unobservable parameters are involved .
the previously mentioned `` weak '' interpretation of the de finetti measure says that a prior distribution can be viewed , in any case and with no distinction between observable and not observable parameters , as an approximation of the law of a frequency distribution or of some functional of it . moreover , prediction can be carried out without relying on the bayes
laplace paradigm : it is enough to specify the system of predictive distributions connected to the exchangeable sequence .
and i have appreciated very much the work by j. pitman which , in the spirit of de finetti s stance , relies on the proposal of systems of predictive distributions that are then proved to be associated to an exchangeable sequence . *
a : * i also guess that a subjectivist would not agree on the notion of posterior consistency as a frequentist validation criterion of bayesian nonparametric methods .
* e : * i have to admit that , besides the bayesian context , i am skeptical on the use of consistency in a frequentist setting as well . on the one hand , these limiting results are very neat and beautiful from a mathematical point of view .
but , on the other , they lack a sensible statistical interpretation .
this is very well discussed in de finetti ( @xcite ) , volume 2 , in the section devoted to the laws of large numbers where he motivates why results , such as consistency , do not represent justifications of statistical procedures under the assumption of stochastic independence .
the same can be said for the glivenko
cantelli theorem .
of course , my position encompasses commonly used frequentist validation criteria adopted in a bayesian framework . a different role
must be attributed to approximation results , like central limit theorems or , also , the `` weak '' interpretation of the de finetti measure , for which these concerns do not apply .
so , what are the kind of asymptotic problems that you think are interesting for bayesians ?
* e : * the kind of results i like are those in the spirit of blackwell and dubins ( @xcite ) where they investigate the phenomenon of the merging of opinions .
this is a very nice finding both from a mathematical and from a statistical point of view . on the one hand
, it is a general result valid even beyond exchangeability . on the other hand , it has a nice statistical interpretation for bayesians since it hints at the predominance of empirical findings over different subjective prior opinions as the sample size or , in other terms , the amount of information , increases .
other results of great interest are those that currently are designated as bernstein
von mises type theorems for the posterior distribution .
it is worth noticing that among the first contributions to this topic there is also an important paper , romanovsky ( @xcite ) , published on
_ giia_. * a : * from what you said up to now on the interplay between bayesian inference and de finetti s interpretation of probability , a valuable research topic would focus on the analysis of the asymptotic behavior of the predictive distributions . *
e : * you are right . in my opinion , an important issue to address is the analysis of the distance between the predictive and the empirical distributions . instead of looking at the limiting behavior , it would be more interesting to analyze how such a discrepancy changes for any sample size @xmath6 and , _ a fortiori _ , as @xmath6 increases .
since the predictive can also be obtained as a functional of the posterior distribution , one can also gain some insight if one relies on convergence theorems , which say that the posterior converges , in some sense , to a distribution concentrated on the limit of the empirical process . in this respect , bernstein
von mises type results are useful .
* i : * you have had a large number of students , and by now also descendants , working in many different universities in italy and abroad . in your opinion , what is the background a statistics student needs to perform well in nowadays research and what are the topics you would suggest to pursue ? * e : * as i said earlier , i see statistics as inductive reasoning under the supervision of probability theory .
therefore , it is natural that i firmly believe that statisticians should have a solid background in probability : the more the better .
however , a statistician must also be able to think through the logical and philosophical aspects of what she / he is doing .
this concerns modeling , the understanding of practical implications yielded by the mathematical formulation that is used , and the interpretation of results .
mathematical skills are not enough , logical and conceptual rigor being a necessary complement .
one needs to be able to handle statistics since it is a powerful instrument , which allows one to make substantial steps forward , compared to traditional deterministic procedures .
statistics can get you close to the best solutions , avoiding overwhelming technical and mathematical difficulties that often arise within deductive deterministic reasoning .
the latter approach lacks the flexibility of a learning mechanism , whereas in the probabilistic framework everything is kept under control : you have a law which governs everything and , unless you change the learning mechanism , it allows one to learn from experience in a way that is transparent and controlled by bayes theorem .
* a : * in modern science the specialization of re - searchers is constantly increasing . even probability and statistics , which have grown in close relationship to each other ,
seem to be drifting apart .
* e : * you can observe the fragmentation of fields all over the place .
this phenomenon also originates from an excessive specialization that characterizes most undergraduate studies .
the situation was , in the past , quite different and there were many scholars with a wide spectrum knowledge and diversified cultural and scientific interests .
de finetti and gini are excellent examples in this respect .
that said , fragmentation in research is unavoidable and it would be unrealistic to try reversing it .
it is just a pity to see that it tends to create duplications and repetitions , whereas a more cohesive scientific community could produce better results in a collective effort . in statistics ,
bayesian statisticians have kept to themselves for some time in reaction to the then mainstream statistics , which was certainly not in favor of bayesian methods . now with bayes statistics well - established , i note that younger generations are more open to interactions with non - bayesian , which in my opinion is certainly beneficial .
a different issue is the specialization in education , which should be contrasted to some extent because it precludes possible paths to future researchers .
as i have already said , every statistician should have a solid background in probability and every probabilist should know the basics of statistics , which is a noble and fascinating , at least to me , field of application of probability . *
i : * how should , in your opinion , a good statistics paper be structured ?
* e : * well , first of all the definition `` good '' is to be considered with reference to the historical period . until some years ago theoretical papers were very appreciated , whereas nowadays applied work plays an increasing role thanks to the computational tools .
however , i think that , in general , different forms of motivation are equally valid : an enrichment of the available tools , an improvement over other existing contributions or a useful application are all fine . however , in all cases it is crucial that the paper is logically sound and coherent with its motivation . this is essential since we write for the scientific community and not for the general public , which is another job . for instance , i do not like methodological papers , to which an illustration has been evidently added only for editorial needs
. a methodological contribution can stand on its own if its motivation is sound .
while i was young i experienced some of the last manifestations of gini s school , which as a rule of thumb required publications to include data , a table and a plot . to me this does not make sense .
i also do not like applied papers in which one sets forth a model , analyzes a couple of data sets and concludes that the model works well .
any so - called empirical validation does not show anything and is not enough to assess the suitability of a model .
indeed , there should also be a sensible methodological motivation in the sense that one should explain which features of a certain model make it more appropriate for the problem at hand . *
a and i : * moving away from statistics and probability , we already mentioned your passion for music
. how did you get fond of music and what else are you interested in when you do not do research ?
* e : * being born in cremona , my passion for music is quite natural : it is 20 km away from the places giuseppe verdi grew up in , melodrama is popular and there is a great tradition .
it is also the hometown of claudio monteverdi and of amilcare ponchielli , two famous composers .
last but not least , it is the town of lute makers , the most renowned being antonio stradivari .
even the general public knows opera very well .
then , starting from opera when i was young , my interest extended to symphonic music .
i have also been fond of visual arts since i was a kid : i loved paintings , architecture and sculpture since i related them to italian history .
i remember having a great teacher at school who used to emphasize links between history , arts and literature .
a peculiar feature of italy is that , if you are interested in any historical aspect , you necessarily end up considering also painting , sculpture and architecture since they are all intimately connected .
we obviously benefited from christian culture that played a fundamental role , after the fall of the roman empire , in preserving the wonders inherited from classical greek and roman traditions and in promoting arts in forms we can today admire while visiting churches , historical buildings , squares and museums .
moreover , during the renaissance , there were a large number of small states and many of them had patrons who liked and could afford being surrounded by artists .
hence , many towns developed their peculiar artistic heritage .
* a and i : * eugenio , thanks a lot for patiently answering all our questions .
* e : * thanks to you for listening to all this !
many thanks to elena di biase for a careful reading of the manuscript and several useful suggestions which improved the presentation . this work is partially supported by miur , grant 2008mk3afz . | eugenio regazzini was born on august 12 , 1946 in cremona ( italy ) , and took his degree in 1969 at the university `` l. bocconi '' of milano .
he has held positions at the universities of torino , bologna and milano , and at the university `` l. bocconi '' as assistant professor and lecturer from 1974 to 1980 , and then professor since 1980 .
he is currently professor in probability and mathematical statistics at the university of pavia . in the periods
19892001 and 20062009 he was head of the institute for applications of mathematics and computer science of the italian national research council ( c.n.r . ) in milano and head of the department of mathematics at the university of pavia , respectively . for twelve years between 1989 and 2006 , he served as a member of the scientific board of the italian mathematical union ( u.m.i . ) . in 2007 , he was elected fellow of the ims and , in 2001 , fellow of the `` istituto lombardo accademia di scienze e lettere . ''
his research activity in probability and statistics has covered a wide spectrum of topics , including finitely additive probabilities , foundations of the bayesian paradigm , exchangeability and partial exchangeability , distribution of functionals of random probability measures , stochastic integration , history of probability and statistics .
overall , he has been one of the most authoritative developers of de finetti s legacy . in the last five years
, he has extended his scientific interests to probabilistic methods in mathematical physics ; in particular , he has studied the asymptotic behavior of the solutions of equations , which are of interest for the kinetic theory of gases .
the present interview was taken in occasion of his 65th birthday . . |
Subsets and Splits